Bioreducible polymer-mediated gene therapy for the treatment of ischemic heart disease

Coronary heart disease, especially myocardial infarction, is a serious and deadly disease that remains the number one cause of death in developed nations. While physicians have a wide array of drugs and tools at their disposal to treat myocardial infarction, there is a scarcity of drugs that effectively prevent the pathological remodeling of the left ventricle that occurs after acute infarction and greatly predisposes the patient to the risk of heart failure. This dissertation focuses on identifying nonviral, bioreducible polymer-based gene therapies to limit infarct expansion, prevent left ventricular remodeling, and retain heart function after myocardial infarction. Bioreducible polymers represent a major advancement in nonviral technology to transfect primary cells, with high efficiency, and low cytotoxicity. We first examined a combined gene/cell therapy-based strategy by implanting VEGF165-transfected primary skeletal myoblasts to the myocardium of infarcted rat hearts. The VEGF-expressing skeletal myoblasts acted as bioreactors, secreting proangiogenic VEGF and inducing new vessel formation in the infarcted hearts. This treatment strategy produced both global and regional improvements in the left ventricle, improving ejection fraction, decreasing cell death, and limiting left ventricular remodeling. A cationic polymer system utilizing arginine-grafted bioreducible polymer (ABP) was also used to efficiently mediate siRNA knockdown of BNIP3, a hypoxia-inducible iv proapoptotic protein. siRNA-mediated BNIP3 knockdown both in vitro and in vivo protected rat primary cardiomyocytes from hypoxic death. The inhibition of BNIP3 in acutely ischemic rat hearts resulted in improved retention of ejection fraction, decreased infarct formation, decreased cellular remodeling, and decreased left ventricular remodeling.

BIOREDUCIBLE POLYMER-MEDIATED GENE THERAPY FOR THE TREATMENT OF ISCHEMIC HEART DISEASE by Arlo Nuttall McGinn A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Pharmacology and Toxicology The University of Utah May 2013Copyright © Arlo Nuttall McGinn 2013 All Rights ReservedThe University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Arlo Nuttall McGinn has been approved by the following supervisory committee members: Sung Wan Kim , Chair 11/14/2012 Date Approved William R. Crowley , Co-Chair 11/14/2012 Date Approved Donald K. Blumenthal , Member 11/14/2012 Date Approved David A. Bull , Member 11/14/2012 Date Approved Philip J. Moos , Member 11/14/2012 Date Approved and by William R. Crowley , Chair of the Department of Pharmacology and Toxicology and by Donna M. White, Interim Dean of The Graduate School. ABSTRACT Coronary heart disease, especially myocardial infarction, is a serious and deadly disease that remains the number one cause of death in developed nations. While physicians have a wide array of drugs and tools at their disposal to treat myocardial infarction, there is a scarcity of drugs that effectively prevent the pathological remodeling of the left ventricle that occurs after acute infarction and greatly predisposes the patient to the risk of heart failure. This dissertation focuses on identifying nonviral, bioreducible polymer-based gene therapies to limit infarct expansion, prevent left ventricular remodeling, and retain heart function after myocardial infarction. Bioreducible polymers represent a major advancement in nonviral technology to transfect primary cells, with high efficiency, and low cytotoxicity. We first examined a combined gene/cell therapy-based strategy by implanting VEGF165-transfected primary skeletal myoblasts to the myocardium of infarcted rat hearts. The VEGF-expressing skeletal myoblasts acted as bioreactors, secreting proangiogenic VEGF and inducing new vessel formation in the infarcted hearts. This treatment strategy produced both global and regional improvements in the left ventricle, improving ejection fraction, decreasing cell death, and limiting left ventricular remodeling. A cationic polymer system utilizing arginine-grafted bioreducible polymer (ABP) was also used to efficiently mediate siRNA knockdown of BNIP3, a hypoxia-inducible iv proapoptotic protein. siRNA-mediated BNIP3 knockdown both in vitro and in vivo protected rat primary cardiomyocytes from hypoxic death. The inhibition of BNIP3 in acutely ischemic rat hearts resulted in improved retention of ejection fraction, decreased infarct formation, decreased cellular remodeling, and decreased left ventricular remodeling. TABLE OF CONTENTS ABSTRACT ....................................................................................................................... iii LIST OF FIGURES .......................................................................................................... vii LIST OF TABLES ............................................................................................................. ix LIST OF ABBREVIATIONS ..............................................................................................x ACKNOWLEDGEMENTS ...............................................................................................xv Chapter 1. INTRODUCTION ...........................................................................................................1 1.1 General Introduction .......................................................................................1 1.2 Study Rationale ...............................................................................................3 1.3 Specific Aims ..................................................................................................6 1.4 References .......................................................................................................9 2. MYOCARDIAL INFARCTION, GENE THERAPY, AND POLYMER GENE CARRIERS: BACKGROUND AND LITERATURE REVIEW ......................................11 2.1 Myocardial Infarction ...................................................................................11 2.2 Left Ventricular Remodeling ........................................................................12 2.3 Surgical and Pharmacological Management of Myocardial Infarction ........14 2.4 Gene Therapy for Recovery of Myocardial Infarction .................................20 2.5 Tissue Oxygen Levels and Hypoxia Response .............................................22 2.6 The Proapoptotic Protein BNIP3 ..................................................................24 2.7 RNA Interference and siRNA .......................................................................29 2.8 Gene Delivery and Carriers ..........................................................................36 2.9 Cell Therapy for Cardiovascular Disease .....................................................56 2.10 Therapeutic Angiogenesis .............................................................................63 2.11 References .....................................................................................................66 vi 3. BIOREDUCIBLE POLYMER-TRANSFECTED SKELETAL MYOBLASTS FOR VEGF DELIVERY TO ACUTELY ISCHEMIC MYOCARDIUM ........................93 3.1 Abstract .........................................................................................................93 3.2 Introduction ...................................................................................................94 3.3 Materials and Methods ..................................................................................97 3.4 Results .........................................................................................................103 3.5 Discussion ...................................................................................................119 3.6 Conclusions .................................................................................................121 3.7 Acknowledgements .....................................................................................121 3.8 References ...................................................................................................122 4. SIRNA-MEDIATED KNOCKDOWN OF BNIP3 IN RAT CARDIOMYOCYTES USING AN EFFICIENT AND NONTOXIC POLYMER GENE CARRIER ................125 4.1 Abstract .......................................................................................................125 4.2 Introduction .................................................................................................126 4.3 Materials and Methods ................................................................................128 4.4 Results .........................................................................................................136 4.5 Discussion ...................................................................................................148 4.6 Conclusions .................................................................................................152 4.7 Acknowledgements .....................................................................................153 4.8 References ...................................................................................................154 5. BIOREDUCIBLE POLYMER-MEDIATED DELIVERY OF BNIP3 SIRNA TO RETAIN HEART FUNCTION IN A RAT MODEL OF ISCHEMIA/ REPERFUSION INJURY ...............................................................................................158 5.1 Abstract .......................................................................................................158 5.2 Introduction .................................................................................................159 5.3 Materials and Methods ................................................................................161 5.4 Results .........................................................................................................168 5.5 Discussion ...................................................................................................180 5.6 Conclusions .................................................................................................187 5.7 Acknowledgements .....................................................................................187 5.8 References ...................................................................................................188 6. CONCLUSIONS, LIMITATIONS, AND FUTURE DIRECTIONS ..........................193 6.1 Conclusions .................................................................................................193 6.2 Limitations ..................................................................................................195 6.3 Future Directions ........................................................................................198 6.4 References ...................................................................................................201 LIST OF FIGURES Figure 1.1 Graphical overview of myocardial infarction therapy strategies ..................................5 2.1 The cellular oxygen response mediated through HIF-1α ...........................................23 2.2 BNIP3-regulated death pathways ...............................................................................27 2.3 The three domains of cationic lipids used in gene delivery ........................................44 2.4 Overview of polyplex internalization, endosomal escape, DNA expression, and siRNA-mediated knockdown .............................................................................................47 2.5 The chemical structures of the prototypical cationic polymers linear polyethyleneimine (LPEI), branched PEI (bPEI), and poly-L-lysine (PLL) ......................49 2.6 Chemical structures of the degradable cationic polymers acid-labile polyethyleneimine (PEI), poly(4-hydroxyl-L-proline ester) (PHP), and poly[α-(4-aminobutyl)-L-glycolic acid] (PAGA) ...............................................................................50 2.7 Bioreducible poly(amido amines) based on cystamine bisacrylamide .......................55 3.1 Human VEGF expression in rat hearts three days after surgery as measured by ELISA ..............................................................................................................................105 3.2 Cardiac parameters measured by MRI four weeks after treatment ..........................107 3.3 Masson's trichrome staining for infarct size, four weeks after surgery ...................112 3.4 Representative immunohistochemistry stains for SM α-actin and CD31 in the infarct border zone, 4 weeks after myocardial infarction ................................................114 3.5 The loss of cardiomyocytes, 4 weeks after surgery as indicated by loss of immunohistochemical staining for cTnT .........................................................................117 3.6 TUNEL staining for apoptotic activity, 4weeks after surgery ..................................118 4.1 Synthesis of arginine-grafted bioreducible polymer .................................................137 viii 4.2 Chemical structures of cationic polymers screened to transfect rat cardiomyocytes ................................................................................................................138 4.3 Luciferase transfection efficiency assay of various cationic polymers on rat cardiomyocytes ................................................................................................................140 4.4 MTT cytotoxicity assay of various polymers on rat cardiomyocytes .......................141 4.5 Gel retardation assay of ABP:siRNA polyplexes at 0.5:1, 1:1, 5:1, 10:1, 20:1, and 40:1 w:w ratios ..........................................................................................................143 4.6 qRT-PCR knockdown of rat BNIP3 in rat cardiomyocytes .....................................146 4.7 siRNA knockdown of BNIP3 protein over a 7-day period .......................................147 4.8 Beating cardiomyocytes after siRNA-mediated knockdown of BNIP3 and exposure to 24 hours hypoxia ..........................................................................................149 4.9 In vitro live/dead microscopy assay .........................................................................150 5.1 Temporary ligation of the rat heart at the left anterior descending coronary artery showing the infarct region of the left ventricle distal to the ligation site ..............169 5.2 Masson's trichrome staining showing infarct formation between all treatment groups and I/R control .....................................................................................................170 5.3 MRI analysis of rat hearts for functional analysis by ejection fraction ....................173 5.4 MRI analysis of rat hearts for global changes in end diastolic and end systolic left ventricle chamber volumes ........................................................................................175 5.5 Temporal wall thickening for I/R only (top), thoracotomy (middle), and siBNIP3:ABP (bottom) ....................................................................................................176 5.6 Cardiomyocyte hypertrophy and density from H&E staining ..................................178 5.7 BNIP3 immunohistochemical staining showing increased BNIP3 expression ........181 5.8 TUNEL analysis for apoptotic activity in the infarct border zone 14 days after surgery.............................................................................................................................182 LIST OF TABLES Table 2.1 Seminal Publications in the History of RNAi Discovery ...........................................31 2.2 Current siRNA and shRNA Clinical Trials ................................................................33 2.3 Relative Comparison of Gene Transfer Methods .......................................................38 