REPRESSING A REPRESSOR: ADENO-ASSOCIATED VIRUS DELIVERED SHORT-HAIRPIN RNA MOLECULES TARGETING PHOSPHOLAMBAN By LOURDES MARIA ANDINO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006
Copyright 2006 by Lourdes MarÃa Andino
To the One, the True, the Good, and the Beautiful.
iv ACKNOWLEDGMENTS The best part of writing a dissertation is thisÂ— acknow ledging people. Over the many years that I have tirelessly engaged in this endeavor, I have come across so many people that I would now like to thank. Fi rst and foremost, my familyÂ—Mami and Dado, Kaki and Pat, Glu and Luis and Mati and Ma rc as well as my wonderful collection of nieces and nephewÂ—they all have been my insp iration. They have all encouraged me to use my energy and my mind to achieve the most that I can with all of my wonderful Godgiven talents. Dr. Lewin has been more than a mentor a nd an advisor to meÂ—he has been a father figure and a role model. I would like to thank him for teaching me what it really means to be a scientist in the broa dest sense of the word. He has opened my eyes to the many beautiful things of the world as he has gen tly guided my path and endlessly encouraged me. I would like to acknowledge all of the amazing support system in our department especially: Joyce the best gra duate secretary ever, and the pa tient and organized ladies of the fiscal officeÂ—they have shown me that women do really run the show! Many thanks go out to all of the people th at patiently taught me technical skills over the years including Stacy Porvasnik, Jeff Kelley, Mard a Jorgenson, Hideko Kasahara, Larry Bish at UPenn, and Julieta and Djamel at Mass Ge neralÂ—without them my work wouldnÂ’t be complete.
v Additionally, I would like to thank my committee members: Ed Scott, Tom Rowe, Peggy Wallace and Barry Byrne for their time and valuable advice. I would also like to acknowledge Ed Chan for letting me use his amazing microscope for endless hours and for letting me be an auxiliary member to his labÂ—to him, I am very grateful. I have had quite possibly the best labmates in the world. These guys have made the years fly by and to them, I am very grateful . I would like to thank them for not only letting me borrow stuff all the time and teach ing me many things but for being my family when I have needed it the most. They have put up with all of my craziness, encouraged me to be me, and participated in many of my wild ideas. I woul d like to thank James, Marina, Mary Ann, Robert, Alan, Lynn, Jia, Ve rline, Lee, Jen, Eun, Edgar, Fredric and our angel Patrick for watching over me, havi ng fun with me, and caring so much for meÂ—I really couldnÂ’t have done it w ithout such an awesome bunch! Now, getting to the heart of the matter ( no pun intended), I would like to especially acknowledge Andrij Jakymiw. Without him, not only would I not have made it all the way to the end, but it wouldnÂ’t have been worthw hile. He has been the real prize that I have been waiting for. Sure the degree will be nice and useful (I hope!), but without someone to share it with, someone that a ppreciates what it has REALLY taken, it would be worthless. He has made this experien ce priceless. I would like to thank him for teaching me so much and being with me every step of the way when it mattered most. I hope someday soon our two diplomas will hang side by side in a home office! Most importantly, to the good Lord who ha s walked with me all the wayÂ—I would like to thank him for the talents, the pa ssion, the ambition and the energyÂ—all of the glory and honor is for him.
vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...............................................................................................................x LIST OF FIGURES...........................................................................................................xi ABSTRACT.....................................................................................................................xi v CHAPTER 1 INTRODUCTION........................................................................................................1 The Heart...................................................................................................................... 1 Cardiac Architecture..............................................................................................1 Cardiac Excitation-Contraction Coupling.............................................................4 The Role of Calcium in Cardiac Disease..............................................................5 Phospholamban.............................................................................................................6 Discovery of Phospholamban................................................................................6 Phosphorylation of Phospholamban......................................................................6 Phospholamban Structure......................................................................................8 Heart Disease................................................................................................................9 Effects on Society..................................................................................................9 Causes of Cardiovascular Disease (CVD)...........................................................10 Current Available Treatments.............................................................................10 Gene Therapy..............................................................................................................12 History.................................................................................................................12 Gene Delivery......................................................................................................14 Retroviruses..................................................................................................15 Adenovirus...................................................................................................16 Herpesvirus...................................................................................................17 Adeno-Associated Virus (AAV)..................................................................17 RNA Interference........................................................................................................22 Short-Interfering RNAs.......................................................................................23 Short-Hairpin RNAs............................................................................................25 Interfering with Disease: Prospects for Therapy.................................................25 Project Overview........................................................................................................26
vii 2 MATERIALS AND METHODS...............................................................................28 Design of Phospholamban Expressing Plasmid.........................................................28 Design and Preparation of siRNAs.............................................................................29 siRNA Design......................................................................................................29 Deprotection and Annealing of siRNA Molecules..............................................30 Desalting the Duplex...........................................................................................30 Quantification of siRNA Duplex.........................................................................31 Cloning of shRNAs into AAV TR Plasmids..............................................................31 DNA Techniques........................................................................................................33 Linearization and Dephosphorylation of Vectors................................................33 Ligation................................................................................................................33 Transformation....................................................................................................35 Plasmid Mini-Preparation....................................................................................35 Large Scale DNA Purificati on using Cesium Chloride (CsCl) Gradients..........35 DNA Sequencing.................................................................................................37 DNA Packaging...................................................................................................37 Primary Neonatal Rat Ventricular Cardiomyocytes (PNRVC)..................................38 Isolation...............................................................................................................38 Percoll Gradient and Final Cell Preparation........................................................40 Preparation of Tissue Culture Vessels.................................................................42 Mitomycin-C Cell Treatment..............................................................................42 Primary Adult Rat Ventri cular Cardiomyocytes........................................................43 Isolation...............................................................................................................43 Infection and Incubation......................................................................................44 Measurements of Intracellular Calcium..............................................................44 RNA Techniques........................................................................................................45 RNA Extraction and Preparation from Mammalian Cells..................................45 RNA Extraction from Cardiac Tissue.................................................................46 Reverse Transcriptase (RT)-PCR........................................................................46 Sybr Green Staining............................................................................................47 Phosphorimager Analysis....................................................................................47 Protein Techniques.....................................................................................................48 Protein Extraction from Cells..............................................................................48 Protein Extraction from Tissue............................................................................48 Protein Quantitation.............................................................................................49 Western Blot Analysis.........................................................................................49 Transfer to Nitroc ellulose Membrane.................................................................50 Blocking of Membrane and Application of Antibodies......................................50 Chemiluminescence and Film Development.......................................................51 Immunofluorescence...........................................................................................51 Microscopy..........................................................................................................52 Co-Transfection of HEK 293 Cells............................................................................52 Co-Transfection with Do uble-Stranded RNA Oligos.........................................52 Co-Transfection with shRNAs Expressed from AAV TR Plasmids...................53
viii Animal Procedures......................................................................................................53 Animal Selection and Ordering...........................................................................53 Injections and Incubations...................................................................................54 Organ Harvesting and Freezing...........................................................................55 Histology.............................................................................................................55 3 RESULTS...................................................................................................................58 Expression of Phospholamban....................................................................................58 Double-Stranded RNA Co-Transfections...................................................................59 Co-Transfections of shRNAs in AAV Expression Plasmids......................................62 Primary Neonatal Rat Ventricular Cardio myocytes Infected with PLN shRNAs......67 In vivo Testing of PLN shRNA-Expressing AAV......................................................74 Temporal Vein Injections....................................................................................75 Sub-Xiphoid Injections........................................................................................76 4 DISCUSSION AND CONCLUSIONS......................................................................83 Reduction of Co-Transfected PL N using both siRNA and shRNA...........................84 Infections in PNRVC usi ng AAV1 Vectored shRNAs..............................................86 Infections in PARVC usi ng AAV1 Vectored shRNAs..............................................88 Animal Studies............................................................................................................90 Future Directions........................................................................................................92 APPENDIX A RIBOZYMES TARGETING PHOSPHOLAMBAN.................................................95 Introduction.................................................................................................................95 Materials and Methods...............................................................................................96 5Â’ End-Labeling of Deprotected Target RNA.....................................................97 In vitro Timecourse Cleavage Reaction..............................................................98 Preparation of Ribozyme Sequences into AAV Vectors.....................................99 Co-Transfection of HEK 293 Cells.....................................................................99 Ventricular Neonatal Rat Card iomyocytes Experiments..................................100 Direct Injection into the Murine Myocardium..................................................101 Results.......................................................................................................................1 01 Discussion and Conclusions.....................................................................................105 B SELF-COMPLEMENTARY AAV MOLECULES IN THE HEART.....................107 Introduction...............................................................................................................107 Materials and Methods.............................................................................................107 Results.......................................................................................................................1 09 Discussion and Conclusions.....................................................................................114
ix LIST OF REFERENCES.................................................................................................118 BIOGRAPHICAL SKETCH...........................................................................................130
x LIST OF TABLES Table page 1-1 Different tropisms for Adeno-associated virus serotypes........................................19 2-1 Alignment of murine and rattus PLN 248 siRNA sites............................................30 2-2 Primers.................................................................................................................... .34 2-3 Recipe for CBF-HH cell isolation buffer.................................................................39 2-4 Recipe for 10X Clear ADS buffer............................................................................40 2-5 Composition of Perc oll gradient layers....................................................................41 2-6 Antibodies used for Western blotting.......................................................................51
xi LIST OF FIGURES Figure page 1-1 Anatomy of muscle....................................................................................................2 1-2 Cardiac excitation-contraction coupli ng events in a ventricular myocyte.................3 1-3 Phospholamban and its homologue, Sarcopilin.........................................................7 1-4 Interactions between th e cardiac signaling pathways................................................8 1-5 Single-stranded and self-complemen tary Adeno-associated virus genomes...........22 1-6 The siRNA pathway.................................................................................................24 2-1 Murine phospholamban expressing plasmid............................................................29 2-2 Sequence of the components of PLN shM248.........................................................32 2-3 Percoll gradient used for the purification PNRVC...................................................41 3-1 Immunoblot showing a time course of PLN expression in transfected HEK 293 cells.......................................................................................................................... .59 3-2 Immunoblot of PLN samples boiled vs. unboiled....................................................59 3-3 Co-transfection of HE K 293 cells with 50 nM of siRNA and 0.5 Âµg of PLN expressing plasmid...................................................................................................60 3-4 Co-transfection of HE K 293 cells with 50 nM of siRNA and 1.0 Âµg of PLN expressing plasmid...................................................................................................61 3-5 Co-transfection of HE K 293 cells with 28 nM of siRNA and 4.0 Âµg of PLN expressing plasmid...................................................................................................62 3-6 The AAV expression plasmids created to express shRNAs....................................64 3-7 Co-transfections with sc and ss AAV plasmids expressing PLN shRNAs lead to reductions in PLN mRNA levels..............................................................................65 3-8 Immunoblot depicting PLN levels of HEK 293 cells co-transfected with shRNA expressing plasmids..................................................................................................66
xii 3-9 Primary neonatal rat ventricular cardiomyocytes infected with AAV.....................68 3-10 RNA levels in primary cardiomyocytes infected with AAV shRNAs.....................69 3-11 Immunoblot depicting PLN protein levels in primary cells infected with AAV.....70 3-12 PLN sc/ss shRNA AAVs decrease the amount of PLN in PNRVC.........................71 3-13 Adult ventricular myoctes infected with sc-shR248................................................72 3-14 Calcium transient (CaiT) measurements in adult ve ntricular rat cardiomyocytes infected with sc-shR248...........................................................................................73 3-15 Effect of PLN down regulation in adult ventricular ra t cardiomyocytes.................74 3-16 Temporal vein injection into 1-day-old CD-1 mouse..............................................75 3-17 Levels of GFP expression 4 weeks pos t temporal vein injection with AAV...........76 3-18 Sub-xiphoid injection into 4-day-old CD-1 mouse..................................................77 3-19 Sub-xiphoid injected CD-1 mouse with LacZ construct..........................................78 3-20 Sub-xiphoid injected CD-1 mouse with pTR-UF11 construct.................................79 3-21 In vivo reduction in PLN protein levels fo llowing infection with PLN shRNAs....81 3-22 No reductions in PLN protein levels are observed following infection with control ss-AAV viruses in vivo ................................................................................82 A-1 Phospholamban ribozymes: Rz 201 and Rz 304......................................................97 A-2 An AAV ribozyme construct ex pressing a PLN Ribozyme and RFP....................100 A-3 Cleavage efficacy of PLN ribozymes.....................................................................101 A-4 PLN mRNA levels in HEK 293 cells co -transfected with PLN expressing plasmid and PLN Ribozymes.................................................................................102 A-5 Immunoblot of PNRVC infected with AAV packaged PLN Ribozymes..............104 A-6 Primary neonatal ventricular ra t cardiomyocytes infected 1x103 vg/cell of AAV 4 days post-infection..............................................................................................105 B-1 The pTR-UF11 plasmid.........................................................................................108 B-2 The scGFP plasmid................................................................................................108 B-3 Murine heart 5 days post-infection with scGFP.....................................................109
xiii B-4 Representative areas of murine liver and lung 5 days post-in fection with scGFP.110 B-5 Murine heart 10 days po st-infection with scGFP...................................................111 B-6 Murine heart 10 days post-infecti on with either scGFP or pTR-UF11..................112 B-7 Murine heart 14 days po st-infection with scGFP...................................................113 B-8 Murine heart 14 days post-infecti on with either scGFP or pTR-UF11..................114
xiv Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy REPRESSING A REPRESSOR: ADENO-ASSOCIATED VIRUS DELIVERED SHORT-HAIRPIN RNA MOLECULES TARGETTING PHOSPHOLAMBAN By Lourdes MarÃa Andino August 2006 Chair: Alfred S. Lewin Major Department: Medical Sciences Â—Molecular Genetics and Microbiology Regulation of calcium flux via the blockade of the -adrenergic receptor affects the phosphorylation state of phospholam ban (PLN), a regulator of the sarcoplasmic reticulum calcium ATPase (SERCA2a). Reducing the ac tivity of PLN enhances cardiac function in some animal models of congenital and ischem ic cardiac disease. Our aim was to use viral vectors to deliver short hairpin R NAs (shRNA) targeting the PLN mRNA through the RNA interference pathway. We used Ade no-associated virus serotype 1 (AAV1) and also a self-complementary version of AAV1 (scAAV1). Two PLN-specific siRNAs, si248 and si750, and a control siRNA were de signed. DNA hairpin versions were each cloned into the AAV vectors also expressing GFP. Co-transfection experiments were performed using ratios of PLN:shRNA plas mids. The shRNA plasmids were then packaged into AAV-1 capsids. These viruses we re used for infection of primary neonatal and adult ventricular cardiomyocytes. We measured PLN mRNA and protein levels using RT-PCR and immunoblots, respectively. Additionally, calcium transients were
xv recorded in infected adult cardiomyocytes. Neonatal mice were injected with the AAV PLN shRNA constructs and subsequently a ssayed using indirect immunofluorescence. Results showed a 19 to 33% reduction in PLN mRNA levels 48 hours post-transfection using the different PLN shRNAs expressed from the AAV plasmids. The reductions in protein levels were much higher (>80%) for bot h shRNAs. In the prim ary cells, levels of RNA and protein were reduced by 46 and 51% , respectively after infection with the scAAV virus. Additionally, increases in calc ium transients were observed in cultured adult myocytes infected with the PLN shRNA scAAV construct. In vivo studies using indirect immunofluorescence dem onstrated reductions in PLN levels in myocytes that had been infected with the PLN shRNA AAV constructs. The RNA interference pathway has proven to be a powerful tool that may be used for targeting cardiac proteins such as PLN. Results in primary cardiomyocytes cu ltures as well as in murine myocardium demonstrated efficacy in reducing PLN levels in these model systems. Further experimentation is needed to warrant the effi cacy of using these constructs in models of cardiac disease.
1 CHAPTER 1 INTRODUCTION When you are inspired by some great purpos e, some extraordinary project, all your thoughts break their bonds; your mind tr anscends limitations, your consciousness expands in every direction, and you find y ourself in a new, great and wonderful world. Dormant forces, faculties and ta lents become alive, and you discover yourself to be a greater person by far than you ever dreamed yourself to be. Â—Patanjali (c. first to third century B.C.) The Heart From the moment it begins beating until the moment it stops, the human heart works tirelessly. Beating over 35 million times per year, this amazing muscle has the ability to circulate 5.6 L of blood thr oughout our body, 3 times every minuteÂ—unless it fails. The goal of this dissertation is to de lineate efforts that have been made toward a therapy aimed at cardiac disease. Cardiac Architecture As fascinating as the functions of the hear t are, probably even more fascinating is the systematic precision with which the musc le is arranged (Figure 1-1). Composed predominantly of fibers, the tissue is arranged in a hierarchical fash ion. First, there are bulky fibers, enrobed in a surface membrane te rmed the sarcolemma, that are composed of dozens of smaller fibers known as myofibrils . Interspersed in th ese fibrils are nuclei and mitochondria that provide the necessary s upport for the cells. One myofibril is made up of many filaments alternati ng between thick filaments, generally myosin, and thin filaments, predominantly actin. The interdig itation of these 2 groups of filaments gives the appearance of striations to cardiac tissue. Many of these fibers are arranged end to
2 end with regions in between the filaments known as Z-disks. The region between one Z-disk to another is cons idered a cardiac cell. Figure 1-1. Anatomy of muscle. Muscle fibe rs are composed of ma ny myofibrils. Each myofibril contains filament s of actin and myosin in terspersed to form an actual muscle cell known as a sarcomere 1. The sarcoplasmic reticulum (SR), one of the organelles of interest in this research project, is the predominant st orage location for calcium (Fi gure 1-2). It is a diffuse membrane structure that surrounds the sarcom ere and closely approaches the t-system. The t-system is a network of t-tubules that are invaginations of the sarcolemma. The t-tubules serve to propagate the sarcolemmal act ion potential deep into the muscle fiber. The membrane of the sarcoplasmic reticulum c ontains primarily two proteins. The first, which comprises 90% of the membrane protei n, is the calcium ATPase known as SERCA and, in cardiac tissue more specifically, the SERCA2a. The other predominant SR protein is the calcium release channel which is close to a nd faces the t-tubule and is known as the ryanodine receptor (RyR).
3 Figure 1-2. Cardiac excitationcontraction coupling events in a ventricular myocyte. Shown are the interactions between the different cardiac signaling pathways. Inset depicts how an action potential resu lts in an increase in intracellular calcium, [Ca]i which ultimately results in the contraction of the cell 2. There are 3 human SERCA genes (SERCA 1Â–3) that may be alternatively spliced into 10 isoforms 3. The SERCA1 isoform comprises at l east two variants and is specific for the SR in fast-twitch skeletal muscle. This ATPase is regulated by a membrane protein known as Sarcopilin. Loss of function mutati ons in SERCA1 have been associated with a rare muscle disorder know n as Brody disease, a condition resulting in stiffness and cramping in the muscles. The SERCA2 protein is responsible for transporting calcium from the cytosol into th e SR, and has been found in the SR of slowtwitch skeletal muscle as well as other tissues . More specifically, SERCA2a is expressed exclusively in the SR of cardiac muscle wher e it controls the levels of calcium in the cytosol thereby regulating cont raction and relaxation of the muscle. In cardiac muscle, SERCA2a is regulated by a protein know n as Phospholamban (PLN), a homologue of Sarcopilin (See Figure 1-3 for structure compar isons). An alternatively spliced isoform of SERCA2 is SERCA2b, which is found pr edominantly in smooth muscle and non-
4 muscle tissues such as the epidermis. Mutations in this gene have been associated with a skin disorder known as Darier disease 4. Finally, SERCA3 has been found co-expressed with SERCA2b in many tissues. Unlik e SERCA2b however, SERCA3 has not been found to have any essential housekeeping functio ns. It is however, believed to be involved in the relaxation of vascular sm ooth muscle as well as regulating insulin secretion. For this reason, mu tations in this gene have b een associated with non-insulin dependent type-II diabetes mellitus 3. Cardiac Excitation-Contraction Coupling When calcium is released from the sarcoplasmic reticulum following membrane depolarization, calcium induced myofilame nt activation occurs and is termed excitation-contraction coupling. A 10-fold tr ansient increase in th e concentration of cytosolic free calcium (from about 0.1 ÂµM to about 1.0 ÂµM) triggers a myocardial contraction. Although some of this additional calcium is a result of the trigger calcium that enters through the L-type calcium channe ls and the sodium-calcium exchangers of the sarcolemma, the majority is released from the SR. The release of calcium from the sarcoplasmi c reticulum is regulated by the calcium release channels known as the ryanodine receptors. These re ceptors are located in the terminal cisternae of the SR and function to bind the trigger calcium that enters the cell, in addition to exporting the calcium from the SR into the cytosol. From the cytosol, calcium serves many functions (Figure 1-4). First, it can be resequestered into the SR through the actions of the SERCA pump a nd its regulatory pr otein Phospholamban (PLN). Second, some calcium may leave th e cell through various pumps including an ATP dependent pump or a sodium-calcium ex changer. Third, calcium could go on to activate contraction by binding to Troponin C (TnC). Calcium binding to TnC on the
5 thin actin filaments causes tropomyosin to di minish its interacti on with actin, exposing the binding site for the myosin head on the thick filament, activating this molecular motor and setting off the cascade of events that re sult in the sliding of thick filaments relative to thin filaments. Fourth, calcium may al so enter the mitochondria through what has recently been found to be the only known calcium-selective channel in the cellÂ—a calcium uniporter 5. In the mitochondria, calcium acts at several levels to stimulate ATP synthesis. The Role of Calcium in Cardiac Disease The mishandling of calcium by cardiomyocytes results in cardiac insult. Impaired calcium release causes decreases in muscle contraction (systo lic dysfunction), and defective calcium removal hampers relaxation (diastolic function) 6. Additionally, irregular calcium cycling can cause cardiac hypertrophy (thick ening of the heart muscle usually associated with stress or with increa sed work load). Elevations in calcium are believed to trigger hypertrophic signaling through the calcium activated phosphatase, calcinurin 6. This ultimately impairs cardiac contractile functions. The re-lengthening of cardiac myofilaments post-contraction is the result of two synergistic mechanisms. The first involves th e rapid resequestration of calcium into the SR, mediated by the SERCA2a. Second, calcium must be expelled through the sarcolemmal membranes. The predominant me thod for this expulsion involves exiting of calcium through a sodium-calcium exchanger (NCX). Conversely, the inward calcium current is mediated by the L-type channels. Because of the interplay between calcium and sodium, it is not surprising th at elevated levels of sodium have been found in patients with cardiac disease 7.
6 Phospholamban Discovery of Phospholamban In the early 1970s, researchers had been e xhaustively studying th e cause and effects of high cAMP levels on the uptake of calcium by the sarcoplasmic reticulum 8. The connection between cAMP and calcium uptak e was clarified by the discovery that cAMP-dependent protein kinase A (PKA) mediated the signal cascade initiated by receptor binding of the -adrenergic agonists such as catecholamines. However, the effector molecule for this process was sti ll elusive. In the mid 1970s, several groups reported that a component of the SR wa s the substrate for the protein kinase 9-11. Then Tada et al. (1975) identified a novel prot ein in the SR that was responsible for sequestering most of -32P in the presence of cAMP dependent PKA 12. Phospholamban was the name assigned to this protein from the Greek Â‘lambanÂ’ meaning to sequester and Â‘ phosphoÂ’ referring to the phosphate group that wa s incorporated into the molecule. Since these findings, studies have taken place to piece together the pathway that leads to contraction and relaxation of cardiac muscle, which is ultimately regulated by the phosphorylation cascade present inside these ce lls coupled with the regulation of ions such as calcium (Figure 1-4) 13. Phosphorylation of Phospholamban As discussed above, the SERCA2 colocali zes in the SR membrane with a unique endogenous inhibitory peptide knows as PLN. The stoichiometry that exists between the two molecules is roughly 1:1 13,14. PLN is a highly conserved 52 amino acid peptide (Figure 1-3) that is found pre dominantly in cardiac tissue and to a lesser extent in slow skeletal muscle 15. PLN contains 2 amino acid residues that are subject to phosphorylation. These are Ser16, a target for Protein kinase A (PKA) and Thr17
7 phosphorylated by calmodulin dependent protein kinase (CaMKII) 13,16. The protein has been found to exist as a 6kDa m onomer or as a 27 kDa pentamer 17. Figure 1-3. Phospholamban and its homologue, Sarcopilin. PLN consists of a cytosolic domain containing the phosphorylatable residues as well as a transmembrane domain within the sarcoplasmic reticulum 13. The phosphorylation of phospholamban is th e result of the activation of the -adrenergic pathway. Binding of agonists such as catecholamines or isoproterenol to the -1 adrenergic receptor in the sarcolemma of myocytes activates adenylate cyclase which leads to increases in the le vels of intracellular cAMP. This, in turn, leads to the activation of protein kinase A (PKA) whic h causes the phosphorylation of proteins involved in calcium regulation su ch as PLN as well as the L-type calcium channel. The phosphorylation of PLN by PKA on Ser 16 and Thr17 by CaMKII allows a
8 conformational change to ensue. This cha nge releases the inhibi tion that PLN normally has on the SERCA and subsequently, calcium can readily flow into the SR allowing the heart to enter diastole. Subsequently, calcium is then released from the SR and the heart enters systole. Figure 1-4. Interactions between the cardiac signaling pathways 13. Phospholamban Structure Originally, PLN had been implicated as solely regulating the activity of SERCA. However, recent studies have fu rther elucidated the role of PLN to encompass that of a channel-like structure for the exchange of calcium ions 18. With the resolution of the unphosphorylated structure of PL N, Oxeniod and colleagues have revealed PLN to be a funnel-like channel that cont ains a hydrophobic neck. The integrity of the symmetric
9 structure is believed to be the result of the supercoiling and bending of the membrane spanning helices of the homopentamer. Mutants such as AFA-PLN (PLN with C36A, C41F, C46A mutations) that are capable only of harboring PL N monomers have been shown to form stable complexes with SERCA that are capabl e of inhibiting the ATPase 14,18,19. It is now believed that the monomers interact with the calcium pump wh ile the pentamer plays a role in monomer storage 13. The current models have the homope ntamer initially recognizing SERCA and forming a transient intermediate. This interac tion is then thought to provide the energy to strip a subunit off of the pentamer allowing the released monomer to form a stable complex with SERCA 18. Heart Disease Effects on Society According to the Heart Disease and Stroke Statistics published in 2006, the estimated direct and indirect costs for car diovascular disease we re 403.1 billion dollars 20. Affecting 71.3 million Americans, that is one in every three Americans, it is not surprising that cardiac disease is the number one killer in our nation. According to the American Heart Association, heart disease has been credited w ith causing 58% and 37.5% of all deaths in 2002 and 2003, respec tively. Since 1900, car diovascular disease (CVD) has been the number one killer in the United States every year except 1918 (Influenza pandemic). Nearly 2500 Americans die of CVD each day, an average of 1 death every 35 seconds. CVD claims more lives each year than the next 4 leading causes of death combined, which are cancer, chronic lower respiratory diseases, accidents, and diabetes mellitus 20.
