Integrating viral vectors as a gene therapy approach for cystic fibrosis

Transcription

1 University of Iowa Iowa Research Online Theses and Dissertations Spring 2018 Integrating viral vectors as a gene therapy approach for cystic fibrosis Ashley L. Cooney University of Iowa Copyright 2018 Ashley L. Cooney This dissertation is available at Iowa Research Online: Recommended Citation Cooney, Ashley L.. "Integrating viral vectors as a gene therapy approach for cystic fibrosis." PhD (Doctor of Philosophy) thesis, University of Iowa, Follow this and additional works at: Part of the Microbiology Commons

2 INTEGRATING VIRAL VECTORS AS A GENE THERAPY APPROACH FOR CYSTIC FIBROSIS by Ashley L. Cooney A thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy degree in Microbiology in the Graduate College of The University of Iowa May 2018 Thesis Supervisors: Associate Research Professor Patrick L. Sinn Professor Paul B. McCray, Jr.

3 Copyright by ASHLEY LEA COONEY 2018 All Rights Reserved

4 Graduate College The University of Iowa Iowa City, Iowa CERTIFICATE OF APPROVAL This is to certify that the Ph.D. thesis of PH.D. THESIS Ashley Lea Cooney has been approved by the Examining Committee for the thesis requirement for the Doctor of Philosophy degree in Microbiology at the May 2018 graduation. Thesis Committee: Patrick L. Sinn, Thesis Supervisor Paul B. McCray, Jr., Thesis Supervisor Wendy J. Maury Aloysius J. Klingelhutz Adam J. Dupuy

5 To my husband and son, Todd and Ethan, thank you for your encouragement and support through this journey. To my mom, thank you for pushing me the best I can be. To my dad and sisters, thank you for your daily encouragement and uplifting pep talks. ii

6 ACKNOWLEDGEMENTS I would like to thank my mentors Dr. Patrick Sinn and Dr. Paul McCray for providing me with excellent training and their incredible knowledge base which inspired me to work hard and think scientifically. I would especially like to thank Dr. Sinn for his patience with me through my learning process and providing a great balance between hands on training and independent learning. I would also like to thank my true first mentor, Dr. Wendy Maury, for giving me this incredible opportunity and being my inspiration to study science. I would like to thank Dr. Aloysius Klingelhutz and Dr. Adam Dupuy for their insightful conversations and project guidance. I would like to thank Chris Wohlford-Lenane for her expertise and friendship in the lab. I would like to thank current lab members, Dr. Brajesh Singh, Dr. Jennifer Bartlett, and Cami Hippee for their support. I would like to thank Dr. David Stoltz and his lab members Linda Powers, Mal Stroik, Drake Bouzek, Nick Gansemer, and Peter Taft and the numerous pig feeders for their time and support through all of the labor-intensive pig experiments. I would like to thank previous lab members Erin Burnight and Shyam Ramachandran for being exemplary influential graduate students. I would like to thank the CF research groups at the University of Iowa including Dr. Ian Thornell, Dr. Mahmoud Abou Alaiwa, Dr. Viral Shah, Dr. Lynda Ostedgaard, Dr. Michael Welsh, and Dr. Joseph Zabner for investing their time in improving my projects. Lastly, I want to thank my family and friends. Thank you to my husband Todd who supported my decision to embark on this journey and start our family at the same time. I would also like to thank my mom for fostering my interest in science from a young age and pushing me through the challenging times. I could not have done this without you. iii

7 ABSTRACT Cystic fibrosis (CF) is the most common autosomal recessive genetic disease in Caucasian populations. CF affects multiple organ systems including pancreas, liver, intestines, sweat glands, and male reproductive organs, however the leading cause of morbidity and mortality in CF patients is chronic lung disease. CF is caused by a mutant cystic fibrosis transmembrane conductance regulator (CFTR) gene which leads to chloride (Cl - ) and bicarbonate (HCO - 3 ) anion dysregulation at the airway surface. Without adequate anion exchange, thick, viscous mucus accumulates at the airway surface allowing bacterial colonization to occur. Complementing CFTR in the appropriate airway cells restores the anion channel activity in CFTR-deficient cells. The ultimate goal for CF gene therapy is to design an integrating vector that would lead to persistent and efficient expression of CFTR in the airways. Performing gene therapy experiments is dependent upon a relevant animal model. The CF pig is a large animal model similar in size, anatomy, and physiology to humans. Importantly, the CF pig recapitulates human lung disease. From the CF pig, we have learned much about CF lung disease and have developed relevant assays to measure anion channel correction. We have learned that loss of CFTR leads to a decreased airway surface ASL ph, bacterial killing ability, and increased mucus viscosity. Standardized assays have been developed to evaluate the change in current by Ussing chambers, ASL ph, bacterial killing in vivo and ASL ph and viscosity on primary airway cultures in vitro. Ultimately, these metrics allow us to make conclusions about the efficiency of CFTR restoration. iv

8 Viral vectors are promising candidates for CF gene therapy. Viral vectors such as adenovirus (Ad), adeno-associated virus (AAV), and pseudotyped lentiviral vectors such as feline immunodeficiency virus (FIV) or human immunodeficiency virus (HIV) can efficiently transduce airway cells and express CFTR. Ad and AAV have both been tested in CF clinical trials, but CFTR expression was transient, if detected at all. Understanding vector biology and overcoming barriers in the lung have allowed us to improve vector delivery to the airways. However, the next major hurdle was achieving persistent expression. Ad and AAV are both transiently expressing vectors, and vector readministration is implausible due to the presence of neutralizing antibodies that develop against the vector. Creating a hybrid nonviral/viral vector in which the integrating nonviral piggybac transposon system is delivered by an Ad or AAV vector has allowed us to achieve persistent expression in mice. In a third integrating vector system, lentiviral vectors have historically been challenging to work with due to low titer levels. However, improvement in vector purification methods have allowed us to validate a lentiviral vector as a viable gene therapy option. In total, we have validated three integrating vector systems by restoring CFTR to CF pigs to correct the phenotypic defect. v

9 PUBLIC ABSTRACT Cystic fibrosis (CF) is a genetic disease that affects multiple organs in the body. People who have CF suffer most from chronic lung disease and mucus plugging in the airways, however the pancreas, liver, intestines, sweat glands, and male reproductive organs are also affected. To date, there is no cure for this disease. Current treatments for CF include nutritional therapy, antibiotics for lung infections, wearing a vibrating therapy vest to loosen airway mucus, and pulmonary rehabilitation. In advanced lung disease, a lung transplant is the only option for improved quality of life. What causes such a debilitating disease? CF is caused by a mutation in a single gene called the cystic fibrosis transmembrane conductance regulator (CFTR). Without this gene, the lining of the airways becomes acidic due to inadequate anion exchange, becoming a breeding ground for bacterial infections and mucus accumulation. Restoring CFTR to the appropriate airway cells corrects the phenotypic defect. The goal of my work was to improve CFTR gene delivery to pig airways. Through this work, we have successfully delivered CFTR and achieved phenotypic correction in CF pigs. We are encouraged that this work could translate to humans and prevent onset of CF lung disease. vi

10 TABLE OF CONTENTS LIST OF TABLES... x LIST OF FIGURES... xi LIST OF ABBREVIATIONS... xiii CHAPTER 1: INTRODUCTION- HISTORY OF CF GENE THERAPY... 1 Introduction: What is cystic fibrosis?... 1 Establishing benchmarks of success and Ad-based gene therapy trials ( )... 2 Alternatives to Ad ( )... 7 New vector design to address barriers to gene transfer ( ) New models for preclinical studies ( ) Lentivirus Transposon system Transposon system carried by a viral vector CHAPTER 2: LENTIVIRAL-MEDIATED PHENOTYPIC CORRECTION OF CYSTIC FIBROSIS PIGS Introduction Results FIV-CFTR corrects the anion channel defect in vitro FIV-CFTR rescues the anion channel defect in CF pig trachea and bronchus Partial rescue of ASL ph and bacterial killing in CF pigs Cultured ethmoid sinus epithelia from FIV-CFTR treated CF pigs show rescue of anion current and ASL ph Discussion Materials and Methods Pigs Vector production Primary cultures of airway epithelia In vivo viral vector administration Electrophysiology of CFTR anion channel Localization of CFTR qpcr ph and bacterial killing assays Statistics CHAPTER 3: HYBRID NONVIRAL/VIRAL VECTOR SYSTEMS FOR IMPROVED PIGGYBAC DNA TRANSPOSON IN VIVO DELIVERY Introduction Results DNA transposition from a piggybac/aav vector into the genome DNA transposition from a piggybac/ad vector into the genome piggybac/ad-mediated CFTR correction in CF primary airway cells vii