2.4 Cell Types Investigated for Cardiac Repair ................................................................58 4.1 siRNA Sequences Tested for BNIP3 Knockdown Potency .....................................144 LIST OF ABBREVIATIONS AAV adeno-associated virus ABP arginine-grafted bioreducible polymer (see also poly(CBA-DAH-R)) ACE angiotensin converting enzyme AGENT angiogenic gene therapy clinical trials AMD age-related macular degeneration ANG-2 angiopoietin-2 ANOVA analysis of variance Ang II angiotensin II ARC apoptosis repressor with caspase recruitment domain ARNT aryl hydrocarbon receptor nuclear translocator, a.k.a. HIF-1ß AT angiotensin receptor ATP adenosine triphosphate BNIP3 BCL2/adenovirus E1B 19 kDa interacting protein 3 bFGF basic fibroblast growth factor bPEI branched polyethyleneimine BSA bovine serum albumin CABG coronary artery bypass grafting CBA cystamine bisacrylamide CD31 cluster of differentiation 31 (capillary marker) CSC cardiac stem cell xi cTnT cardiac troponin-T CUPID calcium up-regulation by percutaneous administration of gene therapy in cardiac disease clinical trial DAB 3-3' diaminobenzidine DICOM digital imaging and communications in medicine image format DI H2O deionized water DMEM Dubellco's Modified Eagle's Medium DMSO dimethyl sulfoxide dsRNA double-stranded RNA EF ejection fraction ELISA enzyme-linked immunosorbent assay EDV end diastolic volume EDTA ethylenediaminetetraacetic acid EDWT end diastolic wall thickness ESV end systolic volume ESWT end systolic wall thickness EtBR ethidium bromide FBS fetal bovine serum FDA U.S. Food and Drug Administration FGF fibroblast growth factor FLASH MRI fast-low-shot-angle FOV MRI field of view GSH reduced glutathione GSSG oxidized glutathione disulfide xii H9C2 secondary rat cardiomyocyte cell line H&E hematoxylin and eosin HIF hypoxia inducible factor HIF-1α hypoxia inducible factor-1α HIF-1ß hypoxia inducible factor-1ß hpf high-powered field HRE hypoxia response element HRP horse radish peroxidase iPSC induced pluripotent stem cell LPEI linear polyethyleneimine LAD left anterior descending coronary artery lPEI linear polyethyleneimine Luc luciferase LV left ventricle MAGIC myoblast autologous grafting in ischemic cardiomyopathy clinical trial MeOH methanol MI myocardial infarction MMP matrix metalloprotease MRI magnetic resonance imaging MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide cytotoxicity assay MWCO molecular weight cut-off NEX MRI number of averages NIH National Institutes of Health xiii NMR nuclear magnetic resonance PAGA poly[α-(4-aminobutyl)-L-glycolic acid)] PBS phosphate-buffered saline PCI percutaneous coronary intervention pCMV cytomegalovirus promoter-containing plasmid PDGF platelet-derived growth factor pDNA plasmid DNA PEG poly(ethylene glycol) PEI polyethyleneimine PHD prolyl hydroxylase PHP poly(4-hydroxy-L-proline ester) PIGF placental growth factor PLL poly-L-lysine poly(CBA:DAH) poly(cystaminebisacrylamide-diaminohexane) poly(CBA:DAH:R) poly(cystaminebisacrylamide-diaminohexane-arginine) pVHL von Hippel-Lindau tumor suppressor protein qRT-PCR quantitative reverse transcriptase PCR RAS renin-angiotensin-system RES reticuloendothelial system RLU relative luminescence units RNAi RNA interference ROS reactive oxygen species rPCM rat primary cardiomyocyte SCID-X1 x-linked severe combined immunodeficiency xiv SEM standard error of the mean SERCA2a sarcoplasmic/endoplasmic reticulum calcium ATPase 2 SHP-1 Src homology domain 2 (SH2)-containing tyrosine phosphatase-1 SMA smooth muscle actin (arteriole marker) siRNA small interfering RNA SS-PAA disulfide-linked poly(amido amine) TAE tris-acetate EDTA TE MRI echo time TFA trifluoroacetic acid TLR toll-like receptor TNF-α tumor necrosis factor α tPA tissue plasminogen activator TR MRI repetition time TUNEL terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling assay UV ultraviolet VEGF vascular endothelial growth factor W/W weight to weight ratio of polymer to DNA or siRNA XIAP x-linked inhibitor of apoptosis protein ACKNOWLEDGEMENTS Completion of this work and project would not have been possible were it not for a number of important and influential people. First and foremost, I would like to thank my professor and mentor, Dr. Sung Wan Kim, for his constant guidance, support, encouragement, and feedback. I would also like to thank my committee co-chair, Dr. William R. Crowley, and all of my committee members, Dr. Donald K. Blumenthal, Dr. Philip J. Moos, and Dr. David A. Bull. Their willingness to meet consistently during my academic career and to offer their support and advice were immensely helpful. In addition to the support and guidance of my committee members, I would like to extend my thanks to the faculty and staff of the Department of Pharmacology, my class professors, and the Graduate Training Committee. My time here at the University of Utah has been one of intellectual and emotional expansion and I thank all of you for your help in making me who I am now. My thanks goes out to all of my lab members, past and present, especially Katie Blevins, Stelios Florinas, Mei Ou, and Young Wook Won for your assistance in troubleshooting, brain-storming, and venting. I consider you all friends in addition to valuable coworkers. To my family and parents, I also extend my thanks for your support, love, and patience throughout this process. And to my wife Megan, I extend the biggest thanks of xvi all. This would not have been possible were it not for the support, motivation, and encouragement you provided. This work was supported through the National Institutes of Health, which provided funding through the National Heart, Lung, and Blood Institute grants HL065477 and HL071541. Chapter 3 is reprinted from Biomaterials, 32/3, A.N. McGinn, H.Y. Nam, M. Ou, N. Hu, C.M. Straub, J.W. Yockman, D.A. Bull, S.W. Kim, Bioreducible polymer-transfected skeletal myoblasts for VEGF delivery to acutely ischemic myocardium, 942-949, (2011), with permission from Elsevier. 1CHAPTER 1 INTRODUCTION 1.1 General Introduction Gene therapy is broadly defined as the use of nucleic acids as pharmacological agents at a cellular level to correct genetic abnormalities and dysfunction or to express a protein to treat a disease. Over the past 40 years, the concept of treating or curing disease at its genetic root has moved from mere speculation to a more immediate reality, with thousands of gene therapy clinical trials performed to date [1, 2]. Gene therapy can take many forms, whether it is used to deliver plasmid DNA, small interfering RNA (siRNA), or sense/antisense oligonucleotides and these techniques have all been used in clinical therapies to treat diseases such as muscular dystrophy, cancer, cardiovascular disease, and even HIV [3-6]. However, despite the promise of gene therapy and our increased understanding of the genetic abnormalities that cause disease, the greatest barrier to achieving efficient gene therapy remains the delivery of genetic material to the population of diseased cells [7, 8]. The earliest attempts to overcome the cellular barrier to gene therapy in humans utilized crude and inefficient calcium phosphate systems to chemically transfect bone marrow cells and reinfuse them into the patients [9-11]. More efficient viral systems were not used for gene therapy until several years later and while they are still the predominant vectors used for gene therapy trials today [2], they are not without significant problems 2 such as immunogenicity, mutagenic potential, toxicity, high expense, and difficulty in manufacture [12, 13]. Nonviral vectors for gene therapy represent a promising alternative to viral vectors for gene delivery as they do not suffer from many of the same limitations [14, 15]. Polymer gene carriers are especially attractive as nonviral vectors because they efficiently protect nucleic acids from degradation, mediate transport across cell membranes, and can be easily modified with shielding agents like polyethylene glycol to increase circulation time and decrease toxicity, or modified with targeting motifs to imbue cell-type specificity to the polymer complex [16, 17]. The work in this dissertation evaluated the application of polymer gene delivery to treat the deleterious effects of pathological left ventricular remodeling after myocardial infarction. Myocardial infarction (MI) is a pathological state characterized by an acute loss of blood supply to the myocardium due to coronary artery occlusion. The obstruction results in decreased oxygen and nutrient delivery to myocardial tissues, causing local tissue death and formation of an infarct. This tissue loss occurs via necrosis, apoptosis, and autophagy and results in a weaker heart that is greatly predisposed to cardiac failure. Significant interest has been shown in developing therapeutic interventions to limit remodeling of the left ventricle (LV), restore heart function, and improve long-term patient prognosis, as the current pharmacological treatments do little to prevent LV remodeling. One technique that has shown potential is antiapoptotic and proangiogenic gene therapy of the LV to restore heart function and limit infarct expansion. While most gene therapy research for treating MI has focused on delivery of recovery factors to ameliorate damage to the myocardium [18-20], current research has suggested that there 3 are some very promising targets that can help cardiomyocytes survive the hypoxic insult caused by MI, such as targeting proapoptotic factors like BNIP3, a hypoxia-inducible cell death switch that has been shown to be highly upregulated in the myocardium post-MI [21-23]. The work in this dissertation focuses on two therapeutic approaches applying polymer-mediated gene delivery to treat myocardial infarction and tests these approaches in animal models. First, we detail a combined cell therapy/angiogenic strategy where primary rat skeletal myoblasts were transfected ex vivo with a VEGF165 plasmid and then implanted directly to the infarcted rat myocardium to act as VEGF bioreactors. Second, we detail the design of small interfering RNA (siRNA) towards BNIP3, a proapoptotic protein specifically upregulated by hypoxia and the validation of this target to ameliorate hypoxic cell death after exposure to hypoxia. This in vitro validation was then further investigated in a rat model of myocardial ischemia/reperfusion injury. The multiple approaches to ameliorating the deleterious effects of left ventricle remodeling postmyocardial infarction in this study lead to the dual hypothesis that 1) implantation of VEGF165-expressing skeletal myoblasts to infarcted myocardium preserves heart function, increases neoangiogenesis, decreases cardiomyocyte cell death, and increases cardiomyocyte viability, and that 2) reduction of cardiomyocyte BNIP3 levels after myocardial infarction in the short term by siRNA therapy results in decreased hypoxia-induced cell death, increased cell survival, and retention of heart function. 1.2 Study Rationale The work in this dissertation focuses on identifying polymers with high biocompatibility and low toxicity, identifying and validating gene targets for the 4 treatment of myocardial infarction, and combining the two components together to successfully treat and limit the negative effects of left ventricular remodeling in animal models of myocardial infarction (Figure 1.1). The chapter following this brief introduction contains background information and a review of the current literature surrounding myocardial infarction, siRNA and plasmid gene therapy for cardiovascular disease, and polymer gene carriers. The work in Chapter 3 is unique as it combines polymer-mediated gene therapy with cell therapy to revascularize and repopulate the myocardium after myocardial infarction. This chapter details the ex vivo transfection of primary rat skeletal myoblasts with a gene expressing VEGF165 and the implantation of these cells in the cardiac wall post-MI. Most angiogenic gene therapy approaches have focused on direct transfection of the endogenous myocardial cell population of cardiomyocytes and fibroblasts. However, these cell populations have proven to be extremely difficult to transfect in situ, which results in low expression of transgenes. By transfecting skeletal myoblasts ex vivo with an efficient and nontoxic bioreducible polymer, we were able to attain high transfection efficiency, combining the benefit of high angiogenic factor expression and secretion to surrounding tissues with the positive benefit of cell therapy with a myocyte progenitor cell. The implanted cells provide further benefit by repopulating the myocardium and by secreting recovery factors into the myocardial milieu. The fourth chapter describes the work that was done in determining the polymer gene carriers that would be used in later studies to directly transfect the myocardium with high efficiency and low toxicity. This chapter also details the work performed to evaluate the siRNA target BNIP3 to reduce hypoxia-induced cell death in the infarcted 5 Figure 1.1. Graphical overview of myocardial infarction therapy strategies. 6 myocardium. An in vitro model of hypoxia was used to determine the effects of BNIP3 knockdown on hypoxia survival. To increase the translatability to subsequent in vivo studies, primary rat neonatal cardiomyocytes were used. Chapter 5 applies the work performed in the previous chapter in an in vivo rat model of ischemia/reperfusion injury. The transient knockdown of BNIP3 immediately after ischemia/reperfusion injury was achieved by delivering BNIP3 siRNA directly to the myocardium with a novel arginine-conjugated bioreducible polymer. Direct myocardial siRNA gene therapy is easier to achieve than myocardial plasmid gene therapy as siRNA needs only reach the cytoplasm to exert its action, whereas plasmids must penetrate the nucleus in order for expression to occur. 1.3 Specific Aims In order to test the aforementioned hypotheses, this dissertation was broken up into three separate specific aims. 1.3.1 Specific Aim 1 Determine if intramyocardial implantation of VEGF165-transfected skeletal myoblasts after acute myocardial infarction in rats results in retained heart function, decreased infarct size, increased cardiomyocyte viability, and increased capillary and arteriole formation. We used a rat model of permanent coronary artery occlusion and randomly assigned animals to one of four groups: 1) sham operation control (thoracotomy only), 2) ligation only, 3) implanted skeletal myoblasts alone, and 4) implanted VEGF165-expressing skeletal myoblasts. Cardiac function and ventricle geometry were measured four-weeks after surgery by MRI. Histology to determine infarct size and 7 immunohistochemistry to measure vessel formation, cardiomyocyte viability, and apoptosis were performed on isolated hearts immediately after MRI acquisition. 1.3.2 Specific Aim 2 Characterize the optimal bioreducible polymer carrier to deliver genetic material to cardiomyocytes and determine the protective effects of siRNA-mediated BNIP3 knockdown on hypoxic cardiomyocytes. Selection of an ideal polymer gene delivery agent for use in vivo is a multifaceted process. We identified that a novel poly(amido amine) polymer synthesized by our lab, arginine-grafted poly(cystaminebisacrylamide diaminohexane) (referred to as arginine-grafted bioreducible polymer; ABP), showed the ideal characteristics of high transfection efficiency, low cytotoxicity, and prolonged stability [24, 25]. Cardiomyocytes (especially those from a primary animal source) have long been known to be notoriously difficult to transfect [26, 27]; thus, the identification of the optimal polymer to obtain high transfection efficiencies with minimal toxicity is paramount. 