10 Causes of Cardiovascular Disease (CVD) The overwhelming cause of death from car diovascular disease is a result of coronary heart disease 20. Responsible for 53% of deaths , coronary heart disease (CHD) also referred to as coronary artery disease (CAD) or coronary atherosclerosis, involves the progressive narrowing of th e arteries that nourish the heart muscle. The narrowing is due to a buildup of fatty plaques (atheroscler osis) along the artery walls. These deposits are composed mostly of choles terol, other lipids, and fibrous tissue, such as collagen. Atherosclerosis of larger coronary arteries can lead to insuffici ent blood supply to the myocardium resulting in ischemia and consequential death of the myocardium 21. Additionally, high blood pressure resulting from atherosclerosis may lead to CVD and has been estimated to be the cause of 62% of stroke instances. This often times crippling condition is the result of a lack of oxygenated blood flowing to the brain 22. High blood pressure may also be the result of di abetes, which can ultimately cause CVD 23. Similarly, in peripheral vascular disease, elevated blood pressure can result in a narrowing of the vessels that carry blood to the legs, arms, stomach or kidneys 24. Aside from CHD, congenital cardiovascular defect s, high blood pressure and stroke account for the majority of incidences of heart failure 25. Once heart failure ensues, the outcome may be irreversible. Current Available Treatments Treatments for cardiac disease do exist. However, usually they function by temporarily ameliorating the problems or requiring a long-term commitment to drug therapy or in certain instances, major surg ery. For example, coronary bypass surgery entails removing a severely blocked artery a nd replacing it with a he althy graft from the patientÂ’s own body. Although bypass surgery gr eatly improves the way most patients
11 feel, it is not a cure for heart disease. Unless other preventive steps are taken, the processes that caused the artery di sease will continue. Additionally, -blocking drugs (which act by blocking the effect of the symp athetic nervous system on the heart) slowing heart rate, decreasing blood pressure, and thereby reduc ing the oxygen demand of the heart have been utilized to tr eat certain cases of cardiovasc ular disease by preventing left ventricular remodeling. In post-infarct rats, the use of some of these drugs such as Carvedilol has been shown to prevent coll agen deposits, while Metoprolol has been shown to reduce myocardial hypertrophy 26. Similarly, calcium channel blockers have also been used as a means of reducing the amou nt of calcium that enters the muscle cells in the coronary artery walls and thus preven ting spasms. However, th ere have also been many risks associated with the use of thes e types of medications , including increased risks of cancer 27-29 as well as intestinal hemorrhage and even death 30. Diuretics have also been used as a means of relieving high bl ood pressure by expell ing salts out of the body. However, as a side effect, some classe s of diuretics also cau se the excretion of potassium from the body and are ther efore risky for certain patients 31. One of the earliest medications that was prescribed for heart disease was nitrog lycerin. Still used today, this medication serves as a vas odilator once the body na turally converts the nitroglycerine into nitrate oxide , a vasodilator. Nitroglycerin has also been utilized as a method to treat and prevent attacks of angina (chest pains due to ischemia) and reduce blood pressure 25. Although the use of -blockers to treat a malfunctioning heart seems counterintuitive, res earchers have shown reductions in mortality 32, improvement in ventricular function, as well as a re versal in pathological remodeling 33. Over 20 years
12 ago, it was established that patients with congestive heart failure have hearts in a hyperadrenergic state 34, despite having a reduc tion in the amount of -1 adrenergic receptors ( -AR). Although the -ARs levels are reduced, hyperphosphorylation of ryanodine receptors has been seen in pa tients with heart failure. Conversely, hypophosphorylation of phospholamban also results 35. Although the mechanisms are not completely understood, these phenomena are reversible with -blocker treatments 36. In order to provide an alternative for these t ypes of treatments, we began to explore the prospects of gene therapy to tr eat this devastating disease. Gene Therapy History Over the past 25 years, the field of gene therapy has not only become established, it has revolutionized the prospects of modern medi cine. The first decade proved to be a bit trying for the field. However, gr eat strides have been made in terms of safety and breadth of application. The field of gene therapy began turbulen tly with the first unauthorized attempt occurring in 1980. Dr. Martin Cline, a genetici st at the University of California in Los Angeles, conducted a recombin ant DNA (rDNA) transfer into the bone marrow cells of two patients with hereditary blood disorders. His research was conducted without the approval of neither the National Institutes of Health (NIH) nor of the Institutional Review Board (IRB) at his university. Consequen tly, Dr. Cline was forced to resign his departmental chairmanship at UCLA and lost several research grants 37. Although neither of his patients surv ived, the door had opened for this new field of therapy. First disease approved for gene therapy treatment Â– In 1990, adenosine deaminase (ADA) deficiency was the first disease a pproved for gene therapy treatment by the
13 federal government. Lacking the ADA enzyme, patients with this disorder have a severely weakened immune system. The initial therapy was performed by Blaese et al. on two young girls 38. It involved removing peripheral T lymphocyte cells from the patientÂ’s body, using recombinant retrovirus gene transf er to introduce the ADA gene into the cells and then replacing the cells into the patient. Following the transfusions, the girls were able to live relatively normal lives an d were reported to have gene-transferred lymphocytes 10 years after the therapy 39. In an alternate study, gene therapy was us ed to attempt to treat 10 children with X-linked severe combined imm unodeficiency disease (X-SCID) 40. In this study, cycling hematopoetic stem cells were transduced ex vivo with a murine leukemia virus vector containing the IL2RG gene (necessary for the production of the c cytokine receptor subunit). Following this treatm ent, the childrenÂ’s T cell leve ls were restored to normal levels 40. Using this therapy, the ch ildren appeared to be comp letely cured of the disease without requiring any additional medication. Un fortunately, three year s later, two of the 10 children developed leukemia af ter the initial treatment. Th e virus integrated into the LMO2 proto-oncogene promoter, leading to aberrant transcription and expression of LMO2, a transcription factor and cen tral regulator of hematopoiesis 41. Without a doubt however, the field has cont inued to progress and the regulations imposed by the government have become more stringent allowing for safer therapies to ensue. Several years ago, two initiatives by the FDA and NIH came into play to address the concerns of gene therapy safety. The first was Â“The Gene Therapy Clinical Trial Monitoring Plan.Â” It was designed to increase the level of scrutiny with additional reporting requirements for study sponsors. A dditionally, a series of Â“Gene Transfer
14 Safety SymposiaÂ” were designed to get researchers to talk to each other, to share their results about unexpected problems and to make sure that everyone knew what the rules and expectations were for safe trials 42. Furthermore, the FDA launched random inspections of 70 clinical trials in more than two dozen gene therapy programs nationwide and instituted new reporting requ irements. Moreover, President Clinton proposed a legislation that would mandate pe nalties for researchers and institutions found to be in violation of regulations governing human clinical trials 42. The continued optimism in the gene ther apy field can be seen by the rapidly growing number of clinical tr ials. In the late 1990s th ere were only about 200 gene therapy trials underway 43; however, that number has now risen to well over 600 ongoing clinical trials across the world 44. These trials have been hi ghlighted with glimpses of success including the trial used to treat pa tients suffering from Hemophilia B. These patients lack the factor IX clotting factor and as a resu lt, were treated with an AAV vector via intramuscular injec tions to re-introduce the ge ne. Although there were only minimal amounts of factor IX detected in tr eated patients, it was enough to show minimal clinical benefits 45. Gene Delivery There are two basic categories for gene de livery vehicles used in gene therapy. The first involves the use of non-viral delive ry. These methods employ agents such as cationic lipid DNA mixtures, delivery of na ked DNA or a delivery involving poly-lysine coated DNA. The downside to these approaches is that the gene of interest is often short-lived, not highly expresse d, or inefficiently delivered. Although there are ways to increase the efficiency of delivery, the transi ent nature of the gene expression remains an obstacle to direct DNA delivery 43.
15 The second category of gene delivery invol ves the use of attenuated viruses. The use of viral vectors has many positive impli cations; most importantly perhaps is the virusÂ’ natural mechanism for entering a cell. Although host cell reactions to viruses have proven to be problematic, mechanisms to cope with these obstacles ar e being studied. An overview of the most commonly used viruses will be discussed below. Retroviruses Several viruses have come to the forefront over the past few years in terms of being the promising Â“vector of choiceÂ” for gene therap y. The first that will be discussed are the retroviruses, which contain an RNA genome. Being the first virus tested for clinical applications in the SCID trials , this virus offers great promis e because of its fairly large packaging capacity (7.5 kb) and stability. Retroviruses can be classified into two categories: simple retroviruses (based on the murine leukemia virus) including the Alpha, Beta and Gamma retrovir uses and the complex retrovir uses such as Lentiviruses (e.g. human immunodeficiency virus [HIV]) , Deltaretroviruses, Spumaviruses and Epsilonretroviruses. Simple retroviruses do not have the ability to infect non-dividing cells and have the ability to integrate into the host chromosome. Complex retroviruses on the other hand, can infect both dividing and non-dividing cells and may also integrate into the host genome. This ability to infect non-dividing cells is one of the reasons that lentiviral vectors are commonly used fo r gene therapy, especially in neurons 46 and human lymphocytes 47. Among the commonalities existing between these RNA viruses are the presence of the gag , pol , and env essential genes flanked by long terminal repeats (LTRs). Additionally, lentiviruses encode 3Â–6 additional viral proteins needed for replication and perseverant in fection. The most extensivel y studied of the lentiviral vectors has been the human immunodefici ency virus type 1 (HIV-1). Although
16 investigators were initially pessimistic about using these vectors in gene therapy because of safety concerns, strides aimed at creating self-inactivating vectors have increased their biosafety. These vectors are also devoid of all the accessory proteins and have minimal amounts of viral sequences making homologous r ecombination to recreate an active viral genome highly unlikely. Adenovirus Several DNA viral vectors are also curren tly in use for gene therapy. Among the most common ones is the Adenovirus (Ad). Th ere are more than 50 serotypes in the Ad contingency and infectivity has been reported in a wide range of tissue types. Most viral vectors, however, have been derived from se rotype 5. The virus consists of 36 kb of double-stranded DNA and codes for over 50 proteins. The genes of Ad can be categorized into 3 groups depending on their ro le in the viral repl ication cycle: early, delayed and the late. The genome is packag ed into a non-enveloped icosahedral capsid. Having a large packaging capacity is one of the more favorable characteristics of this vector. Vectors have been ut ilized that have deletions of the nonessential early genes E1 and E3 providing a total cloning ca pacity of 8.3 kb. However, in vivo studies have been hampered because of the high levels of inflammatory response associated with this virus. Initially, Ad causes a nonspecific host cyto kine response. A secondary response that follows includes cytotoxic T lymphocytes targ eting infected cells displaying the viral peptide antigens. In order to address these concerns, gutless Ad vectors that have a 36 kb packaging capacity have been developed. Th ese viruses are entirely devoid of all coding viral regions except the invert ed terminal repeats and the packaging signal. Although the host immune response is not as robust when using these gutless vectors, systemic
17 delivery of these viruses stil l induces an adaptive humora l response against the vector capsid 48. Herpesvirus Human herpesviruses are an additional cla ss of DNA viral vectors currently in use for gene therapy. Consisting of four compone nts: envelope, tegument, capsid, and viral genome, the entire virion is 20 nm in di ameter. The linear double stranded DNA genome is 152 kb in length and is composed of unique long and unique short regions flanked by terminal repeats. Recombinant HSV-1 vect ors contain a number of mutations in the immediate early genes and can therefore carry large payloads of up to 30 kb in length. Because of the deletions, these vectors are nonreplicating and can be produced in high titers in cell lines that pr ovide the deleted genes in trans . There are three different types of vectors that have been derived from HS V-1. The first are attenuated recombinant vectors, which are replication competent and can only infect actively dividing cells. The second class includes defectiv e recombinant vectors. Th ese are disabled and are therefore replication incompet ent and not able to dissemina te out of infected cells. Lastly, the amplicon vectors are defective, he lper dependant vectors. These take full advantage of the large transgen ic capacity of the virus. There is no other mammalian vector that can deliver such a large payl oad without simultaneously delivering viral genes 49. Adeno-Associated Virus (AAV) The last class of viral vect ors I will discuss is derive d from the Adeno-associated virus. This has become our viral vector of choice for a host of reasons. First, AAV has been shown to transduce both dividing and non-dividing cells 50-52. For the purpose of our studies, this is an essential feature si nce cardiomyocytes do not divide. Secondly,
18 AAV vectors do not contain any viral protein coding genes and, therefore, there has not been any toxicity associated with gene expression 53-55. Although some toxicity has been associated with the capsid proteins 56, the actual viral infectio n has not been proven to have any malignant side-effects because it do es not have the ability to produce any proteins on its own. Additionally, AAV has been demonstrated to have the ability to express for very long periods of time in the heart 57. This characteristic is beneficial when considering it as a form of gene ther apy, since one treatment may be sufficient to achieve the desired levels of gene expr ession. AAV does however, provoke a humoral immune response that may prevent the successful readministration of vectors of the same serotype (see below). In animal studi es, AAV does not stimulate an inflammatory response or cell mediated immunity, maki ng it inherently safer than Ad or HSV 58. Additionally, AAV is also an attractive vector because of the fact th at it can be produced in high titers relatively simply 59. Another suitable characteristic of AAV is th e plethora of serotypes that have been isolated. Although very similar in structure, differences in the capsid structure of the different serotypes have allo wed for varied tissue tropisms due to the fact that they require different types of receptors for cell entry. That being said, targeting specific organs or tissue types no longer have to be li mited to using particular promoters. Instead, the desired DNA sequences can be packaged in to the serotype speci fic for the tissue of interest. Although AAV serotype 8 (AAV8) ha s now been proven to be the preferred serotype for the myocardium because it lead s to rapid and vigorous genetic transduction 60, AAV1 has also proven to be very efficient 61 and was my serotype of
19 choice for this study. The propensity of each serotype for particular tissues has been extensively studied and reviewed 62 and is summarized in Table 1-1. Over 100 different serotypes of AAV ha ve been isolated (reviewed by Wu et al. 63) namely AAV 1Â–9. Although relatively simila r in their capsids, differences have manifested themselves in their tropism. AAV 2, being the first to be isolated, has been the most widely studied and found to have a broad cell tropism. Conversely, AAV5 has been found to have the greatest affinity for the retina and AAV1 has been shown to greatly transduce muscle and liver 64-66. In a study performed by Du and colleagues, AAV1 was found to be the best at transducing murine ca rdiac muscle as well as primary ventricular adult murine cardiomyocytes. However, AAV2 was found to be slightly better at transducing neonatal cultures 61. Different serotypes are no t only useful for infecting different tissue types, but st udies have also shown that i ndividuals may be tolerant to certain serotypes of AAV, particularly AAV2 67. In this case, in fecting with another serotype of AAV allows the virus to evade the humoral response. Table 1-1. Different tropisms for Adeno-associated virus serotypes Serotype Isolation Tissue or Cell Tropism AAV1 Cell line Muscle, eye, liver, lung AAV2 Cell line Muscle, brain, liver, eye AAV3 Cell line Not determined AAV4 Cell line Brain AAV5 Human lesion Brain, muscle, liver, lung, eye AAV6 Cell line Muscle, eye, liver, lung AAV7 Monkey Muscle, liver AAV8 Monkey Liver, heart AAV9 Human tissue Heart, skeletal muscle, pancreas This table is adapted from Hildinger et al. 62 and summarizes were the different AAV serotypes were isolated from and what their tissue preference is. The major impediment imposed by AAV is its relatively small packaging capacity namely, 4.7 kb. However researchers have f ound a way to circumvent this problem by
20 packaging large genes into two separate constructs 68,69 and allowing homologous recombination to recreate the full-len gth product once the cell is infected 70. In reference to wild-type AAV biology, the virus is relatively simple. The virion particle is 20 nm in diameter and contains a single-stranded DNA genome. There are two viral open reading frames (orfs): rep coding for the replication and packaging proteins and cap coding for the 3 structural capsid prot eins (VP1, 2, 3). The rep orf produces 4 alternatively spliced gene products under the expression of the same promoter. These are termed, depending on the molecular weight of their protein products, Rep78, Rep68, Rep52 and Rep40. The capsid proteins VP 1, 2, and 3 are produced from another promoter and are also alternatively spliced pr oducts of the same gene . One viral capsid particle contains the 3 proteins in a 1:1:10 ratio (VP1:2:3). The rep and cap genes are flanked by 150 bp inverted terminal repeats (ITRs ) that are essential for viral packaging, replication and integration. Consequentl y, in recombinant AAV (rAAV) constructs, the ITRs are the only part of AAV that are retain ed. The ITRs fold back on themselves to form hairpin structures that contain the origin for viral re plication. The other genes are provided in trans by helper plasmids allowing a payload of interest to be cloned in their place. Belonging to the dependovirus family, AAV requires co-infection with a helper virus in order to produce a pr oductive infection. In its latent phase, AAV has been pinpointed to integrate into a par ticular site in chromosome 19 71, although this phenomenon has been found to happen infre quently. Because AAV integration requires interactions amongst the chromosomal target site, the AAV TR structures as well as the rep gene (which is only pres ent in wtAAV), integration of r AAV has rarely been seen
21 although, it appears to be random in these serendipit ous occurrences 72. Nonetheless, these concerns are somewhat alleviated because r AAV is generally thought to persist in an episomal state in transduced cells 72. Self-complementary AAV. One of the major downsides to AAV is its inability to express the gene of interest in a rapid fash ion. Research in different tissues has shown that expression from AAV vectors begins at about 2 weeks with maximal expression occurring at about 4 weeks. Both Samulski Â’s and WilsonÂ’s groups have suggested that synthesis of the second DNA strand is the rate limiting step for AAV in the absence of helper viruses such as Ad 73,74. In 2001, McCarty and coll eagues cleverly devised an AAV construct that would address this concern 75. Because AAV contains a singlestranded DNA genome (Figure 1-5 A), synthesis of the s econd strand must take place before expression occurs. This part of AAV biology has been considered to be the ratelimiting step of AAV expression 73,74. To overcome this rate-limiting step, McCarty et al. created a construct (Fig ure 1-5 B) in which one of the AAV terminal repeats was mutated so that it would no longer form the traditional hairpin structure. Furthermore, the vector payload was d ecreased by half, which although is a stepback from some points of views, it is not significant when consid ering the delivery of small molecules. This point will be disc ussed further in a subsequent section. The resulting virion could now encapsulate tw o complementary DNA molecules that would anneal to each other creating a double-strande d genome, eliminating the rate limiting step and enabling a rapid expression of the gene of interest. The resulting molecule has been termed the self-complementary AAV (scAAV) vector.
22 A B Figure 1-5. Single-stranded a nd self-complementary Adeno-associated virus genomes. Adapted from McCarty et al. 75 are schematic diagrams for the A) singlestranded and B) self-complementary AAV genomes. Shown are representative molecules expressing the mouse erythropoietin gene (mEpo) transcribed from the CMV promoter. RNA Interference In 1998 a groundbreaking finding was publis hed delineating the Â“potent and specific genetic interference by double-stra nded RNA in Caenorhabditis elegans,Â” demarcating the birth of RNA interference (RNAi) 76. In this paper, Fire et al. described how double-stranded RNA molecules readily inte rfered with the expression of genes in C. elegans in comparison to either the sense or an tisense strands alone. As it turns out, eight years earlier, the RNAi effect had already been observe d in plants, yet was not fully understood 77. This finding was actually the first re port describing the RNAi mechanism. Napoli et al. proceeded to demonstrate how the in troduction of a gene, that they were hoping would produce more intensely colo red petunias, actually resulted in hypopigmented flowers due to RNAi. Following the Fire et al. report, researchers began to exploit this technology and use it as a means of performing reverse genetics and studying the function of virtually any gene. Soon after, Zamore et al. demonstrated that the RNAi mechanism was mediated by 21 nt dsRNA fragments that were the cleavage
23 products of long double-stranded RNA molecules 78. These Â“short-interferingÂ” RNA molecules (siRNAs), as they are now known, ha ve become a valuable tool for regulating gene expression. Short-Interfering RNAs Several years later, Elbashir et al., devised a sequence specific method of applying this novel technique to cultured mammalian cells 79. In this study, short-interfering RNAs were created so as to bypass the interferon response. The interferon response is most often induced by molecules greater than 30 nt in length. This pathway results in unintended off-target effects leading to global, nonspecific i nhibition of protein translation 79. The findings by Elbashir et al. described 21Â–22 nucleotide molecules that could evade the host innate immune response to RNA because of their small size, while at the same time mediating the site specific cleavage of the targeted mRNA. Molecules larger than this were found to activate an antiviral response lead ing to the non-specific degradation of RNA transcripts and eventu ally a complete shutdown of host cell translation (i.e. interferon response). Conve rsely, molecules 21Â–22 nuc leotides in length with symmetric 2 nucleotide 3Â’-overhangs were found to effectively knock-down gene expression in HEK 293, NIH/3T3, He La and other mammalian cells. Mechanism . The RNAi machinery is ubiquitously expressed in eukaryotic cells. Endogenously, it regulates the expression of ge nes that are involved in cell fate and differentiation through the use of small hairpin RNAs known as pre-microRNAs (pre-miRNAs). Pre-miRNAs are processe d in the cytoplasm into miRNAs, ~21Â–26 nucleotide RNA molecules 80,81. Although the mechanism is still being elucidated, the overall scheme is effectively understood. Initially, double-stranded RNA (dsRNA) enters the cytoplasm of a cell in essentially one of f our ways. The first two ways come from the
24 nucleus in the form of either dsRNA from m obile genetic elements and abnormal genes or pre-miRNA precursors. Additionally, the double-stranded RNA may enter the cell as the genetic material of an RNA virus or as a synthetically synthe sized double-stranded RNA molecule. Once inside th e cytoplasm of the cell, an enzyme known as Dicer cuts these long dsRNA molecules into small 21Â–22 nuc leotide fragments referred to as either short-interfering (siRNAs) or miRNAs (S ee Figure 1-6 for siRNA mechanism). Subsequently, the RNA induced silencing co mplex (RISC) is believed to cleave the passenger strand (the sense strand relative to the mRNA target) and incorporate the guide (antisense) strand 82. The RISC, in conjunction with the single-stranded RNA molecule, then proceeds to bind to its target mRNA. When perfect complementarity arises between the guide strand and its mRNA target, the tr anscript is degraded. However, if the complementarity is imperfect, tr anslational inhibition occurs 83-85. Figure 1-6. The siRNA pathway. The basic mechanism of RNAi is thought to be a multistep process occurring in the cytoplasm of cells 86.
25 Short-Hairpin RNAs Although siRNAs quickly proved to be a useful means for down-regulating genes in tissue culture, their transient nature rest ricted applications in cluding the prospects of using these molecules in gene therapy applications. For this reason, DNA vector-based RNAi technology was developed. In this system, either U6 or H1 RNA pol III promoters have been implemented to drive the expr ession of the short-ha irpin RNA molecules (shRNA) from their complementary DNA seque nces. Many papers have since emerged using this methodology includi ng one of the first by Sui et al. , in which they knocked down both endogenous and transfected genes using shRNAs synthesized from DNA templates 87. The U6 and H1 promoters have proven to be favorable because they have been demonstrated to direct the synthesis of small, non-coding tran scripts whose 3Â’ ends are terminated by a stretch of 4Â–5 uridines 88. Additionally, the cleavage of the transcript occurs at the termination s ite after the second uridine 89 yielding a transcript resembling a synthetically synthesized siRNA. In addi tion to the selection of a promoter, a loop structure is needed to function as a short spac er so that the predicted transcript can fold back on itself and form the double st randed RNA molecule. Brummelkamp et al. were successful in designing an optimal loop sequenc e for this function (i.e. UUCAAGAGA) 90. More recently several groups have expressed fully complementary siRNAs in the contex t of miR-30, and microRNA, using RNA polymerase II promoters. This permits re gulation of siRNA expression using an alternative promoter, not possible with the U6 or H1 promoters 91. Interfering with Disease: Prospects for Therapy With the advent of RNAi, researchers quickly commandeered the technology to be used as a therapeutic method to combat diseas e. Initially, there was concern as to the
26 effective way to deliver these sm all double stranded RNA molecules in vivo due to their short half-life. However, many of the traditional gene therapy viruses and cloning vectors, including the ones mentioned earli er, have quickly adapted their cloning and expression strategies to allow for effective delivery of these powerful molecules. McCaffery and coworkers, for example, demonstrated the efficacy of treating rodents in vivo with these strategies to combat seve ral diseases including Hepatitis B using a shRNA expressing plasmid DNA 92. Furthermore, Xia et al. used AAV to express a shRNA in an animal model of Spinocerebellar ataxia-1 93. More recently, Xu et al. have published the first report demonstrating the use of scAAV to deliver shRNA molecules into cell lines in an attempt to diminish th e multidrug resistance associated with treating cancer patients with chemotherapy 94. Project Overview Combining all of the knowledge that has b een elucidated over the past few years in the field of gene therapy, it was clear to us that a potentially therapeutic way to treat cardiac disease would be to design shRNA mol ecules to target PLN and deliver them to the myocardium using AAV1. In order to reach this goal, I implemented a methodical approach to designing, testi ng, and delivering our PLN shRN A molecules. This study will present initial co-t ransfection experiments performed in HEK 293 cells that validates the activity of chemically synthesized PL N siRNA molecules. Subsequently, I will design AAV vectors that will deliver these siRN A molecules as stable shRNAs expressed from both self complementary and singl e stranded AAV packaging plasmids. The activity of these PLN shRNA-expressing AAV plasmids will then be tested and ultimately packaged into AAV1, where they will be used to infect both neona tal and adult ventricular cardiomyocytes. Using several biochemical assays, I will demonstrate the
27 efficacy of these molecules in reducing the levels of PLN and determine the physiological effect of this silencing in infected cardiomyocytes. This study will ultimately culminate in the delivery of the AAV-PLN shRNA molecules into the myocardium of neonatal mice using subxiphoid injections. Furthermore, due to the efficacy of scAAV infections, a comparison be tween the efficiency of the scAAV and the more traditional ssAAV will also be addre ssed as it relates to the myocardium. In summary, our experiments have demonstr ated that we have been able to design siRNA molecules that are capable of reduc ing exogenous PLN in co-transfected HEK 293 cells. Additionally, these molecules have been delivered as AAV vectored shRNAs into primary cardiomyocyte cultures and have also reduced mRNA and protein levels of PLN. This reduction has led to increases in calcium transients within infected cardiomyocytes suggesting an increase in cardiac function. Finally, using AAV as a delivery vehicle, the shRNA molecules deli vered using a subxiphoid injection into murine myocardium. Here they were found to reduce the amount of endogenous PLN. Due to the importance of the regulation of cal cium in failing myocardium, we feel that these molecules could potentially ameliorate cardiac dysfunction in instances of ischemia and myocardial infarction. Furt her testing will be needed to warrant the efficacy of these molecules in models of cardiac disease.