11 Discussion Materials and Methods Constructs Colony formation assay Integration site recovery for Illumina HiSeq2000 sequencing Ethics statement In vivo delivery and bioluminescence Primary epithelial cultures and electrophysiology studies Immunohistochemistry Statistics CHAPTER 4: CFTR DELIVERY BY A HYBRID PIGGYBAC/AAV VECTOR CORRECTS THE AIRWAY DEFECT IN CYSTIC FIBROSIS PIGS IN VIVO Introduction Results piggybac/aav transposition in vitro Phenotypic correction in CF pigs by piggybac/aav-cftrdr Measuring ASL ph and viscosity in primary cells cultured from treated CF pigs Discussion Materials and Methods Constructs Colony formation assay CF pigs In vivo viral vector administration Ussing chamber studies Tracheal airway surface liquid (ASL) ph and bacterial killing ph ASL measurements Viscosity and dye immobilization CHAPTER 5: WIDESPREAD AIRWAY DISTRIBUTION AND PHENOTYPIC CORRECTION OF CYSTIC FIBROSIS PIGS FOLLOWING AEROSOL DELIVERY OF PIGGYBAC/ADENOVIRUS Introduction Results Distribution of piggybac/ad in pig pulmonary airways Phenotypic correction via piggybac/ad-cftr Discussion Materials and Methods Pigs Constructs and vector production In vivo viral vector delivery Ussing chamber studies Bacterial killing assay Immunofluorescence Quantitative Real-Time PCR Statistics viii

12 CHAPTER 6: DISCUSSION Correcting the appropriate cells Options for integrating gene delivery Lentivirus DNA transposons Hybrid DNA transposon/viral vectors Animal models of cystic fibrosis CF rat CF ferret CF pig Outcome measures Quantitative real-time PCR and CFTR protein expression Functional correction Reduced infection and inflammation Challenges to pulmonary gene transfer with integrating vectors Delivery Insertional mutagenesis Gene transfer and phenotypic correction using a porcine model Conclusions REFERENCES ix

13 LIST OF TABLES Table 1: Six classes of mutations in CFTR that result in CF disease x

14 LIST OF FIGURES Figure 1: Schematics of integrating transgene delivery systems Figure 2: Correction of Cl- transport in CF pig primary epithelia with a lentiviral vector Figure 3: Anion channel correction in tissue explants Figure 4: Tracheal ph and bacterial killing ability in CF pigs Figure 5: FIV-CFTR corrects CF pig ethmoid sinuses Figure 6: piggybac/aav transduction and colony formation assays Figure 7: The heat map summarizes piggybac distributions to genomic features Figure 8: AAV delivered piggybac transposon mediates persistent transgene expression in mouse airways Figure 9: Mapping piggybac/aav or piggybac/ad integrations in mice Figure 10: piggybac/ad transduction and colony formation assays Figure 11: Ad delivered piggybac transposon mediated persistent transgene expression in mouse airways Figure 12: Luciferase expression in immunocompetent mice Figure 13: piggybac/ad-cftr restores Cl - transport in CF tracheal epithelia Figure 14: Schematics of hybrid vectors Figure 15: piggybac/aav forms puromycin-resistant colonies in the presence of the transposase Figure 16: piggybac/aav2/h22-cftrdr corrects the anion channel defect in vivo Figure 17: Primary airway epithelia cultured from piggybac/aav-cftrdr treated pigs exhibits increased ASL ph and decreased viscosity compared to untreated CF pigs Figure 18: Experimental design and quantification of transduced cells in conducting airways Figure 19: Dual positive GFP and mcherry cells in the airways Figure 20: Distribution of GFP in pig lungs xi

15 Figure 21: piggybac/ad-gfp co-localizes with major cell types Figure 22: piggybac/ad-cftr corrects the CF pig phenotype in vivo Figure 23: piggybac/ad-cftr co-localizes with Ad-GFP in vivo Figure 24: Airway epithelia cultured from piggybac/ad-cftr treated CF pigs retains correction in ASL ph and viscosity Figure 25: Gaussia luciferase expression in BAL and serum Figure 26: HDAd transduces trachea and conducting airways in vivo Figure 27: Adenovirus transduces airway epithelial cells, submucosal glands, and corrects the anion channel defect in vivo xii

16 LIST OF ABBREVIATIONS DF508 Deletion of phenylalanine at CFTR position 508 AAV Adeno-associated Virus Ad Adenovirus Amil Amiloride ASL Airway Surface Liquid b-gal b-galactosidase camp cyclic Adenosine MonoPhosphate CF Cystic Fibrosis CFBE Cystic Fibrosis Bronchial Epithelium CFTR Cystic Fibrosis Transmembrane conductance Regulator Cl - Chloride CMVie CytoMegaloVirus immediate early enhancer DIDS 4,4-DiIsothocyanatostilbene-2,2 Disulfonic acid EGTA Ethylene Glycol-bis-Tetraacetic Acid F&I Forskolin and 3-Isobutyl-1-methylxanthine FIV Feline Immunodeficiency Virus FRAP Fluorescence Recovery After Photobleaching xiii

17 G551D Glycine to Aspartic Aciid at CFTR position 551 GFP Green Fluorescent Protein GlyH GlyH-101 HA HemAgglutinin (influenza) HCO 3 - Bicarbonate HIV Human Immunodeficiency Virus DI Change in Current ipb7 Insect piggybac transposase JSRV Jaagsiekte Sheep Retrovirus KGF Keratinocyte Growth Factor LPC LysoPhosphatidylCholine LV LentiVirus M2 Matrix-2 protein of influenza MCT MucoCiliary Transport mmlv murine Moloney Leukemia Virus NPD Nasal Potential Difference PD Potential Difference pfu plaque forming units xiv

18 PIV ParaInfluenza Virus polya Poly Adenylation RSV Respiratory Syncytial Virus RSV Rous Sarcoma Virus SARS Severe Acute Respiratory Syndrome scaav Self-Complementary Adeno-Associated Virus SP-C surfactant protein C SMG SubMucosal Gland ssaav Single-Stranded Adeno-Associated Virus vg vector genomes VSVG Vesicular Stomatitis Virus Glycoprotein V t Voltage xv

19 CHAPTER 1: INTRODUCTION- HISTORY OF CF GENE THERAPY Introduction: What is cystic fibrosis? Cystic fibrosis (CF) is an autosomal recessive disease caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene that encodes a campregulated anion channel. While our knowledge of CFTR function has advanced greatly since the discovery of the gene in 1989, CF remains fatal (1, 2). Although CF is a multiorgan system disease, most people with CF die of progressive lung disease that begins early in childhood and is characterized by chronic bacterial infection and inflammation (2). Nearly 90% of people with CF have at least one copy of the ΔF508 mutation, but there are >2,000 mutations that result in a range of disease severities (2). These mutations can be divided into six classes based on the type and consequence of the mutation (Table 1): class I, no synthesis; class II, defective processing; class III, defective regulation; class IV, altered conductance; class V, reduced synthesis; and class VI, accelerated turnover (3). However, new mutations continue to be identified and one mutation may fit into more than one category by disrupting CFTR transcription, protein trafficking, or protein regulation in more than one way (4). Pharmacologic approaches aimed at activating alternative ion transport pathways (5-8), reducing inflammation (9, 10), and inhibiting or eliminating bacterial infection (11, 12) are active areas of therapeutic development. There is also intense interest in identifying interventions that can restore function to the mutant protein (6, 7, 13, 14). The promise of using a small molecule potentiator to restore function to a mutant protein was recently validated in a clinical trial for the CFTR conductance mutation G551D and other conductance mutations present in ~5% of CF patients (7). However, unlike small molecule potentiator or correctors, which 1

20 modify or repair CFTR function, a CFTR gene replacement approach would be efficacious regardless of the disease-causing mutation and is potentially a single dose, life-long curative therapeutic strategy for a devastating disease. There are a great number of vector options for CFTR gene delivery. Nonintegrating viral vectors (i.e., adenovirus or adeno-associated virus) and non-integrating nonviral vectors (i.e., plasmid DNA or in vitro transcribed RNA) each have important attributes and have resulted in significant advances in the CF gene therapy field (reviewed in (15-17)). However, potential limitations to these episomal expression systems may include gradual decreases in transgene expression over time and limiting host immune responses associated with vector readministration. These potential pitfalls could be avoided if a therapeutic transgene is stably integrated into the genome of a progenitor cell population. Thus, in this thesis, we focus on the use of integrating vectors for gene delivery, although some of the topics covered will be relevant to multiple vector systems. This thesis discusses the established and emerging options for integrating vectors and efforts to deliver integrating vectors to CF animal models. The remainder of this introduction describes the early studies of CF gene therapy, how the field of gene therapy has progressed, and challenges since the discovery of CFTR in Establishing benchmarks of success and Ad-based gene therapy trials ( ) In 1989, the gene responsible for CF was identified as CFTR (18, 19). Sequencing led to the identification of multiple CFTR mutations, most commonly a three-base deletion that results in the loss of phenylalanine at position 508 (18, 20). Cloning CFTR was a major leap for studying CF and quickly launched the concept of gene-based 2

21 therapeutics. Within a year of discovering CFTR, two groups independently demonstrated the proof-of-concept of CF gene therapy by expressing CFTR in CF cells and restoring anion channel activity. Restoring CFTR activity was proposed to be the cure all for CF (21, 22). Soon after, several studies using viral and nonviral approaches to deliver CFTR and correct the camp-mediated Cl - permeability in CF cells included an adenoassociated virus (AAV)-based vector (23), adenovirus (Ad)-based vector (24), plasmid formulated with cationic liposomes in mice (25) or a retroviral vector in CF cells (26). At this time, complementing CFTR in CF patients was considered a near-term goal. To determine the percentage of cells that need to be corrected to be therapeutically beneficial, Johnson et al. performed the first in vitro cell culture studies. By mixing CF and non-cf cells in different ratios, as few as 6-10% of airway cells expressing CFTR achieved non-cf levels of Cl - transport (27). These studies confirmed that CFTR gene delivery was a potential curative strategy and established a common benchmark of success for gene therapy (i.e. transduction of as few as 5% of airway cells). Also during this time, functional CFTR assays and new model systems were being developed rapidly. Experiments were performed on patient-derived immortalized cells termed cystic fibrosis bronchial epithelial (CFBE) cells and bronchial xenografts (23). CFTR levels and activity were measured by mrna, iodide efflux, in situ hybridization, patch clamp, and Ussing chambers (21). Prior to 1992, no animal models existed to test functional gene transfer efficacy in vivo. However, within a short period of time, three groups independently generated CF mice by targeted knockout of endogenous Cftr (28-32). These mice exhibited an increase in steady-state nasal potential difference (NPD) compared to non-cf mice, altered Cl - transport, abnormal mucus accumulation, 3