1.3.3 Specific Aim 3 Determine if a reduction of BNIP3 levels in rat cardiomyocytes after acute myocardial ischemia/reperfusion injury results in increased cell survival, decreased apoptosis, and retention of heart function in an in vivo rat model. We used a rat coronary ligation model of myocardial infarction to evaluate the effectiveness of polymer-mediated BNIP3 mRNA knockdown in increasing cardiomyocyte survival and subsequently retaining heart function. Rats were randomly assigned to one of five groups: 1) sham operated control (thoracotomy only), 2) coronary ligation ischemia/reperfusion control, 3) siBNIP3 delivered intramyocardially via 8 arginine-grafted bioreducible polymer (siBNIP3:ABP), 4) siBNIP3 delivered intramyocardially alone (naked siRNA), and 5) siLuciferase delivered intramyocardially via arginine-grafted bioreducible polymer (siLuc:ABP). Two weeks after surgery, heart function and global and regional ventricle geometry were measured via magnetic resonance imaging (MRI). Additionally, histological and immunohistological analyses were performed on isolated hearts to determine infarct size, cardiomyocyte geometry, cardiomyocyte viability, BNIP3 expression, and degree of apoptosis via TUNEL assay. 9 1.4 References [1] T. Friedmann, A brief history of gene therapy, Nat. Genet., 2 (1992) 93-98. [2] The Journal of Gene Medicine Clinical Trials Database, in: J. Gene Med., John Wiley & Sons, Ltd., 2012. [3] P.R. Clemens, S. Eghtesad, D.P. Reay, Gene therapy for muscular dystrophy reaches human clinical trial, Ann. Neurol., 66 (2009) 267-270. [4] R.S. Bora, D. Gupta, T.K. Mukkur, K.S. Saini, RNA interference therapeutics for cancer: challenges and opportunities (review), Mol. Med. Report, 6 (2012) 9-15. [5] M. Hedman, J. Hartikainen, S. Yla-Herttuala, Progress and prospects: hurdles to cardiovascular gene therapy clinical trials, Gene Ther., 18 (2011) 743-749. [6] S.J. Zeller, P. Kumar, RNA-based gene therapy for the treatment and prevention of HIV: from bench to bedside, Yale J. Biol. Med., 84 (2011) 301-309. [7] M. Nishikawa, L. Huang, Nonviral vectors in the new millennium: delivery barriers in gene transfer, Hum. Gene Ther., 12 (2001) 861-870. [8] D.M. Dykxhoorn, D. Palliser, J. Lieberman, The silent treatment: siRNAs as small molecule drugs, Gene Ther., 13 (2006) 541-552. [9] N. Wade, Gene therapy pioneer draws Mikadoesque rap, Science, 212 (1981) 1253. [10] N. Wade, Gene therapy caught in more entanglements, Science, 212 (1981) 24-25. [11] N. Wade, UCLA gene therapy racked by friendly fire, Science, 210 (1980) 509-511. [12] R. Tomanin, M. Scarpa, Why do we need new gene therapy viral vectors? Characteristics, limitations and future perspectives of viral vector transduction, Curr. Gene Ther., 4 (2004) 357-372. [13] H.S. Zhou, D.P. Liu, C.C. Liang, Challenges and strategies: the immune responses in gene therapy, Med. Res. Rev., 24 (2004) 748-761. [14] J.H. Jeong, S.W. Kim, T.G. Park, Molecular design of functional polymers for gene therapy, Prog. Polym. Sci., 32 (2007) 1239-1274. [15] T.G. Park, J.H. Jeong, S.W. Kim, Current status of polymeric gene delivery systems, Adv. Drug Deliv. Rev., 58 (2006) 467-486. [16] S.C. De Smedt, J. Demeester, W.E. Hennink, Cationic polymer based gene delivery systems, Pharm. Res., 17 (2000) 113-126. [17] D.W. Pack, A.S. Hoffman, S. Pun, P.S. Stayton, Design and development of polymers for gene delivery, Nat. Rev. Drug Discov., 4 (2005) 581-593. 10 [18] T.A. Khan, F.W. Sellke, R.J. Laham, Gene therapy progress and prospects: therapeutic angiogenesis for limb and myocardial ischemia, Gene Ther., 10 (2003) 285-291. [19] D.W. Losordo, P.R. Vale, J.F. Symes, C.H. Dunnington, D.D. Esakof, M. Maysky, A.B. Ashare, K. Lathi, J.M. Isner, Gene therapy for myocardial angiogenesis: initial clinical results with direct myocardial injection of phVEGF165 as sole therapy for myocardial ischemia, Circulation, 98 (1998) 2800-2804. [20] T.K. Rosengart, L.Y. Lee, S.R. Patel, T.A. Sanborn, M. Parikh, G.W. Bergman, R. Hachamovitch, M. Szulc, P.D. Kligfield, P.M. Okin, R.T. Hahn, R.B. Devereux, M.R. Post, N.R. Hackett, T. Foster, T.M. Grasso, M.L. Lesser, O.W. Isom, R.G. Crystal, Angiogenesis gene therapy: phase I assessment of direct intramyocardial administration of an adenovirus vector expressing VEGF121 cDNA to individuals with clinically significant severe coronary artery disease, Circulation, 100 (1999) 468-474. [21] A. Hamacher-Brady, N.R. Brady, S.E. Logue, M.R. Sayen, M. Jinno, L.A. Kirshenbaum, R.A. Gottlieb, A.B. Gustafsson, Response to myocardial ischemia/reperfusion injury involves Bnip3 and autophagy, Cell Death Differ., 14 (2007) 146-157. [22] D.A. Kubli, M.N. Quinsay, C. Huang, Y. Lee, A.B. Gustafsson, Bnip3 functions as a mitochondrial sensor of oxidative stress during myocardial ischemia and reperfusion, Am. J. Physiol. Heart Circ. Physiol., 295 (2008) H2025-2031. [23] T.R. Burton, S.B. Gibson, The role of Bcl-2 family member BNIP3 in cell death and disease: NIPping at the heels of cell death, Cell Death Differ., (2009). [24] T.I. Kim, M. Ou, M. Lee, S.W. Kim, Arginine-grafted bioreducible poly(disulfide amine) for gene delivery systems, Biomaterials, 30 (2009) 658-664. [25] S.H. Kim, J.H. Jeong, T.I. Kim, S.W. Kim, D.A. Bull, VEGF siRNA delivery system using arginine-grafted bioreducible poly(disulfide amine), Mol. Pharm., 6 (2009) 718-726. [26] C. Thiel, M. Nix, Efficient transfection of primary cells relevant for cardiovascular research by nucleofection, Methods Mol. Med., 129 (2006) 255-266. [27] S. Bauer, S.K. Maier, L. Neyses, A.H. Maass, Optimization of gene transfer into neonatal rat cardiomyocytes and unmasking of cytomegalovirus promoter silencing, DNA Cell Biol., 24 (2005) 381-387. CHAPTER 2 MYOCARDIAL INFARCTION, GENE THERAPY, AND POLYMER GENE CARRIERS: BACKGROUND AND LITERATURE REVIEW 2.1 Myocardial Infarction Cardiovascular disease is the most common cause of death in developed nations and accounts for one out of every three deaths in the United States for a total of 811,900 deaths in 2008 [1]. Of the various forms of cardiovascular disease, coronary heart disease is the deadliest, accounting for half of all reported cardiovascular disease-related deaths and 405,300 deaths in 2008 [1]. Myocardial infarction (MI), or heart attack, is a particularly deadly form of heart disease (1 out of 20 deaths, 2008 total of 134,000 deaths [1]) and is characterized by a loss of blood flow to the heart, typically due to coronary artery blockage, resulting in ischemic tissue death. Even when a person survives acute MI, the long-term prognosis is daunting, with an average survival of around six years [1]. The poor prognosis after MI is due to both microscopic cellular and macroscopic geometric damage leading to deterioration in heart function and progression towards heart failure. The deteriorated heart performance is the result of further death of cardiac cells and MI-initiated remodeling of the left ventricle (LV), a process that has been shown to occur immediately after MI and which may continue for the duration of the patient's life [2-4]. The remodeling ultimately results in LV dilation, which has been shown to be a primary determinant of morbidity and mortality [5]. Due to the high 12 incidence of MI and the limited treatment options in preventing progression to heart failure, many researchers are attempting to develop treatments that will limit LV remodeling and retain heart function. 2.2 Left Ventricular Remodeling Left ventricular remodeling after myocardial infarction is a complex, progressive, and pathological transformation whereby the ventricle changes shape, size, and composition after MI, leading to decreased function [4, 6]. The heart's capacity to remodel is a necessary adaptive process for normal growth, development, and compensation for exercise stress, but after MI, the process can become maladaptive, decreasing the performance of the heart. Remodeling of the LV has been shown to begin as soon as minutes after a myocardial infarction occurs and can continue for many months and even years [7]. These changes occur due to pathological alterations in the myocardium, including cardiomyocyte apoptosis, cardiomyocyte hypertrophy, fibroblast proliferation, and increased intercellular fibrosis. Slowing or eliminating remodeling is of the utmost clinical value as the degree of remodeling has been shown since the 1970s to correlate highly with poor prognosis, future risk for heart failure, and increased mortality [5, 8, 9]. The process of LV remodeling is regulated by a number of factors, including genetic, mechanical, and hormonal signals [10]. As a result of these multifaceted signals, remodeling occurs at three distinct levels within the heart: at the ventricular level, involving changes in volume, wall thickness, and shape; at the cellular level, involving changes in cell volume, length, number, and cross-sectional area; and at the molecular level, involving changes in gene expression due to neurohormonal signals. 13 2.2.1 Molecular and Cellular Remodeling Remodeling begins with activation of an array of paracrine, neuroendocrine, and autocrine factors. These factors are quickly upregulated in response to myocardial injury, mechanical wall stress, alteration in hemodynamics and blood pressure, hypoxia, and increased reactive oxygen species (ROS). Hypotension incident to myocardial infarction activates the renin-angiotensin-system (RAS) [11], increases secretion of natriuretic peptides [12], and increases norepinephrine and other catecholamine production by the adrenal medulla [13, 14]. MI also causes activation of inflammation pathways by triggering a cytokine cascade initiated by tumor necrosis factor alpha (TNF-α) [15]. Dysfunction in the myocardial wall activates angiotensin II type (AT) stretch receptors, inducing hypertrophy through angiotensin II (Ang II) [16]. Hypoxia upregulates a large number of genes with hypoxia inducible factor- (HIF) sensitive hypoxia response elements (HREs) and reperfusion causes a large degree of damage due to rapid increases in reactive oxygen species [17-19]. Collagenases and proteins that degrade the extracellular matrix like matrix metalloproteases (MMPs) are also upregulated, causing degradation of the normal connective tissue and remodeling of the extracellular matrix [20, 21]. 2.2.2 Geometric Ventricular Remodeling Ultimately, the cellular and molecular changes that occur above present as global and regional changes in the LV. Due to the changes in cardiomyocyte numbers, size, and utility, the LV begins to alter in shape and function. The walls of the ventricle begin to thin due to increased fibrosis and cardiomyocyte slippage [22]. Loss of cardiomyocytes via necrosis produces an inflammatory response with fibroblast proliferation and collagen 14 deposition to replace the dead cells. The remaining cells are embedded and isolated within the fibrotic scar and the rhythmic contraction of the cardiomyocytes begins to suffer as a result [6, 23]. The thinning walls increase the mechanical strain within the heart, causing the ventricle to progressively deform and expand [24]. This expansion leads to increased chamber volume at both systole and diastole with increased LV diameters, especially through the infarct zone and towards the apex of the heart [25]. Due to the serious implications of LV remodeling following MI and its high correlation to heart failure rates, researchers and clinicians are focused on surgical and pharmacological strategies to limit the extent of remodeling that occurs. 2.3 Surgical and Pharmacological Management of Myocardial Infarction In the event of acute myocardial infarction, the most pressing issue is to eliminate the occlusion, reestablishing blood flow to the myocardium. Rapid restoration of blood flow to the myocardium as measured by time to reperfusion has been consistently demonstrated as the most effect means of increasing patient survival while lessening infarct size [26, 27]. Achieving quick time to reperfusion is a difficult process, however, as there are many steps that must occur rapidly from when a patient first experiences symptoms of an acute MI to when the patient is treated for the condition. A patient suffering an acute MI must first be quickly transported to the hospital, admitted to the emergency department, triaged, and diagnosed, and then finally treated with either thrombolytics or angioplastic percutaneous coronary intervention (PCI) [28]. In complicated cases requiring multivessel replacement, surgical coronary artery bypass grafting (CABG) is performed. 15 2.3.1 Drugs for Early Reperfusion Thrombolytic therapy involves the use of fibrinolytic agents to lyse the thrombus/thrombi occluding the coronary arteries. There are three major classes of thrombolytic drugs in use today: the tissue plasminogen activators (tPAs) alteplase, reteplase, and tenecteplase; the streptokinases anistreplase, and natural streptokinase; and urokinase. All three of these drug classes work on the same target within the same clot formation pathway-plasminogen. Plasminogen is the inactive precursor to the active fibrinolytic plasmin and is activated via cleavage of a signal peptide to liberate active plasmin. The tPAs are all enzyme mimetics of endogenous tPA, which is a proteolytic enzyme released by endothelial cells in response to signals such as hemostasis. The enzyme tPA and its drug analogs function by binding to fibrin where it activates plasminogen that is also bound to the fibrin clot. This localized effect gives the tPA class of fibrinolytics a slight advantage over other fibrinolytic drug classes as it produces more local fibrinolysis with less systemic lytic effect [29]. Alteplase is a recombinant version of the endogenous tPA enzyme but suffers from an extremely short systemic half-life of approximately 5 minutes [30]. Reteplase and tenecteplase are both recombinant mutant forms of tPA with modifications to increase their circulating half-life to 13-16 minutes and 90-130 minutes, respectively [30]. Similar to the tPA class of drugs, streptokinase also acts on plasminogen, but does not possess enzymatic activity to convert plasminogen to plasmin. Instead, it induces a conformational change in plasminogen, which exposes the plasminogen active site, increasing the liberation of plasmin. Since the introduction of the newer tPA class, 16 streptokinase is not commonly used in a clinical setting as tPAs have been shown to be more effective [31]. Urokinase is not commonly utilized for acute MI as it is not indicated for the condition, being used instead to clear pulmonary emboli. The use of thrombolytics for acute MI is very time dependent. After approximately 2 hours, fibrin cross-linking within the clot is extensive and the clot is more compact, making lysis more difficult [26, 32]. Thrombolysis of aged clots requires higher fibrinolytic concentrations and is accompanied by increased risk of hemorrhage. 