28 CHAPTER 2 MATERIALS AND METHODS When you work you are a flute through whose heart the whispering of the hours turns to music. To love life through labor is to be intimate with lifeÂ’s inmost secret. All work is empty save when there is love, for work is love made visible. Â—Kahlil Gibran (1883Â–1931) Design of Phospholamban Expressing Plasmid To date, no cell line that expresses a dete ctable level of Phospholamban has been established. During the course of these studies, se veral cardiac cell li nes including H9C2, a Rattus norvegicus myoblast cell line and A7R5, a Rattus norvegicus fibroblast cell line, tested negative for the presence of Phos pholamban using RNA and protein detection methods. For this reason, a plasmid expr essing Phospholamban was created that would be able to express Phospholamba n in transfected mammalian cells. Cloning PLN into pTR-UF11. In order to construct th e PLN expressing plasmid, total RNA was extracted from a mouse heart as described below. This RNA was used to reverse transcribe the Phospholamban sequen ce using the Â‘PLN 842Â’ antisense primer and conditions listed in Ta ble 2-2. Following revers e transcrip tion using 1st strand cDNA synthesis kit (Amersham-Pharmacia), the PLN cDNA was amplified by the polymerase chain reaction (PCR) using the primers desc ribed in Table 2-2. The insert was then cloned in the pCR2.1-TOPO II pl asmid (Invitrogen). The insert was subsequently cut out of this plasmid with the unique enzymes Not I and Bam HI . These same enzymes were used to cut the GFP and the Neomycin casse ttes out of the pTR-UF11 plasmid. This allowed for directional cloning of the PLN ge ne downstream of the CBA promoter. The
29 plasmid was then sequenced and validated to ha ve the correct insertion and orientation. A map of the final plasmid can be seen in Figure 2-1. Chicken -actin promoter Phospholamban6078 bp AmpR ColE1 ori f1(+) origin TR TR PLN orf CMV ieenhancer Intron PLN Exon1 Chicken -actin promoter Phospholamban6078 bp AmpR ColE1 ori f1(+) origin TR TR PLN orf CMV ieenhancer Intron PLN Exon1 Figure 2-1. Murine phospholam ban expressing plasmid. PL N is expressed downstream of the CBA promoter. The plasmid also contains ampicillin resistance for propagation in bacteria. Design and Preparation of siRNAs siRNA Design The murine Phospholamban sequence was analyzed using Ambion siRNA design software freely available at www.ambion.com. Two candidate siRNAs were selected. The first, si248, led to cleavage within th e PLN coding region after nucleotide 248. The second, si750, led to cleavage after position 75 0 which lies in the 3Â’ untranslated region and had the following sequence: 5Â’AACTA AGTGTTCTTCAGTTCT. A BLAST search of the mouse genome using the two siRNAs se quences found no additional target genes. As a control siRNA, we chose one that targ eted GFP, siGFP (Dharmacon). The target sequence for this siRNA was GGC TAC GTC CAG GAG CGC ACC 95. An alignment was done of the murine PLN sequence with several available Rattus PLN coding regions (Accession numbers listed in Table 2-1). A very high level of homology was found amongst the mouse and ra t PLN sequences; however, the sequences
30 were not identical for the siRNA regions se lected. For si750, the homology was very low and for this reason, a rat version of si750 was not designed. However, for si248, only two nucleotides were different. Therefore, the siRNAs were modified to contain the appropriate rat sequence and were then s ynthesized by Dharmacon. In total, three siRNAs were synthesized: M750Â—targe ting mouse, M248Â—targeting mouse, and R248Â—targeting rat. Table 2-1. Alignment of murine and rattus PLN 248 siRNA sites MurinePLN NM_023129 AAGC A CGTCAGAA T CTCCAGA RattusPLN s95849 AAGC G CGTCAGAA C CTCCAGA RattusPLN 022707 AAGC G CGTCAGAA C CTCCAGA RattusPLN X71068 AAGC G CGTCAGAA C CTCCAGA Deprotection and Annealing of siRNA Molecules In order to deprotect the 2Â’-O-methyl group of the RNA oligos, 200 ÂµL of the 2Â’-deprotection buffer (100 mM acetic acidÂ—TEMED pH 3.8) provided by the manufacturer were added to each single-s tranded complementary RNA strand. The two volumes of the complementary RNA strands were combined, briefly vortexed and incubated at 60oC for 45 minutes. The samples were then removed from the heat and briefly centrifuged. They were then allowed to cool at room temperature for 30 minutes. Desalting the Duplex Ethanol precipitation was pe rformed in order to desalt the siRNA duplex. Briefly, 40 ÂµL of 10M ammonium acetate and 1.5 mL of absolute ethanol were added to the 400 ÂµL siRNA duplex solution. Sa mples were incubated at -20oC overnight. Subsequently, they were centrifuged at 14,000g for 30 minutes at 4oC. The ethanol was then removed and the pellets were rinsed w ith 95% cold ethanol and dried under vacuum.
31 Quantification of siRNA Duplex The dried pellets were resuspended in 1 mL of Dharmacon Universal buffer (20 mM KCl, 6 mM HEPES-KOH pH 7.5, 0.2 mM MgCl2) and the concentration of this stock solution was quantitated using a spectrophotometer and reading the OD260. Cloning of shRNAs into AAV TR Plasmids Based on the effectiveness of the siRNA molecules in cell culture, short hairpin versions of these molecules were synthesi zed. Because the lifespan of naked siRNA molecules in vivo is transient, mammalian expression vectors were created that would direct the synthesis of siRNA-like transcripts over a longer period of time. We designed a 19 nucleotide sequence separated by a short spacer loop followed by the reverse complement of the original sequence. The resulting transcript would form a hairpin structure that upon cleavage by Dicer, would result in an siR NA molecule. The size and sequence of the loop is very important. Ther efore, the loop that we chose was the one recommended by Brummelkamp et al .: 5Â’UUCAAGAGA 90. Additionally, a Bam HI site and a Hind III site were added to the 5Â’-ends of the sense and antisense strands, respectively (see Figure 2-2). The resulting DNA sequence was ordered from Invitrogen and cloned into the pSilencer plasmid from Ambion. This plasmid contains a multiple cloning site downstream of an HI promoter, an RNA pol III promoter. Positive clones were identified by restriction digests, as well as sequencing. Primers (Ambion/ Sal I, see Table 2-2) were then desi gned to amplify the region of the pSilencer construct containing the HI prom oter and the shRNA sequence, including a termination sequence. These prim ers were constructed to contain Sal I sites within, allowing cloning into the unique Sal I site in the pTR-UF11 plasmid. This plasmid already contained the Chicken -actin (CBA) promoter drivin g the expression of the GFP
32 sequence. This newly created construct al lowed us to use GFP as a reporter gene, identifying the cell which is expressing our siRNA of interest. The final clone was verified by DNA sequencing. G GATCCCCGC ACGTCAGAAT CTCCAGA TTCAAGAGA TCTGGAGATTCTGA CGTGC TTTTTGGAAA AGCTT CCTAG GGGCG TGCAGTCTTA GAGGTCT AAGTTCTCT AGACCTCTAAGACT GCACG AAAAACCTTTTCGA A sense antisense loop BamHI HindIII G GATCCCCGC ACGTCAGAAT CTCCAGA TTCAAGAGA TCTGGAGATTCTGA CGTGC TTTTTGGAAA AGCTT CCTAG GGGCG TGCAGTCTTA GAGGTCT AAGTTCTCT AGACCTCTAAGACT GCACG AAAAACCTTTTCGA A sense antisense loop BamHI HindIII Figure 2-2. Sequence of the components of PLN shM248. This fragment is ready for cloning into AmbionÂ’s pSilencer constr uct. Red nucleotides denote those present in the pSilencer construct wh ile black nucleotides represent the sequence being introduced. In addition to the shRNA pictured in Fi gure 2-2, another murine shRNA targeting PLN was created based on si750. This shRNA ta rgeted the 3Â’ untranslated region of the PLN RNA . Additionally, shR248 was also created to target the rat PLN sequence at position 248 as described above. The fourth shRNA that was create d targeted Rhodopsin and was used as an irrelevant shRNA control. The sequence was 5Â’CTTGGCTGTGGCTGACCTC and was kindly provided by Dr. Marina Gorbatyuk. All of the shRNA constructs were engineered in the same manner. Sequencing confirmed their sequences and Sma I digests confirmed the presence of the inverted terminal repeats (TRs) necessa ry for packaging into AAV. Self-Complementary AAV Constructs. The traditional AAV packaging construct has a packaging capacity of 4.7 kb. However, be cause this virus is single stranded, it has been found the rate limiting step for viral e xpression to be the synthesis of the second strand 50,73. For this reason, a second generation of AAV constr ucts has been developed as a way of circumventing this phenomenon. These constructs contain a mutated TR and half of the packaging capacity of their si ngle-stranded counterpa rts. Through these
33 modifications, 2 strands of complementary DNA can be encapsulated into one virion particle enabling a rapid expres sion of the gene of interest (see Chapter 1). These viruses have been named self-complementary AAV viruses (scAAV). We sought out to create scAAV constructs to harbor a GFP expression cassette as well as our shRNA of interest in a similar ma nner to the creation of their single-stranded counterparts. Briefly, shR NA molecules were cloned in to the pSilencer backbone, amplified using Sal I primers and then ligated into the unique Sal I site found in the scAAV backbone kindly provided by Dr. Jude Sa mulski. Constructs were sequenced to verify the insert and Sma I digests confirmed the presence of TRs. DNA Techniques Linearization and Dephosphorylation of Vectors Plasmid vectors were linearized with the appropriate restriction enzymes (Promega) in the supplied reaction buffer for 1 to 2 hours. Dephosphorylation of the 5Â’phosphate on the vector was carried out with shrimp alkaline phosphatase (Promega) at 37oC for 1 to 2 hours. Digested DNA was resolved by gel elect rophoresis in a 1% agarose gel with TBE buffer ((10 mM EDTA, 0.45 M Tris, 0.45 M Bori c Acid)) containing 0.1 Âµg/mL ethidium bromide (EtBr). Bands of DNA were visu alized using long wave UV and excised from the agarose gel. Ligation Ligations were carried out in a volu me of 10 to 20 ÂµL and included 200 ng linearized vector, insert, 0.1 units T4 DNA ligase enzyme (Promega), and buffer provided with the enzyme. Insert to vector molar rati os were either 1:2 or 1:3. Reactions were carried out at 14o to 16oC overnight.
34Table 2-2. Primers PCR Conditions Primer Name Orientation Sequence Anneal Number of Cycles Sense ACC ATC TTC CAG GAG CGC GA GAPDH Antisense GAG CCC TTC CAC GAT GCC AA 55 C 45 sec 18 Sense ATC ACC GAA GCC AAG GTC TC PLN 400 bp Antisense TCG TGA CCC TCA CAA AGC TG 55 C 45 sec 26 Sense TGC CCA GCT AAG CTC CCA TAA PLN 842 bp Antisense GTC ATC TAT GTG AGG ACC CAG TGA G 50 C 45 sec 35 Sense TCA CGA C GT CGA C AA ACG ACG GCC AGT Sal I/Ambion Antisense AAA CAG CTA GTC GAC CAT GAT TAC GC 55 C 30 sec 25 All primers are listed beginning at the 5Â’-end. PCR conditions ar e listed for each primer set. An additional 5 min denaturati on step was added at the beginning of each PCR as well as a 7 minute fi nal extension step at the end. The bolded sequence in the Sal I/Ambion primers denotes the Sal I cut sites.
35 Fast ligation reactions were carried out similarly but instead, a fast ligation buffer (Promega) was utilized. The ligation wa s performed for 5Â–15 minutes at room temperature in a 15Â–20 ÂµL volume. Ligations re actions were then either stored at -20 C or transformed immediately. Transformation E. coli DH5 cells (Invitrogen) were used for the transformation of non-TR containing plasmids while SURE cells (Invitr ogen) were used for the transformation of TR containing plasmids. For both cell type s, transformations were carried out by electroporation at 1.5 volts for 5 msec in 1 mm Fisher cuvettes. Cells were allowed to recover in 1 mL LB for 1 hour at 37oC in a shaking incubator. Subsequently, the cultures were spread on LB agar plates containing am picillin (amp) for selection of the positive transformants. Plates were incubated at 30oC for about 16Â–18 hours taking care not to let the TR containing colonies grow too large. Plasmid Mini-Preparation Three to 5 mL of LB media containing 100 Âµg/mL of ampicillin was inoculated with individual bacterial clones picked from the agar plates. Cells were incubated for 12 to 14 hours at 30oC in a shaking incubator. Plasmids were isolated using either the GenElute HP plasmid miniprep kit (Sigma ) or the Perfect Prep Plasmid Mini kit (Eppendorf). Plasmids were then analy zed by restriction enzyme digestion. TR containing plasmids were additionally an alyzed for the presence of TRs using a Sma I digest. Large Scale DNA Purification using Cesi um Chloride (CsCl) Gradients Plasmids with the appropriate insert and TRs were prepared on a large scale for packaging into AAV or for transfection. Pre-cu ltures were grown in 5 mL of LB medium
36 containing 0.1 mg/mL of ampicillin. They we re inoculated with single colonies and grown overnight at 30oC. Two liter flasks containing 500 mL of LB were each inoculated with 500 ÂµL of pr e-culture and 500 ÂµL of 100 mg/mL ampicillin. The cultures were allowed to grow for 16 hours at 30oC while shaking vigorousl y. All centrifugations were done in a Beckman model J2-21 centrif uge unless otherwise noted. Bacteria were spun down in 500 mL centrifuge bottles for 15 minutes at 3,020 rcf (5,000 rpm). Supernatants were discarded and pellets were resuspended by vortexing in 30 mL TE resuspension buffer (10 mM Tris-HCl, 1mM ED TA). The cell suspension was then lysed with 30 mL lysis solution (0.2 M NaOH and 1% SDS) by gently inverting bottles 5Â–6 times. Samples were then incubated at room temperature for 5 mi nutes. Subsequently, 30 mL of neutralization solution (3 M sodium acetate (NaOAc)) and 200 ÂµL chloroform were added and samples were mixed gently and thoroughly. Samples were placed on ice for 10 minutes and then centrifuged at 12,100 rcf (10,000 rpm) for 15 minutes. The supernatant was then poured through 3 layers of cheesecloth into 250 mL bottles. At this point, the volume was determined to be ~ 88 mL and 0.6 volumes (53 mL) of 2-propanol were added. Samples were mixed thoroughl y and incubated on ice for 30 minutes. Samples were then centrifuged at 12, 100 rcf (10,000 rpm) for 15 minutes and supernatants were discarded. The resulting pellet was rinsed with 20 mL of 80% EtOH and allowed to dry upside down on paper towe ls for 15Â–30 minutes. The pellet was then resuspended in 8 mL of TE. Resuspended sa mples were added to 50 mL conical tubes that contained 8.4 g of CsCl. Tubes were ge ntly rocked until all of the CsCl was in solution. Using a Pasteur pipette, samples were passed into 11.2 mL capacity OptiSealÂ™ polyallomer centrifuge tubes (Beckman). Af ter adding 150 ÂµL of 10 mg/mL EtBr, tubes
37 were sealed with their caps and centrif uged using an NVT 65 Beckman rotor at 300,000 rcf (60,000 rpm) for 12Â–18 hours in an OptimaÂ™ LE-80K ultra centrifuge (Beckman). Plasmid bands were then extracted using a Pa steur pipette and put into a 15 mL conical tube. The EtBr was then removed by extrac ting samples with 2 mL isoamyl alcohol. Phases were allowed to separate for 1Â–5 mi nutes and then the top layer, containing the EtBr, was discarded. The isoamyl extractions were carried out until the EtBr was no longer visible, about 3Â–5 times. The plasmi d DNA was then transferred to a screw cap Oak Ridge tube (Nalgene) and 2.5 volumes of water were added to dilute the CsCl. Afterwards, 2 volumes of EtOH were added. Tubes were then incubated on ice for 30 minutes and spun down at 9,800 rcf (9,000 rpm) for 15 minutes at 4oC. Supernatants were discarded, and the resulting DNA pellets were rinsed with 5 mL of 80% EtOH. Tubes were allowed to air dr y upside down for 15Â–30 minutes. Pellets were resuspended in 500 ÂµL of TE and passed to a 1.5 mL tube . DNA was precipitated by adding 50 ÂµL of 3 M NaOAc and 1 mL EtOH. DNA was pellete d by spinning in a bench top centrifuge for 5 minutes at maximum speed. Supernatant was discarded and pellet was rinsed with 1 mL 80% EtOH. Pellet was then resuspe nded in 0.5Â–1.5 mL of TE. Samples were quantitated by reading the OD260. DNA Sequencing Sequencing of DNA was car ried out by the DNA Seque ncing Core Facility, Interdisciplinary Center for Bi otechnology Research (ICBR), at th e University of Florida. DNA Packaging One cell factory with approximately 1X109 HEK 293 cells were cultured in DulbeccoÂ’s Modified EagleÂ’s Medium supplemented with 5% fetal bovine serum and Gentamycin, 40 Âµ g/mL. A CaPO4 transfection precipitation wa s prepared by mixing a 1:1
38 molar ratio of rAAV vector plasmid and a ps eudotype specific rep-cap helper plasmid. This precipitate was added to 1100mL of cDME M and the mixture was applied to the cell monolayer. The transfection was allowed to incubate at 37oC for 60 hours. The cells were then harvested and lysed, by three freeze/thaw cycles. The crude lysate was clarified by centrifugation and the resulting vector-contai ning supernatant was divided between 4 discontinuous iodixanol (Grein er) step gradients. The grad ients were centrifuged at 350,000 g for 1 hour, and 5 mL of the 60Â–40% step interface was removed and combined. This iodixanol fraction was further pu rified and concentrated by column chromatography. The iodixanol fraction was applied to a 5 mL HiTrap Q Sepharose column using a Pharmacia ATKA FPLC system. This column is a strong anion exchanger with positively charged quate rnary ammonium as the func tional group. The vector was eluted from the column using 215 mM Na Cl pH 8.0 and the AAV peak was collected. This eluted fraction was then concentrated and buffer exchanged in PBS (Cellgro) using a Biomax 100K concentrator (Millipore). The AAV was then aliquoted to be stored at -80o. Primary Neonatal Rat Ventricu lar Cardiomyocytes (PNRVC) Isolation Sprague Dawley lactating rats with 10Â–12 pups each were ordered from Charles Rivers. Animals were specified to be shipped on the same day as their birth so that cells could be prepared from pups as close to 1 day old as possible. Between 24Â–50 pups were used for each isolation procedure. CBF-HH cell isolation buffer (Table 2-3) was then prepared.
39 Table 2-3. Recipe for CBF-HH cell isolation buffer Reagent Final concentration (mM) NaCl137.00 KCl5.36 MgSO4 7H200.81 Dextrose5.55 KH2PO40.44 Hepes (pH 7.4)10.0 Na2HPO4 12H2O0.34 Na3HPO40.32 The solution was prepared in sterile water and stored at 4oC under sterile conditions. Sufficient buffer for multiple isolation procedures was prepared at one time. The entire isolation procedur e, including the preparation of the solutions used for the isolation procedure, was carried out under sterile condi tions. The tissue digestion buffer (CBF-HH/trypsin/DNase) was prepared fresh each time isolation was to be performed. First, 0.1 g of trypsin (Beckt on-Dickson) was dissolved in 20 mL of CBF-HH. This solution was then filter ster ilized using a 0.22 ÂµM por e filter in the hood, after which an additional 80 mL of sterile CBF-HH was added. Additionally, 2 mL of DNase (Ambion) were added to the solution to complete the preparation of the digestion buffer. Three, 6 cm petrie dishes each containing 2 mL PB S (140 mM NaCl, 2.7 mM KCl, 10.1 mM Na2HPO4, 1.8 mM KH2PO4) were laid on ice insi de the hood. Hearts were excised from the pups and placed in the fi rst 6 cm dish. After all of the hearts were excised, the hearts were then passed to the s econd dish as a means of rinsing the blood off. The hearts were then held by the aorta and the upper third of th e heart (atrial region) was cut off and discarded. The hearts were then passed to the third dish for a final rinsing. The PBS was aspirated off and using special scissor that function like tweezers, each heart was gently minced with about 4 clip s. Using 6 mL of the digestion buffer, all of the tissue pieces were pipetted and dispen sed into a small bottle (12 mL capacity) that had a small magnetic stir rod in it. The bottle was then wrapped in foil to maintain
40 sterility and incubated in a 37oC oven on a stir plate set on medium. Every 8 minutes, the supernatant was collected and an additional 6 mL of the digestion buffer was added. The supernatant was then kept in a 50 mL conical tube that had 5 mL of horse serum (100%) in it. The washes were repeated every 8 mi nutes until all of the tissue was digested and the 100 mL of digestion buffer was used. When the digestions were finished, the two, 50 mL conical tubes with the cel l suspensions were spun down for 6 minutes at 1500 rpm at 4oC (Eppendorf centrifuge). Both pellets were th en resuspended in a total of 6Â–12 mL of Red ADS (see recipe in Table 2-4) and s upplemented with an additional 100Â–200 ÂµL of DNase. This solution was used to overlay the Percoll (Amersham-Pharmacia) gradient. Percoll Gradient and Final Cell Preparation The Percoll gradient was comprised of two layers and was utilized to separate the cardiomyocytes from fibroblasts. Two types of ADS buffer were utilized to prepare the gradients: clear and red. Th e recipe for 10X clear ADS can be found in Table 2-4. Red ADS was prepared using 0.2g of Phenol Red diss olved in 1 L of 10X clear buffer. Both buffers were kept at 4oC under sterile cond itions. The Percoll gradient was assembled in 3 to 5, 15 mL conical tubes by first prepar ing the top and bottom la yers as detailed in Table 2-5. Table 2-4. Recipe for 10X Clear ADS buffer Reagent Final Concentration (M) NaCl1.164 Hepes0.200 NaH2PO4 H200.010 NaH2PO40.010 Glucose0.033 KCl0.053 MgSO4 7H200.008
41 Table 2-5. Composition of Percoll gradient layers Top Layer Bottom Layer Percoll 9 mL 10.4 mL 1X Clear ADS 11 mL 1X Red ADS 5.6 mL The ADS buffers utilized in the preparation of th e gradient were diluted to 1X with sterile water. FibroblastsRed ADS Clear ADS Cardiomyocytes Cell debris FibroblastsRed ADS Clear ADS Cardiomyocytes Cell debris Figure 2-3. Percoll gradient used for the purification PNRVC. The Â“topÂ” and Â“bottomÂ” layers were prepar ed independently and then combined to form the gradient. First, 4 mL of Â“topÂ” wa s added into each 15 mL conical tube. Then, 3mL of Â“bottomÂ” was added by gently introduci ng the pipette all the way to the bottom of the tube and very slowly rele asing the Percoll solution bei ng careful not to disturb the layers. When the gradients were ready, 2 mL of the cell suspensi on was gently overlaid onto the gradient. The conical tubes were then centrifuge d at 1,880 g for 30 minutes at 4oC with the acceleration and the brake set at ze ro. The resulting gradient separated the components as seen in Figure 2-3. With a Pasteur pipette, the fibroblast laye r and half of the clear ADS layer were aspirated off and discarded. The cardiomyocyte layer and half of th e red ADS layer were collected into a fresh 50 mL conical tube. Th is fraction was rinsed with 20 mL 1X red ADS to remove traces of Percoll. The sa mple was then centrifuged for 5 minutes at
42 1,500 rpm at 4oC with the brake and acceleration se t a maximum. The red ADS was aspirated off and the wash procedure was repeated. The resulting pellet was then resuspende d in 20Â–50 mL of cardiomyocyte media (DMEM/F12 (Gibco) supplemented with 0.5% In sulin transferin selenium (Gibco), 1% penicillin/streptomycin (Gibco ) and 5% newborn calf serum (G ibco)). Cell viability was assessed using a 1:1 ratio of the cell suspensi on to 0.4% Trypan blue (Gibco). Viable cells appeared white under the microscope a nd were counted using a hemacytometer. Preparation of Tissue Culture Vessels Tissue culture vessels used for the plating of PNRVC we re coated using aseptic techniques with either Collage n Type I (Sigma) diluted 1:10 with water or 0.1% Gelatin Type A from Porcine skin (Sigma) prepared with water and autoclaved. Enough of the coating material was added to coat the bottom of the dish with a thin layer. Vessels were incubated at room temperature for 1Â–4 hours. The coating reagent was then aspirated off and slightly ajar vessels were allowed to dr y in the hood for at least 1 hour before cells were plated on to them. Cells were then plated using 6.5Â–8.5x105 cells per well in a 6 well pl ate, 4.0Â–8.0x104 per chamber in an 8 chamber slide, 2.8Â–3.0x105 per chamber in a 4 chamber slide and 1.0Â–2.0x107 in a 6 cm plate. Mitomycin-C Cell Treatment Antibiotic treatment with Mitomycin-C (Mit-C) followed 18Â–24 hours after plating of the cells. This treatment prevents the proliferation of dividing cells, such as fibroblasts, from overtaking the cardiomyocyt e cultures. Treatments were done by first rinsing the primary cell cultures one time w ith cardiomyocyte media. Cells were then incubated for 3Â–3.5 hours with a final concen tration of 1 Âµg/mL Mit-C diluted in
43 cardiomyocyte media. After the incubation, cells were rinsed 2Â–3 times with PBS to ensure complete removal of Mit-C. Fresh media was replenished and cells were allowed to recover until the next day. Primary Adult Rat Ventricular Cardiomyocytes Isolation Adult Sprague Dawley rats (Charles Rivers) were euthanized with an intraperitoneal inject ion of 0.5 cc of Pentoba rbital (50 mg/mL). The heart was then excised by cutting just above the thymus a nd then suspended through the aorta in a Langendorf perfusion apparatus. Perfusion of the heart was then carried out using Ca2+-free Krebs buffer (590 mM NaCl, 23 mM KCl, 6 mM KH2PO4, 125 mM NaHCO3, 12 mM glucose) until all of the blood had washed out (about 2Â–3 minutes). Subsequently, the perfusion was then perfor med for 20 minutes with Enzyme 1 solution (30 mL of Ca2+-free Krebs buffer supplemented with 0.5 mg/mL each of Collagenase [Worthington] and Hyaluronidase [Sigma]) at a rate of 1 drop/second. Subsequently, the heart was removed from the perfusion apparatus and caref ully cut into very thin longitudinal strips into a 100 mL container containing 20 mL of Enzyme 2 solution (20 mL of Enzyme 1 supplemented with 0.5 mg/m L each Collagenase and Hyaluronidase and 0.023 mg/mL each of Trypsin [Sigma] and DNaseI [Worthington]). Tissue samples were then incubated at 37 C shaking at 150 rpm for 18 minut es with the enzyme solution. A nylon mesh filter was then fitted into a 50 mL co nical tube. First, 10 mL of wash buffer (1:1 mixture of Ca2+-free Krebs buffer with ACCT medium containing DulbeccoÂ’s phosphate buffered saline, including 25 mmol/l HEPES and 25 mmol/L NaHCO3 with L-glutamine, 2 mg/mL BSA, 2 mmol/L L-carni tine, 5 mmol/L creatine, 5 mmol/L taurine with 100 IU/mL penicillin and 100 Âµg/mL stre ptomycin [GIBCO]) was filtered through,
44 and, then, the tissue suspension was pipetted through the filter into the tube. The tube was then centrifuged at 50 g for 90 seconds. The supernatant was then aspirated off to the 5 mL level demarcated on the tube. Sl owly, 10 mL of the wash buffer was then added to the tube and the cells were resusp ended by gentle pipetting. The cells were allowed to settle for 5Â–8 minutes, at which time the supernatant was aspirated off in the same manner and the wash procedure was repeated once more. After removing the supernatant from the second wash, the cell suspension was then transferred to a BSA solution (20 mL ACCT media supplemented with 1.29 mg of BSA). The cells were then allowed to settle for 6Â–8 minutes at which time the supernatant was aspirated off. The cell pellet was then resuspended in ACCT media and the suspension was plated onto tissue culture plates that had been coat ed with Laminin (Invitrogen) (0.60 mg/mL prepared in ACCT media). Infection and Incubation Infections of adult rat ventricular cardiomyocytes were carried out using 1x109 vg/cell of AAV1 containing the shRNAs target ing PLN. Cells were co-infected with Adenovirus using 1 viral particle/cell. Br iefly, immediately followi ng cell isolation, the corresponding AAV and Ad viruses were added dire ctly to the media in the culture dish. Cells were allowed to incubate for 48 hour s at which time calcium transients were measured. Measurements of Intracellular Calcium Measurements of intracellular calcium [Ca2+]i were performed in AAV infected myocytes isolated from adult Sprague Dawley rats. Myocardial cells were loaded with the Ca2+ indicator Fura-2 by incubating th e cells in medium containing 3 Âµ mol/L Fura-2-AM (Molecular Probes) for 15 min. Th e cells were then washed with Krebs
45 Solution (119.1 mM NaCl, 4.7 mM KCl, 1.9 mM MgSO4, 1.2 mM KH2PO4, 25 mM NaHCO3, 11.5 mM Glucose, 0.099 mM Ascorbate) and allowed to equilibrate for 15 min in a light-sealed, temperature controlled chamber (25 C) mounted on an inverted microscope (Olympus). Cells were exposed to 1mM Ca2+ at 1 Hz stimulation rate using an external stimulator (IONOPTIX). A dua l excitation spectrofluorometer (IONOPTIX) was used to record fluorescence emissions (505 nm) elicited from exciting wavelengths 360 nm and 380 nm. [Ca2+]i was calculated according to the formula [Ca2+]i=Kd (R-Rmin)/Rmax-R)B, where R is the ratio of th e fluorescence of the cell at 360 and 380 nm; Rmax and Rmin represent the ratios of Fura-2 fluorescence in the presence of saturating amounts of calcium and effectivel y Â“zeroÂ” calcium respectively, Kd is the dissociation constant of Ca2+ from Fura-2, and B is the rati o of fluorescence of Fura-2 at 380 nm in zero Ca2+ and saturating amounts of Ca2+. The myocytes were imaged using a CCD video camera (IONOPTIX) attached to th e microscope and motion along a selected rastor line segment was quantified by a vide o motion detector system (IONOPTIX). Time course and peak amplitude of the calcium tr ansient from AAV infected and non-infected myocytes were compared. RNA Techniques RNA Extraction and Preparatio n from Mammalian Cells RNA was extracted from mammalian HE K 293 cells or primary PNRVC 48 hours post-transfection/infection using either the GenElute Mammalian Total RNA Miniprep Kit (Sigma) or the RNAqueousÂ®-Micro K it (Ambion) following the manufacturersÂ’ protocols. Samples were eluted off of th e respective columns in 15Â–45 ÂµL of elution buffer. RNA was then treated with DNase I (DNA free [Ambion]) following the
46 manufacturerÂ’s protocol. RNA sample s were quantitated by reading the OD260. The purity was determined using the OD260/OD280 ratio. RNA Extraction from Cardiac Tissue RNA extraction from cardiac tissue was carried out using a PT 2100 homogenizer (Polytron) with a 5 mm generator in Trizol reagent (Invitrogen). One half of a mouse heart cut through the septum wa s minced with a razor blade and added to a 1.5 mL tube containing 1 mL of Trizol. The sample was homogenized on the highest setting cooling the sample by intermittently pl acing it on ice. Subsequently, 200 ÂµL of chloroform was added. Samples were vortexed briefly. Following a 2 minute incubation at room temperature, samples were centrifuged in a desktop centrifuge at 9,500 rpm for 15 minutes at 4oC. The top layer was extracted and passed to a 1.5 mL tube. Precipitation was then carried out using 500 ÂµL of 2-propanol. Sample s were vortexed briefly and incubated at room temperature for 10 minut es. Centrifugation was then performed at 9,500 rpm for 10 minutes at 4oC, after which the 2-propanol was removed. RNA pellets were rinsed in 1 mL of 75% EtOH and spun down at 8,000 rpm for 5 minutes at 4oC. The EtOH was removed and a final quick spin was performed for the complete removal of the EtOH. RNA pellets were resuspende d in 50Â–75 ÂµL of TE or diethyl pyrocarbonate (DEPC) treated water. Samples were DNase treated with DNA free (Ambion) following the manufacturerÂ’s protocol a nd quantitated by reading the OD260 and OD280. Reverse Transcriptase (RT)-PCR RT-PCR was utilized to quantitate the am ount of PLN mRNA by st andardizing it to the amount of GAPDH mRNA. Reactions were carried out for both genes. For the RT reaction, the 1st strand cDNA synthesis kit (AmershamPharmacia) was used. Reactions were carried out in a 15 ÂµL volume. Briefl y, 1 Âµg of RNA was diluted in a final volume
47 of 8 ÂµL using water and incubated at 65oC for 10 minutes. Subsequently, 1 ÂµL DTT, 1 ÂµL 15 pmol/ÂµL antisense primer and 5 ÂµL enzyme mix was added to the sample and then incubated at 37oC for 1 hour. One 15 ÂµL RT reaction was then used for 3 PCR reactions as follows. The PCR reactions were carried out in a 50 ÂµL final volume containing a final concentration of 3.3 pM sense and anti sense primers (PLN 400 bp in Table 2-2), 2 mM MgCl2 (Promega), PCR buffer (50 mM KCl, 10 mM Tris-HCl pH 9.0) (Promega), 2.5 mM dNTPs and 5U of Taq polymerase (Promega) following the conditions in Table 2-2. Sybr Green Staining In order to visualize and quantitate the final RT-PCR products, SYBR Green I (Invitrogen) staining wa s employed. Briefly, 7% polyacrylamide (19:1, acrylamide:bis-acrylamide) gels were prepared using 8M urea. Approximately 5Â–7 ÂµL of PCR product were loaded on gels and run for a bout 2 hours taking care not to let the gels warm up more than 37 C. Gels were then staine d using a 1:10,000 dilution of SYBR green in TBE (10 mM EDTA, 0.45 M Tris, 0.45 M Boric Acid) for 10 minutes gently shaking. Phosphorimager Analysis Wet gels were placed in between two tr ansparencies and then scanned using the Storm860 Phosphorimager (Molecular Dymanics ) with the fluorescen ce mode set at 450 nm. Analysis of the scanned bands was carri ed out using the ImageQuant software 1D gel program (Amersham) using the rubber band method option in the software to subtract background. PLN mRNA levels were e xpressed as a ratio with GAPDH mRNA.