22 and disease-related changes in the lung and reproductive tract but mice did not develop lung disease. Similar to humans, evidence of intestinal obstruction was also reported (33). Consistent with in vitro experiments, in a CF mouse model, transduction of as few as 5% of cells expressing CFTR was equivalent to 50% wild-type Cl - secretion (34). Multiple in vivo experiments examined Ad-based lung gene transfer in various models. Repeat doses of Ad-LacZ to cotton rats or nonhuman primates showed that Ad transduced cells within the proximal bronchi and bronchioles including ciliated, secretory, undifferentiated, basal cells (35), and even submucosal glands (36). The safety of repeated Ad administration was evaluated and studies concluded that Ad delivery was safe, exhibiting little to no immune response upon repeat administration (37-40). These and other studies (38, 41-47) suggested that Ad was a promising vector for CF gene therapy. In 1993, Zabner et al. published results from the first clinical trial for CF gene therapy (48). Three CF patients received an Ad vector (serotype 2) carrying CFTR (49). The vector was applied to nasal turbinates for 30 minutes and then removed by suction. One or three days later, one side was biopsied and the contralateral side was used to measure NPD. In retrospect, airway injury may have facilitated delivery. The authors observed a decreased voltage in the nasal epithelia and a restored response to a camp agonist; however, CFTR mrna and protein were undetectable. Independently, three additional Ad-based clinical trials ensued in efforts to restore CFTR function in CF patients. Crystal et al. performed a phase I dose-escalation safety study in 4 people with CF and concluded that up to 2x10 9 pfu of Ad2 led to CFTR cdna expression in airway epithelium. One subject that received 2x10 9 pfu to the right lower lobe experienced 4

23 transient systemic inflammation in the first 24 hours, but the symptoms resolved and a 6 month follow up showed no long-term effects. The NPD results in this study were reported as inconclusive but Ad delivery to the nose and lungs appeared to be safe (50). Knowles et al. used an Ad serotype 5 vector in a larger study of 12 CF participants. Here, Ad-mediated delivery of CFTR did not correct the functional defects, perhaps due to inflammatory responses (51-53). Hay et al. delivered Ad-CFTR to nasal epithelia and reported that NPD decreased toward normal compared to the contralateral untreated nostril (54-56). In general, these studies suggested that Ad-based vectors can partially correct the Cl - transport defect in CF airway epithelia; however, the effects were transient and inflammatory responses were observed. Although several studies showed CFTR correction in vitro (57-59), questions about safety and efficacy in humans remained. The immune response remained an obstacle to achieving long-term and efficient gene transfer with Ad-based vectors. A nonhuman primate study reported increased alveolar inflammation with high doses of Ad vector. Although Ad efficiently delivered genes to the lung, transient expression meant that repeat administration would be required (60, 61). Studies of the immune response indicated that preexisting neutralizing antibodies may prevent serotype-specific transduction (62). These serotype specific neutralizing antibodies against Ad viral antigens were found to activate CD8 T cells in mice (63). Consequent transgene elimination by MHC-II presentation of viral antigens generated formation of neutralizing antibodies and prevented effective repeat administration (64). Efforts to facilitate repeat administration included: blocking IgA antibodies (65), neutralizing antibodies (66), and a 5

24 non-depleting hcd4 antibody (67), none of which substantially improved repeat Ad administration. To determine if Ad-vector readministration would be effective to achieve longterm CFTR expression in people with CF, clinical trials tested safety and effectiveness of repeat Ad administration. A trial in 1996 tested dose-escalation with 5 repeated doses up to pfu in nasal epithelium of CF subjects. They concluded that the vector partially corrected the Cl - transport defect but observed significant variability between subjects. Importantly, they found less CFTR-correction with each administration, likely due to the immune responses (68-70). Another Ad clinical trial by Bellon et al. reported no toxic effects using doses of pfu and detectable CFTR DNA and RNA (71). Other clinical trials occurring during this time using Ad with additional E3 or E4 deletions reported transient expression after a second administration (72) or low efficiency in the lower respiratory tract with short duration of expression (73). In total, repeated doses of Ad was not an effective therapeutic option. In 1999, in a clinical trial for ornithine transcarbamylase (OTC) deficiency, an Ad vector carrying the OTC gene was delivered to an 18-year-old male named Jesse Gelsinger. He received a dose of 3.8x10 13 viral particles via femoral artery for delivery to the liver (74-76). Four days after delivery, he died from multiple organ failure and associated cytokine storm. Although this disease is unrelated to CF, it tragically demonstrated that high systemic doses of Ad can be fatal. In total, there were 9 clinical trials for CF using Ad as the delivery vehicle, the last taking place in 2001 (77, 78). In the first decade of CF gene therapy, scientists and clinicians learned that complementing a non-functional CFTR with a corrected cdna was more challenging 6

25 than originally anticipated. CF clinical trials using Ad-based vectors generally supported that gene therapy had potential therapeutic benefit, but improvements in vector design were necessary to correct CF lung disease. Alternatives to Ad ( ) The immunogenicity and transient expression of Ad-based vectors was recognized early on. In parallel to many of the aforementioned studies, development of other viral and non-viral vectors were underway. Three delivery platforms that received the most attention were 1) AAV-based vectors, 2) non-viral (ie. plasmid) vectors, and 3) retroviral- (or lentiviral)-based vectors. These vectors were evaluated for the ability to transduce airway cells in vivo and in vitro. Vector modifications and delivery agents were screened with the goal of improving titer, tropism, transduction efficiency, expression, and stability. The barriers to pulmonary gene transfer were examined in greater detail and better measurements for successful gene transfer were invented. In this era, research focused on improving gene delivery, measuring persistent expression, and understanding the host immune response to viral vectors and encoded transgenes. Similar to Ad, recombinant AAV-based vectors transduce terminally differentiated and non-dividing cells (79). AAV persists episomally and does not integrate in the absence of exogenous Rep protein (80). In preclinical studies, AAV (serotype 2) expressing CFTR persisted in rhesus macaques for 3 months (81) and a single delivery to the lung was shown to be safe (82). In the newborn rabbit, AAV transduced alveolar epithelial cells, tracheobronchial and ciliated lung cells (83) and vector DNA was detectable for up to 6 months (84). Functional CFTR activity was 7

26 demonstrated in cultured cells by an iodide efflux assay (85). Loss of expression was attributed to the dilution of transduced cells through lung growth and lack of stem cell transduction (86). AAV could be readministered through transient immunosuppression with an anti-cd40 ligand antibody and soluble CTLA4-immunoglobulin fusion protein and persist for 8 months (87). Thus, a single dose of AAV transduced airways and conferred stable expression for many months. A limitation of AAV for CF gene therapy is its small carrying capacity of ~4.6 kb. The AAV inverted terminal repeat (ITR) was shown to possess promoter activity and mediate stable, but low levels of CFTR expression in vitro (85). Still, the addition of an internal promoter and polyadenylation signal was necessary and exceeded the carrying capacity (88-90). Deletion analysis showed that removing a portion of the R-domain shortened CFTR by nearly 300 base pairs and still retained anion channel function while addressing AAV packaging constraints (91). Additional studies to shorten other expression cassette components led to the generation of a shortened CMVie promoter, polya (92) and a hybrid promoter (93). AAV vectors with these modifications delivered a functional CFTR and corrected the anion channel defect. During this time period, AAV vectors were the forefront of CF gene therapy. Many studies showed their safety and efficacy in repeat dosing to nonhuman primates (94, 95) and mice, even in the presence of neutralizing antibodies (96). Serotype studies in ferrets and pigs showed that AAV1, 2, and 5 transduction patterns closely mirrored human airway epithelia, but mice had a different profile (97-99). Delivering CFTR with AAV helped uncover features of CFTR biology. For example, AAV-mediated CFTR delivery to mouse airways exposed a correlation between gene transfer and low mrna 8