2.3.2 Angioplastic Intervention for Reperfusion The other common method to restore myocardial blood flow and remove coronary occlusion is primary percutaneous coronary intervention (PCI). PCI involves inserting a guided catheter through a peripheral (usually femoral) artery to gain direct access to the coronary artery blockage. Meta-analyses have consistently shown better long-term outcomes for PCI versus thrombolytic therapy, reducing short-term mortality, reinfarction, stroke, and long-term death [33]. PCI was first used for revascularization in the 1970s by inserting a catheter with a balloon tip through the vasculature and through the coronary blockage then expanding the balloon to expand the artery, restoring blood flow [34]. While percutaneous balloon angioplasty revolutionized the way coronary artery diseases were treated, the effectiveness of the procedure was limited due to side effects such as thrombus formation and rapid coronary artery reocclusion. To improve upon balloon angioplasty, other devices were developed that removed the plaque blockage or physically left a device such as a stent to hold the vessel open. Bare metal stents were introduced in 1986 [35] and were further improved in 2003 by coating bare metal stents with a drug-eluting polymer 17 to form drug-eluting stents [36, 37]. The drug-eluting stents release pharmacological agents to limit neointimal hyperplasia, a process where the damage to the endothelial wall causes continued growth and division of endothelial and smooth muscle cells into the intimal space of the vessel, eventually restricting blood flow and occluding the vessel [38, 39]. Current drug-eluting stents contain antiproliferative agents such as sirolimus and paclitaxel. Early trials showed excellent prevention of restenosis, with 0% restenosis using a sirolimus-eluting stent versus 26.6% restenosis with a bare metal stent at 6 months [40]. Similarly, some of the early trials with paclitaxel-eluting stents showed the same 0% 6-month restenosis rate versus a 10% restenosis rate with standard stents [41]. While the short-term success of drug-eluting stents is extremely high, long-term use of the devices is accompanied by risks that are not seen in patients who receive bare metal stents. Some of these complications include late and very-late thrombosis [42, 43], hypersensitivity immune response [44, 45], delayed healing [46], and endothelial dysfunction [47]. Recently conducted meta-analyses of trials using drug-eluting stents confirms that their use is not without unique risks, but shows that the benefits are still in favor of using the newer stents over bare metal stents [48, 49]. Next generation drug-eluting stents are already in development and are currently being tested in humans. Improvements include new biodegradable polymers with enhanced biocompatibility that do not induce as serious an inflammatory response as the current nonbiodegradable polymers [50], updated antiproliferative drugs, and polymer-free drug-eluting stents [51, 52]. 18 2.3.3 Pharmacological Management Post-MI Patients managing the after-effects of myocardial infarction can be placed on a variety of pharmacological agents designed to lessen the risk factors for heart failure. These drugs positively affect the heart by either increasing the contractile force of remaining cardiomyocytes or by reducing the workload on the heart. These drugs include angiotensin-converting enzyme (ACE) inhibitors, ß-blockers, diuretics, angiotensin receptor blockers, aldosterone antagonists, hydralazine and isosorbide dinitrate, and digitalis glycosides. The use of inotropic agents in heart failure is usually managed with care as most inotropic agents (with the exception of digitalis glycosides) have shown increased mortality rates in heart failure patients [53, 54]. However, while the use of inotropes may shorten the patient's lifespan, the quality of life of patients on these drugs is dramatically improved. Inotropic agents such as digoxin, phosphodiesterase inhibitors (milrinone), and ß-receptor agonists (dopamine and dobutamine) exert their positive hemodynamic effects by increasing cardiac output while reducing, to a lesser degree, neurohormonal activation. Angiotensin receptor blockers and ACE inhibitors are some of the most common drugs used post-MI and have demonstrated their ability to both slow, stabilize, and even mildly reverse pathological remodeling [55]. Due to neurohormonal activation after MI, the sympathetic nervous system and RAS are activated, leading to high levels of angiotensin, aldosterone, and norepinephrine in the myocardium. Increased myocardial levels of norepinephrine are especially problematic as norepinephrine is proapoptotic to the myocardium [56, 57]. Large multicenter trials have demonstrated increased survival, retention of function, fewer hospitalizations, and retained heart geometry in MI patients 19 receiving ACE inhibitors [58-61]. Efficient inhibition of the RAS using angiotensin-receptor blockers has also been shown to provide a survival benefit and limit remodeling [62, 63]. ß-blockers are used to counteract the effects of sympathetic nervous system activation and are well known to reduce remodeling and limit progression to heart failure. The beneficial mechanism of these drugs comes from negative chronotropic effects on heart rate, resulting in reduced intramyocardial neurohormone levels, decreased oxygen demand, and decreased infarct expansion [64]. Clinical studies have repeatedly demonstrated this benefit with long-term and short-term improvement seen in infarct expansion, LV ejection fraction, and myocardial dysfunction [65-67]. Aldosterone antagonism is hypothesized to produce beneficial effects in post-MI patients through inhibition of myocardial fibrosis [68, 69]. Studies evaluating aldosterone antagonists in combination with angiotensin receptor blockers showed improvement in LV mass, end diastolic/systolic volumes, and ejection fraction [70]. Patients who have undergone some form of PCI receive prophylactic treatment to prevent recurrence of atherosclerosis and thrombosis. Most guidelines recommend dual antiplatelet therapy with life-time, low-dose aspirin, and thienopyridine prescribed for one year. Statin drugs are also prescribed to help reduce serum cholesterol levels, ß-blockers are used to decrease cardiac workload and arrhythmias, and lifestyle management changes are recommended to lower the risk for recurrence. The current first-line therapies of angiotensin and ß receptor blockade, and ACE inhibition, have clearly been demonstrated to aid in patient outcome and limit remodeling; however, pathological remodeling still continues in most patients. Because 20 of this, novel targets and treatment strategies such as antiapoptosis, antiinflammation, matrixmetalloprotease inhibition, cell-based, and proangiogenic therapies are currently being evaluated to prevent remodeling and increase LV function and patient survival. 2.3.4 Decreasing Infarct Size at Primary Intervention Several strategies to reduce infarct size by limiting the extent of injury caused by reperfusion are actively being investigated. Some success has been observed in trials using adjuvant adenosine to limit infarct size by reducing free radical formation and neutrophil accumulation. The AMISTAD I trial found that administration of adenosine reduced anterior infarct size by 33% but had no effect on inferior infarctions [71]. Further investigation in the AMISTAD II trial found a nonsignificant trend towards smaller infarct sizes with low-dose adenosine but significantly smaller infarct sizes with high doses [72]. Some progress has been made in using therapeutic hypothermia to reduce the degree of reperfusion injury. Therapeutic hypothermia requires lowering the temperature of the heart either locally or by cooling the whole body to approximately 33 °C. This approach has demonstrated cardioprotective capabilities in animal models [73, 74] and in human studies [75]. However, it suffers from several limitations, including uncontrollable shivering, decreased drug metabolism, and the prolonged period required to achieve adequate cooling (approximately 1 hour) [76]. 2.4 Gene Therapy for Recovery of Myocardial Function Significant gains have been made in recent years in pharmacological therapies of the infarcted myocardium. While these gains have improved post-MI life expectancy and helped limit progression to heart failure, the therapeutic regimens are mainly focused on 21 decreasing the burden on the injured heart. In order to more fully treat the infarcted heart, therapies are actively being sought that can regenerate functional myocardium or prevent the loss of cardiomyocytes caused at the time of injury and afterwards. Gene therapy is one technique that is being investigated as a method for protecting the myocardium and preventing the loss of cardiomyocytes. To date, there have been over 1,800 gene therapy clinical trials performed in humans [77]. Successful gene transfer in the cardiovascular system was first seen as long ago as 1989 [78]. Since that time, the popularity of cardiovascular diseases as a target for gene therapy has continued with cardiovascular diseases second only to cancer in popularity for gene therapy indications. Of all the almost 2,000 gene therapy clinical trials performed to this point, 155 or 8.4% of all trials have been indicated for the treatment of cardiovascular diseases [77]. While many gene targets have been investigated as potential strategies to treat cardiovascular disease in animal models, thus far, the overwhelming majority of human trials have utilized angiogenic factors to treat the disease [79]. The only other target evaluated in humans is sarcoplasmic/endoplasmic reticulum calcium ATPase 2 (SERCA2a). This target was recently evaluated in the Calcium Up-Regulation by Percutaneous Administration of Gene Therapy in Cardiac Disease (CUPID) trial and showed functional improvement after treatment with an adeno-associated virus 1 AAV1-SERCA2a vector [80, 81]. In addition to the angiogenic targets used in human clinical trials, a wider array of targets have been evaluated in animal models and shown to prevent left ventricular remodeling. Other gene therapy approaches include targeting extra-cellular matrix 22 degradation, via TGF-ßR [82], decorin [83], and integrin-linked kinase expression [84]. Other strategies utilize gene transfer to minimize cell death and include targeting antiapoptosis or prosurvival factors after ischemia/reperfusion injury. These targets include soluble Fas [85], heme oxygenase-1 [86], targets such as midkine growth factor [87], Bcl-2 [88], and the TNF-α receptor-1 [89]. Additionally, previous work in our laboratory has investigated VEGF expression in a plasmid under control of a hypoxia-specific RTP-801 promoter [90-92], using a novel cholesterol-conjugated cationic polymer and a terplex system composed of DNA, low density lipoprotein, and stearyl-poly-L-lysine [92]. In addition to plasmid-based approaches, several groups have utilized gene therapy to deliver siRNA to the myocardium to lower mRNA levels of deleterious proteins. Some of the siRNA targets investigated include PHD2 [93-95], SHP-1 [96], ALOX5 [97], and PTP-1B [98]. While these groups have mainly focused on the delivery of prosurvival, protective, and general antiapoptotic genes, no one has been able to demonstrate the utility of a system to protect from more specific hypoxia-induced apoptosis. 2.5 Tissue Oxygen Levels and Hypoxia Response Cellular response to oxygen is a complex and tightly regulated process, mediated by a myriad of proteins (Figure 2.1). One important protein for regulating tissue O2 response is the transcription factor hypoxia inducible factor-1 (HIF-1). HIF-1 is a heterodimeric protein consisting of a HIF-1α and HIF-1ß subunit (HIF-1ß is also known as the aryl hydrocarbon receptor nuclear translocator, ARNT). While HIF-1ß is constitutively expressed and remains localized within the nucleus regardless of O2 levels 23 Figure 2.1. The cellular oxygen response mediated through HIF-1α. 24 [99-101], HIF-1α levels are kept low under normoxic conditions via posttranslational, oxygen-dependent degradation [102]. During normoxia, the 402/564 prolyl residues of HIF-1α are hydroxylated by the oxygen-dependent prolyl hydroxylases (PHD1, PHD2, and PHD3) [103]. The enzymatic activity of these hydroxylases is dependent on both iron and molecular O2 availability. After HIF-1α has been hydroxylated, it binds to the von Hippel-Lindau tumor suppressor protein (pVHL) [104]. Ubiquitination of HIF-1α is mediated by the pVHL complex, targeting HIF-1α for proteasomal degradation. Under hypoxic conditions, the PHDs lack the molecular O2 necessary for their enzymatic action, resulting in decreased HIF-1α degradation. HIF-1α then translocates to the nucleus where it is free to heterodimerize with HIF-1ß to form the active transcription factor HIF-1. This transcription factor induces the upregulation of genes containing hypoxia response elements (HREs) in their promoters, allowing for genetic adaptation to decreased oxygen levels [105]. While the primary physiological response to lowered O2 levels is an adaptive and prosurvival response, if normal levels of O2 are not restored, the cell will ultimately undergo apoptosis, autophagy, or necrosis. 2.6 The Proapoptotic Protein BNIP3 The myocardium has historically been found to be highly resistant to apoptosis with an average observed apoptotic rate of 1 in 100,000 cardiomyocytes [106]. This low level of observed apoptosis is thought to be at least partially due to high myocardial expression levels of caspase inhibitors such as the apoptosis repressor with caspase recruitment domain (ARC) and X-linked inhibitor of apoptosis protein (XIAP) [107]. In diseased hearts, however, the rate of apoptosis has been shown to dramatically increase over 200-fold for both ischemic cardiomyopathies and idiopathic dilated 25 cardiomyopathies [106]. While cardiomyocytes are capable of spontaneous regeneration after myocardial infarction, the degree to which they regenerate is insufficient to compensate for the increased rate of cell loss [108, 109]. With the balance of cell death and regeneration tipped towards cell death, the heart gradually loses more and more cardiomyocytes and progresses in a path towards heart failure. Apoptosis and cell death following acute myocardial infarction has been shown to be readily detectable in humans especially within a one-week time frame after MI. Studies have shown that apoptosis peaks within the first few days, decreasing rapidly after four days to a much lower level [110]. An important cellular mediator of hypoxia-inducible cell death is Bcl-2/adenovirus E1B nineteen-kilodalton interacting protein 3 (BNIP3). BNIP3 is an HRE-containing proapoptotic protein belonging to the Bcl-2 family and is one of the most highly upregulated proteins in response to hypoxia [111-113]. Human BNIP3 is a 194-amino acid protein with several domains that dictate its structure. BNIP3 has a transmembrane domain at residues 164-184, which allows it to insert into the mitochondrial membrane and homodimerize with other BNIP3 proteins in the membrane to form an ion channel [114, 115]. BNIP3 also has a Bcl-2 homology domain 3 (BH3) at residues 104-119, which allows it to bind with the antiapoptotic Bcl-2 [116]. Finally, BNIP3 also has an evolutionarily conserved cysteine residue at amino acid 64, which is thought to increase BNIP3 stability under conditions of oxidative stress [117]. Under oxidative conditions such as after MI when high levels of ROS are present in the myocardium, the cytosolic compartment switches from a reductive environment to an oxidative environment. When this occurs, disulfide bonds become more favorable and 26 stable. The BNIP3 cysteine, which is located on the cytosolic side of the protein when it inserts into the mitochondrial membrane, forms a stabilizing disulfide bonds with its homodimer, increasing its activity and depolarizing the mitochondrial membrane [117]. While BNIP3 has been studied extensively, the exact mechanisms by which it brings about cell death are not yet completely understood. However, it has been shown that cells where BNIP3 is upregulated can undergo one of at least three pathways that can lead to death: apoptosis, autophagy, or necrosis (Figure 2.2). As BNIP3 plays such a pivotal role in the regulation of the hypoxia death response, it is a prime candidate for gene therapy to selectively disable hypoxia-inducible cell death [118]. 2.6.1 BNIP3-Regulated Apoptosis It is postulated that BNIP3 induces cell death through localization to the mitochondrial membrane and depolarizing the membrane [114]. In some cell types, the localization of BNIP3 to the mitochondria causes cytochrome c release, initiating the caspase cascade and committing the cell to apoptosis [119]. However, in other cells such as neurons, BNIP3 induces endonuclease G release from the mitochondria but does not cause cytochrome c release, causing caspase-independent cell death [120]. 