48 Protein Techniques Protein Extraction from Cells Mammalian cells were rinsed one time with PBS to remove residual media. Cells were treated with tryps in for 3Â–5 minutes. Cells were th en collected in a 15 mL conical tube and spun down for 5 minutes at 640 g in a Beckman J6-HC centrifuge. The resulting supernatant was discarded. The pe llet was resuspended in 1 mL PBS to rinse away the trypsin and transferred to a 1.5 mL tube. The cells were then spun down in a desktop centrifuge for 5 minutes at 600 g. The PBS wash was discarded and the cell pellet was resuspended in 67 ÂµL PBS. A 1 ÂµL aliquot of two proteas e inhibitor cocktails prepared in either water or DMSO (100X st ock) were then added to each sample. The protease inhibitor cocktails (PIC) were prepared as follows. The stock DMSO cocktail contained 0.5 M PMSF, 5 mg/mL Peps tatin A, 1 mg/mL Chymostatin and 950 ÂµL DMSO. The stock PIC prepared in wate r contained 1 M t-aminocaproic acid, 1 M p-aminobenzamidine, 1 mg/mL Leupeptin, 2mg/mL Aprotinin and 950 ÂµL water. Both stocks were diluted 100-fold in the final solutions. Samples were then sonicated for 30 seconds using a Vibra Cell (Sonics & Materials Inc.) with the amplitude set at 30 and the pul ser set for 3 seconds. When sonication was completed, 33 ÂµL of 3-fold concentrated Laemmli buffer (0.167 M Tris pH 6.8, 6.6% SDS, 0.007 M EDTA, 0.03% Brom ophenol blue, 33% glycerol ) was then added to the samples. Protein Extraction from Tissue Mice hearts were cut in half along the septum and then minced with a razor bade and added to a 1.5 mL tube containing 750 ÂµL PBS supplemented with 7.5 ÂµL each PIC-W and PIC-D. The samples were homogenized using the PT 2100 homogenizer
49 (Polytron) set at the highest setting taki ng care not to let the samples warm up by intermittently setting them on ice. The samples were then sonicated as described above to sheer the chromosomal DNA. Additionall y, 250 ÂµL of the 3-fold concentrated Laemmli buffer was added to each sample. Protein Quantitation Protein quantitation was carri ed out using the DC Protein Assay (Biorad) on a 96 well plate. Samples were each measured in tr iplicate and the values were then averaged in order to ensure the most accurate readi ng possible. A standard curve was prepared with 5 samples of BSA (Promega) ranging fr om 0.125 Â– 2.0 mg/mL. Linear regression on Microsoft Excel was utilized to find th e final concentration of the samples. Western Blot Analysis Porzio gels. These gels resulted in a more effective separation resolution of myofibrillar proteins. Porzio gels 96 with an 8% final concentr ation of acrylamide (30% Acrylamide/1.5% Bis [20:1]) were utilized to separate protein fragments. Porzio gel buffer (2-fold concentrated) was made us ing 0.2 M Tris base, 0.6 M Glycine, 0.12 mM EDTA, 0.2 % SDS, 10 % glycerol. A standa rd 8 % Porzio gel consisted of 17.5 mL Porzio buffer, 9.35 mL acrylamide, 7.93 mL water, 58 ÂµL TEMED, 233 ÂµL 10 % APS. Using 18 x 16 cm glass plates (Hoefer) fitted with either 1.0 or 1.5 mm spacers, gels were poured and allowed to polymerize for at leas t 30 minutes. Porzio running buffer (4X) was utilized to run the gel. This buffe r contained 0.21 M Tris base, 0.60 M glycine and 0.4 % SDS. The top reservoir of the apparatus was filled with 1X Porzio running buffer while the bottom was filled with one half con centration of the running buffer. The gels were run at about 300 volts for 2 hours. A pre-stained molecular weight marker, SeeBlueÂ® Plus2 (Invitrogen) was used to track the migration of the proteins in the gel as
50 well as the efficiency of the tr ansfer. Between 10 Â– 60 Âµg of protein were loaded for each sample. Laemli gels. These gels were used for sample s not containing PLN or as a pre-gel to the Porzio gels. Precast 1 mm, 12% Tris-H Cl gels (Bio-Rad) were also utilized for protein electrophoresis. The running buffer contained 0.2 M Tris, 1.9 M Glycine, and 0.03 M SDS. Gels were loaded and run as described above. Transfer to Nitrocellulose Membrane Proteins were transferred to ProtranÂ® BA79, 0.1 Âµm Nitrocellulose membranes (WhatmanÂ® Schleicher & Schuell). Th e transfer buffer was prepared in H20 and contained 42 mM sodium phosphate dibasi c, anhydrous (Fisher) and 7mM sodium phosphate monobasic, anhydrous (Sigma). For 1 mm thick gels, transfers were conducted overnight using 0.3 amps or 2 am ps for 90 minutes. Once transfer was completed, efficiency of transfer was conf irmed by staining the membrane with 0.1 % Naphtol Blue/Black diluted 1:10 with water for 2Â–5 minutes. Destai ning was carried out in water. Membranes were cut at this time with a scalpel if necessary. Blocking of Membrane and App lication of Antibodies Membranes were blocked for 30Â–60 minutes with 5 % milk (Carnation) prepared with PBS containing 0.1 % TweenÂ® 20 (U nited States Biochemical Corp.) (PBS/Tween/milk). Primary antibodies were diluted in PBS/Tween/milk as described in Table 2-6 for 1 hour. Membranes were wa shed three times with PBS/Tween for 5 minutes. The membrane was then incubate d with secondary antibodies, diluted in PBS/Tween/milk described in Table 2-6 for 1 hour at room temperature.
51 Table 2-6. Antibodies used for Western blotting Antigen Vendor Dilution -PLN (monoclonal) Upstate 1:3000 -MyBP-C Mathias Gautel 1:2000 -Actin Sigma 1:5000 actin Sigma 1:5000 -mouse Amersham 1:5000 -rabbit Caltag 1:5000 Chemiluminescence and Film Development Both secondary antibodies used were c onjugated to Horseradish peroxidase and membranes were incubated with 2Â–5 mL of ECL plus Western Blotting Detection System (Amersham Biosciences) according to the ma nufacturerÂ’s protocol. Membranes were then exposed for the appropriate amount of time to X-ray film (RPI) for autoradiography/chemiluminescence. Immunofluorescence Primary cardiomyocytes were seeded on 4 well glass chamber slides (Lab-TekÂ® II Chamber SlideÂ™ System from Nalgene Nunc In ternational) that had been coated with collagen or gelatin as described above. Infections were carri ed out as described above. Two days post-infection, cells were rinsed one time with PBS. Cells were then fixed with 4% paraformaldehyde diluted in PBS for 15 minutes. Slides were rinsed 3 times in Copland jars filled with PBS for 5 minutes per ri nse. Slides were then either stored in fresh PBS for a maximum of 5 days or direct ly used. Cells were permeabilized using 1 mL of 0.5 % TritonX-100 for 5 minutes. Slides were then washed with PBS 3 times in Copland jars (5 minutes per wash). Upst ate PLN antibody was used at a 1:1000 dilution in combination with a myocyte marker, My BP-C at a 1:200 dilution. Slides were incubated with antibodies for 1 hour and then rinsed 3 times with PBS. Secondary antibodies were then added. A Cy-5 conjugated goat -rabbit antibody (Jackson Immuno
52 Research) at a 1:100 dilution was used to detect the MyBP-C. Similarly, an Alexa FluorÂ® 568 conjugated goat -mouse antibody (Molecular Pr obes) at a 1:200 dilution was used to detect PLN. Slides were incuba ted with antibodies for 1 hour and then rinsed 3 times with PBS. Slides were then m ounted with VectashieldÂ® mounting medium containing DAPI (Vector Labs). Microscopy A Zeiss Axiovert 200M microscope was used to obtain fluorescent images. Images of fixed cells were taken using 20X, 0.75 NA objectives. The Zeiss AxioVision image acquisition software was utilized to obtain four-color images. Co-Transfection of HEK 293 Cells Co-Transfection with Double-Stranded RNA Oligos For the first set of co-transfection ex periments, the two murine PLN specific siRNAs (si248 and si750) were used. As a c ontrol, an siRNA targeting GFP (siGFP) was also used. Approximately 5x105 HEK 293 cells were seeded onto NUNC (Denmark), 6 well tissue culture plates. On the followi ng day, cells were about 95% confluent and transfected with Lipofectamine 2000 (Invitrogen) as follows. Briefly, 50 nM of each of the siRNA duplexes were combined with either 0.13 nM or 0.25 nM of the PLN expressing plasmid. The DNA/RNA combinati on was diluted to a final volume of 250 ÂµL with OptiMem serum free media (Gibco). In a separate tube, a master mix was prepared to contain the nece ssary amount of transfection re agents for 6 reactions. For each transfection, 3 ÂµL of Lipofectamine 2000 was diluted with 247 ÂµL of OptiMem Transfections were then carried out accord ing to the manufacturerÂ’s recommendations. The cells were rinsed once with 1mL of OptiM em before adding the transfection mixture. Once the transfection mixture was added, the cell s were allowed to incubate for 4 hours.
53 The transfection media was then removed and 2mL of fresh growth media (DMEM supplemented with 10% Fetal Bovine Serum an d 1% Pennicillin/Streptomycin) was then added. Co-Transfection with shRNAs Ex pressed from AAV TR Plasmids Approximately 5x105 HEK 293 cells were seeded on 6 well NUNC plates. For this experiment, 0.25 pmol (1 Âµg) of the PLN expressing plasmid was combined with 1.0 pmol of the plasmid expressing either of th e murine PLN specific short hairpins, shM248 or shM750, or a short hairpin ta rgeting Rhodopsin as a control. Two sets of short hairpin AAV plasmids were utilized Â– either the tr aditional single stranded AAV (5.1 Âµg) or the self-complementary AAV (5.0 Âµg). The co-tra nsfected plasmids were combined in a 1:4 molar ratio (PLN:shRNA). The transfection procedure was carried out as previously described. The amount of Lipofectamine 2000 used for each transfection was 5 ÂµL. Animal Procedures Animal Selection and Ordering Suitable animals were selected based on thei r heartiness as a resu lt of out breeding as well as large litter size and reasonable price. CD-1 mice from Charles Rivers were selected because they suited all of our needs in these regards. An imals were ordered to be shipped on the same day as their birth. L itters of at least 10 a nd up to 14 animals were available with the runt rarely dying. Mo thers were good caretake rs and as a result, animals were never lost because of her negligence. On the day before a scheduled injection, animals were anesthetized on ice. A 28 gauge 0.5Â” needle (Becton Dickinson) was us ed to inject diluted India ink into the footpads of the mice. Different marking patt erns were used to di fferentiate different
54 injection cohorts. Booster inj ections with additional ink were administered in subsequent days if fading was observed. Injections and Incubations Subxiphoid injections. On the 4th day of life, CD-1 mice were anesthetized on ice until movement was barely visible. Viru s aliquots were prepared to contain 2.5x1011 total particles of AA V1 or scAAV1 in a 50 ÂµL final volume of PBS. Animals were injected using a 250 ÂµL luer-tip syringe (Hamilton) fitted with a 33-gauge metal hub needle (Hamilton) with tygon tubing around it to expose only 2 mm of the tip. The needle was inserted underneath the lowest ri b to the right of the xiphoid process at an approximately 15o angle. The virus was slowly deliv ered into the pericardial sac. Successful injections resulted in no bulges underneath the skin and no blood return after the removal of the needle. Animals were allowed to incubate for 2Â–4 weeks before organ harvesting. Temporal vein injections. One to two day old CD-1 mice were anesthetized in a beaker of ice water carefully maintaining th e animalÂ’s head just above water. Once visible movement ceased, animals were injected into the left temporal vein with 50 ÂµL of 5x1011 viral particles of AAV1 using a 28-gauge needle. The tip of the needle was positioned to point towards the tip of the animalÂ’s nose and the virus was slowly released. In the same manner, a consecutive injecti on was performed on the following day into the contra-lateral side with 5x1010 viral particles for a total of 1x1011 total particles per animal. Successful injections resulted in a brief clearing of the blood in the immediate vasculature and blood return to th e area upon removal of the needle.
55 Organ Harvesting and Freezing Animals were euthanized with 0.05 Â– 0.1cc of 50 mg/mL Pentobarbital (Abbott Laboratories). Chests were opened and the hear ts were extracted. The hearts were then briefly rinsed in PBS before slicing into thir ds with a sharp razor blade. The resulting donuts were then used for the isolation of RNA and protein or fixed for indirect immunofluorescence. For fixing tissue, samples were placed in plastic cassettes with lids (CMS Tissue Path) that were submerged overnight in 80 mL graduated formalin containers (Fisher) which cont ained 40 mL of 4% paraformal dehyde prepared in PBS. Organs were then rinsed 2 times for 15 mi nutes in 50 mL of PBS. Subsequently, cassettes with hearts in them were incubate d in containers that contained 20% sucrose prepared in PBS for 12Â–24 hours. The next day, heart segments were frozen down in cryomolds (Tissuetek) containing OCT us ing 2-Methylbutane (Sigma) and liquid Nitrogen. Histology Indirect immunofluorescence. Five micron cardiac cross sections were cut from frozen blocks using a Leica CM1850 cryostat, and thaw mounted onto Superfrost plus slides. Slides were permitted to air dry overnight to improve tissue adhesion. Following a five minute submersion in -20oC acetone, tissue sections we re once again air dried. Slides were placed in 0.1 mM sodium citrat e buffer, pH 6.0 and antigen retrieved in a microwave oven for a total of 25 minutes. Af ter rinsing in water and equilibrating in TBS-T buffer (DAKOCytomation), the secti ons were blocked for endogenous biotin using a kit (Vector Labs). Slides were additionally blocked with normal horse serum (diluted at 1:75) for 20 minutes at room temp erature. Excess block was blotted from the slides and 1:1000 rabbit anti-GFP (Abcam ) was applied for an overnight 4oC incubation.
56 Following a 5 minute wash with TBS-T at room temperature, mouse anti-PLN antibody (Upstate Signaling) was applied at a 1:100 dilution for 20 minutes at room temperature. To avoid reactivity of anti-mouse secondary antibody with endogenous mouse immunoglobulin, the PLN antibody was fi rst biotinylated us ing the ARKÂ™ kit (DAKOCytomation). Briefly, mouse primary antibody (PLN) was labeled with biotinylated anti-mouse immunoglobulin be fore adding a blocking reagent containing normal mouse serum. After the antibody incubation, slides were washed for 5 minutes with TBS-T at room temperature. The fluor escent secondary anti bodies were combined, placed on the slides and incubated in the dark for one hour at room temperature. PLN was detected using Streptavidin-Alexa 594 (M olecular Probes) at a 1:200 dilution. GFP was detected with donkey rabbit Alexa Fluor 488, diluted 1:200. Sections were then washed 3 times for 3 minutes with TBS-T at room temperature. Subsequently, cover slips were mounted using VectashieldÂ® mounting medium with DAPI (Vector). Staining was documented on an Olympus BX51 microscope equipped with an Optronics digital camera and Magnafire software. Direct immunofluorescence of GFP. Tissues analyzed exclusively for GFP expression were prepared by cutting 5 Âµm cr oss sections from frozen blocks using a Leica cryostat. Samples were thaw mounted onto Superfrost plus slides and mounted using VectashieldÂ® mounting medium with DAPI (Vector). Staining was documented on an Olympus BX51 microscope equipped with an Optronics digital camera and Magnafire software. -Galactosidase immunohistochemistry. Briefly, 4 Âµm cryosections were thawed and post-fixed in 4% paraformaldehyde at room temperature (RT) for 15 minutes.
57 Permeabilization was accomplished using 0.25% TritonX prepared in 1X PBS for 5 minutes at RT. Slides were subsequently rinsed in 1X PBS and incubated with 10% normal goat serum (DakoCytomation) for 20 minutes at RT. The sections were blocked with avidin for 15 minutes, followed by biotin for 15 minutes (Vector La boratories Inc.). Slides were rinsed in 1X TBS followed by addition of rabbit anti-Galactosidase (Rockland Immunochemicals, Inc.) for 1 hour at RT. After sections were rinsed in1X TBS, the rabbit ABC-AP (Vector Laboratori es Inc., Burlingame, CA) secondary was applied for 30 minutes at RT. Sections were subsequently incubated in the rabbit-ABCAP tertiary complex for 30 minutes at RT . The chromogenic reaction was performed using Vector Blue (Vector Laboratories Inc.) for 5 minutes at RT. Slides were mounted and coverslipped with aquam ount (Lerner Laboratories).
58 CHAPTER 3 RESULTS The greater danger for most of us is not t hat our aim is too high and we miss it, but that it is too lo w and we reach it. Â—Michelangelo (1475Â–1564) Expression of Phospholamban In an attempt to reduce the levels of PL N in cardiac tissue, se veral intermediate steps were taken to approach this goal syst ematically. Initially, the entire PLN sequence was amplified from murine cardiac tissue and cloned into an expression vector. Due to the lack of cell lines expre ssing this gene, we proceeded to express PLN in human embryonic kidney 293 cells (HEK 293) using transf ection techniques. As seen in Figure 3-1, high levels of the 25 kDa pentameric a nd 6 kDa monomeric forms of PLN were able to be detected by Western blot analysis in transfected HEK 293 cells. Due to the strength of the CBA promoter in driving the expression of PLN, low levels of this protein could be detected as early as one day with much higher levels appearing after two days. Expression was also detected at 4 days post-transfection. One characteristic of this protein is its ability to disassemble when heated 17. As seen in Figure 3-2, boiling a pr otein homogenate from a rat heart in 12% SDS resulted in the disassembly of the pentameric form of PLN. However, maintaining the samples on ice preserved the pentameric form.
59 D a y 1 D ay 2 D ay 3 D ay 4 +D a y 1 D ay 2 D ay 3 D ay 4 ++Figure 3-1. Immunoblot showi ng a time course of PLN e xpression in transfected HEK 293 cells. HEK 293 cells were transfected with 5 Âµg of the PLN expressing plasmid. Cells were harvested and protei n lysates were prepared for days 1Â–4. A primary neonatal cardiomyocyte cell lysate was used as a positive control (+) and non-transfected HEK 293 cells were used as a negative control (-). Both the PLN monomer and pentamer ar e visible as early as one day with higher levels of expression by day 2 and out to day 4. 28 kDa 14 kDa 17 kDa 6 kDaBoil -+ 28 kDa 14 kDa 17 kDa 6 kDaBoil -+ 28 kDa 14 kDa 17 kDa 6 kDaBoil -+ Figure 3-2. Immunoblot of PLN samples boile d vs. unboiled. Protein homogenate from a rat heart was either boil ed (+) or left unboiled (-) and then loaded on a 10Â– 20% gradient Laemmli gel. Boiling caused the disassociation of the 25 kDa pentamer into the 6 kDa monomer. Double-Stranded RNA Co-Transfections With an assay in hand, I next attempted to knockdown phospholamban expression by using two specific siRNAs, si248 and si750. For these experiments, differing amounts of PLN target plasmid and siRNAs were co-t ransfected into HEK 293 cells and incubated for 48 hours, after which the RNA was extracted. The first experiment involved cotransfecting 0.5 Âµg (0.13 nM) of the PLN expre ssing plasmid with 50 nM of either of the
60 siRNAs targeting PLN or an irrelevant control siRNA targeting the non-endogenous gene, GFP. Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* ** p<0.05Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* *Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* ** p<0.05Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* ** p<0.05Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* *Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* ** p<0.05Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* ** p<0.05Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* ** p<0.05Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* *Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* *Standardized PLN mRNA levels 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH 0 0.2 0.4 0.6 0.8 1 1.2 1.4si248si750siGFPmRNA Ratio of PLN:GAPDH Non-standardized PLN mRNA levels 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units 02000000 4000000 6000000 8000000 10000000 12000000si248si750siGFPPI units A.) B.)* ** p<0.05 A B Figure 3-3. Co-transfection of HEK 293 cells with 50 nM of siRNA and 0.5 Âµg of PLN expressing plasmid. (A) RT-PCR anal ysis of PLN mRNA levels after co-transfection of cells with siRNAs ta rgeting either PLN, si248 and si750, or GFP, siGFP. Values are non-standard ized and expressed as Phosphorimager units (PI units). Three PCR reactions were run per experiment. (B) Normalized PLN values. For each replicate, the PLN PI units were normalized to GAPDH and the ratios were averaged and compared to the control, siGFP. *p<0.05 vs. siGFP trea ted group. Data are mean Â± S.D. After RNA extraction and RT-PCR amplif ication of the PLN mRNA, the PLN mRNA levels were quantitated either without standardization (Fi gure 3-3 A) or with standardization to GAPDH levels (Figure 3-3 B). As seen in Fi gure 3-3, si248 and si750 were both successful at significantly reducing th e mRNA levels of PLN, as compared to control siRNA. Without standardization, si 248 and si750 were determined to decrease
61 the amount of PLN mRNA by 47% and 29%, re spectively. Standardizing to GAPDH levels resulted in a similar decrease with a 55% and 34% decreas e seen with si248 and si750, respectively. These values were calculat ed to be statistica lly significant with a p-value < 0.05. 0 500000 1000000 1500000 2000000 2500000 3000000 3500000 4000000 4500000 5000000si248si750siGFPPI unitsNon-standardized PLN mRNA levels A.) 0 0.5 1 1.5 2si248si750 siGFPPLN:GAPDH mRNAStandardized PLN mRNA levels B.)* ** p<0.01 0 500000 1000000 1500000 2000000 2500000 3000000 3500000 4000000 4500000 5000000si248si750siGFPPI unitsNon-standardized PLN mRNA levels A.) 0 500000 1000000 1500000 2000000 2500000 3000000 3500000 4000000 4500000 5000000si248si750siGFPPI unitsNon-standardized PLN mRNA levels A.) 0 0.5 1 1.5 2si248si750 siGFPPLN:GAPDH mRNAStandardized PLN mRNA levels B.)* ** p<0.01 0 0.5 1 1.5 2si248si750 siGFPPLN:GAPDH mRNAStandardized PLN mRNA levels B.)* ** p<0.01 A B Figure 3-4. Co-transfection of HEK 293 cells with 50 nM of siRNA and 1.0 Âµg of PLN expressing plasmid. (A) RT-PCR anal ysis of PLN mRNA levels after transfection of cells with siRNAs targeting either PLN, si248 and si750, or GFP, siGFP. Values are non-standard ized and expressed as Phosphorimager units (PI units). Three PCR reactions were run per experiment. (B) Normalized PLN values. For each replicate, the PLN PI units were normalized to GAPDH and the ratios were averaged and compared to the control, siGFP. *p<0.01 vs. siGFP trea ted group. Data are mean Â± S.D. I then wanted to see whether increasing th e amount of PLN target and maintaining the concentration of the siRNAs constant would result in a d ecrease in the efficiency of the siRNAsÂ’ activity. For this experiment , I doubled the amount of PLN target and
62 retained the siRNA concentrati on at 50 nM. As can be seen by Figure 3-4, increasing the amount of target to 1.0 Âµg (0.25 nM) did not decrease the efficacy of the siRNAs. Without standardization, si248 and si750 were still able to reduc e the amount of PLN message by 51% and 31% respectively. St andardizing these values to endogenous GAPDH levels did not affect these results . The amount of re duction using si248 and si750 after standardization was calculated to be 54% and 29%, respectively with a statistically significant p-va lue < 0.01 (Figure 3-4 B). Because the activity of the siRNAs did not seem to be affected with increasing amounts of target at the RNA level, I tested whether a further increase in target would lead to reduced protein expres sion as a consequence of RNAi. In this experiment, I s i G F P s i 7 5 0 C a r d i a c e x t r a c t s i 2 4 8ÃŸ-actin PLNs i G F P s i 7 5 0 C a r d i a c e x t r a c t s i 2 4 8ÃŸ-actin PLN Figure 3-5. Co-transfection of HEK 293 cells with 28 nM of siRNA and 4.0 Âµg of PLN expressing plasmid. Immunoblot analys is of PLN protein levels after transfection of cells with siRNAs targeting either PLN, si248 and si750, or GFP, siGFP. Murine cardiac extrac t was used as a positive control. increased the amount of PLN target co-transf ected into HEK 293 cells up to 4 Âµg (1 nM) and additionally decreased the amount of the siRNAs added to 28 nM. Following
63 co-transfection of the siRNA and the PLN plasmid, proteins were extracted, and an immunoblot was performed fo r both PLN (25 kDa) and -actin (43 kDa), a loading control. Murine heart extract expres sing actin, and not -actin, was used as a positive control. As seen in Figure 3-5, both si248 and si750 successfully reduced the amount of PLN protein 48 hours post-transfection. The control siRNA targeting GFP did not cause a decrease in the amount of PLN. Great er than 85% knock-down was observed using both of these siRNAs. Co-Transfections of shRNAs in AAV Expression Plasmids Having successfully designed two very activ e siRNAs, I wanted to devise a method with which to express these molecules in ce lls from a plasmid based system in order to obtain long-term expression and silencing. In this system, the siRNA would be cloned into the plasmids as a shRNA-expressi ng DNA molecule, which upon transcription and processing by the enzyme Dicer would ge nerate siRNAs. Because AAV has been extensively studied by many laboratories and ha s been found to have distinct advantages as described in Chapter 1, I decided to e ngineer our shRNAs to be expressed in AAV constructs. Therefore, for this part of the project, I created 2 type s of constructs. The first would be packaged into the traditional single-stranded (ss) AAV. This vector would contain the shRNA cassette including the HI promoter driving th e expression of the shRNA and additionally it w ould include the chicken -actin promoter to express GFP as a reporter gene (Figure 3-6 A). The second set of constructs would be packaged as self-complementary (sc) versions inside AAV. These constructs, (Figure 3-6 B) would contain the exact same shRNA cassette and G FP reporter gene except that the GFP would be expressed from a CMV promoter. These two vector systems used are expected to behave similarly upon transfecti on. However, when the plasmids are packaged into AAV
64 M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalIA.) GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R B.) M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalIA.) M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalI M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalIA.) GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R B.) GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R B.) A B M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalIA.) GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R B.) M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalIA.) M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalI M siRNA +GFP AmpR ColE1 ori f1(+) origin TR TR GFPh CMV ieenhancer Intron HI promoter + sh M 248 or sh M 750 SV40 poly(A) Chicken ÃŸ-actinpromoter Exon1 SalI SalI ssAAV M siRNA +GFP6302bp ) SalIA.) GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R B.) GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R GFP SV40 SD/SA TR delta TR HI promoter + sh M 248 or sh M 750 SV40 polyA CMVie SalI SalI scAAVM siRNA +GFP6120bp Amp-R B.) A B Figure 3-6. The AAV expression plasmids creat ed to express shRNAs. A) Traditional single-stranded AAV plasmid. B) Self-complementary AAV plasmid. and delivered using the virus, the onset of gene expre ssion from self-complementary should be more rapid, thereby enhancing the e ffect of the shRNA and reducing the target
65 sooner (see Appendix B for comparisons between the self-complementary vs. single-stranded AAV vector systems). Upon generating our PLN shRNA expressing plasmids, I tested their activities in cell culture. An irrelevant sh RNA targeting rhodopsin was used as a control, since this protein is not expressed in HEK 293 cells. This shRNA was expressed using only the ssAAV vector system. Cells were co-transf ected with the PLN expressing plasmid as well as the shRNA-expressing plasmids. Co-tra nsfections were performed at a 1:4 molar ratio of PLN plasmid to shRNA plasmid and incubated for 48 hours. Following incubation, RNA was extracted a nd protein lysates were also prepared. Levels of PLN mRNA were standardized to levels of GAPDH. As can be seen in Figure 3-7, both the 0 0.5 1 1.5 2 2.5sc-shRNA ss-shRNA ControlPLN:GAPDH mRNA * * * *Standardized PLN mRNA levels * p<0.05 0 0.5 1 1.5 2 2.5sc-shRNA ss-shRNA ControlPLN:GAPDH mRNA * * * * 0 0.5 1 1.5 2 2.5 0 0.5 1 1.5 2 2.5sc-shRNA ss-shRNA Control shRNAPLN:GAPDH mRNA shM248 * * * *Standardized PLN mRNA levels* p<0.05shM750 0 0.5 1 1.5 2 2.5sc-shRNA ss-shRNA ControlPLN:GAPDH mRNA * * * *Standardized PLN mRNA levels * p<0.05 0 0.5 1 1.5 2 2.5sc-shRNA ss-shRNA ControlPLN:GAPDH mRNA * * * * 0 0.5 1 1.5 2 2.5 0 0.5 1 1.5 2 2.5sc-shRNA ss-shRNA Control shRNAPLN:GAPDH mRNA shM248 * * * *Standardized PLN mRNA levels* p<0.05shM750 Figure 3-7. Co-transfections with sc and ss AAV plasmids expressing PLN shRNAs lead to reductions in PLN mRNA levels. HEK 293 cells were co-transfected with 1.0 nM shRNA expressing plasmid and 0.25 nM of PLN expressing plasmid. Shown are RT-PCR analyses of PLN mRNA levels after transfection of cells with shRNAs targeting either PLN, shM248 and shM750, or a control shRNA using either the sc or ss vector syst em. Three PCR reactions were run per experiment and the PLN values were normalized to GAPDH values. The resulting ratios were then averaged an d compared to the control. *p<0.05 vs. control shRNA treated group. Data are mean Â± S.D.