27 levels which suggested a regulatory role of CFTR as an activator of other chloride channels (100). Multiple in vitro and in vivo studies suggested AAV could efficiently restore CFTR expression in airway epithelia. In 1999, the first AAV-CFTR (serotype 2) vector clinical trial tested single and two dose treatments, reporting successful dose-dependent gene transfer to the maxillary sinus with little to no cytopathic or host immune response (101, 102). In the midst of designing and testing new vectors, two additional AAV clinical trials took place (103, 104) and a third clinical trial tested repeat administration of AAV-CFTR (105). The last AAV clinical trial for CF took place in 2007 (106). In general, AAV delivery was safe; however, the resultant levels of CFTR expression were typically below the limit of detection. Despite low CFTR expression levels, results were encouraging because they demonstrated the safety of the vector. As an alternative to viral vectors, others investigated a nonviral strategy for CFTR delivery were investigated. Nonviral vector delivery is an enticing approach because there are no size constraints; however, the challenge is efficient delivery to cells. This challenge is most commonly addressed by formulating plasmid DNA with a cationic lipid. Many cationic lipids tested were reported to be safe plasmid delivery vehicles for gene transfer ( ). In the first nonviral clinical trial by Caplen et al., a cationic lipid complexed with CFTR showed partial correction of camp-mediated Cl - transport in nasal epithelium of 6 people with CF, 20% of non-cf levels as measured by NPD ( ). Hyde et al. performed a repeat administration trial with a nonviral complex delivered to the nose and reported no loss of efficacy with repeat administration (113). Four additional nonviral clinical trials were conducted. All of which reported partial, 9

28 transient expression not lasting longer than 4 weeks ( ). After a wave of nonviral clinical trials in 1997, there was a lull in trials for nearly a decade to focus on developing a vehicle to enhance gene transfer. In perhaps the most ambitious CF gene therapy clinical trial to date, Alton et al. tested lipid formulated with plasmid CFTR delivery to the lung and nose of patients. Modest Cl - correction and reduced bacterial adherence was observed (119). However, despite these encouraging early results, a nonviral approach to delivering CFTR was not enough to achieve clear phenotypic correction. In the late 1990s, gene therapy studies uncovered important aspects about airway cells while improving viral delivery methods. Since MuLV-based vectors transduce only dividing cells, keratinocyte growth factor (KGF) was found to stimulated proliferation of airway epithelial cells and improve MuLV gene transfer efficacy. Additionally, EGTA enhanced transepithelial permeability and gene transfer with MuLV (120) and AAV (121). Other studies using retroviral vectors (122) reported enhanced gene transfer in mice after injury with sulfur dioxide (123). A similar phenomenon was reported with Ad (124). It became clear that Ad preferred basolateral entry due to receptor-mediated entry and EGTA enhanced Ad gene transfer in mice epithelia (125) by opening tight junctions (126). The Coxsackievirus and Adenovirus receptor (CAR) was localized to the basolateral surface and redirecting it to the apical surface with a GPI linkage was sufficient to support apical gene transfer (127). These were the first studies to shed light on how receptor polarity and access influences transduction efficiency. Lentiviral vectors were welcomed as promising therapeutic options because genomic integration could provide long term expression. Lentiviral vectors quickly replaced retroviral vectors as they could effectively transduce nondividing airway cells 10

29 (128). Specifically, FIV vectors were shown to transduce nondividing airway cells, persistently express a transgene of interest, and correct the CFTR defect (129). Further modifications to FIV vectors also improved gene transfer (130). Although typically pseudotyped with VSVG, its basolateral preference hindered efficient airway gene transfer(131). Envelopes such as RSV (132), Marburg and Ebola virus (133, 134) HA- M2 of influenza (135), SARS spike protein (136) JSRV (137) and baculovirus GP64 (138, 139) were evaluated for apical entry capabilities. Fronted by these studies, the use of lentiviruses was on the rise and had become the new interest in the field of CF gene therapy. Many lentivirus studies proved lentiviral vectors to be promising gene therapy candidates. VSVG-pseudotyped HIV-CFTR with a lysophosphatidylcholine (LPC) pretreatment showed persistent CFTR expression for at least 12 months in CF-null mice (140) and LPC was shown to enhance lentiviral gene transfer (141). New vectors PIV (142), SV40 (143), and RSV (144) were also evaluated as gene transfer vectors. Additionally, the F/HN proteins from Sendai virus were used to pseudotype a SIVderived lentiviral vector carrying CFTR was shown to correct the CF defects in vitro (145, 146). Meanwhile, GP64-FIV was shown to support gene transfer in a pig model (147). HIV-based lentiviruses also transduced marmoset lungs (148) and achieved longterm correction in CF mice (149). New vector design to address barriers to gene transfer ( ) Seemingly, vector modifications alone were not enough to overcome immunemediated clearance of transgene expression from viral vectors. In further efforts to 11

30 enhance viral gene transfer, several vehicle formulations were evaluated for their ability to reduce immunostimulation upon entry to increase the efficiency of transduction. Viral vector formulation with polycations neutralized the negative charge of the membrane glycoproteins which increased transgene expression. Examples included polybrene, protamine, DEAE-dextran, poly-l-lysine, and cationic lipids (150, 151). Methylcellulose formulations also improved apical gene transfer for more than one vector class (152). Agents such as the natural airway surfactant LPC or EGTA were employed to transiently disrupt the tight junctions. This improved VSVG-HIV entry in vivo (153) and lipid and peptide complexes in human airway epithelia (125). Precipitation of Ad using calcium phosphate was shown to enhance gene transfer by receptor-independent endocytosis ( ). Formulating Ad with dexamethasone reduced inflammation (158), using polidocanol to perturb cell membranes (159), and using a u7-peptide to improve apical binding (160) were all method that enhanced gene transfer. Many of these vehicles have since become widely used in preclinical studies and have played a major role in enhancing gene transfer. Additional efforts to improve transgene delivery and understand CFTR expression were explored through vector design. Selection of the appropriate promoters and polya signals with the appropriate strength might improve CFTR expression and increase anion channel activity, potentially achieving therapeutic levels of correction with few cells transduced. The PGK, EF1a and CMV promoters were evaluated in context with bovine growth hormone (BgH) or SV40 poly A signals and compared by camp-mediated Cl - transport used as an endpoint and required prolonged incubation time (161, 162). The inclusion of inducible promoters regulated by transgene expression levels and modified 12

31 CFTR expression cassettes were novel ways to regulate gene expression through vector design ( ). Selective transgene expression in epithelia was first documented by showing that K18-CFTR drives expression in airway cells (166). Several advances helped AAV gene transfer was steadily move forward. Codelivery of AAV with proteasome inhibitors doxorubicin and LLnL increased expression in airway cells by preventing the vector from being proteasomal degradation (167, 168). Directed evolution of AAV on human airway epithelial cells led to the development of AAV2.5T capsid with improved apical transduction properties to human airway cells (169). Novel AAV capsid variants derived from AAV1, 6, and 9 were generated by directed evolution to improve CFTR delivery (170). In a dual reporter gene study in chimpanzee airways, AAV1 was shown to transduce 20-fold higher than AAV5 at 90 days post-delivery (171). Other studies using AAV showed that incorporating peptide motifs into the AAV capsid improved gene transfer in human airway epithelial cells (172). Due to the size limits in the AAV packaging capacity, the development of a minimal synthetic promoter F5Tg83 facilitated improved CFTR expression with a strong polya within the size constraints (173). To circumvent the harmful immunostimulatory properties of Ad, a helperdependent Ad (HDAd) (also known as delta-rad or gutted Ad) was developed, expressing no viral proteins. This was a strategy helped reduce immune responses to viral antigens, and may provide a safer vector option for CF gene therapy (174). Efforts to understand the innate immune response to Ad advanced. Systemic IL-6 levels were elevated following multiple routes of vector instillation (175), however use of HDAd led to reduced levels of inflammation (176). Interestingly, using Ad-CFTR or HDAd-CFTR 13

32 both decreased lung bacterial infections through restoration of CFTR to the airways in mice ( ). HDAd shows promise for future studies. It is encouraging that it can be readministered to the mouse lung (181) and be used to deliver genes to modulate the innate immune response and prevent cytokine imbalance or airway remodeling (182). Alternative vector systems to Ad have shaped the way we think about gene transfer. Lentiviral vectors were shown to be less immunogenic than adenoviral vectors and could be readministered to the respiratory tract without loss of transgene expression or blocking immune responses (183). Efforts to overcome physical barriers continued, but increasing Ad vector access to necessary receptors using agents to disrupt tight junctions in the airways or preventing premature proteasomal degradation of AAV markedly improved gene transfer. Additionally, further evaluating the prevalence of neutralizing antibodies for Ad and AAV vectors was eye-opening and prompted genetic engineering of viral vector capsids. With an expanded toolbox of delivery options and more focused approaches to important problems, continued improvements in gene transfer seemed closely within reach. New models for preclinical studies ( ) In this recent era of CF gene therapy, it was evident that an animal model that developed lung disease was needed to advance studies of lung disease pathogenesis as a preclinical model to study delivery and efficacy. Because the CF mouse does not develop lung disease similar to people with CF, a model that more closely represents human lung disease would be an advancement. Studies from new models could lead to a better understanding at the basic level of the relationship between loss of CFTR and 14

33 development of lung disease. With this new information, new strategies to develop gene therapy vectors led to enhanced transduction efficiency and relevant animal models to validate these improved vectors. We currently have a variety of CF animal models. In 2008, the development of a CF pig was reported (184, 185). The pig was chosen because it is a large animal model with many pulmonary, anatomical, physiological, and biochemical similarities to humans. Development of this model has changed the way we think about CF gene therapy. One limitation of the CF pig was the occurrence of meconium ileus in all animals which required surgical intervention for long term survival. Creating a gut corrected CF pig expressing CFTR in the gastrointestinal tract by the fatty acid binding protein (FABP) promoter alleviated meconium ileus, making longitudinal studies more feasible (186). In 2014, the CF ferret was published as another large animal model (187). Similar to CF pigs, CFTR-knockout ferrets recapitulated many disease similarities to humans including defective airway Cl - transport, meconium ileus, and development of spontaneous bacterial lung infections. Studies in non-cf ferrets show that raav2/1 and lentiviral vectors facilitate gene transfer in newborn airways, and thus hold promise as vector platforms for CF studies. Other newer CF animal models include the CF rat, zebrafish, and rabbit. The CF rat was created in 2014 and exhibits an NPD and bioelectric properties similar to human levels. Additionally, histological abnormalities in the ileum parallel intestinal complications seen in other animal models. Efforts to generate other animal models such as rabbit and sheep are ongoing (188, 189). Overall, an animal model that develops lung 15