2.6.2 BNIP3-Regulated Autophagy Autophagy is a catabolic cellular process by which proteins and subcellular organelles are compartmentalized, and broken down by lysosomal degradation. In the heart and other tissues, autophagy is a normal process allowing cells to eliminate damaged proteins and organelles while scavenging many of the nutrients contained in the digested structures. Unlike apoptosis and necrosis, however, initiation of autophagy does 27 Figure 2.2. BNIP3-regulated death pathways. Adapted from [115] 28 not necessarily end in cell death. Autophagy is a normal cellular process that is observed in some healthy cell types [121-123]. The process allows for recycling of dysfunctional organelles, and defective cytoplasmic proteins without committing the cell to a death fate. In some animal models of repetitive ischemia/reperfusion, cardiac injury autophagy has been observed as a precursor to eventual apoptosis, with autophagic cell death occurring at a later time than apoptotic cells [124]. This observation would suggest a protective mechanism, giving cardiomyocytes extra time to acclimate to the disturbed oxygen and nutrient status post-ischemia/reperfusion and to attempt to avoid apoptosis by removing damaged intracellular structures. The autophagic role of BNIP3 and its related family member Nix/BNIP3L in cardiomyocytes has received a great deal of attention over the past years. The role of BNIP3 in inducing autophagy is thought to be due to its interaction with the antiapoptotic Bcl-2. This antiapoptotic protein plays a role in sequestering beclin-1, a protein which is known to induce autophagy. Since BNIP3 has a BH3 domain through which it can bind Bcl-2, BNIP3 can act as a competitive inhibitor of Bcl-2 for beclin. In periods of high BNIP3 expression such as hypoxia, BNIP3 displaces beclin by binding to Bcl-2, causing beclin to initiate autophagy [125, 126]. 2.6.3 BNIP3-Regulated Necrosis Unlike the death pathways listed above which are energy- and ATP-dependent, necrosis is a passive process. Necrosis usually occurs as a result of overwhelming injury, which causes cellular swelling, dysregulation of organelle function, and ultimately leads to rupture of the cell membrane. This rupture rapidly releases cellular contents to the extracellular environment, damaging neighboring cells and causing inflammation. 29 BNIP3 has been shown to play a role in the necrosis of neurons [127, 128], tumor cells [129-131], and cardiomyocytes [117, 132]. BNIP3-mediated necrosis is thought to occur through interactions at the mitochondrial membrane similar to its site of action for inducing autophagy and apoptosis. However, over-activation of BNIP3 at the mitochondrial membrane is thought to rapidly push the cell to a necrotic as opposed to autophagic fate. In addition to the conserved transmembrane and BH3 domain, BNIP3 also has a conserved cysteine residue at position 64 located on the cytosolic side where BNIP3 inserts into the mitochondrial membrane [117]. When the cytoplasmic compartment shifts from a predominantly reductive environment to a predominantly oxidative environment under oxidative stress situations such as is experienced during reperfusion injury in cardiomyocytes, this cysteine residue forms a disulfide bond with the cysteine residue on another BNIP3 protein forming a covalent homodimer [116, 117]. This increase in activity on the mitochondrial membrane is thought to push the cell towards a necrotic as opposed to autophagic cell death [133]. 2.7 RNA Interference and siRNA RNA interference (RNAi) refers to the endogenous cellular pathway by which double-stranded RNA (dsRNA) is processed into small interfering RNA (siRNA) that can target mRNA transcripts for gene silencing. The RNAi phenomenon was originally observed by Guo and colleagues in 1995. While performing gene function studies in C. elegans, they detected a sense strand sequence as the complementary antisense sequence for repressing gene expression [134]. This observation was further investigated by Fire et al. in 1998 and they published a seminal paper demonstrating that introduction of double-stranded RNA sequences 30 produced a more potent and specific interference effect than either the sense or antisense strand alone [135]. This discovery initiated a period of rapid research further elucidating the mechanism and potential impact of RNAi. A summary of many important publications highlighting the discovery and development of RNAi can be seen in Table 2.1. Using siRNA to target a specific protein is particularly attractive because of the ability of siRNA to knockdown one target while essentially sparing all other proteins. Careful design of siRNA constructs is performed by using algorithms that optimize siRNA efficiency, analyzing selected constructs for lack of shared homology to other proteins by utilizing NCBI Blast searches, and by screening several siRNA constructs to determine the sequence with the highest knockdown efficiency and fewest off-target effects. Additionally, published sequences that have been shown to effect efficient knockdown of the target can be used [136-138]. Many researchers use siRNA as a laboratory tool for the analysis of gene function and to easily and rapidly decrease specific protein levels. However, many people have also been looking at siRNA as a potential therapeutic. While siRNA possess many attributes that make them ideal candidates for therapeutics, there are inherent issues with employing siRNA in vivo. Nucleotides have extremely short extracellular half-lives (seconds to minutes in the circulation [139]). Also, their large size (average 13.5 kDa), charged phosphodiester backbone, and hydrophilicity impede their ability to cross cell membranes. Therefore, a delivery vehicle is needed to protect the nucleotides from degradation and aid in gaining access to the cellular compartment. 31 Table 2.1. Seminal Publications in the History of RNAi Discovery Year Author Organism Discovery Reference 1995 Guo et al. C. elegans RNA sense strand interferes with gene expression [134] 1998 Fire et al. C. elegans Double-stranded RNA reduces gene expression more potently than sense or antisense strands alone [135] 2000 Zamore et al. D. melanogaster Dicer protein processes double-stranded RNA to 21-23 nt fragments [140] 2001 Elbashir et al. Mammals First observation of RNAi in mammalian cells [141] 2002 Paddison et al. Mammals RNAi through expressed short hairpin RNA (shRNA) [142] 2004 Acuity Pharmaceuticals H. sapiens First human Phase I clinical trial using siRNA 2010 Davis et al. H. sapiens Targeted systemic polymer nanoparticle siRNA delivery demonstrated dose-dependent target accumulation [143] 32 2.7.1 siRNA Therapy and Clinical Trials Over forty clinical trials utilizing siRNA or shRNA have been performed or are in progress (Table 2.2). While many of these trials are still early-stage (Phase I and II), at least one trial is in late-stage Phase III trials. Despite the large number of clinical trials, no candidates have yet gained FDA approval. The first clinical trials investigating RNAi were performed with naked, unmodified siRNA and were delivered either by local injection to the therapeutic site or via inhalation to the lungs. Even when the first trials began, the limitations of poor siRNA bioavailability were understood. Therefore, ocular diseases with a separated physiological compartment that allowed easy access and protection from rapid degradation in the circulation were used. However, the knockdown even in these isolated compartments has not been very high and while some benefit was observed in early trials targeting age-related macular degeneration (AMD), it has been hypothesized that the observed benefit may be from activation of toll-like receptors (TLRs) and not target knockdown [146, 147]. While the clinical success of siRNA over the past decade has been disappointing, there is still hope for its therapeutic potential. A better understanding of the limitations of siRNA therapy, including off-target effects, short effect duration, low efficacy, and TLR activation has led to marked overall improvements over the years. New siRNA technologies, including improved siRNA chemical modifications to increase availability and decrease degradability, and development of efficient viral and nonviral delivery systems to increase siRNA delivery to target cells, are currently being investigated in clinical trials with more promising results [145, 148]. 33 Table 2.2. Current siRNA and shRNA Clinical Trials Drug Target Delivery Vehicle Disease Company/Sponsor Phase Status Clinicaltrials.gov Identifier Start Date Cancers Bcr-Abl siRNA Bcr-Abl Anionic liposome CML U of Duisburg I Completed N/A 2010 FANG vaccine furin/ TGFß1 TGFß2 Autologous ex vivo electroporated GMCSF cells Solid cancers Gradalis, Inc. I Recruiting NCT01061840 Dec '09 Advanced cancer Gradalis, Inc. I Recruiting NCT01505153 Feb '12 Melanoma Gradalis, Inc. II Recruiting NCT01453361 Oct '11 Ovarian cancer Gradalis, Inc. II Recruiting NCT01551745 Mar '12 Ovarian cancer Gradalis, Inc. II Recruiting NCT01309230 Feb '11 Colon cancer Gradalis, Inc. II Recruiting NCT01505166 Mar '12 siRNA-transfected dendritic cells LMP2 LMP7 MECL1 Ex vivo transfected dendritic cells Melanoma Duke University I Recruiting NCT00672542 Jan '08 TKM-080301 PLK1 Liposome (cationic and fusogenic lipids with PEG) Cancer Tekmira I Recruiting NCT01262235 Dec '10 Primary/secondary liver cancer NCI I Completed NCT01437007 Aug '11 ALN-VSP02 KSP VEGF Liposome (mixture of cationic and fusogenic lipids with PEG) Liver cancer Alnylam I Completed NCT00882180 Mar '09 Solid tumors Alnylam I Ongoing NCT01158079 Jul '10 Atu027 PKN3 lipoplex Solid tumors Silence Therapeutics I Recruiting NCT00938574 Jun '09 CALAA-01 RRM2 cyclodextrin and PEG nanoparticle with transferrin-targeting Solid tumors Calando I Ongoing NCT00689065 May '08 siG12D LODER KRASG12D LODER biodegradable polymer matrix Pancreatic cancer Silenseed 0/I Recruiting NCT01188785 Jan '11 Pancreatic cancer Silenseed II Not yet open NCT01676259 Sep '12 CEQ508 ß-catenin shRNA in E. coli Colon cancer Marina I On-hold N/A 2010 siRNA-EphA2-DOPC EphA2 Neutral DOPC liposome Advanced cancers MD Anderson I Not yet open NCT01591356 Nov '12 34 Table 2.2. Continued. Drug Target Delivery Vehicle Disease Company/Sponsor Phase Status Clinicaltrials.gov Identifier Start Date Ocular disorders Bevasiranib VEGF Naked siRNA Wet AMD Opko Health I Completed NCT00722384 Aug '04 II Completed NCT00259753 Jul '05 III Terminated NCT00499590 Aug '07 III Withdrawn NCT00557791 Nov '09 Diabetic macular edema Opko Health II Completed NCT00306904 Jan '06 AGN211745 (Sirna-027) VEGF-R1 Naked siRNA AMD Allergan/Sirna I/II Competed NCT00363714 Nov '04 AMD Allergan II Terminated NCT00395057 Jan '07 PF-04523655 RTP-801 Naked siRNA Wet AMD Quark/Pfizer I Completed NCT00725686 Feb '07 Diabetic retinopathy Quark/Pfizer II Terminated NCT00701181 Jun '08 Wet AMD Quark/Pfizer II Completed NCT00713518 Nov '09 Diabetic macular edema, retinopathy, choroidal neovascularization Quark II Recruiting NCT01445899 Feb '12 QPI-1007 Caspase-2 Naked siRNA Nonarteritic anterior ischemic optic neuropathy (NAION) Quark I Ongoing NCT01064505 Feb '10 SYL040012 Adrenergic receptor ß2 Naked siRNA Glaucoma, ocular hypertension Sylentis I Completed NCT00990743 Sep '09 I/II Recruiting NCT01227291 Oct '10 SYL1001 TrpV1 Naked siRNA Dry eye, ocular pain Sylentis I Completed NCT01438281 Jul '11 Inhaled formulations ExcellairTM Syk kinase Unknown Asthma ZaBeCor II Unknown N/A 2009 ALN-RSV01 RSV nucleocapsid Naked siRNA RSV infection Alnylam II Completed NCT00496821 Jul '07 II Completed NCT00658086 Apr '08 IIb Completed NCT01065935 Feb '10 35 Table 2.2. Continued. Drug Target Delivery Vehicle Disease Company/Sponsor Phase Status Clinicaltrials.gov Identifier Start Date Miscellaneous disorders PRO-040201 ApoB Liposome (mixture of cationic and fusogenic lipids with PEG) Hypercholest-erolemia Tekmira I Terminated NCT00927459 Jun '09 ALN-TTR01 TTR Liposome (mixture of cationic and fusogenic lipids with PEG) Transthyretin-mediated amyloidosis Alnylam I Completed NCT01148953 Jun '10 I5NP P53 Naked siRNA AKI after cardiovascular surgery Quark I Completed NCT00554359 Aug '07 AKI after cardiovascular surgery Quark I Terminated NCT00683553 May '08 Delayed graft function in kidney transplantation Quark I/II Recruiting NCT00802347 Dec '08 TD101 Keratin 6aN171K mutant mRNA Naked siRNA Pachyonychia congenital Transderm Ib Completed NCT00716014 Jan '08 pHIV7-shI-TAR-CCR5RZ treated CD4 cells HIV Tat/rev, TAR, CCR5 Ex vivo lentiviral-transfected autologous T-cells HIV City of Hope 0 Teminated NCT01153646 Apr '10 CML = chronic myelogenous leukemia; GMCSF = granulocyte macrophage colony stimulating factor; LMP = low molecular mass protein; MECL1 = multicatalytic endopeptidase complex subunit 1; PLK1 = polo-like kinase 1; NCI = National Cancer Institute; KSP = kinesin spindle protein; VEGF = vascular endothelial growth factor; PEG = polyethylene glycol; PKN3 = protein kinase N3; RRM2 = ribonucleotide reductase subunit M2; VEGF-R1 = VEGF receptor 1; TrpV1 = capsaicin receptor; Syk = spleen tyrosine kinase; RSV = respiratory syncytial virus; ApoB = apolipoprotein B; TTR = transthyretin; p53 = tumor protein 53; AKI = acute kidney injury; HIV = human immunodeficiency virus; TAR = HIV trans-activation response element;CCR5 = CC chemokine receptor type 5. All data from www.clinicaltrials.gov, press releases, corporate websites, Vaishnaw et al. [144], and Burnett et al. [145]. 36 2.8 Gene Delivery and Carriers Naked DNA and RNA alone are highly inefficient drugs. In order for them to be successfully utilized in research or clinical settings, a carrier must be used or chemical modifications made to overcome the many barriers that prevent uptake of genetic material. In order to successfully introduce genetic material in the form of DNA or RNA to a cell, there are several barriers that must be surmounted. These barriers include: 1) rapid degradation in the circulation or extracellular space; 2) nonselective delivery to nontarget cells; 3) crossing the cell membrane; 4) escape from degradative compartments (i.e., the endosome and lysosome); 5) dissociation of cargo from the carrier; and 6) in the case of DNA, successful nuclear localization of the gene [149, 150]. Several methods for gene delivery which overcome these obstacles are currently available, but most techniques can generally be classified within two classes-viral and nonviral [151-154]. As their name implies, viral methods utilize viruses that have been attenuated to be nonreplicating and nonpathogenic and have been loaded with a gene for delivery. While viral gene vectors have evolved over millions, if not billions, of years to efficiently overcome the above-listed hurdles to transduce cellular organisms, they create safety concerns due to their immunogenicity and the potential for insertional mutagenesis. In order to circumvent these problems and generate delivery systems that can efficiently and selectively deliver genetic material to target tissues, a great deal of research has been done on developing nonviral methods for gene therapy. Nonviral methods consist of physical methods of transfection and the use of synthetic chemicals-37 most commonly cationic lipids and polymers. A comparison of the most common viral and nonviral delivery vectors can be seen in Table 2.3. 2.8.1 Viral Gene Delivery Viral methods are extremely efficient, but their clinical usefulness is hampered by several major restrictions. Viral delivery is limited by problems with potential insertional mutagenesis [157, 158], difficulty and high expense incurred in large-scale manufacturing and quality control [159, 160], limited DNA cargo capacity, and immunogenicity. In 1999, an 18-year-old patient enrolled in an adenoviral gene therapy clinical trial for ornithine transcarbamylase deficiency experienced a fatal systemic inflammatory response against the adenoviral vector [161]. Later in 2000, a French group reported successful treatment of a group of children suffering from X-linked severe combined immunodeficiency (SCID-X1). However, in 2002, it was reported that two of the ten children in the trial had developed a form of leukemia [162]. Further analysis of the children led to the discovery that the retrovirus vector had integrated near a proto-oncogene promoter, causing the development of leukemia [163]. 