66 self-complementary and the single-stranded versions of shM248 and shM750 reduced the levels of PLN mRNA as compared to the c ontrol shRNA. The shM248 were calculated to result in 19% and 33% reductions when e xpressed from the ss and sc AAV plasmids, respectively. The shM750 reduced the PL N mRNA by 29% and 22%, respectively when expressed from the ss and sc AAV plasmids. Th ese results were statistically significant (p<0.05) when compared to the irrelevant shRNA control. PLN C o n t r o l s h R N A ÃŸ-actins s s h M2 4 8 s c s h M2 4 8 C a r d i a c E x t r a c t PLN C o n t r o l s h R N A ÃŸ-actins s s h M2 4 8 s c s h M2 4 8 C a r d i a c E x t r a c t Figure 3-8. Immunoblot depict ing PLN levels of HEK 293 ce lls co-transfected with shRNA expressing plasmids. Co-transf ection of HEK 293 cells with 1.0 nM of either ss or sc shM248 and 0.25 nM of PLN expressing plasmid. An irrelevant shRNA expressing ssAAV plas mid was used as a control and PLN values were standardized to -actin levels. Murine cardiac tissue extract was used as a positive control for the antibody reaction. Interestingly, although a moderate reducti on was observed in the levels of PLN mRNA, the PLN protein response in the HE K 293 cells treated with the PLN shRNAs was quite dramatic. Figure 3-8 is a represen tative immunoblot of HEK 293 cells treated with shM248 expressing self-complementary and single-stranded AAV constructs. The blot shows representative bands for th e 25 kDa PLN pentamer as well as for -actin (43 kDa) used as a loading control. Cells tran sfected with the shM248 expressing sc or ss
67 plasmid silenced virtually all the PLN prot ein. In comparison, cells treated with the control shRNA expressing plasmid had high le vels of PLN. Similar results were observed for HEK 293 cells treated with shM750 expressing sc or ss plasmids (data not shown). Primary Neonatal Rat Ventricular Cardio myocytes Infected with PLN shRNAs Since the PLN shRNA expressing plasmids effectively silenced PLN in transfected HEK 293 cells, my next objective was to reduce the amount of PLN in cells endogenously expressing PLN. Therefore, I cultivated primary ne onatal rat ventricular cardiomyocytes (PNRVC) from the hearts of 1 day old rats. The cardiomyocytes were co-infected w ith AAV and Adenovirus (Ad). Although AAV alone readily infects cardiomyocytes, co -infection with Ad permits a more rapid expression of genes from AAV. Ad aids in expediting the synthesis of the second strand of AAV, as well as in exporting the newl y double-stranded DNA out of the nucleus and into the cytoplasm where translation and e xpression of the payload can occur very rapidly 97. For these reasons, co-infecting with Ad would somewhat mask the advantages of using the self-complementary construc ts in this cell culture context. Initially, I synthesized the PLN shRNAs to target the murine coding region. Although the rat and mouse PLN sequences are highly homologous, the shM248 required two nucleotides to be altered whereas the shM750 required multiple changes. Therefore, I created a rat version only of shM248 term ed shR248, since the rat version of sh750 may not be as effective as its mouse counterpart and would require addi tional testing in 293 cells. The shR248 constructs were cloned into both the sc and ss AAV constructs. Figure 3-9 shows representative images of PNRVC co-infected with Ad and AAV expressing the PLN shRNAs. Both sc a nd ss PLN shRNA AAV and the control shRNA
68 AAV infected the cells very efficiently. The pe rcent of infected cells was estimated to be greater than 70% after two days in culture. Control shRNA sssh R 248 sc-sh R 248 GFPPhaseMerge Control shRNA sssh R 248 sc-sh R 248 GFPPhaseMerge Figure 3-9. Primary neonatal rat ventricular cardiomyocytes infected with AAV. Cells were infected with 1,000 vector genomes (vg) per cell of AAV and were photographed 48 hours post-infection. The degree of infection was monitored by GFP expression. Bright field (Phase) was used to depict all cells in the field. Merged images were used to estimate the percent of AAV infected cells. After determining the percent of AAV inf ected cells, the RNA and protein were harvested. Using RT-PCR, the PLN and GAP DH mRNAs were amplified and a ratio of PLN:GAPDH was calculated for PNRVC 2 days post-infection (Figure 3-10). Compared to the control shRNA, both the sc and the ss PLN shRNA AAV viruses reduced the amount of PLN mRNA significantly (p<0.01). In this experiment, in the presence of Ad, the single-stranded version appeared to be more efficient at reducing the PLN mRNA.
69 The ss-shR248 was calculated to reduce PLN mRNA by 85%, while the sc-shR248 caused only a 46% decrease. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35sc-sh R 248ss-sh R 248PLN:GAPDH mRNA 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35sc-sh R 248ss-sh R 248 Control shRNAPLN:GAPDH mRNA* * *p< 0.01 Standardized PLN mRNA levels 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35sc-sh R 248ss-sh R 248PLN:GAPDH mRNA 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35sc-sh R 248ss-sh R 248 Control shRNAPLN:GAPDH mRNA* * *p< 0.01 Standardized PLN mRNA levels Figure 3-10. RNA levels in primary cardiom yocytes infected with AAV shRNAs. Cells were infected with 1,000 vg using the ss and sc AAV constructs. Shown are RT-PCR analyses of PLN mRNA levels after the infection of cells with shR248 in either the sc or ss vector co mpared to control shRNA. Three PCR reactions were run per experiment and th e PLN values were normalized to the GAPDH values. The resulting ratios were then averaged and compared to the control. *p<0.01 vs. control shRNA tr eated group. Data are mean Â± S.D. Analysis of the protein lysates demonstr ated a more comparable decrease between the two viral vectors. Figure 3-11 is a repr esentative immunoblot of PNRVC lysates two days post-infection with either sc or ss-s hR248 AAV. The blot shows protein bands for both Actin (43 kDa) and the 25 kDa PLN pe ntamer. Viral treatment with sc-shR248 resulted in a 51% PLN decrease, compared to the shRNA control. Similarly, the ssshR248 decreased PLN by 44%, as compared to control shRNA. Cumulatively, these data provided confidence in AAV delivered sh RNA constructs in be ing able to silence PLN within a cardiac environment.
70 s c s h R 2 4 8 PLN ActinC o n t r o l s h R N A C a r d i a c e x t r a c t s s s h R 2 4 8 s c s h R 2 4 8 PLN PLN ActinC o n t r o l s h R N A C a r d i a c e x t r a c t s s s h R 2 4 8 Figure 3-11. Immunoblot depic ting PLN protein levels in pr imary cells infected with AAV. Cells were infected with 1,000 vg per cell of AAV constructs. Shown are PLN protein levels 48 hours after inf ection of cells with shR248 in either the sc or ss vector or control shRNA. Murine cardiac extract was used as a positive control for PLN. Equal loading of samples was monitored by Actin levels. As an alternative method to confirm th e decrease in PLN protein levels we performed indirect immunofluorescen ce studies on PNRVC infected with AAV (co-infected with Ad). Two days post-infec tion, PLN protein levels were assessed within infected cells. This technique is useful b ecause it enables one to directly observe the level of PLN within an infected cell. Gr een cells indicate AAV infection, due to the expression of GFP. To differentiate betw een myocytes and cont aminating fibroblasts, which lack endogenous PLN, the cells were st ained with MyBP-C antibody. MyBP-C is a cardiac protein responsible in part for the movement of myosin during a contraction. Using an antibody specific to PLN we were able to demonstrate that PLN shRNA infected cardiomyocytes had reduced levels of PLN as demonstrated by the decreased levels of perinuclear staining of the PLN protein (Figure 3-12).
71 Figure 3-12. PLN sc/ss shR NA AAVs decrease the amount of PLN in PNRVC. PNRVC were infected with 1,000 vg/cell of AAV containing either ss-shR248 (A-C), sc-shR248 (D-F) or control shRNA (G-I) for 48 hours. Cells expressing the AAV constructs are positive for GFP (A, D, G). Anti-PLN antibody denotes the presence of PLN (B, E, H). Ce lls were double-stained with PLN and MyBP-C antibodies and the nuclei were counterstained with DAPI. The merged 4 color images are seen in pa nels C, F, I. Arrows demonstrate representative infected cells that can be seen to have a reduction in PLN due to the PLN shRNA expression (B, E). Arrows in (H) demonstrate non-reduced levels of PLN in control shRNA infected cells. Although the use of neonatal cells provided us data s uggesting that there was a silencing effect on PLN due to the presence of our PLN shRNA molecules, these cells were not capable of providing any physiological data. Therefore, to be able to measure a physiological effect upon PLN silencing, w ith assistance of Julieta Palomeque and Djamel Lebeche of the Cardiovascular Re search Center at Massachusetts General Hospital, I resorted to using primary adu lt ventricular rat cardiomyocytes (PAVRC).
72 These cells, unlike the neona tal cells, do not spontaneously beat in culture and thus, can be artificially stim ulated to contract allowing one to measure the amount of calcium that is involved in each contraction ev ent. Additionally, the rod shaped nature of the adult cells (Figure 3-13), as opposed to th e fibroblast-like shape of the neonatal cells, permits the use of specific software that ha s been specially designe d to measure cellular parameters involved in each contraction ev ent. Therefore, PAVRC were cultured and infected with the ss and sc-shR248 PLN AAV constructs (co-infected with Ad). However, after 48 hours in culture, only the sc-shR248 virus infected cells were observed to express the GFP reporter gene. As a re sult, the effects of only this construct on calcium handling were compared to non-infected myocytes. sc-shR248Bright field sc-shR248Bright field Figure 3-13. Adult ventricular myoctes infect ed with sc-shR248. Ventricular myocytes from adult rats were isolat ed and infected with 1x109 vg/cell of sc-shR248 for 48 hours. A representative image of the infection in the myocytes is shown. Green cells have been infected with AAV. The brightfield image is used to depict all of the ce lls in the field. A control myocyte (Figure 3-14 A) has decreased intracellular calcium during a contraction, as measured by fluorescence amplitude, compared to a myocyte that has been infected with the sc-shR248 AAV (Figur e 3-14 B). When the amplitude of a single wave form from the sc-shR248 is normalized to a wave form from a control myocyte and
73 subsequently superimposed (Figure 3-14 C), it is clear that the rate of relaxation was increased in the cell that has been infected with the sc-shR248 AAV. This parameter is measured as the time to Â½ relaxation (t1/2). Fura-2 fluorescence (Normalized) t1/2sc-shR248 t1/2controlCaiTsc-shR248 CaiTcontrol Fura-2 fluorescence (Normalized) t1/2sc-shR248 t1/2controlCaiTsc-shR248 CaiTcontrol Control myocyte222324252627282930 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1.00 Fura-2 fluorescence (360/380)Time (sec)Control myocyte222324252627282930 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1.00 Fura-2 fluorescence (360/380)Time (sec)222324252627282930 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1.00 Fura-2 fluorescence (360/380)Time (sec)sc-shR248 myocyte656667686970717273 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 Fura-2 fluorescence (360/380)Time (sec)sc-shR248 myocyte656667686970717273 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 Fura-2 fluorescence (360/380)Time (sec)656667686970717273 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 Fura-2 fluorescence (360/380)Time (sec)A.)B.)C.) A B C Fura-2 fluorescence (Normalized) t1/2sc-shR248 t1/2controlCaiTsc-shR248 CaiTcontrol Fura-2 fluorescence (Normalized) t1/2sc-shR248 t1/2controlCaiTsc-shR248 CaiTcontrol Control myocyte222324252627282930 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1.00 Fura-2 fluorescence (360/380)Time (sec)Control myocyte222324252627282930 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1.00 Fura-2 fluorescence (360/380)Time (sec)222324252627282930 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1.00 Fura-2 fluorescence (360/380)Time (sec)sc-shR248 myocyte656667686970717273 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 Fura-2 fluorescence (360/380)Time (sec)sc-shR248 myocyte656667686970717273 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 Fura-2 fluorescence (360/380)Time (sec)656667686970717273 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 Fura-2 fluorescence (360/380)Time (sec)A.)B.)C.) A B C Figure 3-14. Calcium transient (CaiT) measurements in adult ventricular rat cardiomyocytes infected with sc-s hR248. Continuous recording of CaiT (Fura-2 fluorescence) depict ed by representative cardi omyocytes stimulated at 1Hz in a Krebs-buffered so lution containing a 1mM Ca2+. Recordings represent control (A) and a PLN silen ced (B) cardiomyocytes, respectively, cultured for 48 hr either without infe ction or with sc-shR248-AAV. (C) An individual CaiT from a non-infected myocyte in bl ue and an infected cell with sc-shR248 in red, normalized and supe rimposed for better comparison. As expected, the cardiomyocyte with PL N knockdown presented a shorter time to half relaxation (t1/2). When these calcium transient (CaiT) values were graphica lly represented (Figure 3-15), the peak amplitude (Figure 3-15 A), re presenting the amount of calcium within the cell during a contraction was increased in th e sc-shR248 treated cells as compared to a control myocyte. Similarly, the percent fluorescence (Figure 3-15 B) representative of the calcium values was also increased in th e sc-shR248 treated cells . An average of a representative group of calcium transients fo r the sc-shR248 infected cells (Figure 3-14 A and B) had a significant (p<0. 05) reduction in the amount of time required to achieve t1/2 relaxation compared to control noninfected cells (Figure 3-15 C).
74 Peak % fluorescence t1/2relaxation (sec)sc-sh R 248sc-sh R 248sc-sh R 248 controlcontrolcontrol 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1.1 0 1 2 3 4 5 6 7 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45*A.)B.) C.)Peak % fluorescence t1/2relaxation (sec)sc-sh R 248sc-sh R 248sc-sh R 248 controlcontrolcontrol 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1.1 0 1 2 3 4 5 6 7 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45*A.)B.) C.) A B C Figure 3-15. Effect of PLN down regulation in adult ventricular rat cardiomyocytes. Cells were either infected with sc-s hR248 for 48 hours or left uninfected. Cells were then loaded with Fura -2 and the calcium transient (CaiT) parameters were measured. Clearly, th e infected myocytes showed increased CaiT peak (A), CaiT amplitude (B) and faster relaxation (C). Data are mean Â± S. E. (p<0.05). In brief, the biochemical and indirect immunofluorescence data obtained from the neonatal cells, showing reduction in PLN, in addition to the physio logical data obtained from the adult cells, showing an effect on the handling of calcium, corroborated the effectiveness of the PLN shRNAs in enha ncing cardiac function and made us more confident to progress to the next stageÂ—te sting AAV-mediated PLN silencing in a live murine model. In vivo Testing of PLN shRNA-Expressing AAV Since the immune systems of neonata l mice are not fully developed, the myocardium is still in development and it is easy to obtain litters of pups, we decided to use neonatal mice for our in vivo studies. By using this an imal model system, we would not have to worry about any adverse immunological effects, we could readily order large numbers of animals at a time, the injected anim als would be able to be compared to their littermates reducing variability and we would potentially ne ed smaller amounts of virus to inject neonatal animals. Because we wanted to achieve the highest levels of AAV
75 expression in the heart, two different trans duction techniques were experimented with initially to determine which would be the most suitable technique. Temporal Vein Injections Temporal vein injections were performe d using 1 day old CD-1 mice. These mice were injected with 5.0x1010 vg into the left temporal vein (Figure 3-16). On the following day, these animals were re-injected using the same procedure, this time, however, into the right temporal vein br inging the total number of vector genomes injected into each animal to 1.0x1011. After the animals were incubated for 4 weeks, they were sacrificed and direct im munofluorescence was performed to detect the extent of the infection. Figure 3-16. Temporal vein in jection into 1-day-old CD-1 mouse. Animals are briefly anesthetized on ice and then injected into the tempor al vein using a 28-gauge insulin needle and 5.0x1010 vg of AAV. A second injection into the contra-lateral side was performed on th e next day bringing the total number of vg injected into each animal to 1.0x1011. Initially we tested pTR-UF11, an AAV cons truct that only expresses GFP, so that we would be able to assess the efficacy of the infection. Figure 3-17 demonstrates the extent of the transduction in the murine myo cardium using this construct. Each of the
76 series shown represents a diffe rent cross-section of the h eart in which GFP expression was detected. Low magnification (100x) images are shown to orient th e viewer as to the extent of the infection while the higher magnification (400x) images emphasize the transduced filaments. Alt hough this technique provided so me evidence of transduced fibers, the level of transduction was not high (< 0.1% of myocytes transduced). Based on these low transduction levels , this technique would most likely not produce effective silencing and thus an alternative approach was needed. Figure 3-17. Levels of GFP expression 4 week s post temporal vein injection with AAV. CD-1 mice were injected with a total of 1x1011 vg of pTR-UF11. Animals were incubated for 4 weeks at which time they were sacrificed and their hearts excised. Hearts were fixed and then frozen. Subsequently, 5 Âµm sections were cut, stained with DAPI and mounted for direct visualization of GFP. Each series represents a different section that was found to have a GFP positive myocyte. Images of the same region were then taken at low (100x) and high (400x) magnifications. Sub-Xiphoid Injections The sub-xiphoid injection technique i nvolves introducing a 33-gauge Hamilton needle into the thoracic cavity of a 4 day old CD-1 mouse. The thinking is that a young animal does not yet have all of the cell junctions properly formed to lock cells rigidly into
77 place. The virus, therefore, is then ab le to diffuse throughout the cells of the myocardium. resulting in a high degree of infection. Figure 3-18 demonstrates the technique used to hold the animals. The animal is held on its lateral sides so as to make the skin over the thoracic cavity taut. A th in piece of tygon tubing is then placed over the needle to allow only about 2 mm of the tip to en ter the cavity. The needle is inserted just beneath and to the anatomical left of the xi phoid process and the vi rus is then injected into the cavity. Figure 3-18. Sub-xiphoid injection into 4-da y-old CD-1 mouse. Animals are briefly anesthetized on ice and th en held to expose their chests. A sub-xiphoid injection is then performed using a 33 -gauge needle encased in a piece of tygon tubing allowing only the 2 mm tip to enter the thoracic cavity of the animal. Initial results using this technique proved to be very successful using a LacZ expressing AAV construct. Briefly, animals were injected using 2.5 x 1011 vg of AAV expressing the LacZ gene under the control of a CMV promoter. After 4 weeks of incubation, animals were sacrificed and their hearts extracted, cut into thirds laterally, frozen down and then sectioned. Th e sections were then stained for -galactosidase. As can be seen in Figure 3-19, all three regions of the heart (i.e. the apex, the central region
78 and the base) stained positive for -galactosidase almost over the entire myocardium. This finding suggested that this technique would potentially be successful in delivering the PLN shRNA expressing AAV constructs th roughout the entire ca rdiac tissue. Control LacZInjectedApexCenter Base Control LacZInjectedApexCenter Base Figure 3-19. Sub-xiphoid injected CD-1 mous e with LacZ construc t. CD-1 mice were injected with 2.5x1011 vg of AAV containing the LacZ gene. Hearts were excised 4 weeks post-injection and cut in to thirds laterally before freezing. Upon sectioning, each section would cont ain a representative area from the base, the center and the apex of the heart. Staining for -galactosidase is indicated by the blue chromagen (X-gal ). Non-infected hearts (Control) stained negative for the presence of -galactosidase. Images were taken using a 10X objective. Having demonstrated the effec tiveness of this technique, we then proceeded to try this injection technique us ing a GFP expressing construc t, pTR-UF11. In this experiment, we injected 2.25x1011 vg of AAV. The animals were incubated for 3 weeks at which time they were sacrificed and thei r hearts excised and prepared as described above. As seen in Figure 3-20, GFP positive myocytes are visible, although to a lesser extent than the -galactosidase positive myocytes. None theless, we proceeded to use this injection technique to validate the silenci ng ability of our PLN shRNA expressing AAV constructs in live animals.
79 Apex CenterBase Apex CenterBase Figure 3-20. Sub-xiphoid injected CD-1 m ouse with pTR-UF11 construct. CD-1 mice were injected with 2.25x1011 vg of AAV containing the GFP gene. Hearts were excised 3 weeks post-injection, fixed and cut into thirds laterally before freezing. Upon sectioning, each section would contain a representative area from the base, the center and the apex of the heart. Sections were stained with DAPI and visualized directly for GFP fluorescence. Images were taken using a 10x objective. In order to confirm the silencing abilit y of our PLN shRNA viral constructs, we injected 4-day-old CD-1 mice using the sub-xiphoid injection technique and 2.25x1011 vg of AAV. Mice were injected with 5 di fferent constructs: sc-shM248, ss-shM248, ss-shM750, the control ss-shRNA, and pTRUF11. Mice were incubated for 3 weeks after which they were sacrifi ced, their hearts excised, fixed and frozen for sectioning. Sections were then stained using antibodies against PLN and GFP. Unfortunately, the viral preparation for sc-shM248 was suboptimal and extensive infection comparable to that seen with pTR-UF11 was not achieved. However, infections with the four other constructs were satisfactory and we able to perform indirect immunofluorescence to assess the effect of the differe nt viral constructs on PLN levels. Figure 3-21 shows two representative images (A) and (B) of car diac cells infected with ss-shM248 and ss-shM750 respectively. The arrows in the PLN panels denote cells that have been infected with either ss-shM248 or ss-s hM750 AAV virus (verified by GFP expression) and show reduced amounts of PLN while th e arrowheads show ne ighboring cells that
80 were not infected and have a higher level of PLN expression. DAPI was used to stain the nuclei, and merged images are shown. As controls, I used two different viruses. The first, seen in Figure 3-22 A, contains the ss-shRNA control. The second control virus, pTR-UF11, only contains the GFP expression cassette and no shR NA (Figure 3-22 B). As seen in both of these controls, there appears to be no difference in PLN levels in either cells that have been infected with the virus (arrows) or cells that were uninfected (arrowheads). The presence of the virus was observed by the expression of the GFP repor ter gene. Sections were stained with DAPI and the merged images are shown. Thes e results indicated that the control viruses were unable to effect the changes in PLN le vels that were seen using the PLN shRNA viruses.