34 disease in a pattern similar to humans would be useful to studying various stages of disease onset in CF. An important milestone in understanding CF disease pathogenesis was elucidated using the CF pig by Pezzulo et al. in Here, they provided evidence that airway surface liquid ph reduced the bacterial killing ability in CF pigs (190). From these studies, relevant assays were available to measure CFTR correction in vivo. In 2016, manuscripts by Cooney et al. (Chapter 2) and Steines et al. used these assays to show that delivery of CFTR by FIV (191) and AAV (192) to CF pig airways increased the ASL ph, bacterial killing ability, and mucociliary transport defect. Current knowledge suggests that for gene therapy to be successful, the vector selected must either integrate into a basal cell or other cell type with progenitor capacity or be readministered. While AAV, Ad, and HDAd are generally considered to be nonintegrating, development of a hybrid nonviral transposon/viral integrating vector system led to persistent expression in mice for at least a year. Previously, the DNA transposon piggybac has been shown to promote persistent gene transfer in mice (193). In this system, the piggybac/aav (Chapter 3, 4) and piggybac/ad (Chapter 3, 5) vector system leads to persistent expression in mouse airways only in the presence of the transposase (194). Promising work evaluating HDAd suggests that this vector efficiently transduces airway cells (195), including submucosal glands (196), and can be readministered with use of an immunosuppressive (197, 198). Perhaps an integrating piggybac/hdad could provide long-term, efficient correction and be readministered if necessary. Currently, there are exciting new developments in CF gene therapy. A new vector, human bocavirus, has been identified and has been shown to transduce ferret 16

35 airways (199), PEGylated nanoparticles/pei and PLGA have shown efficacious for lung delivery (200) and the UK group is gearing up to perform the first in-man lenti CF clinical trial using a lentiviral vector (201). The field of gene therapy has learned valuable lessons from the many viral and nonviral clinical trials to date and is now taking careful, calculated approaches to develop a vector that can efficiently and persistently correct the CFTR anion channel defect. With the development of animal models that recapitulate human lung disease and methods to improve receptor access, further preclinical safety studies will be able to more closely predict a successful gene therapy vector which in turn will hopefully lead to a successful clinical trial. Developing the therapeutic cure all for CF has encountered many unexpected hurdles, but may now feasibly be within reach. With newborn screening for CF mandatory in all 50 states, a gene therapy application prior to disease onset could mean that the treated individual would need minimal to no treatments have a substantially improved quality of life. This thesis describes engineering and delivering of the following integration systems to achieve persistent CFTR expression: Lentivirus Lentiviruses are defined by the ability to reverse transcribe their RNA genome and integrate proviral DNA into the genome of the host cell (202). (Chapter 2) FIV production systems are divided into three expression plasmids (Figure 1A): o 1) a plasmid containing the transgene of interest flanked by the HIV long terminal repeats (LTRs); o 2) a packaging plasmid expressing structural and enzymatic proteins 17

36 o 3) an envelope glycoprotein expression plasmid. Multiple viral proteins, such as nef, vif, vpu, env, vpr, and much of the U3 region of the 3 LTR. Transposon system Recombinant DNA transposons, such as piggybac, used for gene transfer applications are comprised of a two-part integrating system; one encodes terminal repeats (TRs) flanking a transgene of interest (transposon), and the other a catalytic protein responsible for transposition (transposase) (Figure 1B). Transposon system carried by a viral vector Recombinant DNA piggybac transposon delivered by non-integrating viral vectors, termed hybrid vectors, combine the advantage of efficient transduction of the viral vector with persistent expression of the transposon, creating an integrating vector. Adeno-associated virus (AAV) (203) (Chapter 3, 4) Adenovirus (Ad) (194, 204) (Chapter 3, 5), have been investigated as delivery tools for DNA transposons (Figure 1C). 18

37 Class ~Frequency Mutation Type Common Representative CFTR Protein Outcome I 10% Nonsense, G542X No CFTR splice II 70% Missense DF508 Defective Processing III 2%-3% Missense G551D Defective Regulation IV <2% Missense R117H Altered Conductance V <1% Missense, KB Reduced Synthesis splice VI <1% Missense N287Y Accelerated Turnover Table 1: Six classes of mutations in CFTR that result in CF disease. As described in (3, 205, 206). 19

38 Figure 1: Schematics of integrating transgene delivery systems. (A) FIV-based lentiviral vectors are produced by three plasmid transfection. The gag/pol plasmid supplies structural and enzymatic proteins; the env plasmid supplies the envelope glycoprotein (GP64). The gene of interest (goi) is flanked by the long terminal repeats (LTRs) and driven by a heterologous promoter. Only the genetic material flanked by the LTRs is packaged and integrated into the host genome. (B) Recombinant DNA piggybac transposons are typically delivered as a two-part plasmid system. Transposase catalyzes the transposition of the genetic material flanked by the appropriate terminal repeats (TRs) from the plasmid and into the host genome. (C) DNA piggybac transposons can also be delivered by viral vectors to improve delivery efficiency. One viral vector carries the DNA transposon and the other carries the transposase. The goi is flanked by the transposon TRs, which in turn are flanked by the viral vector inverted terminal repeats (ITRs). Once inside the cell, transposition functions as described in B. 20

39 CHAPTER 2: LENTIVIRAL-MEDIATED PHENOTYPIC CORRECTION OF CYSTIC FIBROSIS PIGS Introduction Cystic fibrosis (CF) is a common autosomal recessive disorder caused by mutations in the CF transmembrane conductance regulator (CFTR) gene (1, 18). The most common CFTR mutation is a phenylalanine deletion at amino acid position 508 (ΔF508), however there are over 2,000 different disease-associated CFTR mutations. CF affects multiple organ systems, yet progressive lung disease, characterized by recurrent bacterial infections and inflammation, is the leading cause of CF morbidity and mortality (207). CFTR mutations result in impaired anion transport. This contributes to lung disease, in part, through reduced airway surface liquid ph (ASL ph), defective bacterial killing, and reduced mucociliary transport (MCT) ( ). While new CFTR potentiator (7, 206) and corrector (211) therapies are now in the clinic, there remains a great need to develop treatments for all people with CF. Gene therapy is a mutation agnostic approach to restore CFTR activity. Lentiviral vectors transduce both dividing and nondividing cells, integrate into the host genome, and provide long-term transgene expression (212, 213). A feline immunodeficiency virus based (FIV-based) viral vector pseudotyped with the baculovirus envelope protein GP64 transduces epithelial cells at the apical surface and persistently expresses a transgene of interest in mouse airways (147, 214). We previously demonstrated that FIV pseudotyped with the GP64 envelope from baculovirus efficiently transduces both human and pig airway epithelia (147, 215). Here, we use a lentiviral vector to deliver a CFTR expression cassette to pig airways in vivo. 21

40 At birth, CF pigs manifest physiologic defects associated with loss of CFTR function, including a reduced ASL ph, impaired bacterial killing, and reduced mucociliary transport (MCT) (190, 210, 216, 217). In this single time point pilot study, we hypothesized that delivery of GP64-FIV-CFTR to the sinuses and lower respiratory tract of newborn CF pigs would correct the anion channel defect. We show that delivery of GP64 pseudotyped FIV-CFTR to newborn CF pigs (186) can achieve partial physiological correction of the defective anion transport and its host defense consequences in vivo. Results FIV-CFTR corrects the anion channel defect in vitro Multiple studies have established that CFTR complementation restores the anion channel defect in vitro (21, 26, 129, 218). Before we delivered vector to CF pigs, we first confirmed that lentiviral-mediated porcine CFTR delivery corrects the anion defect in primary cultures of CF airway epithelia. Well-differentiated airway epithelial cultures derived from CF pig ethmoid sinuses were transduced apically with transducing units (TU) of GP64-pseudotyped FIV-CFTR (MOI = 5). Four days after transduction, the bioelectric properties of epithelia were analyzed in Ussing chambers. FIV-CFTR treated cells demonstrated a significant increase in transepithelial Cl - current in response to forskolin and 3-isobutyl-1-methylxanthine (IBMX, F&I) (Figure 2, A and B). The current was inhibited by the CFTR channel blocker, GlyH-101 (Figure 2, A and C). CFTR subcellular localization was examined using IHC. CFTR protein localized to the apical membrane in both ciliated and nonciliated epithelial cells (Figure 2D) and was absent in 22