2.8.1.1 Adenoviral Gene Therapy Adenoviruses were one of the first viral vector types used for gene therapy applications. Adenoviruses are a nonenveloped, double-stranded DNA-type virus with an approximately 36 kb genome encoding over 50 polypeptides. The adenovirus has a capacity of approximately 8.5 kb for foreign DNA, which is sufficient to accommodate most therapeutic genes. Some of the benefits of adenovirus vectors include their high efficiency, stability, nonintegrative nature, ability to infect both dividing and quiescent cells, and the lack of viral sequences in the progeny of dividing cells [164, 165]. 38 Table 2.3. Relative Comparison of Gene Transfer Methods Vector Gene Transfer Efficiency Immunogenicity Duration of Expression Loading Capacity Ease of Manufacturing Safety Concerns Viral Adenovirus ++++ ++++ +++ ++ + Lentivirus +++ +++ +++ ++ + HIV origins Retrovirus ++++ ++ ++++ ++ + Insertional mutagenesis Adeno-associated virus ++++ ++ +++ + + Insertional mutagenesis Nonviral Naked + ++ + ++++ ++++ Liposome ++ ++ ++ ++++ +++ Polymer ++ + ++ ++++ ++++ References [155, 156]. 39 However, most humans have been exposed to several adenoviral serotypes and have formed antibodies against the virus. As a result, when adenovirus vectors are administered to humans, almost 90% of the virus is eliminated within 24 hours. In addition to the rapid immune system clearance of adenoviral vectors, the capsid coat on the vectors has been known to stimulate acute and dangerous immune responses [161, 166, 167]. The last disadvantage of using adenoviral vectors is the short duration of gene expression achieved in transfected cells. Typical expression times for adenoviral-transfected cells are between 1-8 weeks [168, 169]. 2.8.1.2 Retroviral Gene Therapy Retroviruses are small RNA viruses (7-11 kb genome) that replicate via a DNA intermediate and were also some of the first viral constructs applied towards gene therapy [170]. While the total genome is much smaller than the adenovirus genome, due to the simplistic nature of the virus, a roughly 8 kb therapeutic gene can be delivered without compromising the function of the virus [171]. The major benefit of retrovirus use is its ability to effect long-term, stable expression of therapeutic genes due to host chromosome integration. Limitations of the use of retroviral vectors include: relatively low transfection efficiency compared to other viruses, low viral stability, the inability to transduce nondividing cell types, and of course the risk of insertional mutagenesis [172]. 2.8.1.3 Adeno-associated Viral Gene Therapy The adeno-associated viruses (AAV) are human parvoviruses that normally require the help of another virus such as adenovirus to infect cells. They possess a small 4.7 kb genome of single-stranded DNA, which is converted to double-stranded DNA 40 after infection. The AAV can chromosomally integrate but shows less potential for insertional mutagenesis due to its preferential insertion to a specific site on the human chromosome 19 [173, 174]. The AAV family is one of the most actively investigated viral vectors currently due to the multiple advantages of the system. AAV vectors produce long-term expression, can efficiently transduce a wide variety of host cells with tissue-specific targeting capabilities, are a nonpathogenic virus, and they evoke a minimal immune response [175-177]. Several of the major limitations of AAV vectors include their small transgene packaging capacity, the still present risk of insertional mutagenesis, and difficulty in producing high viral titers sufficient for clinical trials and large-scale manufacturing [178, 179]. 2.8.1.4 Other Viruses Other viruses that are commonly used for gene therapy include the lentiviruses, and the herpes simplex viruses. Lentivirus vectors are useful for their ability to transfect nondividing cell populations and maintain stable gene expression for extended periods [180]. Vectors based on herpes simplex contain a large genome of 152 kb of double-stranded RNA, allowing for the insertion of large single transgenes or even multiple therapeutic sequences. Less common viral vectors include plus-strand RNA viruses such as Sindbis virus, hepatitis A, and polio. These viruses are more commonly investigated for viral immunization strategies since they produce high, yet transient, gene expression to evoke an immune response against the transduced cells. 41 2.8.2 Nonviral Delivery Systems The nonviral delivery systems consist of methods such as injection of naked DNA [181]; delivery by gene gun [182]; by electroporation [183]; by ultrasound [184]; and by transfer to the highly perfused organs by hydrodynamic delivery [185, 186]. While these methods are capable of inserting genetic material into cells, they are neither practical nor efficient delivery systems. To overcome the limitations of the physical approaches above, chemicals such as lipids and polymers are used as quasi-viral mimetics. These chemicals allow for DNA protection from nuclease activity, relatively efficient uptake by cells, and tissue specificity when conjugated with a targeting motif. 2.8.2.1 Naked Polynucleotide Delivery Some of the first methods used to transfer genetic materials to living tissue utilized unmodified DNA. Delivery of naked nucleotides without the assistance of any physical system is extremely inefficient due to the hydrophilic nature of DNA, its high molecular weight, and its negatively charged phosphate backbone. Thus, the dense negative charges are repulsed by the negatively charged cell membrane. However, delivery of naked nucleic acids has seen some clinical application in closed organ systems such as the eye, treating age-related macular degeneration (AMD) [187, 188]. 2.8.2.2 Electroporation Gene Delivery Electroporation uses an electric field to temporarily increase permeability through the cell membrane. The electric pulses generated by this technique generate transient pores that allow polynucleotides to pass through. This technique has been used to effect gene transfer both in vitro and in vivo since 1982, and 1991, respectively [189, 190]. To perform in vivo electroporation, the DNA is injected first to the target area then an 42 electric field with varied pulse duration, and voltage is applied from two electrodes. This technique can achieve transfection efficiencies similar to those seen from viruses [191], but is best suited to models where localized gene transfer is desirable, the target is accessible by electrodes, and where electric pulses will not disrupt normal organ function (e.g., the heart). 2.8.2.3 Ultrasound Gene Delivery Ultrasound-mediated delivery of polynucleotides functions in a similar manner to electroporation, but uses sound waves instead of electrical pulses to induce cell membrane permeability [192]. Recent advances have combined ultrasound gene therapy with microbubbles to increase transfer efficiencies [193]. The technique employs gas-filled microbubbles stabilized with lipids, polymers, or proteins which stabilize the bubbles to allow for systemic administration. The target site is then exposed to ultrasound and when the microbubbles pass through the sonic cone, they oscillate, and cavitate, releasing their cargo at a high velocity and disrupting cell membranes. Ultrasonic gene delivery systems are more versatile and practical than their electroporation counterparts, but still require a solid tissue target site that is accessible via probe [194]. 2.8.2.4 Hydrodynamic Gene Delivery Hydrodynamic gene transfer was first reported in vivo in 1999 [185]. Gene transfer by this method is achieved by injecting a very large volume of a DNA-containing solution into the tail vein of small rodents. The process of injecting a large volume (usually 8-12% of body weight) over a short period of time (3-5 seconds) causes the endothelial lining of the cardiovascular system to leak, allowing the DNA to penetrate into the tissue [195]. Hydrodynamic gene transfer allows for fast and easy transfection of 43 highly perfused organs such as the liver, kidney, lungs, and heart, but is limited to rodent models. The process of scaling this technique up to the volumes required to achieve the same effect in humans is not feasible. However, advances using catheters to perform local hydrodynamic gene transfer have been made, allowing for this technique to be used in large animals [196]. 2.8.2.5 Gene Gun Delivery Finally, the last physical nonviral gene delivery method commonly utilized is gene transfer via gene gun. The gene gun system was first developed and used to transfect plants in 1987 [197]. Genetic material is coated onto metal particles such as gold, silver, or tungsten and the particles are delivered at a high velocity, powered by a compressed gas such as helium, into the target. This technique has been successfully used to deliver DNA to dermis, muscle, and tumor tissues and has even been used in five clinical trials on gene therapy to date [77]. 2.8.3 Lipid-mediated Gene Delivery Transfer of genetic material using lipids was first accomplished in 1987 by Felgner et al. [198]. Of the nonviral methods, transfection using cationic lipids (i.e., lipofection) is the most popular and extensively studied, with thousands of publications examining hundreds of distinct variations of the basic shared lipid structure [199]. Cationic lipids used for gene delivery share three common features: a positively charged head group, a hydrophobic tail, and a linker region that joins the two regions together (Figure 2.3). Cationic lipids function by condensing polynucleotides through electrostatic interaction between the positively charged lipid head and the negatively charged 44 Figure 2.3. The three domains of cationic lipids used in gene delivery: a hydrophobic tail and a hydrophilic head, joined by a linker. The above structure is that of 1,2-dioleoyloxy-3-trimethylammonium propane (DOTAP), a commonly used cationic lipid for gene delivery. 45 phosphodiester backbone of the DNA or RNA. The resultant tightly bound complex is termed a lipoplex and serves to protect the genetic cargo while aiding in cell penetration. Transfection efficiency and toxicity depend on a myriad of customizable factors, such as the composition of each of the hydrophilic and hydrophobic portions, and the DNA/lipid ratio. The most effective lipoplex formulations typically have a residual positive charge on the complex. This allows for association with cell membranes through interaction with the negatively charged membrane surface and causes the lipoplex to be endocytosed by the cell. In order to escape the endosomal compartment, it is hypothesized that the cationic lipids interact with anionic phospholipids in the wall of the endosome, facilitating concurrent disruption of the endosomal wall and dissociation of the polynucleotides from the lipids, releasing the cargo to the cell interior [200]. Cationic lipoplexes are the most commonly used nonviral chemical transfection system in human trials. To date, they have been used in 111 clinical trials, composing 5.9% of all gene therapy trials completed [77]. Limitations of cationic lipids for human therapy include the immune response they can elicit, acute toxicity, and the short duration of expression they mediate. Since cationic lipid lipoplexes form relatively loose complexes with DNA [201] with some externally exposed cargo [202], the patient's immune system can recognize unmethylated CpG sequences from the plasmid DNA and mount a TLR9-mediated immune response increasing cytokine production [203]. Conjugation of polyethylene glycol (PEG) to cationic lipids has been shown to reduce both acute toxicity and ameliorate the immune response, but reduces transfection efficiency [204]. 46 2.8.4 Polymer Gene Delivery The other nonviral chemical method utilizes cationic polymers to mediate transfection and protect polynucleotides. Polycations are a particularly useful class of polymer for gene therapy. Similar to cationic lipids, these polymers form complexes with genetic material through electrostatic interactions between positively charged groups (usually amines) on the polymer and negatively charged phosphates on the nucleic acid backbone. These polymer:DNA complexes (termed polyplexes) mediate gene delivery primarily through bulk endocytosis into the cell; however, targeting moieties can be attached to increase cell specificity by receptor-mediated endocytosis. Once inside the endosomal compartment, most polyplexes employ an entirely different escape mechanism from the lipoplex escape mechanism, termed the "proton sponge effect" described below. An overview of polyplex entry mechanisms, endosomal escape, and DNA expression or siRNA knockdown can be seen in Figure 2.4. 2.8.4.1 Polyethyleneimine Gene Delivery The polycation polyethyleneimine (PEI) was first shown to effect transfection of cells in 1995 by Boussif et al. [205] and is still used as a comparative benchmark in many experiments today. As a polymer, PEI demonstrated the unique ability to avoid lysosomal degradation by escaping from the endosome within cells. PEI can be synthesized as both a linear or branched polycationic polymer with primary, secondary, and tertiary amine groups in the ratio 1:2:1. The array of amine groups with their varying pKa values gives the polymer the ability to buffer over a wide range of pH values. At the physiological blood pH of 7.4, the amine groups in PEI are approximately 20% protonated, but when 47 Figure 2.4. Overview of polyplex internalization, endosomal escape, DNA expression, and siRNA-mediated knockdown. 48 lowered to pH 5, such as that seen in the endosome, the degree of protonation more than doubles to 45% [206]. The "proton sponge effect" postulates that as protons are pumped into the endosome to lower the pH and aid in degradation, PEI and other cationic polymers buffer against the change, causing more protons and chloride counterions to be pumped in to lower the pH. Due to the much higher ionic strength within the endosome, water diffuses across the endosomal membrane via osmosis, causing the endosome to swell and rupture [206, 207]. The high concentration of counterions and the increased internal polymer electrostatic repulsions caused by the protonated amines is thought to help decomplex the polymer from the DNA/RNA cargo, releasing it to the cytoplasm [208, 209]. PEI comes in two common forms: branched and linear (Figure 2.5). Branched PEI (bPEI) is the most commonly used form for gene delivery due to its higher amine density and its wider buffering range. While the positive charge imbued by the amine groups is necessary for its DNA/RNA-complexation and endosomal-escaping functions, it also contributes to a high degree of toxicity seen when using PEI in vivo. The charged nature of the polymer leads to interaction with serum proteins, causing aggregation, and leads to unfavorable cell membrane destabilization [210]. Additionally, a great deal of PEI toxicity is due to its nondegradable nature [211]. Strategies to improve upon PEI transfection efficiencies while decreasing cellular toxicity include: synthesis of a novel acid-labile PEI (Figure 2.6) [212], conjugation with PEG to reduce toxicity [213], and linking low-molecular PEI via degradable or reducible bonds [214, 215]. Early work performed in our lab investigated conjugation of low-molecular weight PEI, which has traditionally demonstrated improved toxicity compared 49 Figure 2.5. The chemical structures of the prototypical cationic polymers linear polyethyleneimine (LPEI), branched PEI (bPEI), and poly-L-lysine (PLL). 