81 A.)PLNMerged GFP DAPI B.)PLNMerged GFP DAPI A.)PLNMerged GFP DAPI A.)PLNMerged GFP DAPI B.)PLNMerged GFP DAPI B.)PLNMerged GFP DAPI A B A.)PLNMerged GFP DAPI B.)PLNMerged GFP DAPI A.)PLNMerged GFP DAPI A.)PLNMerged GFP DAPI B.)PLNMerged GFP DAPI B.)PLNMerged GFP DAPI A B Figure 3-21. In vivo reduction in PLN protein levels fo llowing infection with PLN shRNAs. CD-1 mice were injected using 2.25x1011 vg of ss-shRNAs in AAV and incubated for 3 weeks. Hearts were then excised fi xed, sectioned and stained using antibodies against GFP and PLN. Sections were c ounterstained with DAPI. Repr esentative images from mice injected with ss-shM248 (A) or ss-shM 750 (B) are shown. Cells infected with AAV are denoted by the GFP staining. Arrows denote cells that have been infect ed and show a reduction in PLN levels, while arrowheads denote cells that have not been infected and show higher levels of PLN. Nuclei are c ounterstained with DAPI and the merged images are shown.
82 A B Figure 3-22. No reductions in PLN protein levels are observed following infection with control ss-AAV viruses in vivo. CD-1 mice were injected using 2.25x1011 vg of a control shRNA AAV (A ) as well as an AAV virus only expressing GFP, pTR-UF11 (B). Animals were incubated for 3 weeks at which time hearts were excised, fixed, secti oned and stained using antibodies against GFP and PLN. Sections were c ounterstained with DAPI. Depicted are infected cells that are GFP positive and show no reduction in PLN levels. Represen tative infected cells are demarcated us ing arrows while noninfected cells are indicated using arrowheads. Sections were counter stained with DAPI and the merged images are shown.
83 CHAPTER 4 DISCUSSION AND CONCLUSIONS The world is round and the place which may seem like the end may also only be the beginning. Â—Ivy Baker Priest Following the creation of the PLN knoc kout mouse in 1998 by the Kranias group 98 research on cardiac diseas e took on a whole new look. Suddenly, investigators began looking more closely at the coordina tion of calcium in the myocardium as a potential treatment for cardiac disease. These PLN knockout mice were found to have enhanced myocardial performance and no delete rious side effects. Furthermore, Chien et al. showed that mice deficient in the Z-disk protein, Lim, and consequently suffering from myocardial dysfunction could be rescued by additionally ablating PLN 99. These were some of the first findings that suggest ed one could potentially devise a therapy to reduce the levels of PLN and consequently en hance contractility in failing myocardium. Several groups have attempted to eith er upregulate the expression of SERCA in an attempt to override the repression of PL N by changing the stoichiometry that exists between the two proteins 100,101. Conversely, ChienÂ’s group has utilized a pseudophosphorylated form of PLN (S16E PLN) to activate SERCA and enhance cardiac function in genetic models of heart failure 102, as well as in mode ls of acquired heart failure 103. In our present study, however, we em ployed RNAi technology to develop a potentially clinically releva nt treatment for myocardial dysfunction by silencing PLN. Our work demonstrates that siRNA mol ecules, as well as AAV vectored shRNA molecules can effectively knockdown exogenous, as well as endogenous levels of PLN in
84 both cell culture and animals. Furthermor e, silencing of PLN shows a trend towards increase in calcium transients and incr ease in relaxation velocity in adult rat cardiomyocytes. Reduction of Co-Transfected PLN using both siRNA and shRNA With the advent of RNAi, one of the earlie st papers utilizing this technology to target a cardiac protein was reported by Watanabe et al. 104. In this report , they described siRNA molecules that were designed to e fficiently reduce the PLN mRNA and protein levels in rat cardiomyocytes. Although the au thors effectively suppr essed the levels of PLN, chemically synthesized siRNA molecule s generally exhibit a short half-life and their uptake by mammalian cells can be a limiting factor 105. Furthermore, the use of these molecules could prove difficult in targeting them specifically to heart tissue. Thus, in order to improve upon this strategy and to incur more tissue-spec ific and longer-term gene silencing, we felt that utilizing DNA AA V vector constructs to express PLN shRNA molecules would help overcome these pitfal ls. Upon proper administration, these shRNA molecules would be expressed within the ta rgeted cell and theref ore avoid degradation from nucleases, in order to allow more c onsistent and longer-te rm gene silencing. Prior to designing and utilizing the shR NA molecules, we first tested the knockdown capability of siRNAs specific to PL N. Should the chemically synthesized siRNAs prove effective in silencing PLN, th e next step would be to design shRNAs based on the target sequence used for the siRNA. Following the co-transfection of siRNAs targeting PLN and a plasmid expres sing the PLN gene into HEK 293 cells, we were able to observe decreases in th e amount of PLN mRNA relative to GAPDH in comparison to cells that had been treated w ith a control siRNA ta rgeting GFP. Although there were significant reductions in RNA levels after 48 hours, they were modest
85 compared to the amount of reduction observe d at the protein level. One possible explanation for this type of discrepanc y was that we were observing a phenomenon characteristic of microRNA (miRNA) medi ated translational inhibition. MiRNAs function using the same pathwa y as siRNAs. Although siRNAs have complete homology to their targets, microRNA s usually have several mismatched base pairs. This level of mismatch has been observed to lead to high levels of translational repression 85. Both types of molecules can be f unctionally equivalent , however; their biological origin is what differs 81. MiRNAs are produced from transcripts that form stem-loop structures, whereas siRNAs ar e produced from long double-stranded RNA precursors, which can either be endo genously produced or exogenously provided 81. The primary miRNA (pri-miRNA) transcripts are pr ocessed within the nucleus by an enzyme known as Drosha into 70 nucleotide transcri pts (pre-miRNAs). Subsequently, the premiRNAs are exported into the cytoplasm wher e they enter the siR NA processing pathway and are further processed by Dicer and assembled into the RISC complex. Since we artificially expressed the PL N target in the HEK 293 cells we were confident that the sequence of the PLN siRNAs had high fidelity to the target. Initially, there was speculation that the synthesized si RNA molecule might ha ve contained a minor error made by the manufacturer. However, wh en we created shRNA molecules, cloned them into AAV expressing plasmids and co-transfected them with the same PLN expressing plasmid, we again observed the sa me phenomenon (i.e. modest, yet significant decreases in mRNA levels with overwhelm ing translational inhibition). For these reasons, it is possible that a mechanism for tr anslational inhibition related to siRNAs is
86 not yet fully understood and could explain w hy we are seeing mode st transcriptional decay and robust translational inhibition. An additional explanation for the contrast in RNA and protein levels observed upon siRNA-mediated PLN silencing is that there might be an error in our sampling of RNA via RT-PCR. Because PCR is a method that amplifies DNA, it is possible that we are amplifying our cDNA to artifici ally represent more messenger RNA than what is actually present. Therefore, in order to more accu rately measure transcript levels, one would potentially have to use a method of R NA detection that does not involve any amplification, such as a Northern blot or RN ase protection assay. Regardless, the results of these initial studies demonstrated the eff ectiveness in the design of our siRNA and the shRNA-expressing AAV plasmid constructs in silencing exogenously expressed PLN. Infections in PNRVC using AAV1 Vectored shRNAs Similar to our groupÂ’s approach, other res earchers have also tried to reduce the amount of endogenous PLN in cultured cardio myocytes using alternative approaches. For example, antisense oligonucleotid es have been designed towards PLN 106. Although effective, these molecules were delivered into cultured human cardiomyocytes using Adenovirus. This mode of delivery would be detrimental in a clinic al setting due to the Adenovirus being known to elicit severe responses from the immune system. Furthermore, Adenovirus vectors are limited due to their short-lived expression. Subsequently, another report demonstrated th e use of AAV to deliver these PLN specific antisense oligonucleotides 107. One of the advantages of using AAV, and the reason we also chose this vector system, is its lo wer immunogenicity. Although the AAV-mediated delivery of antisense oligonucleotides appeared promising, RNAi has been reported to be much more efficacious than antisense oli gonucleotides. In fact , reports comparing
87 siRNAs to other small nucleic acids, such as antisense oligonucleotides or ribozymes have shown that the concentr ation of siRNAs required for effective gene silencing is several orders of magnitude lower 79,108. Furthermore, in one head-to-head comparison, siRNAs were demonstrated to knock dow n gene expression about 100Â–1000 fold more efficiently than anti sense oligonucleotides 109. Even our own testing of ribozymes targeting PLN in cultures of PNRVC demonstrated an inability to suppress PLN transcript and protein levels (Appendix A) . For these reasons, we chose to employ shRNA-expressing AAV vector c onstructs to silence PLN in physiologically relevant cardiac environments (i.e. PNRVC). The ability to culture PNRV C, albeit for a short amount of time, has facilitated testing different experimental strategies before moving onto an animal model system. A report published by Du et al. 61 compared expression effi ciencies of different AAV serotypes that were available at the time in adult and neonatal mice cardiomyocytes. This report demonstrated that AAV2 was more e fficient at infecting neonatal cells, while AAV1 was better at infecting adult cells. Our own experimentation demonstrated negligible differences between these two se rotypes in neonatal rat cardiomyocytes. However, because our future experimental plans included testing these constructs in cultures of adult rat cardio myocytes, and due to the high homology between mice and rats, we decided to use the AAV1 serotype for the packaging of our constructs. For these sets of experiments in PNRVC, we used the rat version of shM248, shR248. This shRNA was cloned into both si ngle-stranded (ss) and self-complementary (sc) vectors of AAV in order to see if one construct was more effective than the other. Although PLN mRNA reduction occu rred using both vector cons tructs, surprisingly, we
88 saw a greater level of reduc tion using the ss-AAV. This finding could be possibly attributed to the potential e rror associated with our met hod of quantitating mRNAs as described earlier. In reference to PLN prot ein reduction, we did not see an overwhelming silencing, as was observed in the co-transfe cted HEK 293 cell experime nts. Additionally, we only detected a moderate increase in PLN protein reduction by We stern blot analysis using the sc-AAV, as compared to the ss-AAV. It is possible that the helper functions provided by the Adenovirus (Ad) that we used to co-infect AAV were responsible for this phenomenon. It has been reported that th e E4A gene (orf6) of Ad enhances AAV transduction and aids in the conversion of single-stranded to d ouble-stranded AAV DNA (reviewed by Geoffroy and Salvetti 97). This could also expl ain why in the indirect immunofluorescent studies, th e levels of GFP expressi on observed using the two constructs were fairly similar, as were th e levels of PLN protei n reduction observed by western blotting. An additional set of experiments that could be performed to enhance these findings would be to perform an infecti on time course in these cells. This would elucidate the amount of silencing that coul d be obtained using longer and shorter time points. Similarly, a dose response study c ould determine the minimal amount of virus needed to affect silencing. Overall, the resu lts of these studies demonstrate the successful application of RNAi-mediated gene thera py using AAV to silence both PLN mRNA and protein levels in an in vitro cardiac environment. Infections in PARVC using AAV1 Vectored shRNAs Although the neonatal rat cells were a suitable environm ent for us to assess the biochemical efficacy of our shRNA molecules, we needed to employ the use of adult rat cells (PARVC) to assess the physiological impact of thes e small molecules. Adult myocytes can be artificially stimulated and thus contractility and calcium transients can
89 be measured under various conditions. By reducing the levels of PLN using RNAi, we could theoretically affect SERCA-mediated Ca2+ uptake levels and measure the differences between infected and non-infected cells. In infected adult myocytes, we were able to observe a trend that was indicative of increases in calcium transients as well as decreases in relaxation times (t1/2) as compared to non-infected myocytes. However, a more comprehensive study will be needed. Measurements recording the effects over several calcium concentrations, using at le ast 3 and preferably 6 cells per calcium concentration will be required to determine the statistical significance of our preliminary studies. To estimate the impact on the reduced repression by PLN on SERCA, one could also measure increases in the activity of the ATPase using well established assays such as caffeine pulse experiments. Caffeine acts as a stimulant for the release of calcium from the SR resulting in increases in SR ca lcium transients. By incubating varying concentrations of caffeine with primary car diomyocytes, one could measure changes in the amount of calcium stored in the SR as a result of reducing PLN. One would expect increases in the calcium stored in the SR of infected cells beca use the calcium would have a greater affinity for the SR once the regulator, PLN, is reduced. One aspect that makes these experiments t echnically challenging is the fragility of the adult myocytes in tissue culture. We have found that the healthy morphology of the cells is significantly affected after two days. In order to address this concern, it would be of interest to conduct these experiments us ing a higher number of vector genomes per cell. This could potentially lead to a more rapid expressi on and allow us to conduct the experiment at an earlier time point. Anothe r technical hurdle that we encountered was
90 that the ss-AAV constructs did not express the GFP repor ter gene until 3 days post infection. At this time point, the viability of the cells was significantly reduced and it was virtually impossible to subject them to calcium measurements. Perhaps using higher numbers of vector genomes, one could addres s this issue. Furthermore, to improve upon the viability of these cells, other protocols i nvolving the cultu ring of these cells need to be explored. Regardless, the si gnificance of these results is that PLN silencing appears to increase calcium transients in cardiomyocytes as well as decrease the time to half relaxation. Animal Studies In order to inhibit PLN activity, some groups have resorted to a more non-traditional knockdown approach, by using a PLN targeted antibody 110. A recombinant, intracellularly e xpressed antibody was used to ta rget the cytoplasmic region of PLN thereby inhibiting PLNÂ’s repressive effect on SERCA. The treatment was implemented in mice that were induced to de velop diabetes. This induction of diabetes, typically results in a signifi cant decrease in heart relaxa tion rates. However, antibody treatments of these mice resulted in increas es in both heart relaxation and contraction rates. Although this form of treatment appe ars promising, the authors admitted that they could not exclude interactions of this anti body with other potential ta rgets. Additionally, the delivery vehicle used for this treatmen t was Adenovirus which would not be effective for long-term treatments and would also pose potential immu nological concerns. In an attempt to circumvent these short comings, our treatment uses shRNA molecules delivered by AAV. As mentioned previously, AAV allows expression of the cloned gene of interest (i.e. PLN shRNA) for extended periods of time without the danger of eliciting a host immune response. Add itionally, using a BLAST search demonstrated
91 that our shRNA molecules do not target any other gene, thereby avoiding most off-targeting effects. It is possible, howeve r, that expressing a hi gh level of an shRNA could lead to non-specific effect s by interfering with miRNA pr ocessing or transport (i.e. exportin-5) 111. These phenomena could be evaluated by examining liver function in treated animals. Using the PLN shRNA AAV molecules in an in vivo setting allowed us to assess their efficacy in live animals. We were able to detect a reduction in PLN protein levels in cells that had been infected with PLN shRNAexpressing AAV. Cells that were infected with either of the two control viruses, sh RNA control or pTR-UF11, containing only the GFP cassette, did not show any changes in PL N levels in either infected or uninfected cells. This suggested that our PLN shRNA molecules were effective at reducing protein levels of PLN. However, further experiment ation is required in order to confirm the consequential effects of utilizing these molecules in an in vivo setting. For these reasons, measuring physiological parameters such as calcium transients and SERCA activity would aid in understanding the actual effects that these molecules are having in vivo. It is interesting to note that the levels of transduction of cardiac tissue observed with the LacZ AAV construct were much more robust than with any of the AAV constructs expressing GFP that I tested. Experimental differen ces could potentially account for these differences. First, the LacZ gene was expressed using the CMV promoter while the GFP gene was under the co ntrol of the CBA promoter. While these two promoters have been widely used to expr ess transgenes in myocardium, it is possible that they do not express their respective genes equally. Add itionally, the LacZ infected heart was allowed to incubate for 4 weeks, while the longest time point measured using
92 the GFP AAV vector constructs was 3 weeks. It is also possible that the differences observed in the two constructs could have been due to the University of Florida Vector Core which produced them. Although the two constructs were pack aged at the same core, the time between the preparations of the viral preps was several years. During this time, it is possible that th e personnel as well as producti on and tittering protocols may have changed. In addition, using the subxiphoid injection techni que, we have observed variability from one animal to another as a re sult of this mode of injection. This variability has even been obs erved during injections involvi ng animals from the same litter with the same viral pr eparations. Although all of th ese concerns are valid, it is possible that the diffusion of the -gal chromophore from cell to cell could exaggerate the extent of infection that is actually taking place in the myocardium after injection. Overall, the results of these studies dem onstrate the feasibility of combining RNAi technology and AAV-mediated forms of gene th erapy in silencing PLN to potentially treat cardiac disease. Future Directions The advent of RNAi has rapidly created a powerful tool for regulating gene expression. Only a few years after the eluc idation of siRNA technol ogy, several clinical trials are already underway (reviewed by Beal 112). In order to aid in the understanding of our AAV PLN shRNA constructs in animal models, additional experimentation with the AAV shRNA molecules targeting PLN will be needed. Becaus e we were only able to test the shRNA expr essing ss-AAV construct in vivo (initial sc-AAV prep aration was not optimal), it would definitely be of value to re-test the PLN shRNA expressing sc-AAV vector and compare these molecules with thei r ss-AAV counter part. This would provide insight as to benefits of us ing the shRNA expressing sc-AAV construct. Only one paper
93 has been published to date using sh RNA molecules in the sc-AAV context 94. In this report, Xu et al. used a shRNA expressing sc-AAV to ta rget the P-glycoprotein gene in cell culture models. They demonstrated supe rior efficacy of the shRNA molecules using the sc-AAV compared to the ss-AAV. Their data suggested that the sc-AAV provided a greater inhibitory effect and required a lower number of viral particle s per cell to produce a more rapid onset of gene expression. Shoul d we be successful in infecting animals with PLN shRNA expressing sc-AAV, and demonstr ate a greater silencing efficacy using the sc-AAV, we would be the first to report the use of this vector system in an in vivo setting. This could potentially enhance AAV-medi ated shRNA gene therapy treatments. Newly identified serotypes of AAV, namely AAV8 60 and AAV9 113, are now available and have been reported to have a greater cardiac tropism than any of the other previously isolated serotypes. Using thes e serotypes, investigat ors have shown that simple tail vein injections into adult and neonatal mice can result in high levels of expression of their gene of interest in the myocardium. This would be advantageous, not only because the injection procedure is relativ ely simple, but also, we would be able to test our PLN shRNA molecules in adult mice a nd rats, in which infection of the heart has been difficult. Testing our PLN shRNA mo lecules in adult rode nts would represent another step forward in devel oping a therapy for cardiac diseas e, since all of the current models of infarction are done in adult animals. Another interesting test of whether PLN shRNA expr essing AAV are capable of silencing PLN protein in animal models would be to inject animals with the PLN shRNA expressing AAV constructs, allow them to inc ubate for several week s and then harvest the hearts for primary cardiomyocyte isolation. This type of experiment would allow us
94 to directly assess the PLN prot ein levels of infected myo cardium on a cell-by-cell basis using indirect immunofluorescence studies. Lastly, it has been noted that some geneti c models of cardiac disease, both human 114,115and rodent 116,117 do not benefit entirely from reductions in PLN levels. This is an important consideration when dealing with a developing treatment for cardiac disease. It is possible that extended periods of time wit hout PLN expression could be detrimental to a patient suffering from myocardial dysfunction. For this reason, it would be of interest to create a shRNA molecule whose expression is regulated by an inducible promoter. Amar et al. have created such a system using a lentiv irus vector that regulates expression from an inducible modified U6 promoter 118. In this system, the promoter is turned on in the presence of doxycycline, resu lting in the expression of an shRNA that is capable of silencing an exogenous protein, GFP, as well as an endogenous protein, p53. These studies were performed in th e context of HEK 293T cells. Small hairpin RNAs can also be expressed in the context of miR-30, a microRNA precursor, enabling them to be expressed from regulatable pol II promoters 91. In conclusion, this study has characteri zed the potential usef ulness of applying RNAi technology in the form of gene therapy to treat cardiac disease. By delivering shRNA molecules using a self-complementa ry and single-stranded AAV system, thereby reducing the levels of PLN in cell culture, as well as in an animal model one can increase cardiac function, in order to combat cardiac infarctions. Although a feasible therapy will require additional extensive experimentation, we believe that the modulation of calcium through the controlled expression of PLN could potentially result in a form of therapy that would be beneficial to many patients suffering from a malfunctioning myocardium.
95 APPENDIX A RIBOZYMES TARGETING PHOSPHOLAMBAN Introduction Although the bulk of this di ssertation has been devoted to highlighting RNAi technology that was used to target PLN, I w ould be remiss if I did not at least briefly describe the bread and butter of the Lewin labor atory before the advent of RNAi, that is to say: ribozymes. This appendix will describe two ribozymes created to target PLN. Ribozymes are RNA molecules with the abil ity to catalyze the cleavage and joining of RNA molecules. They consist of 3 base paired stems surrounding a catalytic central core of 15 conserved nucleotides. The cleavag e reaction requires divalent cations such as Mg2+. Hammerhead ribozymes, so called because of their 2-D structure, will cleave the 3Â’ triplet sequence NUX where N is any base a nd X is any base but G. Along with their counterparts, hairpin ribozymes, hammerhead ribozymes have been used extensively for gene therapy trials 119. Hairpins differ in that th eir recognition sequence is YNGUC where again N is any nucleotide and Y must be a pyrimidine. Since hammerheads have a less restricted sequence requirement, they have been used more commonly and will be the focus of the following studies. The major challenge for using ribozymes in vivo has been achieving the continuous expression of the ribozyme sequences in a particular tissue. Nevertheless, some ribozymes have been shown to have a substa ntial biological effect in experimental animals. For example, a hammerhead ribozyme delivered by an adenoviral vector system led to a 96% reduction in human growth hormone mRNA in transgenic mice 120.
96 Ribozymes designed to reduce the expression of oncogenes have frequently been tested in vivo. These experiments have demonstrated the efficacy of ribozymes in animals and serve as a paradigm for the therapeutic use of ribozymes. More recently, combinatorial strategies using different type s of RNA molecules are being devised such as a study using primary hematopoietic cel ls infected with HIV 121. In this study, th e authors designed a lentiviral vector delivery system consisting of an anti-HIV shRNA, anti-CCR5 ribozyme, and a nucleolar-localizing TAR decoy. Their results indicated that this type of combinatorial strategy inhibited HIV replica tion more so than either of the three strategies alone. Our PLN ribozymes were not designed as part of a combinatorial strategy but perhaps one day should be considered as such. Similar to the aforementioned PLN shRNAs, the PLN ribozymes were also designe d to be delivered us ing AAV constructs as a means of developing a gene therapy to am eliorate the effects of cardiac disease. Materials and Methods Two hammerhead ribozymes (Rz) were desi gned to target the coding region of PLN after nucleotide 201 and 304 of the mouse mRNA. Hence, they were so named Rz 201 and Rz 304. As pictured in Figure A-1, th e mRNA target is depicted in red while the actual hammerhead ribozyme is in black. Using a program (MFOLD) freely available from the Rensselaer Institute of Technology (http://www.bioinfo.rpi.edu/applicati ons/mfold/), the structure of the target was deemed to have a minimal amount of s econdary structure and therefore, capable of being accessed by a ribozyme molecule. Fo llowing this, both the RNA coding for the ribozyme as well as the RNA coding for the target sequence were ordered from Dharmacon Research Inc. (Lafayette, CO) at the 50 ÂµM scale. The RNA oligos were
97 chemically synthesized with an acid-labile orthoester protecting group (preventing degradation) on the 2Â’-hydroxyl. The protecting group was removed using the manufacturerÂ’s supplied buffer at pH 3.8 at 60oC following the manufact urerÂ’s protocol. The deprotected RNA molecules were resuspen ded at a 300 pmol/ÂµL final concentration. Rz201Rz304 Rz201Rz304 Figure A-1. Phospholamban ribozymes: Rz 201 and Rz 304. The red nucleotides represent the target sequence of th e ribozyme. The black nucleotides demarcate the sequence of the hammer head ribozyme in a two dimensional configuration. 5Â’ End-Labeling of Deprotected Target RNA The target RNA was labeled on the 5Â’ end with -32P ATP (ICN) using T4 polynucleotide kinase (Promega) as described below. For each r eaction, 20 pmol of target were combined with 1 ÂµL of 10-times concentrated PNK Buffer (700 mM Tris-HCl pH 7.6, 100 mM MgCl2, 1 ÂµL RNasin (Promega), 1 ÂµL 0.1 M DTT, 3 ÂµL H2O, 1 ÂµL [-32P]-ATP, and 1 ÂµL T4 PNK. Re actions were incubated at 37oC for 30 minutes and then 90 ÂµL of H2O were added. The mixtur e was then incubated at 65oC for 5 minutes to inactivate the T4 PNK. Two phenol:chloroform:isoamyl alcohol extractions were then performed and 90 ÂµL of the aqueous phase we re then purified over a Sephadex G-25 spin column (Pharmacia) to separate the labeled target molecule from the unincorporated radionucleotides. The resulting final target solution concentration was 0.2 pmol/ÂµL.