41 the untreated CF epithelia (Figure 2E). These data suggest that apical delivery of FIV- CFTR to primary airway epithelia can correct the anion channel defect in vitro. FIV-CFTR rescues the anion channel defect in CF pig trachea and bronchus To test the efficacy of FIV-CFTR in vivo, we aerosolized TU of concentrated FIV-CFTR formulated with methylcellulose (152) into the ethmoid sinus and trachea of 3 newborn gut-corrected CF pigs. Gut-corrected CF pigs express CFTR in intestinal tissues but lack CFTR expression in the airways (186). Two weeks after vector delivery, tissues were collected and analyzed for CFTR correction. Tissues from untreated CF pigs showed little response to either F&I or GlyH-101 (Figure 3, A D). In freshly excised tracheal tissues from the CF pigs treated with FIV-CFTR, a significant increase in camp-activated Cl - current was observed to near WT levels (Figure 3A). This current decreased in response to GlyH-101 (Figure 3B). A similar result was obtained for freshly excised bronchus tissue (Figure 3, C and D). We also quantified CFTR mrna abundance in treated and untreated animals. In FIV-CFTR treated trachea and bronchus tissues, we observed CFTR mrna levels that were increased significantly over the average background levels detected in untreated CF pig tissue (Figure 3E). Together, these findings confirm that delivery of a CFTR expression cassette by a lentiviral vector can partially correct the anion defect in the trachea and bronchus of CF pigs in vivo. We collected genomic DNA from tracheal and bronchial tissues from FIV-CFTR treated and untreated animals. The tissues included multiple cell types, including surface epithelia, basal cells, submucosal glands, basal lamina, and smooth muscle. Thus, not all cells sampled were surface epithelial cells. Using quantitative PCR (qpcr), we attempted to quantify the levels of DNA integration; however, the levels were below the limit of 23

42 detection, i.e., less than 1 copy per 10 cells (data not shown). Because the tissues represent a mixed cell population, including many that received no vector, detecting CFTR transgene copy number in airway epithelial cells is challenging. However, these results suggest that, in the conducting airways, partial CFTR anion channel correction can be achieved by integrating a transgene in a small percentage of cells. We postulate that overexpressing CFTR with a heterologous promoter in a small subset of airway cells may help compensate for low overall gene transfer efficiency. Partial rescue of ASL ph and bacterial killing in CF pigs The loss of bicarbonate transport through CFTR results in acidification of the ASL and subsequent inhibition of antimicrobial factors, thereby impairing bacterial killing (190). Therefore, we next determined if CFTR complementation by FIV-CFTR increased the ASL ph and improves airways surface liquid (ASL) antimicrobial activity. Two weeks after vector delivery, we measured tracheal ph (190, 219). Compared with untreated animals, average tracheal ASL ph increased from 7.0 to 7.2 (Figure 4A); however, this trend did not reach statistical significance. The average ASL ph of the FIV-CFTR treated group was nearly identical to the non-cf pig group. To assess ASL antimicrobial activity, we interrogated individual bacteria attached to gold grids, as previously described (190). As shown in Figure 4B, the ASL from CF pigs that received FIV-CFTR had an increased bacterial killing ability as compared with untreated CF pigs. Bacterial killing activity measurements in a third CF pig that received FIV-CFTR failed due to technical complications; thus, statistical comparisons were not performed. We concluded that complementation of the anion channel defect in vivo by lentiviral vector transduction has the potential to increase ASL ph and bacterial killing. 24

43 Cultured ethmoid sinus epithelia from FIV-CFTR treated CF pigs show rescue of anion current and ASL ph The CF pigs in this study received vector to both the nasal and pulmonary airways. Two weeks after delivery, the ethmoid sinuses were harvested and epithelia were enzymatically dispersed, cultured on collagen-coated filters, and grown at an airliquid interface. After differentiation in culture, ethmoid sinus epithelia were mounted in Ussing chambers to measure CFTR anion channel activity. Culturing ethmoid epithelial cells was necessary because ethmoid tissue cannot be directly mounted in the Ussing chambers. As shown in Figure 5A and B, in cells from FIV-CFTR treated animals, we observed an increase in transepithelial CFTR-dependent Cl - conductance. This current was blocked by GlyH-101 (Figure 5A and C). No change in current was observed in cells from untreated animals. In addition to functional Cl - channel correction, we observed a significant increase in ASL ph in epithelia cultured from the pigs that received FIV- CFTR (Figure 5D). Together, these results provide evidence of functional correction of CF anion transport and ASL ph defects by a lentiviral vector in the ethmoid sinus in vivo. Discussion Here, we show functional CFTR complementation in a large animal model. Two weeks after aerosolized FIV-CFTR delivery to CF pig lungs, freshly excised tracheal and bronchial tissues exhibited partial restoration of camp-stimulated anion conductance. The response was blocked by the CFTR inhibitor, GlyH-101. We also observed an increase in CFTR mrna levels in treated tissues. Loss of CFTR function leads to the development of chronic bacterial lung infections (220, 221). In CF pigs, loss of CFTR- 25

44 mediated bicarbonate transport leads to acidic airways and impaired bacterial killing (190). The CF pigs that received FIV-CFTR in our experiments showed a trend toward an increase in both tracheal ASL ph and bacterial killing. Cultured nasal epithelial cells from the pigs receiving FIV-CFTR also demonstrated partial restoration of anion channel activity and a significant increase in ASL ph comparable with WT levels, even after 2 weeks in culture. Together, these data provide the first evidence of CFTR correction in a large-animal model by a lentiviral vector. Three well-studied candidate viral vector platforms for CF gene therapy include adenovirus, adeno-associated virus, and lentivirus (16). In addition, nonviral approaches are under study and in clinical trials (222, 223). Each viral vector has its own unique attributes and limitations. In the presence of reagents that disrupt tight junctions, adenovirus robustly transduces airway epithelia. However, adenoviral-mediated transgene expression wanes quickly due to the robust immune-mediated response (224). Adeno-associated virus persists long-term but has a relatively small carrying capacity for a transgene as large as CFTR. In a manuscript by Steines and colleagues (192), this limitation is circumvented by using a previously described partial R-domain deleted CFTR (CFTRΔR) (91). Lentiviral vectors are promising candidates for CF gene therapy because they can persistently express a transgene of interest in the lung and potentially be readministered without immune suppression (146, 183). Lentiviral vectors evaluated for CF gene therapy include human immunodeficiency virus (HIV), simian immunodeficiency virus (SIV), and FIV, as well as equine infectious anemia virus (EIAV) (145, 147, 183, ). Typically, retroviral and lentiviral vectors are pseudotyped with vesicular stomatitis virus 26

45 (VSV-G), but its preference for basolateral entry requires disruption of the tight junctions for efficient transduction (126, 133, 141). Alternate pseudotyping envelope proteins that confer tropism to the airway include baculovirus GP64 (152), influenza hemagglutinin (HA) (227), and the Sendai F and HN proteins (145). Lentiviruses pseudotyped with these envelope glycoproteins efficiently transduce airway epithelia in vivo and persist through the lifetime of a mouse. Integrating vectors that transduce airway epithelial cells and express long-term are favorable candidates to achieve efficient and persistent phenotypic correction after a single administration. Progress for CF gene therapy has been hindered in part due to the lack of animal models that share similar lung disease phenotypes as humans. Though preclinical CF gene therapy studies show correction of CFTR anion channel defects both ex vivo and in vivo in mice (228, 229), phenotypic correction of animal models that recapitulate human CF lung disease is desirable to evaluate the potential of viral vectors as therapeutics. The development of the CF pig and ferret has opened the door to study CF disease pathogenesis, perform translational studies, and develop a standard of metrics to quantify endpoints for CF gene therapy. The sinuses are nearly universally involved in CF (230, 231). Sinus disease pathogenesis has been characterized in the CF pig model (218, 232). In welldifferentiated primary airway epithelial cultures from CF pig sinuses, delivery of CFTR by an adenoviral vector corrects the defective anion transport phenotype, suggesting that the CF pig is a good preclinical model for sinus gene therapy studies (218). In the present studies, we show that lentiviral-mediated delivery of CFTR in vivo complements the 27

46 anion channel defect in sinus epithelia and increases the ASL ph of epithelial cells cultured from a FIV-CFTR treated pig. We acknowledge that this study has limitations. We emphasize that this is a shortterm proof-of-principle study that does not assess the ability to treat or prevent disease, and we have not yet tested the ability to readminister the vector in pigs (183, 233). However, evidence of partial in vivo CFTR correction in this pilot study is encouraging. These findings set the stage for future studies of lentiviral vector gene therapy efficacy, persistence, and safety. Goals for future studies will include efforts to increase the transduction efficiency (e.g., modify envelope pseudotype and vector production), engineer the transgene cassette, and investigate the feasibility of repeated vector administration. Studies of longer duration are needed to assess the persistence of gene expression and the ability to target cells with progenitor capacity. This study achieved important milestones regarding the feasibility of lentiviral-mediated gene therapy for CF. The CF pig spontaneously develops lung disease similarly to humans with CF, experiencing the development of acidic ASL ph, bacterial infections, inflammation, impaired MCT, and airway remodeling (216). This allows us to measure CFTR correction in a large-animal model that recapitulates key features of CF in humans. Sinus and airway epithelia are easily accessible for vector delivery, and these tissues are anatomically available for in vivo and ex vivo analysis. Importantly, the endpoints that we developed for this study allowed us to identify and quantify CFTR-dependent gene transfer events. Historically, CFTR function has been assessed by measuring anion transport. The assays of ASL ph and bacterial killing provide additional end points for phenotypic correction. Future directions will also focus on safety. Study goals include the 28