50 Figure 2.6. Chemical structures of the degradable cationic polymers acid-labile polyethyleneimine (PEI), poly(4-hydroxyl-L-proline ester) (PHP), and poly[α-(4-aminobutyl)-L-glycolic acid] (PAGA). Adapted from Kim et al. [212]. 51 to high molecular weight but severely decreased transfection efficiency, with cholesterol to form a water-soluble lipoprotein [216]. 2.8.4.2 Poly-L-Lysine Gene Delivery Another common cationic polymer used for gene delivery is poly-L-lysine (PLL) (Figure 2.5). PLL acts as a cationic gene carrier due to its large number of ε-amino groups, allowing for electrostatic condensation and complexation of polynucleotides. This polymer was one of the first cationic polymers observed to demonstrate gene transfection capabilities [217]. However, its transfection efficiency is highly variable between cell types, and typically is not as high as that seen with other cationic polymers like PEI. Additionally, because the amine groups on PLL have much higher pKa values than other cationic polymers, it does not efficiently escape the endosomal compartment. To increase transfection efficiency with PLL, chloroquine (an endosomolytic agent) is often delivered with PLL [218, 219]. Unfortunately, the addition of chloroquine increases toxicity and is not feasible for in vivo applications. 2.8.4.3 Degradable Polymers for Gene Delivery One of the major limitations of the early generation of cationic polymers was their nondegradability. Once the polymer carrier has penetrated and delivered its nucleic acid cargo inside of a cell, the inability of the cell to clear the polymer carrier results in delayed toxicity (7-9 hours after transfection) [211, 220]. To overcome this limitation in the early cationic polymers, carriers with degradable linkers were developed to allow the cell to break down the high molecular weight polymers into smaller fragments, aiding in their clearance and reducing their toxicity. Some of the major chemical moieties used to 52 allow for degradation of polymers include: hydrolyzable ester linkages, acid-labile imine linkers, and bioreducible disulfide linkages. The first degradable polymer developed for gene delivery was the polyester agent poly(4-hydroxy-L-proline ester) (PHP) (Figure 2.6). The polymer is degradable due the incorporation of hydrolyzable ester bonds as the monomer linker groups. The rate of hydrolysis of PHP is enhanced by the presence of the amine groups and the polymer is readily degradable, hydrolyzing in less than 24 hours [221]. Another early biodegradable polymer for gene delivery, poly[α-(4-aminobutyl)-L-glycolic acid] (PAGA) (Figure 2.6), was developed in this lab and was shown to a safe and nontoxic alternative to polymers such as PLL [222]. While polymers incorporating ester or other hydrolyzable bonds reduce delayed toxicity due to intracellular accumulation of polycation carriers and their resultant disruption of cellular function, the nonspecific nature of hydrolyzable bond degradation leads to limited circulatory and extracellular in vivo half-lives. In order to develop polymers with the ideal characteristics of resistance to degradation in the extracellular space, but which degrade inside the cell, releasing their cargo and allowing for clearance, polycationic polymers with reducible disulfide bonds were developed. 2.8.4.4 Bioreducible Polymers for Gene Delivery Disulfide bonds are an attractive linker for polymeric gene carriers due to their stability in extracellular environments, but their susceptibility to reduction in the intracellular, particularly cytoplasmic, space through thiol-disulfide exchange [223, 224]. Many proteins, especially those which are secreted, form disulfide bonds between 53 cysteine residues as a way to lend stability to the protein, aid in protein folding, and maintain the folded structure. Exposed disulfide bonds are particularly susceptible to reduction in the cytoplasm due to the high concentration of the thiol-containing molecule glutathione. The cytoplasm is a highly reductive environment, owing to a high concentration of glutathione (0.5-10 mM intracellular versus 15 μM extracellular) [225-227], and because the cytoplasm maintains a ratio of reduced glutathione/oxidized glutathione disulfide (GSH/GSSG) of up to 100/1 [228]. The reduced glutathione pool is maintained through the action of the NADH-dependent flavoenzyme glutathione reductase to convert GSSG back to the reduced GSH form to maintain the reductive gradient. Glutathione stores are also maintained by synthesis of new glutathione from glutamate, cysteine, and glycine by γ-glutamylsynthetase followed by glutathione synthetase. When GSH levels are high, glutathione reduces disulfide bonds via thiol-disulfide exchange. In this reaction, the thiolate anion on GSH attacks a sulfur atom contained in a disulfide bond. The primary disulfide bond is broken and a new disulfide bond between glutathione and the attacked sulfur group is formed [223]. Due to the large difference between intracellular and extracellular glutathione levels, disulfide bonds have proven to be excellent linkers in gene delivery polymers, imbuing stability in the circulation and extracellular milieu while rapidly degrading once the polymer is internalized to the cytoplasm. This results in increased transfection efficiencies, with decreased toxicity and cellular accumulation. The first disulfide-containing polymer used for gene delivery was a disulfide-linked PEI [229]. Gosselin et al. found that their disulfide-crosslinked PEI significantly 54 increased transfection efficiency in Chinese hamster ovary cells compared to the nonreducible parent low molecular weight PEI. The levels of transfection were almost equivalent to that seen with 25 kDa bPEI. Other groups investigated different disulfide cross-linkers to form bioreducible PEI and also found transfection efficiencies similar to 25 kDa bPEI with decreased or very similar cytotoxicities [215, 230]. Other classes of bioreducible polymers beyond disulfide-linked PEI have also been synthesized and include disulfide poly(amido amine)s (SS-PAAs), and poly(ß-amino ester)s (Figure 2.7). SS-PAAs employ Michael addition-type reactions between primary amines and cystamine bisacrylamide. Some of the first polymers synthesized from this family demonstrated that they can condense nucleic acids to positively charged and nano-sized polyplexes (+20 mV ζ-potential and <200 nm, respectively) [231, 232]. These SS-PAAs also demonstrated greater buffering capabilities than even the amine-rich PEI polymers, aiding in escape from the endosomal compartment via the proton sponge effect. SS-PAAs are also capable of highly efficient transfection even in the presence of serum with little cytotoxicity even at high polymer:nucleic acid weight ratios [232]. Finally, many SS-PAAs contain pendant amine groups that are amenable to modification, allowing grafting of targeting moieties to the polymer. 2.8.5 Targeted Polymer Gene Carriers In order to enhance the efficiency and especially specificity of polymer gene carriers for target cells, targeting groups can be attached to the carrier. Nontargeting polymer gene carriers interact nonspecifically with cells through interaction with negatively charged surface molecules like glycosaminoglycans and enter the cell via adsorptive endocytosis [233, 234]. Also, when given systemically, charged polyplexes 55 Figure 2.7. Bioreducible poly(amido amines) based on cystamine bisacrylamide. DAH = diaminohexane, TETA = triethylenetetramine, EDA = ethylene diamine 56 are bound by serum proteins, resulting in rapid clearance by the reticuloendothelial system (RES) [235]. Conjugation of shielding moieties such as PEG have been shown to increase circulatory half-life, but decrease transfection efficiency by interfering with cell surface interaction, and do not increase target cell specificity [236, 237]. To allow polyplexes to selectively transfect specific cell types, a variety of functional groups and targeting moieties can be conjugated to the polymer, including peptides, sugars, antibodies/antibody fragments, and proteins [238]. The specific interaction of these targeting groups with cell membrane components leads to receptor-mediated endocytosis, allowing polyplex entry to the cell [239]. By combining targeting ligand polymer conjugation to increase target cell uptake with PEG shielding to reduce nonspecific cellular interactions, polymers with high tissue specificity and transfection efficiency can be synthesized. 2.9 Cell Therapy for Cardiovascular Disease With the surge of interest in stem cells as a therapeutic strategy in the late 1990s, it was not long before attempts were made to apply stem cell therapy to MI. Since the loss of viable tissue in the ischemic region after myocardial infarction has long been known to decrease the functionality of the myocardium, the potential for stem cells to repopulate the heart and form contractile cardiomyocytes was quickly investigated [240, 241]. A variety of cell sources have been used in studies of cell therapy for ischemic heart disease, but they can be classified as allogeneic (from another donor) or autologous (from the patient). The earliest studies evaluating stem cells for cardiac repair used allogeneic stem cells such as embryonic stem cells, which were also one of the first stem 57 cell populations discovered. As the research around stem cell classes advanced, more populations of stem cells were discovered and evaluated for cardiac repair (Table 2.4). 2.9.1 Mesenchymal Stem Cells Adult mesenchymal stem cells are most commonly derived from bone marrow or adipose tissue. In the area of cardiac cell therapy, however, bone marrow-derived mesenchymal stem cells have been the most thoroughly investigated. Mesenchymal stem cells are multipotent in nature, differentiating to cardiomyocytes, adipocytes, osteoblasts, chondrocytes, and many other cell types, though not as many as pluripotent stem cells [243, 244]. One of the reasons for the intense interest of using mesenchymal and other bone marrow-derived stem cells for cardiac therapy is the ease of obtaining the cells from bone marrow aspirates. Stem cells can rapidly be isolated from bone marrow and quickly expanded to a sufficient number to allow for autologous donation. Though autologous therapy is relatively simple with this stem cell lineage, mesenchymal stem cells are less immunogenic than other cell types, making allogeneic cell therapy a possibility [245]. Allogeneic therapy eliminates or greatly reduces the delay necessary when autologous cells are isolated, expanded, and prepared for implantation. Mesenchymal stem cells, while capable of differentiating to cardiomyocytes, appear to do so only at a very low rate when injected in vivo [245, 246]. While this limits the repopulation of the myocardium with functional cardiomyocytes, implanted mesenchymal stem cells have been shown to have a positive effect on resident cells through the secretion of paracrine growth factors [243, 247]. 58 Table 2.4 Cell Types Investigated for Cardiac Repair Allogenic Stem Cells Autologous Stem Cells Embryonic stem cells Cardiac stem cells Fetal cardiomyocytes Adipose-derived stem cells Human umbilical cord-derived stem cells Skeletal myoblasts Bone marrow-derived stem cells Mononuclear/CD34+ fraction Mesenchymal stem cells Endothelial progenitor cells Multipotent adult progenitor cells Induced pluripotent stem cells Adapted from Mozid et al. [242]. 59 2.9.2 Cardiac Stem Cells The heart has long been viewed as a static organ, with little to no cardiomyocyte turnover. Recently, however, this view has been disproven with the discovery of resident populations of cardiac stem cells (CSCs) within the myocardium [248]. While these cells make up less than 2% of the total cells within the myocardium, they have been shown to contribute to the growth of new cardiomyocytes [249, 250] and coronary vessels [249], especially during pathological states [248, 251]. Importantly, these resident stem cells have demonstrated they play a role in regeneration of cardiomyocytes and vascular cells after myocardial infarction [249]. Despite their low population levels within the myocardium, CSCs can be harvested from patients, proliferated ex vivo, and used for autologous implantation [252]. At least two clinical trials evaluating CSCs in human patients suffering from ischemic cardiomyopathy [253] and MI [254] have been completed with encouraging results. Cells were delivered by intracoronary infusion and in both trials provided improvement in infarct size and LV function. While CSCs have shown promising results thus far, because the field is still nascent, they are still one of the least studied stem cells for cardiac repair. Little is known about their long-term fate in the heart and if the positive results seen in short-term studies will vary over to the long term. 2.9.3 Embryonic Stem Cells Human embryonic stem cells were first isolated in 1998 and are a pluripotent stem cell capable of differentiating into almost all cell lineages [255]. Importantly for use in cardiac cell therapy, embryonic stem cells have been shown to differentiate into 60 contracting cardiomyocytes [256, 257]. The use of embryonic stem cells is still the source of passionate ethical debate due to their source with the resulting regulatory restrictions limiting their adoption. While embryonic stem cells have often been considered the gold standard source for stem cell therapy, their use is not without limitations. Since embryonic stem cells are derived from allogenic sources, they can trigger immunogenicity in the recipient [258]. Additionally, embryonic stem cells have shown predisposition to form teratomas when they are injected in vivo [259]. Finally, while embryonic stem cells are known to differentiate to cardiomyocytes, there is some debate whether the stem cells implant with sufficient structural and electromechanical efficiency to contribute to functional recovery of the myocardium [260]. 