98 In vitro Timecourse Cleavage Reaction Separate dilutions of ribozyme and targ et were prepared as follows. For the ribozyme tube, 13 ÂµL of 400 mM Tris-HCl (p H 7.45) were combined with 1 ÂµL of 2 pmol/ÂµL ribozyme and 80 ÂµL of H20. The target tube contained 1 ÂµL of radiolabeled target RNA (diluted to 0.2 pmol/ÂµL), 1 ÂµL of cold target RNA (diluted to 20 pmol/ÂµL) and 8 ÂµL of H2O. The ribozyme tube was heated to 65oC for 2 minutes and then cooled to room temperature for at least 10 minutes. To this tube, 13 ÂµL of a 1:10 RNasin: 0.1 M DTT mix as well as 13 ÂµL of 50 mM MgCl2 were added. This yielded a final MgCl2 concentration of 5 mM. This tube wa s then allowed to equilibrate to 37oC for at least 10 minutes. Stop buffer (90% formamide, 50 mM EDTA , 0.05% bromophenol blue, and 0.05% xylene cyanol) was prepared to terminate the re action at the appropriate time intervals. For the cleavage reaction, 10 ÂµL of the ta rget mix was added to the entire ribozyme cocktail tube and this was considered to be the initiation of the reaction. At this point, a 10 ÂµL aliquot was removed from the reaction tube and combined with 10 ÂµL of the stop buffer. This timepoint was considered to be time Â“0Â”. The remaining mix continued to incubate at 37oC for the remaining timepoints: 1, 5, 10, 15, 30, 60, 180, 1200 minutes. All of the timepoints were stopped in the same way using 10 ÂµL of the stop buffer. Subsequently, the reaction from each time point was heated to 95oC for five minutes and then quickly cooled on ice for 5 minutes. The samples were then loaded on a 10% polyacrylamide/8M urea gel and separate d via electrophoresis. The gel was then fixed in a solution containing 10% me thanol, 10% acetic acid and 80% H2O for 30 minutes, dried on a vacuum drier for 2 hours and exposed to radioanalytic phosphorescent screens. Finally, the images were analyzed using a Molecular Dynamics
99 Storm 800 system and ImageQuant software (GE Healthcare). The percentage of substrate cleaved in each sample was determin ed from the ratio of radioactivity in the 5Â’-end labeled cleavage product (P) to the sum of the radioactivity in the 5Â’-end labeled cleavage product plus the subs trate band (S): % Cleavage=P/(P+S). Using Microsoft Excel, the percentage of substrate cleaved was then plotted as a function of time to generate a graph representing the rate of cleavage. Preparation of Ribozyme Sequences into AAV Vectors Once the ribozymes were deemed to be catalytically active, DNA oligonucleotides encoding both the sense and antisense stra nd were manufactured by Invitrogen. In addition to the ribozyme sequence, thes e oligos were synthesized with a HindIII cut site at the 5Â’ end as well as an SpeI cut site on the 3Â’ end. These terminal sequences permitted cloning in their respective sites wi thin a plasmid containing the CBA promoter as well as the CMV immediate early (ie) enhancer upstream of the ribozyme. Additionally, the plasmid also contained an IRES site followed by the Red Fluorescent Protein (RFP) gene downstream of the PLN ri bozyme. The resulting constructs can be seen in Figure A-2. The resulting constructs were sequenced for verification. Subsequently, large scale DNA preparations were done as described in Chapter 2. The DNA was then packaged into AAV1 by the Ophthalmology Vector Core. Co-Transfection of HEK 293 Cells HEK 293 cells were co-transfected with a plasmid expressing PLN as well as the AAV plasmids co-expressing RFP and the PLN ribozym es. Co-transfections were carried out using standard protocols as de scribed in Chapter 2 using eith er a 1:1 or 1:4 ratio of PLN
100 plasmid:Rz plasmid. Transfections were conducted with Lipofectamine2000 for 48 hours at which time the cells were harvested and RNA was prepared. HindIII (1925) PLN Rz201/304+RFP7609 bp ApR ColE1 ori f1(+) origin TR TR IRES CMV ieenhancer Intron Hairpin Rz RFP PLN Rz201 SV40 poly(A) ChikenÃŸ-actinpromoter Exon1 Nsi I (2044) Spe I (1961) HindIII (1925) PLN Rz201/304+RFP7609 bp ApR ColE1 ori f1(+) origin TR TR IRES CMV ieenhancer Intron Hairpin Rz RFP PLN Rz201 SV40 poly(A) ChikenÃŸ-actinpromoter Exon1 Nsi I (2044) Spe I (1961) PLN Rz201/304+RFP7609 bp ApR ColE1 ori f1(+) origin TR TR IRES CMV ieenhancer Intron Hairpin Rz RFP PLN Rz201 SV40 poly(A) ChikenÃŸ-actinpromoter Exon1 Nsi I (2044) Spe I (1961) Figure A-2. An AAV ribozyme construc t expressing a PLN Ribozyme and RFP. Ribozymes were cloned into the HindIII/SpeI sites. The CBA promoter expresses the ribozyme and there is an IRES from which RFP is expressed. The region to be packaged is flanked by the two AAV2 TRs. The RNA was reverse transcribed and PCR amplified to assess the levels of PLN mRNA after treatment with the ribozyme expressing plas mids. RT-PCR was conducted for both the PLN mRNA, as well as for the in ternal control, GAPDH, as described in Chapter 2. A ratio of PLN:GAPDH was then calculated and graphed using Microsoft Excel. Ventricular Neonatal Rat Cardiomyocytes Experiments Ventricular neonatal rat cardiomyocytes were prepared as described in Chapter 2. Cells were seeded on a 6-well plate at a hi gh density. Infections using AAV1 containing either of the ribozyme constructs were done in duplicate. As a control, pTR-UF11 virus, expressing GFP, was used to assess infection efficiency. Infections were carried out
101 using 1,000 vg/cell and allowed to incubate fo r 3 days before harv esting. Protein was extracted and run on a Laemlli gel for immunoblotting as described in Chapter 2. Additionally, indirect immunof luorescence experiments were conducted on these cells using the PLN antibody (Upstate) following th e protocols described in Chapter 2. Direct Injection into the Murine Myocardium Briefly, B6/129 mice were anesthetized w ith inhaled isofluorane. The mouse was then intubated and put on a ventilator. Th e chest cavity was opened exposing the apex of the heart. Two, 35 ÂµL injections containing a total of 7x1011 vg of AAV1 were delivered to the apex and the myocardium surrounding th e left ventricle. The animal was then sutured and allowed to incubate for 2Â–4 weeks. Results The first set of experiments involved an in vitro time course to assess the efficacy of the two PLN ribozymes. This experiment was carried out using a 13 nt target and 5mM MgCl2. This concentration of MgCl2 is considered to be st ringent in selecting for the more active ribozymes; and I found that bo th ribozymes appeared to be very active in vitro. As seen in Figure A-3, Rz 201 achieved a 20% cleavage at about 1 minute; while Rz 304 was even faster reaching th e same endpoint in 0.5 minutes. PLN Rz 201 5mM MgCl0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 024681012141618202224262830 Time (minutes)Fraction of target cleaved PLN Rz 304 5mM MgCl0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 01234567891011121314151617181920 Time (minutes)Fraction of target cleaved Figure A-3. Cleavage efficacy of PLN ribozym es. The fraction of target cleaved is represented on the y-axis while time in minutes is on the x-axis.
102 Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control Rz Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl A.) Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control Rz Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control Rz Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control RzB.) Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control Rz Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl Standardized Levels of PLN0 2 4 6 8 10 12 1:1 Ratio Target:Rz1:4 Ratio Taget: RzRatio PLN:GAPDH Rz201 Rz304 RzControl A.) Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control Rz Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control Rz Non standardized PLN levels0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07 1:1 Ratio Target:Rz1:4 Ratio Taget: RzPLN levels Rz201 Rz304 Control RzB.) A B Figure A-4. PLN mRNA leve ls in HEK 293 cells co-trans fected with PLN expressing plasmid and PLN Ribozymes. A 1:1 or 1:4 molar ratio of PLN: Rz was cotransfected into the cells. We used either 0.25 pmol of PLN and Rz (1:1 Ratio Target:Rz) or 0.25 pmol of PLN a nd 1.0 pmol of Rz (1:4 Ratio Target: Rz). (A) RT-PCR analysis of PLN mR NA levels after transfection of cells with ribozymes targeting either PLN, Rz 201 and Rz 304, or a control Rz. Three PCR reactions were run per expe riment. For each replicate, the PLN Phosphorimager units were normalized to GAPDH. The resulting rations were then averaged and compared to th e control treated gr oup. (B)Values are non-standardized and expressed as PLN levels. Once we knew that the catalytic molecu les were highly active, the cloned DNA versions of the PLN ribozymes in the AAV vectors were used for a cell culture
103 experiment involving co-transfection. In th is experiment, the co-transfections were carried out using either a 1:1 or a 1:4 molar ratio of PLN expressing plasmid to ribozyme expressing plasmid. Cells were ha rvested and RNA was extracted 48 hours post-transfection. Following RT-PCR and pol yacrylamide gel electrophoresis, gels were stained with SYBR Green and the Phosphorimag er was utilized to numerically assess a value for each set of experimental conditi ons. PLN mRNA levels were left either unstandardized, (i.e., graphing th e raw PLN values) or expressed as a ratio with GAPDH. Unfortunately, the PLN mRNA levels did not show any reduction as compared to a control ribozyme (see Figure A-4). Although we were unsuccessful at reduci ng PLN mRNA levels by co-transfection with the AAV plasmids, we still decided to te st the efficacy of the viral constructs on primary neonatal ventricular rat cardiomyocytes (PNRVC). These cells were seeded at a high density and infected with the Rz-AAV constructs using 1x103 vector genomes per cell. The pTR-UF11 plasmid expressing GFP was used as a control for the efficiency of infection in addition to servi ng as a control plasmid. The extracted proteins were run on a Laemlli gel and an immunoblot was used to assess any changes in PLN protein levels (Figure A-5). Lanes 1Â–2 correspond to two sets of cells that were tr eated with Rz 201. Similarly, infections were done in duplicate with Rz 304 (Lanes 3Â–4) as well as with the GFP only virus used as a control (Lanes 5Â–6) . As seen in Figure A-5, there was no detectable reduction in the am ount of PLN protein present in these infected myocytes 3 days post-infection. Seeing that the ribozymes did not appear to be active in cells, we decided to test the efficiency of the prepared AAV viruses at in fecting the myocytes, very high levels of
104 45 KDa 97 KDa 30 KDa PLN Rz 201 PLN Rz 304GFP Actin PLN 45 KDa 97 KDa 30 KDa PLN Rz 201 PLN Rz 304GFP Actin Actin PLN123456 45 KDa 97 KDa 30 KDa PLN Rz 201 PLN Rz 304GFP Actin PLN 45 KDa 97 KDa 30 KDa PLN Rz 201 PLN Rz 304GFP Actin Actin PLN123456 Figure A-5. Immunoblot of PNRVC infect ed with AAV packaged PLN Ribozymes. PNRVC were infected with 1x103 vg/cell of AAV1 containing either PLN Rz 201, Rz 304 or pTR-UF11. Cells were harvested 3 days post-infection and equal loading of protein lysate was monitored by Actin levels. AAV were utilized to infect PNRV C for four days. Following infection with 1x104 vg/cell, indirect immunofluor escence was carried out to determine the extent of the infection. Using an RFP only virus, seen in Figure A-6 panel A, we were able to observe a very small percentage of cells expressing th e RFP protein. This construct has the RFP gene driven by the robust CBA promoter. Ho wever, even infecting with such a high number of vector genomes, the Rz 304 plasmid expressing RFP through an IRES sequence did not result in any red cells even after 4 days of in fection. Possibly, for this reason, we were not able to see any decrea se in the amount or intensity of the PLN staining (Figure A-6 panel B, green). Cells that appear to be lacking staining for PLN were later confirmed to be contaminating fi broblasts and hence, l acking PLN naturally. In vivo studies were also disappointing. Upon harvesting the hearts of the mice that had received a direct injection, we observed a major immune response in virtually all animals assessed. All of the constructs injected, Lactated RingerÂ’s, an AAV virus containing the LacZ gene, or an AAV virus containing the PLN ribozymes resulted in a similar finding.
105 PLNDAPIRFPMergedA B PLNDAPIRFPMerged PLNDAPIRFPMergedA B PLNDAPIRFPMerged Figure A-6. Primary neonatal ventricu lar rat cardiomyocytes infected 1x103 vg/cell of AAV 4 days post-infection. A) RFP onl y virus contains the RFP gene under the expression of the CBA promoter. B) Rz 304 virus contains the Rz driven by the CBA promoter/CMV ie enha ncer followed by an IRES-RFP. Discussion and Conclusions We were able to design two PLN ribozym e constructs that had robust activity in vitro. However, translating the success of th ese ribozymes into cell lines, primary cells, and animals proved to be a huge obstacle. Although the ribozymes were extremely active in the cell free system against oligonucleotid e targets, they were inactive against full length mRNA in cells. This lack of function could be due to secondary structure of the mRNA that was not predicted by MFOLD or it could be due to low activity of the ribozyme under cellular conditions. In addition, evidence began to ar ise in the literature that would suggest a certain amount of toxi city associated with the red fluorescent protein. We did see evidence for an immune response in mouse hearts that had been treated with our constructs. However, sim ilar if not worse infiltration was seen with constructs expressing LacZ and even some animals injected with Lactated RingerÂ’s solution. This leads us to believe that the injection technique itself might have been too harsh for the animals and that the needle tr acts (2 for each heart) might have been the
106 primary sight of inflammation. It is also possible that the volume injected into each animal (2, 35 ÂµL injections) was not suitabl e for an adult animal with a developed immune system. Upon transfection with the DNA construc ts expressing RFP and the ribozymes, high levels of RFP could repeatedly be visu alized. However, following the packaging of these constructs into AAV partic les, no RFP was ever observed. After much debate as to the causes of these inconsistencies, it was brought to our attention that there had been substantial setbacks in the V ector Core where our virus was being produced. It seems that problems in the titering system led to ex aggerated particle titer s. It is therefore possible that this could be the reason for the lack of RFP expression seen under any conditions in our experiments. In conclusion, it would definitely be of interest to revisit these molecules and perhaps package the ribozyme cassettes into the self-complementary AAV viruses. These constructs could potentially provide th e desired expression of the ribozyme, while at the same time express in a rapid time frame.
107 APPENDIX B SELF-COMPLEMENTARY AAV MOLECULES IN THE HEART Introduction As mentioned in the intr oduction (Chapter 1), the rate limiting step for AAV expression is the synthesis of the second st rand of the genome and therefore, McCarty and co-workers have devised a se lf-complementary version of AAV 122. Although several groups have already begun to harness these so-called self complementary vectors 123,124, most of them utilize the CMV promoter to drive the expression of the gene of interest. Here we describe a vector that has been created to c ontain the CBA promoter with the CMV immediate early (ie) enhan cer followed by the humanized GFP (GFPh) gene. No such self-complementary vector ha s been reported in the literature for cardiac tissue. Rapid expression of specific genes in the heart could tremendously aid the efforts of researches in designing gene therapies for cardiac disease. Materials and Methods Two different GFP expressing ve ctors were utilized for this study. The first is the traditional pTR-UF11 vector containing 4.3 kb of material to be packaged into the AAV capsid pictured in Figure B-1. This vector contains the -actin exon and the corresponding 924 bp intron. Downstream of this is the GFPh gene followed by a 1099 bp cassette flanked with SalI cut sites. This cassette is used for Neomycin resistance. The second plasmid is very similar to pTR-UF11 in that it also expresses the GFPh gene under the expression of the CBA promoter with the CMV ie e nhancer pictured in Figure B-2. This vector has been termed sc-trs -SB-smCBA-hGFP or scGFP. The construct has
108 had the SalI cassette removed and has a minimal intron containing only the essential splice donor/splice acceptor si tes necessary for mRNA expo rt. In comparison to the intron found in the pTR-UF11 vector, the chim eric intron found in the scGFP vector is only 202 bp. This brings the total region to be packaged to a mere 2,438 bp making it suitable to be packaged as a double-stra nded or self-complementary molecule. ApR ColE1 ori f1(+) origin TR TR GFPh neoR PYF441 enhancer CMV ieenhancer Intron SV40 poly(A) bGHpoly(A) HSV-tk ChikenÃŸ-actinpromoter Exon1 Sal I (2861) Sal I (3960) ApR ColE1 ori f1(+) origin TR TR GFPh neoR PYF441 enhancer CMV ieenhancer Intron SV40 poly(A) bGHpoly(A) HSV-tk ChikenÃŸ-actinpromoter Exon1 Sal I (2861) Sal I (3960) Figure B-1. The pTR-UF11 plasmid. This plasmid contains a CBA promoter, an enhancer, an exon and a 924 bp intron. Following this sequence is a 1099 bp cassette used to establish Neomycin resistance. Amp R GFPh CMV i.e. enhancer Chimericintron TR TR bGH-Poly(A) SV40 poly(A) LftTR-seq-for ChikenÃŸ-actinpromoter Exon1 Amp R GFPh CMV i.e. enhancer Chimericintron TR TR bGH-Poly(A) SV40 poly(A) LftTR-seq-for ChikenÃŸ-actinpromoter Exon1 Figure B-2. The scGFP plasmid. This plas mid contains a CBA promoter, an enhancer, an exon and a 202 bp chimeric intron . This construct lacks the 1099 bp Neomycin cassette and cont ains a defective 5Â’ TR. Both of these vectors were prepared and packaged by the Ophthalmology vector core at the University of Florida. S ub-xiphoid injections we re performed using 1.81x1011 vg per animal into 4 day old CD-1 mice from Charles Rivers following the protocol described in Chapter 2. Animals were allowed to incubate for 5, 10 or 14 days.
109 Hearts, lungs and livers were harvested and prepared for endogenous GFP fluorescence as described in Chapter 2. Results As a pilot study, several animals were in jected to study the ex pression pattern of the pTR-UF11 virus compared to the scGFP viru s. The first animal injected with the scGFP virus showed high levels of GFP expressi on after only 5 days. As seen in Figure B-3, the GFP expression was detected from th e apex to the most superior regions (base) of the heart. Several sections from the same heart (Series IÂ–III) were taken at least 50 Âµm Figure B-3. Murine heart 5 days post-infecti on with scGFP. The heart of an injected mouse was excised and cut into thirds (Base, Center, Apex). Subsequently, the 3 pieces were frozen in one block a nd sectioned. Each series represents a different section, at least 50 Âµm apart, from the same block.
110 apart to assess whether the GFP expressi on was a localized event or a general phenomenon. As seen in Figure B-3, each of the series depicted shows an equal amount of GFP expression with the highest levels obser ved in the apex and the atria in the region of the base. At this same timepoint, vigorous expression of the GFP vector was also seen Lung Liver Lung Liver Figure B-4. Representative areas of murine liver and l ung 5 days post-infection with scGFP. AÂ–B show representative GFP expression in both the liver and the lung. CÂ–D are the same sections counterstained with DAPI. A B C D
111 in the liver while only traces of GFP expression could be seen in the lung (Figure B-4). The sections were counterstained with DAPI to verify the presence of healthy tissue. The second timepoint assessed was 10 days post-injection. At this timepoint (Figure B-5), the scGFP animal had even higher levels of expression th an seen at day 5. The GFP expression in this animal was mo re evenly distributed among the 3 lateral regions of the heart while the apical region still displayed the hi ghest levels of GFP fluorescence. Just as in the day 5 animal, the day 10 animal showed a propensity for the virus to associate with the cells of the atria. Figure B-5. Murine heart 10 da ys post-infection with scGFP. The heart of an injected mouse was excised and cut into thirds (Base, Center, Apex). Subsequently, the 3 pieces were frozen in one block a nd sectioned. Each series represents a different section, at least 50 Âµm apart, from the same block.
112 scGFPpTR-UF11 A B C D Figure B-6. Murine heart 10 days post-infec tion with either scGFP or pTR-UF11. The heart of an injected mouse was excised and cut into thirds. Subsequently, the 3 pieces were frozen in one block and s ectioned. Using confocal microscopy, the apical regions of inje cted animals were compared. The mouse injected with scGFP expressed robust amount s of GFP (A) while the pTR-UF11 animal only had a few cells expressing GFP (B). The bottom panels are the DAPI staining of the same sections (CÂ–D). A head to head comparison at this time point between pTR-UF11, the traditional single-stranded AAV and scGFP, highlight the remarkable diffe rences that exist between the 2 viruses. Whereas the pTR-UF11 virus only showed a handful of cells expressing by day 10, the scGFP infected heart was saturate d with GFP expression. Figure B-6 shows
113 representative areas of the h earts of these two animals ta ken by confocal microscopy. Interestingly, in data not shown, the amount of GFP expression seen in the liver at this timepoint was greatly reduced in the scGFP an imal as compared to the 5 day timepoint. The last timepoint that wa s assessed was 14 days post-infection. Although the expression of GFP was still strong using the sc GFP (Figure B-7), the levels appeared to have diminished slightly from the 10 day tim epoint. As can be seen in Figure B-7 (superior region) the levels of GFP expression seen in the at rial appendage were amongst the highest observed to date. The levels of GFP in the liver and lung, however were still comparable to that seen at 10 days post-infection. Apex Center BaseSeries ISeries IISeries III Apex Center BaseSeries ISeries IISeries III Figure B-7. Murine heart 14 da ys post-infection with scGFP. The heart of an injected mouse was excised and cut into thirds (Base, Center, Apex). Subsequently, the 3 pieces were frozen in one block a nd sectioned. Each series represents a different section, at least 50 Âµm apart, from the same block.
114 scGFPpTR-UF11A.B. C.D. scGFPpTR-UF11AB CD scGFPpTR-UF11A.B. C.D. scGFPpTR-UF11AB CD Figure B-8. Murine heart 14 days post-infec tion with either scGFP or pTR-UF11. The heart of an injected mouse was excised and cut into thirds. Subsequently, the 3 pieces were frozen in one block and s ectioned. Using confocal microscopy, the apical regions of inje cted animals were compared. The mouse injected with scGFP expressed robust amount s of GFP (A) while the pTR-UF11 animal only had a few cells expressing GFP (B). Panels C and D are the DAPI staining for the same sections. The comparison of pTR-UF11 and scGFP at the 14 day timepoint using confocal microscopy can be seen in Figure B-8. A lthough the GFP signal is a bit reduced in the
115 scGFP panels as compared to day 10, there is still a greater amount of expression than what was obtained with the si ngle-stranded pTR-UF11 virus. Discussion and Conclusions The use of small molecules, such as siRNAs and ribozymes, is increasingly becoming more and more popular as potential avenues for ther apy. However, one of the major limitations holding back this Â“small-molecule revolutionÂ” is a means to efficiently deliver these molecules. For a long time, AAV has been a favored vector for therapy because of its lack of pathogenicity, its abil ity to infect dividing and non-dividing cells, and its ability to express for long periods of time. However, its lengthy lead time before the onset of expression has been a setback for developing new therapies in the laboratory. We believe that the scAAV constructs are a major step forward in dealing with these setbacks. Despite its small payload (2.4 kb), scAAV has become an option for a specialized group of researcher s. This construct can easily accommodate a reporter gene cassette as well as an shRNA expression cassette. In this study, we performed a comparison between the expression time required for the traditional single-stranded AAV c onstruct and that required for the self-complementary AAV. Although expression of recombinant prot eins in cardiac tissue has long been difficult to accomplish, we feel that the robust expre ssion levels that the self-complementary virus provides after only 5 days is very promising. The expression of the scGFP virus markedly increased from da y 5 to day 10. On average, the amount of expression from the scGFP virus seen at da y 10 covered at least 40% of the entire myocardium. This is an unprecedented amount of expression for the heart using an AAV-GFP construct.
116 Although this study only monitored the expres sion levels up to 14 days, we suspect that based on the long-term expression of AAV routinely reported in the literature, this self-complementary construct would have sim ilar properties. The slight decrease that was seen at day 14 as compared to day 10 coul d be attributed to the small sampling that was used for this study. Further studies are warranted with larger cohorts of animals for each time point. One potential concern is the level of G FP expression that was seen in the liver following injections into the th oracic cavity. This should be a maintained as a point of vigilance should these studies continue forward. Extremel y high levels of virus were observed in the liver after only 5 days. Nearly all of the hepatocytes observed at this time point appeared to be expressing GFP at very hi gh levels. This subs tantially decreased to about 40% on days 10 and 14. Therefore, care should be taken for future studies involving the use of this vector system. Neither of the time points revealed any elevated levels of GFP within the lung. Interestingly, for reasons unknown, all of th e time points expresse d very high levels of GFP in the structure re presenting the atrial appenda ge. Although the scGFP virus definitely had extremely high levels, especi ally by day 14, the pTR-UF11 virus also showed a preference for this region of the ti ssue. Perhaps viral tropism needs to be addressed more in depth not onl y for the differences that o ccur in different tissues but also relating to the variability within different regions of the same ti ssue. Additionally, it is possible that when the atri a are in diastole th e virus remains there long enough to allow for a more efficient infection to occur in these specialized cardiac cells.
117 In conclusion, the scGFP construct was s uperior to the traditional single-stranded AAV vector system in delivering a gene of in terest, GFP. Although further studies need to be conducted to determine when the highest level of expression o ccurs and the extent of the non-specific tissue tropism by the viru s, we believe that scAAV viruses will be very useful not only to the cardiac research co mmunity, but to gene therapists at large.
118 LIST OF REFERENCES 1. Lodish,H. 86 A.D. Molecular Cell Biology. W.H. Freeman, New York, NY. 2. Bers, D.M. 2002. Cardiac excitation -contraction coupling. Nature 415:198-205. 3. Calcium pumps: http://www.ebi.ac.uk/interpro/potm/2004_3/Page2.htm. European Bioinformatics Institute. 2006. 4. Sakuntabhai, A., V. Ruiz-Perez, S. Carter, N. Jacobsen, S. Burge, S. Monk, M. Smith, C.S. Munro, M. O'Donovan, N. Cra ddock, R. Kucherlapati, J.L. Rees, M. Owen, G.M. Lathrop, A.P. Monaco, T. Strachan, and A. Hovnanian. 1999. Mutations in ATP2A2, encoding a Ca2+ pum p, cause Darier disease. Nat.Genet. 21:271-277. 5. Kirichok, Y., G. Krapivinsky, and D.E. Clapham. 2004. The mitochondrial calcium uniporter is a highly selectiv e ion channel. Nature 427:360-364. 6. Marks, A.R. 2003. Calcium and the heart: a question of life and death. J.Clin.Invest 111:597-600. 7. Hoshijima, M. 2005. Gene therapy target ed at calcium handling as an approach to the treatment of heart failure. Pharmacol.Ther. 105:211-228. 8. Katz, A.M. and D.I. Repke. 1973. Calcium-membrane interactions in the myocardium: effects of ouabain, epin ephrine and 3',5'-cyclic adenosine monophosphate. Am.J.Cardiol. 31:193-201. 9. Morkin, E. and P.J. LaRaia. 1974. Biochemical studies on the regulation of myocardial contractility. N.Engl.J.Med. 290:445-451. 10. Schulze, W., E.G. Krause, and A. Wollenberger. 1972. Cytochemical demonstration and localization of adenyl cyclase in skeletal and cardiac muscle. Adv.Cyclic.Nucleotide.Res. 1:249-260. 11. Wray, H.L., R.R. Gray, and R.A. Olsson. 1973. Cyclic adenosine 3',5'monophosphate-stimulated protein kinase a nd a substrate associ ated with cardiac sarcoplasmic reticulum. J.Biol.Chem. 248:1496-1498. 12. Tada, M., M.A. Kirchberger, and A .M. Katz. 1975. Phosphorylation of a 22,000dalton component of the cardiac sarcopl asmic reticulum by adenosine 3':5'monophosphate-dependent protein ki nase. J.Biol.Chem. 250:2640-2647.
119 13. Maclennan, D.H. and E.G. Kranias. 2003. Phospholamban: a crucial regulator of cardiac contractility. Nat.Rev.Mol.Cell Biol. 4:566-577. 14. Mueller, B., C.B. Karim, I.V. Negrashov, H. Kutchai, and D.D. Thomas. 2004. Direct detection of phospholamban and sarcoplasmic reticulum Ca-ATPase interaction in membranes using fluor escence resonance energy transfer. Biochemistry 43:8754-8765. 15. Song, Q., K.B. Young, G. Chu, J. Gu lick, M. Gerst, I.L. Grupp, J. Robbins, and E.G. Kranias. 2004. Overexpression of phospholamban in slow-twitch skeletal muscle is associated with depressed c ontractile function and muscle remodeling. FASEB J 18:974-976. 16. Chu, G., J.W. Lester, K.B. Young, W. Luo, J. Zhai, and E.G. Kranias. 2000. A single site (Ser16) phosphoryl ation in phospholamban is su fficient in mediating its maximal cardiac respons es to beta -agonists. J.Biol.Chem. 275:38938-38943. 17. Simmerman, H.K., D.E. Lovelace, and L.R. Jones. 1989. Secondary structure of detergent-solubilized phospholamban, a phosphorylatable, oligomeric protein of cardiac sarcoplasmic reticulum. Biochim.Biophys.Acta 997:322-329. 18. Oxenoid, K. and J.J. Chou. 2005. The structure of phospholamban pentamer reveals a channel-like arch itecture in membranes. Pr oc.Natl.Acad.Sci.U.S.A 102:10870-10875. 19. Kimura, Y., K. Kurzydlowski, M. Tada, and D.H. Maclennan. 1997. Phospholamban inhibitory function is activated by depolymerization. J.Biol.Chem. 272:15061-15064. 20. Thom, T., N. Haase, W. Rosamond, V.J. Howard, J. Rumsfeld, T. Manolio, Z.J. Zheng, K. Flegal, C. O'Donnell, S. Kittn er, D. Lloyd-Jones, D.C. Goff, Jr., Y. Hong, R. Adams, G. Friday, K. Furie, P. Gorelick, B. Kissela, J. Marler, J. Meigs, V. Roger, S. Sidney, P. Sorlie, J. Stei nberger, S. Wasserthiel-Smoller, M. Wilson, and P. Wolf. 2006. Heart disease and str oke statistics--2006 upda te: a report from the American Heart Association Statis tics Committee and Stroke Statistics Subcommittee. Circulation 113:e85-151. 21. Myocardial Ischemia: http://www.americanheart.org/presen ter.jhtml?identifier=251. 2006. 22. Stroke: http://www.niapublications .org/agepages/stroke.asp. U.S.National Institutes of Health. 12-20-2005. 23. Diabetes and heart disease: http://diabetes.niddk.nih.gov/dm/pubs/comp lications_heart/#3. National Diabetes information clearinghouse. 2006.