47 measurement of innate and adaptive immune responses to vector application and transgene expression, the development of neutralizing antibodies, mapping of integration sites, and dose-escalation studies. This study and a study by Steines and colleagues using AAV to deliver CFTR to CF pigs (192) are the first gene therapy studies to our knowledge to quantify anion transport, ASL ph, and bacterial killing in a large-animal CF model. Materials and Methods Pigs CF pigs were generated by homologous recombination in fibroblasts as previously described (185). Gut-corrected CF pigs were generated by somatic cell nuclear transfer cloning (186). These CF pigs were housed at the University of Iowa throughout the study. For viral vector delivery, newborn pigs were anesthetized using 2% isofluorane while oxygen levels, pulse, and respiratory rate were monitored. For the bacterial killing assay, pigs were i.m. anesthetized with ketamine (20 mg/kg) and xylazine (both Akorn Animal Health; 2 mg/kg), and anesthesia was maintained with propofol (Fresenius Kabi USA;1 mg/kg) i.v. Animals were euthanized via i.v. Euthasol (Virbac AH Inc.; 90 mg/kg) after ph and bacterial killing were measured. Vector production The FIV vector used in this study was produced by the Indiana University Vector Production Facility in collaboration with the NHLBI Gene Therapy Resource Program. FIV expressed a codon-optimized porcine CFTR cdna (accession no. KT184306) under control of the Rous sarcoma virus (RSV) promoter, and viral vector particles were 29

48 pseudotyped with the baculovirus Autographa californica multicapsid nucleopolyhedrovirus GP64 envelope by transient transfection as described previously (234). Virus was concentrated 250-fold for in vitro studies and 1,000-fold for in vivo studies by overnight centrifugation at 9,000 x g and resuspended in α-lactose buffer. Virus was titered by real-time PCR by the University of Iowa Viral Vector Core (130) ( Primary cultures of airway epithelia Airway epithelial cells from CF pigs were isolated by enzymatic digestion, seeded onto semipermeable filters, and grown at the air-liquid interface as previously described (235). Cultures were maintained in media supplemented with Ultro-ser G (USG) and the following antibiotics: penicillin (50 units/ml), streptomycin (50 µg/µl), gentamicin (50 µg/ml), fluconazole (2 µg/ml), and amphotericin B (1.25 µg/ml). To transduce cells in vitro, FIV-CFTR concentrated lentiviral vector supernatants were applied to the apical surface of cultured primary airway epithelia overnight. In vivo viral vector administration FIV-CFTR was concentrated 1,000-fold, and 1.5 ml of vector was mixed at a 1:1 ratio with 2% methylcellulose (Methocel A4C; Dow Chemical Company) (152). Vector (1 ml) was delivered to nasal passageways by a bolus dose using a 24G Jelco catheter (Smiths Medical). For intratracheal delivery, a MADgic atomizer (LMA) was passed through the vocal cords under direct observation, and 2 ml of vector formulated with 1% methylcellulose was instilled. 30

49 Electrophysiology of CFTR anion channel CFTR correction was measured in Ussing chambers. Well-differentiated primary cultures or tissue explants were mounted into Ussing chambers (236). The apical and basolateral chambers were bathed in symmetrical Ringers solution (135 mm NaCl, 5 mm HEPES, 0.6 mm KH2 PO4, 2.4 mm K2HPO4, 1.2 mm MgCl2, 1.2 mm CaCl2, 5 mm Dextrose). CFTR Cl current was measured using a previously described protocol (236). After baseline transepithelial currents were measured, Amiloride (Sigma Aldrich) (100 µm) was used to inhibit Na+ channels, followed by 4,4 -Dilsothiocyano-2,2 -stilbenedifulonic acid (DIDS) (Sigma Aldrich) (100 µm) to inhibit calcium-activated Cl channels. Once current stabilized, we replaced the apical solution with a low Cl solution as previously described (194). We next applied the camp agonists Forskolin (Cayman Chemical) (10 µm) and 3-isobutyl-1-methylxanthine (IBMX) (Sigma Aldrich) (100 µm). After the current stabilized, GlyH-101 was added to block CFTR-mediated Cl current. Transepithelial voltage (Vt) was maintained at 0 to measure transepithelial current (I) and maintained under the voltage clamp. Localization of CFTR Primary airway epithelia were fixed in 10% Neutral Buffered Formalin (Leica Biosystems), paraffin embedded, and sectioned vertically. CFTR IHC was performed using mouse anti-cftr monoclonal antibody (769, CFFT) as previously described (237). qpcr CFTR mrna abundance was quantified by qpcr using SYBR green (Thermo Fisher Scientific). Total RNA was collected from freshly excised tissue using the Direct-zol RNA MiniPrep protocol (Zymo Research) and reverse transcribed using the High 31

50 Capacity cdna Reverse Transcription Kit (Applied Biosystems). Primer sequences: codon-optimized pig CFTR: Forward: 5ʹ-ACAGGTTCAGCAAAGACATCG-3ʹ, Reverse: 5ʹ- CAGTGGCGAGGAAGATGTAAG-3ʹ. RPL4 was used as a housekeeping gene: Forward: 5ʹ- AGCGCTGGTCATGTCTAAAG-3ʹ, Reverse: 5ʹ- TCCAGGCCTTAAGCTTCTTC-3ʹ. Cycle conditions: 48 C for 30 min; 95 C for 10 min; C for 15 sec, and 60 C for 1 min. ph and bacterial killing assays Tracheal ASL ph was measured by placing a noninvasive dual lifetime referencing planar optode, a ph-sensitive foil, directly onto the tracheal surface as previously described (190, 219). Bacterial killing assays were performed as previously described (190). As reported in these studies, S. aureus isolate SA43 was cultured to log-phase growth and conjugated to gold electron microscopy grids, which were coated with streptavidin and biotin. Bacterial-coated grids were then placed on the airway surface for 1 minute, rinsed with PBS, and immersed in SYTO9 and propidium iodide (Invitrogen) to determine bacterial viability (Live/Dead Bacterial Viability Assay, Invitrogen). Grids were analyzed via confocal microscopy and quantified by Image J. Statistics Statistically significant differences were calculated as indicated using one-way ANOVA comparison or Mann-Whitney nonparametric t test (Graphpad Prism). All data are presented as mean and ±SEM. P < 0.05 was considered statistically significant. (Originally published in JCI Insight(191).) 32

51 Figure 2: Correction of Cl- transport in CF pig primary epithelia with a lentiviral vector. Feline immunodeficiency virus based viral vector expressing cystic fibrosis transmembrane conductance regulator (FIV-CFTR; MOI = 5) was delivered to the apical surface of well-differentiated primary cultures of airway epithelial cells from CF pigs. Transepithelial Cl currents were measured in Ussing chambers. (A) An example of CFTR-dependent Cl current is shown. The average change in Cl current upon addition of forskolin/3-isobutyl-1-methylxanthine (F&I) (B) and GlyH-101 (C) in treated and naive cultures are shown. n = 8 epithelial sheets/treatment (collected from 3 donor pigs). *P < 0.01, Mann-Whitney nonparametric t test. (D) IHC using a CFTR antibody reveals apical localization of CFTR protein in ciliated cells (arrows). c, ciliated cells; nc, nonciliated cells. (E) IHC using a CFTR antibody on untreated CF primary airway epithelia. Scale bar: 250 µm (D and E). Staining performed by David Meyerholz. 33

52 Figure 3: Anion channel correction in tissue explants. Feline immunodeficiency virus based viral vector expressing cystic fibrosis transmembrane conductance regulator (FIV-CFTR) was delivered to the lung of newborn cystic fibrosis (CF) pigs. Two weeks after infection, freshly excised tracheal tissues were mounted into Ussing chambers to measure a change in transepithelial current in response to (A) forskolin/3-isobutyl-1-methylxanthine (F&I) or (B) GlyH-101, and freshly excised bronchus tissues were also tested for their response to (C) F&I or (D) GlyH-101. Diamonds indicate tracheal tissue, and circles indicate bronchus tissue from individual animals. n = 3 pigs; each data point represents 3 replicates/pig. *P < 0.05, one-way ANOVA comparison test. (E) RNA was harvested from trachea and bronchus to measure levels of CFTR by real-time PCR. n = 10 or 8 lung tissue samples from 3 treated pigs. The dotted line indicates the background CFTR mrna level in CF tissue. *P < 0.05, Mann-Whitney nonparametric t test. 34

53 Figure 4: Tracheal ph and bacterial killing ability in CF pigs. Feline immunodeficiency virus based viral vector expressing cystic fibrosis transmembrane conductance regulator (FIV-CFTR) was delivered to the lung of newborn CF pigs. (A) Tracheal ph measurements were taken prior to performing the bacterial killing assay. The tracheal window was removed, and tracheal ph was measured in situ using the noninvasive dual lifetime referencing planar optode. (B) Bacterial-coated grids were placed on the airway surface of the trachea and incubated for 1 minute. Immediately following, grids were subjected to a live/dead stain, imaged by confocal microscopy, and quantified using Image J. Bacterial killing represented as % of total number of bacteria. Diamonds represent data from individual pigs. *P < 0.05, **P < , one-way ANOVA comparison test. n = 2 6 pigs. The CF and non-cf control pigs (A and B) were shared with a manuscript by Steines, et al. Tracheal ASL ph and bacterial killing data were collected by Mahmoud Abou Alaiwa. 35