2.9.4 Skeletal Myoblasts Skeletal myoblasts are not true stem cells as they are limited in their ability to differentiate past a myocyte fate. They are derived from skeletal muscle satellite cells that normally are located under the basal membrane in a quiescent state [261]. When injury occurs, the satellite cells mobilize, proliferate, and fuse to existing muscle cells in order to regenerate muscle tissue. Skeletal myoblasts are satellite cells that are in the process of proliferating but have not yet fused with skeletal myocytes. Skeletal myoblasts were one of the first cell populations considered for cardiac repair because of several clinically attractive attributes they possess. Some of these attributes that make them an ideal cell source for cardiac cell therapy include: a high resistance to hypoxia [262]; ease of isolation [242, 263]; rapid expansion capabilities (a small biopsy can provide over one billion cells in a 2-3 week time period and the process 61 can be automated) [264]; and finally, since they are easily derived from an autologous origin, they do not provoke an immune response [265]. Additionally, the limited differentiation capabilities of skeletal myoblasts may be considered a limitation as they do not directly transdifferentiate to cardiomyocytes [266], but it is also a benefit since there is no risk of teratoma formation. Even though skeletal myoblasts do not directly differentiate to cardiomyocytes, experimental studies in both small and large animal studies have shown they can be successfully implanted to ischemic hearts to provide functional left ventricular benefit [267-272]. Early uncontrolled human trials evaluating the therapeutic benefit of skeletal myoblast implantation in chronic heart failure patients showed promising results in the form of improved global and regional contractility, and cell viability [273-276]. However, subsequent expanded and controlled human clinical trials have produced more mixed results. The myoblast autologous grafting in ischemic cardiomyopathy (MAGIC) trial was a multicenter, controlled, randomized, and double-blinded Phase II trial evaluating autologous skeletal myoblasts. Cells were expanded ex vivo for 3 weeks then implanted intramyocardially to the ischemic border zone during bypass surgery. Unfortunately, in this trial, there was no significant improvement in the primary endpoints of global and regional LV function [277]. The longest trial of the impact of skeletal myoblast implantation for cardiac repair comes from Dib et al. In a nonrandomized, uncontrolled Phase I trial, patients showed improvements in LV ejection fraction and myocardial viability with no observed adverse consequences of skeletal myoblast therapy 4 years after implantation [278, 279]. Additionally, histological 62 analysis of implanted skeletal myoblasts in humans has shown that the cells form striated myotubes in the scar area and engraft in-line with endogenous cardiomyocytes [278, 280]. There was initially some concern as implantation of skeletal myoblasts to the heart was reported to increase the risk of arrhythmias [273, 275, 279, 280]. However, the MAGIC trial did not observe any significant increase in arrhythmias in patients receiving skeletal myoblasts but did note a trend toward significance [277]. In response to the proarrythmogenic concerns of skeletal myoblasts, Chachques et al. hypothesized that the arrythmogenicity of the cells was due to an inflammatory and antigenic response against the fetal bovine serum used to expand the isolated cells. In a study to test this hypothesis, isolated cells were expanded using serum derived from the donor. This study detected no incidence of arrhythmias over a 1-2 year period [281]. Our lab has demonstrated previously that primary skeletal myoblasts transfected with a human isoform of VEGF165 express and secrete high levels of VEGF for up to one week [282] and that these cells promote angiogenesis in vivo as reported in Chapter 3 of this dissertation [283]. 2.9.5 Induced Pluripotent Stem Cells In light of the recent awarding of the 2012 Nobel Prize in Physiology or Medicine jointly to John B. Gurdon and Shinya Yamanaka for the discovery that differentiated cells can be reprogrammed to become pluripotent stem cells [284, 285], a brief overview of recent advances applying this technology to MI is warranted. Induced pluripotent stem cells (iPSCs) are still most commonly derived from a fibroblast cell source, which can be harvested from an autologous skin biopsy. Once dedifferentiated to a pluripotent state, 63 iPSCs can be differentiated to a cardiomyocyte fate [286, 287]. Murine and porcine studies have shown that iPSCs administered to the heart improve cardiac function and attenuate LV remodeling [288, 289]. While the investigation of iPSC use for myocardial repair is still in its infant stages, they have great potential for cardiovascular applications. iPSCs can be generated from donor biopsies and have excellent differentiation capabilities. However, limitations include rapid teratoma formation and the time required (over 30 days) to generate replicating iPSCs from fibroblasts. 2.10 Therapeutic Angiogenesis Therapeutic angiogenesis refers to the treatment strategy of establishing new vasculature to cells and tissues that are not adequately served by the circulatory system [290]. Over the years, more clinical success in blocking angiogenesis in the cases of cancer and ocular diseases have been achieved, but a great deal of research is being performed investigating the potential benefit of establishing blood supply to ischemic tissue. The formation of new vessels, whether arterial or venous in nature, is a complex physiological process and is mediated by several large families of proteins that promote directed vessel growth and vessel maturation. Some of the major families with therapeutic applications include the VEGF, PDGF, and FGF families. 2.10.1 VEGF The most popular family of angiogenic factors for establishing new blood vessels is the vascular endothelial growth factor (VEGF) family. The mammalian VEGF family consists of 7 factors, VEGF-A, -B, -C, -D, -E, -F, and placental growth factor (PIGF). All members of the VEGF family produce their action by interacting with the tyrosine kinase 64 receptors VEGFR-1/Flt1, VEGFR-2/Flk1/KDR, and VEGFR-3/Flt2. These receptors are found in high abundance on the surface of endothelial cells. VEGF-A is the most active family member and stimulates vessel growth by signaling through VEGFR-2 [291]. VEGF-A is known to be one of the most important angiogenic forms of VEGF and VEGF-A deficiency on even one allele is embryonically lethal, showing severely abnormal blood vessel development [292]. VEGF-A has five isoforms containing 121, 145, 165, 189, or 206 amino acids. Of the five isoforms, VEGF165 is the most promising candidate for therapeutic use due to its ability to create the necessary growth factor gradients needed for efficient and functional vessel growth, and because it is 100× more potent than VEGF121 [293]. VEGF165 acts through interaction with the VEGFR-1 and VEGFR-2 receptors to stimulate endothelial cell growth and vessel sprouting. Endothelial tip cells sense the levels of VEGF-A and rapidly guide capillary migration towards the source of VEGF-A expression [294]. 2.10.2 Other Angiogenic Factors Two additional proangiogenic factors investigated for therapeutic vessel growth are platelet-derived growth factor (PDGF), and fibroblast growth factor (FGF). The PDGF family plays an important role in angiogenesis but aids more in the maturation of vessels, allowing them to function properly than in their initial formation [295]. The PDGF growth factors help recruit pericytes to budding and growing vessels [296, 297]. Without proper pericyte recruitment, the formed vessels are immature, leaky, display a high degree of tortuosity, decreased perfusion, and do not always remain viable after angiogenic therapy has ceased [298, 299]. Due to the important role PDGF plays in 65 producing highly functional vessels, PDGF therapy is often paired with VEGF therapy, combining the potency of VEGF at stimulating rapid vessel formation, with PDGF to convert the newly formed vessels into stable and mature vessels [300, 301]. FGF is another family of angiogenic proteins with 23 members and similar function as the VEGF family. The FGF proteins promote angiogenesis by interacting with 4 unique cell membrane-bound tyrosine kinase receptors [302]. Unlike VEGF, FGF receptors are expressed on many different cell types, including smooth muscle cells and fibroblasts, and can act on these cells causing them to secrete additional angiogenic factors [303]. Heart studies have shown that FGF signaling induces hedgehog, angiopoietin-2 (ANG-2), and VEGF-B release, and aids in supporting long-term vascular integrity [304]. Initial testing of FGF in the Angiogenic Gene Therapy (AGENT1) trial showed improved exercise tolerability in patients after 4 weeks [305]. However, the followup study in the AGENT 3/4 trial was prematurely interrupted as the study was not going to be able to reach the primary end point [306]. 66 2.11 References [1] V.L. Roger, A.S. Go, D.M. Lloyd-Jones, E.J. Benjamin, J.D. Berry, W.B. Borden, D.M. Bravata, S. Dai, E.S. Ford, C.S. Fox, H.J. Fullerton, C. Gillespie, S.M. Hailpern, J.A. Heit, V.J. Howard, B.M. Kissela, S.J. Kittner, D.T. Lackland, J.H. Lichtman, L.D. Lisabeth, D.M. Makuc, G.M. Marcus, A. Marelli, D.B. Matchar, C.S. Moy, D. Mozaffarian, M.E. Mussolino, G. Nichol, N.P. Paynter, E.Z. Soliman, P.D. Sorlie, N. Sotoodehnia, T.N. Turan, S.S. Virani, N.D. Wong, D. Woo, M.B. Turner, Heart disease and stroke statistics--2012 update: a report from the American Heart Association, Circulation, 125 (2012) e2-e220. [2] E. Palojoki, A. Saraste, A. Eriksson, K. Pulkki, M. Kallajoki, L.M. Voipio-Pulkki, I. Tikkanen, Cardiomyocyte apoptosis and ventricular remodeling after myocardial infarction in rats, Am. J. Physiol. Heart Circ. Physiol., 280 (2001) H2726-2731. [3] S.R. Tiyyagura, S.P. Pinney, Left ventricular remodeling after myocardial infarction: past, present, and future, Mt. Sinai J. Med., 73 (2006) 840-851. [4] R.G. McKay, M.A. Pfeffer, R.C. Pasternak, J.E. Markis, P.C. Come, S. Nakao, J.D. Alderman, J.J. Ferguson, R.D. Safian, W. Grossman, Left ventricular remodeling after myocardial infarction: a corollary to infarct expansion, Circulation, 74 (1986) 693-702. [5] H.D. White, R.M. Norris, M.A. Brown, P.W. Brandt, R.M. Whitlock, C.J. Wild, Left ventricular end-systolic volume as the major determinant of survival after recovery from myocardial infarction, Circulation, 76 (1987) 44-51. [6] B.I. Jugdutt, Ischemia/Infarction, Heart Fail. Clin., 8 (2012) 43-51. [7] J.A. Erlebacher, J.L. Weiss, M.L. Weisfeldt, B.H. Bulkley, Early dilation of the infarcted segment in acute transmural myocardial infarction: role of infarct expansion in acute left ventricular enlargement, J. Am. Coll. Cardiol., 4 (1984) 201-208. [8] D.L. Mann, Mechanisms and models in heart failure: A combinatorial approach, Circulation, 100 (1999) 999-1008. [9] L.W. Eaton, J.L. Weiss, B.H. Bulkley, J.B. Garrison, M.L. Weisfeldt, Regional cardiac dilatation after acute myocardial infarction: recognition by two-dimensional echocardiography, N. Engl. J. Med., 300 (1979) 57-62. [10] J.L. Rouleau, J. de Champlain, M. Klein, D. Bichet, L. Moyé, M. Packer, G.R. Dagenais, B. Sussex, J.M. Arnold, F. Sestier, J.O. Parker, P. McEwan, V. Bernstein, T.E. Cuddy, G. Lamas, S.S. Gottlieb, J. McCans, C. Nadeau, F. Delage, P. Hamm, M.A. Pfeffer, Activation of neurohumoral systems in postinfarction left ventricular dysfunction, J. Am. Coll. Cardiol., 22 (1993) 390-398. [11] M.G. Sutton, N. Sharpe, Left ventricular remodeling after myocardial infarction: pathophysiology and therapy, Circulation, 101 (2000) 2981-2988. 67 [12] A.M. Richards, M.G. Nicholls, T.G. Yandle, C. Frampton, E.A. Espiner, J.G. Turner, R.C. Buttimore, J.G. Lainchbury, J.M. Elliott, H. Ikram, I.G. Crozier, D.W. Smyth, Plasma N-terminal pro-brain natriuretic peptide and adrenomedullin: new neurohormonal predictors of left ventricular function and prognosis after myocardial infarction, Circulation, 97 (1998) 1921-1929. [13] R.A. Little, K.N. Frayn, P.E. Randall, H.B. Stoner, C. Morton, D.W. Yates, G.S. Laing, Plasma catecholamines in the acute phase of the response to myocardial infarction, Arch. Emerg. Med., 3 (1986) 20-27. [14] C. Hall, Interaction and modulation of neurohormones on left ventricular remodeling, in: M.G. Sutton (Ed.) Left Ventricular Remodelling After Acute Myocardial Infarction, Sciene Press Ltd, London, 1996, pp. 89-99. [15] N.G. Frangogiannis, C.W. Smith, M.L. Entman, The inflammatory response in myocardial infarction, Cardiovasc. Res., 53 (2002) 31-47. [16] J. Sadoshima, S. Izumo, The cellular and molecular response of cardiac myocytes to mechanical stress, Annu. Rev. Physiol., 59 (1997) 551-571. [17] D.M. Yellon, D.J. Hausenloy, Myocardial reperfusion injury, N. Engl. J. Med., 357 (2007) 1121-1135. [18] P. Saikumar, Z. Dong, J.M. Weinberg, M.A. Venkatachalam, Mechanisms of cell death in hypoxia/reoxygenation injury, Oncogene, 17 (1998) 3341-3349. [19] B.J. Zimmerman, D.N. Granger, Reperfusion injury, Surg. Clin. North Am., 72 (1992) 65-83. [20] F.G. Spinale, Myocardial matrix remodeling and the matrix metalloproteinases: influence on cardiac form and function, Physiol. Rev., 87 (2007) 1285-1342. [21] M.L. Lindsey, MMP induction and inhibition in myocardial infarction, Heart Fail. Rev., 9 (2004) 7-19. [22] G. Olivetti, J.M. Capasso, E.H. Sonnenblick, P. Anversa, Side-to-side slippage of myocytes participates in ventricular wall remodeling acutely after myocardial infarction in rats, Circ. Res., 67 (1990) 23-34. [23] P. Whittaker, D.R. Boughner, R.A. Kloner, Role of collagen in acute myocardial infarct expansion, Circulation, 84 (1991) 2123-2134. [24] M.A. Pfeffer, Left ventricular remodeling after acute myocardial infarction, Annu. Rev. Med., 46 (1995) 455-466. [25] G.F. Mitchell, G.A. Lamas, D.E. Vaughan, M.A. Pfeffer, Left ventricular remodeling in the year after first anterior myocardial infarction: a quantitative analysis of 68 contractile segment lengths and ventricular shape, J. Am. Coll. Cardiol., 19 (1992) 1136-1144. [26] Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Fibrinolytic Therapy Trialists' (FTT) Collaborative Group, Lancet, 343 (1994) 311-322. [27] B.R. Brodie, T.D. Stuckey, T.C. Wall, G. Kissling, C.J. Hansen, D.B. Muncy, R.A. Weintraub, T.A. Kelly, Importance of time to reperfusion for 30-day and late survival and recovery of left ventricular function after primary angioplasty for acute myocardial infarction, J. Am. Coll. Cardiol., 32 (1998) 1312-1319. [28] A. Barbagelata, E.R. Perna, P. Clemmensen, B.F. Uretsky, J.P. Canella, R.M. Califf, C.B. Granger, G.L. Adams, R. Merla, Y. Birnbaum, Time to reperfusion in acute myocardial infarction. It is time to reduce it!, J. Electrocardiol., 40 (2007) 257-264. [29] H.R. Lijnen, D. Collen, Fibrinolysis and the control of hemostasis, in: G. Stamatoyannopoulos, P.W. Majerus, R.M. Permutter, H. Varmus (Eds.) The Molecular Basis of Blood Diseases, W.B. Saunders Co., Philadelphia, 2001, pp. 740-763. [30] C.F. Lacy, L.L. Armstrong, M.P. Goldman, L.L. Lance, A. American Pharmaceutical, Drug information handbook, 2003-2004, Lexi-Comp ; American Pharmaceutical Association, Hudson, Ohio; [Washington, D.C.], 2003. [31] The effects of tissue plasminogen activator, streptokinase, or both on coronary-artery patency, ventricular function, and survival after acute myocardial infarction. The GUSTO Angiographic Investigators, N. Engl. J. Med., 329 (1993) 1615-1622. [32] E. Boersma, A.C. Maas, J.W. Deckers, M.L. Simoons, Early thrombolytic treatment in acute myocardial infarction: reappraisal of the golden hour, Lancet, 348 (1996) 771-775. [33] E.C. Keeley, J.A. Boura, C.L. Grines, Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials, Lancet, 361 (2003) 13-20. [34] A. Gruntzig, Transluminal dilatation of coronary-artery stenosis, Lancet, 1 (1978) 263. [35] U. Sigwart, J. Puel, V. Mirkovitch, F. Joffre, L. Kappenberger, Intravascular stents to prevent occlusion and restenosis after transluminal angioplasty, N. Engl. J. Med., 316 (1987) 701-706. [36] J.E. Sousa, M.A. Costa, A.C. Abizaid, B.J. Rensing, A.S. Abizaid, L.F. Tanajura, K. Kozuma, G. Van Langenhove, A.G. Sousa, R. Falotico, J. Jaeger, J.J. Popma, P.W. Serruys, Sustained suppression of neointimal proliferation by sirolimus-eluting stents: 69 one-year angiographic and intravascular ultrasound follow-up, Circulation, 104 (2001) 2007-2011. [37] M. Degertekin, P.W. Serruys, D.P. Foley, K. Tanabe, E. Regar, J. Vos, P.C. Smits, W.J. van der Giessen, M. van den Brand, P. de Feyter, J.J. Popma, Persistent inhibition of neointimal hyperplasia after sirolimus-eluting stent implantation: long-term (up to 2 years) clinical, angiographic, and intravascular ultrasound follow-up, Circulation, 106 (2002) 1610-1613. [38] P.W. Serruys, P. de Jaegere, F. Kiemeneij, C. Macaya, W. Rutsch, G. Heyndrickx, H. Emanuelsson, J. Marco, V. Legrand, P. Materne, et al., A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. Benestent Study Group, N. Engl. J. Med., 331 (1994) 489-495. [39] D.L. Fischman, M.B. Leon, D.S. Baim, R.A. Schatz, M.P. Savage, I. Penn, K. Detre, L. Veltri, D. Ricci, M. Nobuyoshi, et al., A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. Stent Restenosis Study Investigators, N. Engl. J. Med., 331 (1994) 496-501. [40] M.C. Morice, P.W. Serruys, J.E. Sousa, J. Fajadet, E. Ban Hayashi, M. Perin, A. Colombo, G. Schuler, P. Barragan, G. Guagliumi, F. Molnar, R. Falotico, A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization, N. Engl. J. Med., 346 (2002) 1773-1780. [