120 24. Peripheral Vascular Disease: American Heart Association. 2006. http://www.americanheart.org/pr esenter.jhtml? identifier=4692. 25. Cardiac, coronary artery disease: Columbia University Medical Center. http://www.columbiasurgery.org/pat/cardiac/chd.html. 2006. 26. Sun, Y.L., S.J. Hu, L.H. Wang, Y. Hu, and J.Y. Zhou. 2005. Effect of betablockers on cardiac function and calcium handling protein in postinfarction heart failure rats. Chest 128:1812-1821. 27. Pahor, M., J.M. Guralnik, L. Ferrucci, M.C. Corti, M.E. Salive, J.R. Cerhan, R.B. Wallace, and R.J. Havlik. 1996. Calcium-channel blockade and incidence of cancer in aged populations. Lancet 348:493-497. 28. Fitzpatrick, A.L., J.R. Daling, C.D. Fu rberg, R.A. Kronmal, and J.L. Weissfeld. 1997. Use of calcium channel blockers and breast carcinoma risk in postmenopausal women. Cancer 80:1438-1447. 29. Rosenberg, L., R.S. Rao, J.R. Palmer, B.L. Strom, P.D. Stolley, A.G. Zauber, M.E. Warshauer, and S. Shapiro. 1998. Ca lcium channel blockers and the risk of cancer. JAMA 279:1000-1004. 30. Pahor, M., J.M. Guralnik, C.D. Furberg, P. Carbonin, and R. Havlik. 1996. Risk of gastrointestinal haemo rrhage with calcium antagonist s in hypertensive persons over 67 years old. Lancet 347:1061-1065. 31. The Four Groups of Useful Diuretic s: Vocably Innovative Vocabulary Builder. http://self-help.vocaboly.com/archives /384/the-four-groups-o f-useful-diuretics/ . 32. Foody, J.M., M.H. Farrell, and H.M. Krumholz. 2002. beta-Blocker therapy in heart failure: scientific review. JAMA 287:883-889. 33. Olsen, S.L., E.M. Gilbert, D.G. Re nlund, D.O. Taylor, F.D. Yanowitz, and M.R. Bristow. 1995. Carvedilol improves left ventricular function and symptoms in chronic heart failure: a d ouble-blind randomized study. J.Am.Coll.Cardiol. 25:1225-1231. 34. Cohn, J.N., T.B. Levine, M.T. Olivari, V. Garberg, D. Lura, G.S. Francis, A.B. Simon, and T. Rector. 1984. Plasma nor epinephrine as a guide to prognosis in patients with chronic congestive heart failure. N.Engl.J.Med. 311:819-823. 35. Frank, K.F., B. Bolck, K. Brixius, E.G. Kranias, and R.H. Schwinger. 2002. Modulation of SERCA: implicati ons for the failing human heart. Basic Res.Cardiol. 97 Suppl 1:I72-I78. 36. Gilbert, E.M., W.T. Abraham, S. Olsen, B. Hattler, M. White, P. Mealy, P. Larrabee, and M.R. Bristow. 1996. Comp arative hemodynamic, left ventricular
121 functional, and antiadrenergic effects of ch ronic treatment with metoprolol versus carvedilol in the failing heart. Circulation 94:2817-2825. 37. Vargo, D. Gene Therapy: A Brief Hist ory: Citizen Link Forcus on Social Issues. http://www.family.org/cforum/fosi/bioethi cs/genetics/a0032608.cfm. 6-23-2004. 38. Blaese, R.M., K.W. Culver, A.D. Miller , C.S. Carter, T. Fleisher, M. Clerici, G. Shearer, L. Chang, Y. Chiang, P. Tolstosh ev, J.J. Greenblatt, S.A. Rosenberg, H. Klein, M. Berger, C.A. Mullen, W.J. Ramsey, L. Muul, R.A. Morgan, and W.F. Anderson. 1995. T lymphocyte-directed gene therapy for ADASCID: initial trial results after 4 years. Science 270:475-480. 39. Onodera, M., T. Ariga, N. Kawamura, I. Kobayashi, M. Ohtsu, M. Yamada, A. Tame, H. Furuta, M. Okano, S. Matsum oto, H. Kotani, G.J. McGarrity, R.M. Blaese, and Y. Sakiyama. 1998. Successf ul peripheral T-lymphocyte-directed gene transfer for a patient with severe combined immune deficiency caused by adenosine deaminase deficiency. Blood 91:30-36. 40. Cavazzana-Calvo, M., S. Hacein-Bey, B.G. de Saint, F. Gross, E. Yvon, P. Nusbaum, F. Selz, C. Hue, S. Certain, J.L. Casanova, P. Bousso, F.L. Deist, and A. Fischer. 2000. Gene therapy of hu man severe combined immunodeficiency (SCID)-X1 disease. Science 288:669-672. 41. Hacein-Bey-Abina, S., C. Von Kalle, M. Schmidt, M.P. McCormack, N. Wulffraat, P. Leboulch, A. Lim, C.S. Os borne, R. Pawliuk, E. Morillon, R. Sorensen, A. Forster, P. Fraser, J.I. C ohen, B.G. de Saint, I. Alexander, U. Wintergerst, T. Frebourg, A. Aurias, D. Stoppa-Lyonnet, S. Romana, I. RadfordWeiss, F. Gross, F. Valensi, E. Delabesse, E. Macintyre, F. Sigaux, J. Soulier, L.E. Leiva, M. Wissler, C. Prinz, T.H. Ra bbitts, F. le Deist, A. Fischer, and M. Cavazzana-Calvo. 2003. LMO2-associated clonal T cell proliferation in two patients after gene th erapy for SCID-X1. Science 302:415-419. 42. Thompson, L. Human Gene Th erapy Harsh Lessons, High Hopes. FDA Consumer Magazine. 2000. 43. Verma, I.M. and N. Somia. 1997. Gene therapy -promises, problems and prospects. Nature 389:239-242. 44. Verma, I.M. and M.D. Weitzman. 2005. Gene therapy: twenty-first century medicine. Annu.Rev.Biochem. 74:711-738. 45. Kay, M.A., C.S. Manno, M.V. Ragni, P.J. Larson, L.B. Couto, A. McClelland, B. Glader, A.J. Chew, S.J. Tai, R.W. Herzog, V. Arruda, F. Johnson, C. Scallan, E. Skarsgard, A.W. Flake, and K.A. High. 2000. Evidence for gene transfer and expression of factor IX in haemophilia B patients treated with an AAV vector. Nat.Genet. 24:257-261.
122 46. Blomer, U., L. Naldini, T. Kafri, D. Trono, I.M. Verma, and F.H. Gage. 1997. Highly efficient and sustaine d gene transfer in adult neurons with a lentivirus vector. J.Virol. 71:6641-6649. 47. Douglas, J., P. Kelly, J.T. Evans, and J.V. Garcia. 1999. Efficient transduction of human lymphocytes and CD34+ cells vi a human immunodeficiency virus-based gene transfer vectors. Hum.Gene Ther. 10:935-945. 48. Muruve, D.A., M.J. Cotter, A.K. Zaiss, L.R. White, Q. Liu, T. Chan, S.A. Clark, P.J. Ross, R.A. Meulenbroek, G.M. Mael andsmo, and R.J. Parks. 2004. Helperdependent adenovirus vectors elicit inta ct innate but attenuated adaptive host immune responses in vivo. J.Virol. 78:5966-5972. 49. Epstein, A.L. 2005. HSV-1-deri ved recombinant and amplicon vectors for preventive or therapeutic gene transfer: an overview. Gene Ther 12 Suppl 1:S153. 50. Fisher, K.J., K. Jooss, J. Alston, Y. Yang, S.E. Haecker, K. High, R. Pathak, S.E. Raper, and J.M. Wilson. 1997. Recombinan t adeno-associated virus for muscle directed gene therapy. Nat.Med. 3:306-312. 51. Flotte, T.R. 2005. Recent developments in recombinant AAV-mediated gene therapy for lung diseases. Curr.Gene Ther. 5:361-366. 52. Loiler, S.A., T.J. Conlon, S. Song, Q. Tang, K.H. Warrington, A. Agarwal, M. Kapturczak, C. Li, C. Ricordi, M.A. Atki nson, N. Muzyczka, and T.R. Flotte. 2003. Targeting recombinant adeno-associ ated virus vectors to enhance gene transfer to pancreatic islets and liver. Gene Ther. 10:1551-1558. 53. Bennett, J. 2003. Immune response fo llowing intraocular deli very of recombinant viral vectors. Gene Ther. 10:977-982. 54. Conrad, C.K., S.S. Allen, S.A. Afi one, T.C. Reynolds, S.E. Beck, M. Fee-Maki, X. Barrazza-Ortiz, R. Adams, F.B. Askin, B.J. Carter, W.B. Guggino, and T.R. Flotte. 1996. Safety of single-dose admini stration of an adeno-associated virus (AAV)-CFTR vector in the primate lung. Gene Ther. 3:658-668. 55. Hernandez, Y.J., J. Wang, W.G. Kearns , S. Loiler, A. Poirier, and T.R. Flotte. 1999. Latent adeno-associat ed virus infection elicits humoral but not cellmediated immune responses in a nonhuman primate model. J.Virol. 73:85498558. 56. Chirmule, N., W. Xiao, A. Truneh, M.A. Schnell, J.V. Hughes, P. Zoltick, and J.M. Wilson. 2000. Humoral immunity to adeno-associated virus type 2 vectors following administration to muri ne and nonhuman primate muscle. J.Virol. 74:2420-2425.
123 57. Kaspar, B.K., D.M. Roth, N.C. Lai, J.D. Drumm, D.A. Erickson, M.D. McKirnan, and H.K. Hammond. 2005. Myocardial ge ne transfer and long-term expression following intracoronary delivery of adeno-associated virus. J.Gene Med. 7:316324. 58. Bessis, N., F.J. GarciaCozar, and M.C. Boissier. 2004. Immune responses to gene therapy vectors: influence on vector function and effector mechanisms. Gene Ther. 11 Suppl 1:S10-S17. 59. Zolotukhin, S., M. Potter, I. Zolotukhin, Y. Sakai, S. Loiler, T.J. Fraites, Jr., V.A. Chiodo, T. Phillipsberg, N. Muzyczka, W.W. Hauswirth, T.R. Flotte, B.J. Byrne, and R.O. Snyder. 2002. Pr oduction and purification of serotype 1, 2, and 5 recombinant adeno-associated viral vectors. Methods 28:158-167. 60. Wang, Z., T. Zhu, C. Qiao, L. Zhou, B. Wang, J. Zhang, C. Chen, J. Li, and X. Xiao. 2005. Adeno-associated virus sero type 8 efficiently delivers genes to muscle and heart. Nat.Biotechnol. 23:321-328. 61. Du, L., M. Kido, D.V. Lee, J.E. Ra binowitz, R.J. Samulski, S.W. Jamieson, M.D. Weitzman, and P.A. Thistlethwaite. 2004. Differential myocardial gene delivery by recombinant serotype-specific ad eno-associated viral vectors. Mol.Ther. 10:604-608. 62. Hildinger, M., A. Auricchio, G. Gao, L. Wang, N. Chirmule, and J.M. Wilson. 2001. Hybrid vectors based on adeno-asso ciated virus serotypes 2 and 5 for muscle-directed gene transfer. J.Virol. 75:6199-6203. 63. Wu, Z., A. Asokan, and R.J. Samulski . 2006. Adeno-associated Virus Serotypes: Vector Toolkit for Human Gene Therapy. Mol.Ther. 64. Chao, H., Y. Liu, J. Rabinowitz, C. Li, R.J. Samulski, and C.E. Walsh. 2000. Several log increase in therapeutic transgen e delivery by distinct adeno-associated viral serotype vectors. Mol.Ther. 2:619-623. 65. Rabinowitz, J.E., F. Rolling, C. Li, H. Conrath, W. Xiao, X. Xiao, and R.J. Samulski. 2002. Cross-packaging of a si ngle adeno-associated virus (AAV) type 2 vector genome into multiple AAV sero types enables transduction with broad specificity. J.Virol. 76:791-801. 66. Yang, G.S., M. Schmidt, Z. Yan, J. D. Lindbloom, T.C. Harding, B.A. Donahue, J.F. Engelhardt, R. Kotin, and B.L. Davi dson. 2002. Virus-mediated transduction of murine retina with adeno-associated virus: effects of viral capsid and genome size. J.Virol. 76:7651-7660. 67. Peden, C.S., C. Burger, N. Muzyczka, and R.J. Mandel. 2004. Circulating antiwild-type adeno-associated virus type 2 (AAV2) antibodies inhibit recombinant AAV2 (rAAV2)-mediated, but not rAAV5-media ted, gene transfer in the brain. J.Virol. 78:6344-6359.
124 68. Nakai, H., T.A. Storm, and M.A. Kay. 2000. Increasing the size of rAAVmediated expression cassettes in vi vo by intermolecular joining of two complementary vectors. Nat.Biotechnol. 18:527-532. 69. Yan, Z., T.C. Ritchie, D. Duan, and J.F. Engelhardt. 2002. Recombinant AAVmediated gene delivery using dua l vector heterodimerization. Methods Enzymol. 346:334-357. 70. Duan, D., Y. Yue, and J.F. Engelhardt. 2003. Dual vector expansion of the recombinant AAV packaging capacity. Methods Mol.Biol. 219:29-51. 71. Kotin, R.M., M. Siniscalco, R.J. Samulski, X.D. Zhu, L. Hunter, C.A. Laughlin, S. McLaughlin, N. Muzyczka, M. Rocchi, and K.I. Berns. 1990. Site-specific integration by adeno-associated virus. Proc.Natl.Acad.Sci.U.S.A 87:2211-2215. 72. Schnepp, B.C., R.L. Jensen, C.L. Chen, P.R. Johnson, and K.R. Clark. 2005. Characterization of adeno-associated viru s genomes isolated from human tissues. J.Virol. 79:14793-14803. 73. Ferrari, F.K., T. Samulski, T. Shenk, and R.J. Samulski. 1996. Second-strand synthesis is a rate-limiting step for e fficient transduction by recombinant adenoassociated virus vectors. J.Virol. 70:3227-3234. 74. Fisher, K.J., G.P. Gao, M.D. Weitzman, R. DeMatteo, J.F. Burda, and J.M. Wilson. 1996. Transduction with recombinant adeno-associated virus for gene therapy is limited by leading-strand synthesis. J.Virol. 70:520-532. 75. McCarty, D.M., P.E. Monahan, and R.J. Samulski. 2001. Self-complementary recombinant adeno-associated virus (scAAV) vectors promote efficient transduction independently of DNA synthesis. Gene Ther. 8:1248-1254. 76. Fire, A., S. Xu, M.K. Montgomery, S.A. Kostas, S.E. Driver, and C.C. Mello. 1998. Potent and specific genetic in terference by double-stranded RNA in Caenorhabditis elegans. Nature 391:806-811. 77. Napoli, C., C. Lemieux, and R. Jo rgensen. 1990. Introduction of a Chimeric Chalcone Synthase Gene into Petunia Results in Reversible Co-Suppression of Homologous Genes in trans. Plant Cell 2:279-289. 78. Zamore, P.D., T. Tuschl, P.A. Sharp, and D.P. Bartel. 2000. RNAi: doublestranded RNA directs the ATP-depende nt cleavage of mRNA at 21 to 23 nucleotide intervals. Cell 101:25-33. 79. Elbashir, S.M., J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, and T. Tuschl. 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494-498.
125 80. Bartel, D.P. 2004. MicroRNAs: genomi cs, biogenesis, mechanism, and function. Cell 116:281-297. 81. Valencia-Sanchez, M.A., J. Liu, G.J. Hannon, and R. Parker. 2006. Control of translation and mRNA degrada tion by miRNAs and siRNAs. Genes Dev. 20:515524. 82. Rand, T.A., S. Petersen, F. Du, and X. Wang. 2005. Argonaute2 cleaves the antiguide strand of siRNA during RISC activation. Cell 123:621-629. 83. Doench, J.G., C.P. Petersen, and P.A. Sharp. 2003. siRNAs can function as miRNAs. Genes Dev. 17:438-442. 84. Olsen, P.H. and V. Ambros. 1999. The lin-4 regula tory RNA controls developmental timing in Caenorhabditis elegans by blocking LIN-14 protein synthesis after the initiation of translation. Dev.Biol. 216:671-680. 85. Saxena, S., Z.O. Jonsson, and A. Dutta. 2003. Small RNAs with imperfect match to endogenous mRNA repress translation. Implications for offtarget activity of small inhibitory RNA in mammalian cells. J.Biol.Chem. 278:44312-44319. 86. Stevenson, M. 2003. Dissecti ng HIV-1 through RNA interference. Nat.Rev.Immunol. 3:851-858. 87. Sui, G., C. Soohoo, e.B. Affar, F. Gay, Y. Shi, W.C. Forrester, and Y. Shi. 2002. A DNA vector-based RNAi technolo gy to suppress gene expression in mammalian cells. Proc.Natl.Acad.Sci.U.S.A 99:5515-5520. 88. Bogenhagen, D.F., S. Sakonju, and D. D. Brown. 1980. A control region in the center of the 5S RNA gene directs specific initiation of transcription: II. The 3' border of the region. Cell 19:27-35. 89. Baer, M., T.W. Nilsen, C. Costigan, and S. Altman. 1990. Structure and transcription of a human gene for H1 RNA, the RNA component of human RNase P. Nucleic Acids Res. 18:97-103. 90. Brummelkamp, T.R., R. Bernards, and R. Agami. 2002. A system for stable expression of short interfering RNAs in mammalian cells. Science 296:550-553. 91. Zeng, Y., X. Cai, and B.R. Cullen. 2005. Use of RNA polymer ase II to transcribe artificial microRNAs. Methods Enzymol. 392:371-380. 92. McCaffrey, A.P., H. Nakai, K. Pande y, Z. Huang, F.H. Salazar, H. Xu, S.F. Wieland, P.L. Marion, and M.A. Kay. 2003. Inhibition of hepatitis B virus in mice by RNA interference. Nat.Biotechnol. 21:639-644.
126 93. Xia, H., Q. Mao, S.L. Eliason, S.Q. Ha rper, I.H. Martins, H.T. Orr, H.L. Paulson, L. Yang, R.M. Kotin, and B.L. Davidson. 2004. RNAi suppresses polyglutamineinduced neurodegeneration in a mo del of spinocerebellar ataxia. Nat Med 10:816820. 94. Xu, D., D. McCarty, A. Fernandes, M. Fisher, R.J. Samulski, and R.L. Juliano. 2005. Delivery of MDR1 small inte rfering RNA by self-complementary recombinant adeno-associated virus vector. Mol Ther 11:523-530. 95. Novina, C.D., M.F. Murray, D.M. Dykxhoorn, P.J. Beresford, J. Riess, S.K. Lee, R.G. Collman, J. Lieberman, P. Shankar, and P.A. Sharp. 2002. siRNA-directed inhibition of HIV-1 infection. Nat.Med. 8:681-686. 96. Porzio, M.A. and A.M. Pearson. 197 7. Improved resolution of myofibrillar proteins with sodium dodecyl sulf ate-polyacrylamide gel electrophoresis. Biochim.Biophys.Acta 490:27-34. 97. Geoffroy, M.C. and A. Salvetti. 2005. Helper functions requ ired for wild type and recombinant adeno-associated virus growth. Curr.Gene Ther. 5:265-271. 98. Chu, G., D.G. Ferguson, I. Edes, E. Kiss, Y. Sato, and E.G. Kranias. 1998. Phospholamban ablation and compensatory responses in the mammalian heart. Ann.N.Y.Acad.Sci. 853:49-62. 99. Minamisawa, S., M. Hoshijima, G. Chu, C.A. Ward, K. Frank, Y. Gu, M.E. Martone, Y. Wang, J. Ross, Jr., E.G. Krania s, W.R. Giles, and K.R. Chien. 1999. Chronic phospholamban-sarcoplasmic reti culum calcium ATPase interaction is the critical calcium cycling def ect in dilated cardiomyopathy. Cell 99:313-322. 100. Hajjar, R.J., J.X. Kang, J.K. Gwathme y, and A. Rosenzweig. 1997. Physiological effects of adenoviral gene transfer of sarcoplasmic reticulum calcium ATPase in isolated rat myocytes. Circulation 95:423-429. 101. He, H., F.J. Giordano, R. HilalDandan, D.J. Choi, H.A. Rockman, P.M. McDonough, W.F. Bluhm, M. Meyer, M.R . Sayen, E. Swanson, and W.H. Dillmann. 1997. Overexpression of the rat sarcoplasmic reticulum Ca2+ ATPase gene in the heart of transgenic mice accelerates calcium transients and cardiac relaxation. J.Clin.Invest 100:380-389. 102. Hoshijima, M., Y. Ikeda, Y. Iwanaga, S. Minamisawa, M.O. Date, Y. Gu, M. Iwatate, M. Li, L. Wang, J.M. Wilson, Y. Wang, J. Ross, Jr., and K.R. Chien. 2002. Chronic suppression of h eart-failure progression by a pseudophosphorylated mutant of phospholam ban via in vivo cardiac rAAV gene delivery. Nat.Med. 8:864-871.
127 103. Iwanaga, Y., M. Hoshijima, Y. Gu, M. Iw atate, T. Dieterle, Y. Ikeda, M.O. Date, J. Chrast, M. Matsuzaki, K.L. Peterson, K.R. Chien, and J. Ross, Jr. 2004. Chronic phospholamban inhibition prevents progressive cardiac dysfunction and pathological remodeling af ter infarction in rats. J.Clin.Invest 113:727-736. 104. Watanabe, A., M. Arai, M. Yamazaki, N. Koitabashi, F. Wuytack, and M. Kurabayashi. 2004. Phospholamban ab lation by RNA interference increases Ca2+ uptake into rat cardiac myocyte sarcoplasmic reticulum. J.Mol.Cell Cardiol. 37:691-698. 105. Shankar, P., N. Manjunath, and J. Li eberman. 2005. The prospect of silencing disease using RNA interference. JAMA 293:1367-1373. 106. Del Monte, F., S.E. Harding, G.W. Dec, J.K. Gwathmey, and R.J. Hajjar. 2002. Targeting phospholamban by gene transfer in human heart failure. Circulation 105:904-907. 107. Li, J., S.J. Hu, J. Sun, Z.H. Zhu, X. Zheng, G.Z. Wang, Y.M. Yao, N.Y. Chen, and X.Y. Zhao. 2005. Constructi on of phospholamban antisense RNA recombinant adeno-associated virus vector and its effects in rat cardiomyocytes. Acta Pharmacol.Sin. 26:51-55. 108. Miyagishi, M., M. Hayashi, and K. Ta ira. 2003. Comparison of the suppressive effects of antisense oligonucleotides and siRNAs directed against the same targets in mammalian cells. Antisense Nucleic Acid Drug Dev. 13:1-7. 109. Bertrand, J.R., M. Pottier, A. Vekris, P. Opolon, A. Maksimenko, and C. Malvy. 2002. Comparison of antisense oligonucleotid es and siRNAs in cell culture and in vivo. Biochem.Biophys.Res.Commun. 296:1000-1004. 110. Meyer, M., D.D. Belke, S.U. Trost, E. Swanson, T. Dieterle, B. Scott, S.P. Cary, P. Ho, W.F. Bluhm, P.M. McDonough, G. J. Silverman, and W.H. Dillmann. 2004. A recombinant antibody increases cardiac contractility by mimicking phospholamban phosphorylation. FASEB J. 18:1312-1314. 111. Grimm, D., K.L. Streetz, C.L. Joplin g, T.A. Storm, K. Pandey, C.R. Davis, P. Marion, F. Salazar, and M.A. Kay. 2006. Fa tality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature 441:537-541. 112. Beal, J. 2005. Silence is golden: can RNA interference therapeutics deliver? Drug Discov.Today 10:169-172. 113. Inagaki, K., S. Fuess, T.A. Storm, G.A. Gibson, C.F. McTiernan, M.A. Kay, and H. Nakai. 2006. Robust systemic transduction with AAV9 vectors in mice: efficient global cardiac gene tran sfer superior to that of AAV8. Mol.Ther. 14:4553.
128 114. Haghighi, K., K.N. Gregory, and E.G. Kranias. 2004. Sarcoplasmic reticulum Ca-ATPase-phospholamban interactions and dilated cardiomyopathy. Biochem.Biophys.Res.Commun. 322:1214-1222. 115. Haghighi, K., F. Kolokathis, A.O. Gramolini, J.R. Waggoner, L. Pater, R.A. Lynch, G.C. Fan, D. Tsiapras, R.R. Pa rekh, G.W. Dorn, D.H. Maclennan, D.T. Kremastinos, and E.G. Kranias. 2006. A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy. Proc.Natl.Acad.Sci.U.S.A 103:1388-1393. 116. Song, Q., A.G. Schmidt, H.S. Hahn, A.N. Carr, B. Frank, L. Pater, M. Gerst, K. Young, B.D. Hoit, B.K. McConnell, K. Ha ghighi, C.E. Seidman, J.G. Seidman, G.W. Dorn, and E.G. Kranias. 2003. Rescue of cardiomyocyte dysfunction by phospholamban ablation does not prevent ventricular failure in genetic hypertrophy. J.Clin.Invest 111:859-867. 117. Janczewski, A.M., M. Zahid, B.H. Lems ter, C.S. Frye, G. Gibson, Y. Higuchi, E.G. Kranias, A.M. Feldman, and C.F. McTiernan. 2004. Phospholamban gene ablation improves calcium transients but not cardiac function in a heart failure model. Cardiovasc.Res. 62:468-480. 118. Amar, L., M. Desclaux, N. Faucon-Bi guet, J. Mallet, and R. Vogel. 2006. Control of small inhibitory RNA le vels and RNA interference by doxycycline induced activation of a minima l RNA polymerase III promoter. Nucleic Acids Res. 34:e37. 119. Rossi, J.J. 1997. Therapeutic a pplications of catalytic antisense RNAs (ribozymes). Ciba Found.Symp. 209:195-204. 120. Lieber, A. and M.A. Kay. 1996. Ad enovirus-mediated expr ession of ribozymes in mice. J.Virol. 70:3153-3158. 121. Li, M.J., J. Kim, S. Li, J. Zaia, J.K. Yee, J. Anderson, R. Akkina, and J.J. Rossi. 2005. Long-term inhibition of HIV-1 infec tion in primary hematopoietic cells by lentiviral vector delivery of a triple combination of anti-HIV shRNA, anti-CCR5 ribozyme, and a nucleolar-localizing TAR decoy. Mol.Ther. 12:900-909. 122. McCarty, D.M., H. Fu, P.E. Monahan, C.E. Toulson, P. Naik, and R.J. Samulski. 2003. Adeno-associated virus terminal repeat (TR) mutant generates selfcomplementary vectors to overcome the rate -limiting step to transduction in vivo. Gene Ther. 10:2112-2118. 123. Fu, H., J. Muenzer, R.J. Samulski, G. Breese, J. Sifford, X. Zeng, and D.M. McCarty. 2003. Self-complementary ade no-associated virus serotype 2 vector: global distribution and broad dispersion of AAV-mediated transgene expression in mouse brain. Mol.Ther. 8:911-917.
129 124. Zhong, L., L. Chen, Y. Li, K. Qing, K.A. Weigel-Kelley, R.J. Chan, M.C. Yoder, and A. Srivastava. 2004. Self-complemen tary adeno-associated virus 2 (AAV)-T cell protein tyrosine phosphatase vectors as helper viruses to improve transduction efficiency of conventional single-stra nded AAV vectors in vitro and in vivo. Mol.Ther. 10:950-957.
130 BIOGRAPHICAL SKETCH Lourdes MarÃa Andino was born in Miami, Florida on April 8th 1977 to Gloria and Rafael Andino. Her parents immigrated to the United States in 1967, escaping the horrors of the communist regime in Cuba. H oping to give their child ren a better way of life and the best possible education, they labo red tirelessly and selflessly. Lourdes MarÃa (the fourth of their well-educated and succe ssful children) is the first to realize a doctorate degree in her exiled family. Her di ligent efforts and unwavering determination are a testament to the dedicat ed upbringing of her parents an d siblings. Truly she is living the classic American dream. She graduated with top honors from Saint Brendan High School in Miami, Florida, and then continued her higher education at Fl orida International University. Here she completed her bachelorÂ’s degree in biology a nd minored in chemistry. Although her love for science began at an early age (picking up mice and lizards bare-handed in her home garden), it wasnÂ’t until her eyes were opene d to the wonders of mo lecular biology in Dr. Rene HerreraÂ’s genetics laboratory that she realized where her future lay. At his lab, she worked as an undergraduate researcher for almost 4 years. Inspired to pursue graduate school by the desire to learn more about science and all of its intricacies and about herself, Lourdes has had an amazing experience in Gainesville. Developing her scientific mind, her soulful spirit, and her artistic heart has been her passion for the last 6 and a half year s. She is currently seeking employment that will challenge and stimulate her love for learning.