54 Figure 5: FIV-CFTR corrects CF pig ethmoid sinuses. (A) Well-differentiated ethmoid cultures from CF pigs treated with feline immunodeficiency virus based viral vector expressing cystic fibrosis transmembrane conductance regulator (FIV-CFTR) were mounted into Ussing chambers, and bioelectric properties were measured. The change in transepithelial current was measured in response to low chloride (Cl ), (B) forskolin/3-isobutyl-1-methylxanthine (F&I), or (C) GlyH-101. Black diamonds indicate individual animals. n = cultured epithelial cells from 3 pigs; each data point represents 3 replicates/pig. *P < 0.05, Mann-Whitney nonparametric t test. (D) Cultured ethmoid sinuses from CF, non-cf, and CF pigs that received FIV-CFTR were treated with seminaphtharhodafluor (SNARF) + dextran and imaged by confocal microscopy to measure airway surface liquid (ASL) ph. **P < , n = 9, one-way ANOVA comparison. ASL ph data was collected by Viral Shah. 36

55 CHAPTER 3: HYBRID NONVIRAL/VIRAL VECTOR SYSTEMS FOR IMPROVED PIGGYBAC DNA TRANSPOSON IN VIVO DELIVERY Introduction Our goal for gene therapy vector development for life-long genetic diseases such as cystic fibrosis (CF) is to create a vehicle with the ability to efficiently, safely, and persistently express a transgene in the appropriate cell types (238, 239). There are multiple viral-based vectors for delivering genes to airway epithelia. Each system has its advantages and disadvantages. Nonviral vectors provide an expanded tool-set for gene transfer to cells. For example, research with the cut-and-paste DNA transposon Sleeping Beauty pioneered the use of a recombinant transposon and transposase to achieve genomic integration of a transgene (240). Sleeping Beauty-mediated gene transfer resulted in functional correction of coagulation factor deficiencies ( ), lysosomal storage disease (244), as well as in cancer therapeutics (245). PiggyBac is also a DNA transposon and a promising alternative to Sleeping Beauty. Similar to Sleeping Beauty, piggybac is highly active when introduced into mammalian cells ( ), mediates long-term expression in vivo (249, 250), and is a potential therapeutic agent for multiple genetic diseases such as CF (193, 251). Recombinant piggybac transposon and transposase are typically co-delivered by plasmid transfection; however, the greatest barrier of delivering naked DNA is inefficient delivery to somatic cells in vivo. Yant et al. (204) incorporated the Sleeping Beauty integration machinery into Ad vectors. They observed that the Ad genome needed to circularize before releasing the Sleeping Beauty transposon (252, 253). They employed a Flp recombination system to 37

56 drive circularization of the Ad genome in target cells. Here, we explore the potential utility of adenoviral (Ad)- or adeno associated virus (AAV)-based vectors to deliver piggybac components to airway epithelia without an additional recombination step. For these studies, a second adenoviral vector expressing hyperactive insect piggybac transposase (ipb7) is co-delivered to achieve transposase-mediated genomic integration of the transposon. These novel hybrid vector systems provide valuable additional tools for in vivo gene transfer. Results DNA transposition from a piggybac/aav vector into the genome Hybrid piggybac/aav vectors expressing the mcherry reporter and puromycin resistance genes were generated (shown schematically, Figure 6A). For these studies, ipb7 was co-delivered with an Ad5 vector (Ad-iPB7). To determine if ipb7 mobilized the DNA cargo from the piggybac/aav vector and catalyzed integration into the genome, colony formation assays were performed in a mammalian cell line (Figure 6B). Colony formation assays are a widely used indirect measure of transposition efficiency. PiggyBac/AAV was delivered to HeLa cells at multiplicity of infections (MOIs) ranging from 100 to 100,000. Ad-iPB7 was simultaneously delivered at MOIs ranging from 0 to 100. Following vector transduction, cells were selected for puromycin resistance for ~2 weeks and colonies counted. Puromycin-resistant colony formation is indicative of a successful integration event. In the absence of Ad-iPB7, we observed a dose-dependent increase in puromycin resistant colonies, suggesting that there is baseline level of 38

57 integration from an AAV vector alone, consistent with previous observations (254). However, in the presence of Ad-iPB7, the colony numbers increased dramatically. As an additional control to confirm that the presence of Ad vector does not enhance piggybac/aav conferred colony formation in our assay, we substituted AdiPB7 with Ad-GFP (Figure 6B). In the presence of Ad-GFP, the number of puromycinresistant colonies closely resembled no transposase group. These data suggest that ipb7 is necessary to achieve an increase in puromycin resistant colony formation. In general, the levels of transposition were dose-dependent; however, the piggybac/aav (MOI 10,000)/Ad-iPB7 (MOI 100) condition (ratio 100:1) resulted in fewer colonies than the piggybac/aav (MOI 10,000)/Ad-PB7 (MOI 10) condition (ratio 1000:1). This result is perhaps counterintuitive because one would expect that if the transposon is constant and the transposase is increased, increased transposition should occur. Yet, we observed that a 1,000:1 ratio consistently resulted in the best fold-increase of colonies over background. This may be the result, in part, of cellular toxicity associated with high MOI delivery of adenoviral vector or overexpression inhibition. To determine if high MOIs of adenoviral vector might lead to decreased colony counts, we delivered Ad-GFP to HeLa cells at MOIs of 0, 1, 2, 10, 20, 100, and 200 and used flow cytometry to quantify live cells 24 hours later. There was very little variation in the percentage of live cells in all groups. The range was from 94.8 to 97.5%. Interestingly, the percentage of GFP-positive cells plateaued at an MOI of 20 (91.1%) and dropped off at MOI 200 (78.7%). These data suggest that the highest nontoxic MOI of adenoviral vector may not be the optimal dose for maximal gene expression. 39

58 To confirm transposase-mediated genomic integration, cellular DNA was purified and libraries were generated using LAM-PCR as described in Materials and Methods. The libraries were shotgun-cloned and Sanger sequenced. Of the 54 sequences, ~2/3 (37/54) were bona fide transposase-mediated genomic integrations and the remaining ~1/3 (17/54) were recovered vector. A transposase-mediated integration event is defined by the tell-tale sign of a precise junction between the transposon terminal repeat and genomic DNA occurring at a TTAA. A no transposase control was not included because not enough puromycin-resistant colonies could be recovered to yield enough DNA to generate the libraries. Based on the sequencing data, non-transposase-mediated genomic integrations cannot be distinguished from episomal vectors. All but three of the genomic integrations occurred at canonical TTAA sites. The remaining three integrations occurred at either CTAA (1 integration) or TTAG (2 integrations) sites. Low level NTAA or TTAN integration is consistent with plasmid delivered piggybac transposon (193, 255). These data strongly suggest that delivery with AAV allows for transposasemediated integration of the piggybac transposon. Next, we deep sequenced the libraries using the Illumina HiSeq 2000 platform and recovered 19,059 reads from piggybac/aav-transduced HeLa cells and compared the results to 31,078 reads recovered from HeLa cells that were transfected with standard piggybac transposon and ipb7 expression plasmids. For these analyses, we only included verified genomic integrations that included the TTAA junction between the piggybac terminal repeat and mappable human genomic sequence. Multiple metrics were used to determine if the piggybac integration patterns were altered following delivery by an AAV vector. Metrics included distance from transcription start sites, oncogenes, DNase 40

59 hypersensitive sites, CpG islands, gene density, gene expression, and GC content (Figure 7). We observed only subtle differences in the integration profile between AAV- and plasmid-delivered piggybac transposons or computationally selected matched random controls. There was a very modest increase in integrations near gene dense regions or CpG islands and piggybac/aav had a modest decrease in integrations near transcriptional start sites. To further test the utility of this hybrid system in vivo, piggybac/aav vector expressing firefly luciferase was evaluated for its ability to deliver and transpose into the genome of murine conducting airways. A dose of vector genomes (vg) formulated with 1% methylcellulose (152) was delivered via nasal instillation to immunocompetent Balb/c mice. Ad-iPB7 was co-delivered at a dose of pfu, achieving the 1,000:1 ratio as determined in vitro. These doses were chosen as a maximal titer that could be delivered in a 50 μl volume. Control mice were untreated (naive) or received Ad-Empty in place of Ad-iPB7. Beginning 4 days post-transduction, luciferase expression was measured at the indicated intervals using a Xenogen CCD camera imaging system (Figure 8A, B). Both in the presence and absence of ipb7, a decline in nasal bioluminescence was observed between 1 week and 4 weeks post-delivery. The subsequent luciferase expression in animals that received ipb7 stabilized at ~30% initial levels, whereas, expression in animals without ipb7 stabilized at ~6% initial expression. After 12 weeks, the surface epithelial cells were ablated by two consecutive treatments with the detergent 2% polidocanol (256). Monitoring of the bioluminescent expression resumed 2 days following the second polidocanol treatment. Ablation of the surface epithelia is followed 41

60 by a rapid proliferation and repopulation phase(257). This procedure helps distinguish between stable integration events in progenitor cells from episomal expression or expression only in mature epithelium. The luciferase expression in animals that received ipb7 restabilized at ~10% of starting levels and the expression in animals without ipb7 dropped to ~1% of the first timepoint (Figure 8A). These data suggest that the transposase catalyzed piggybac/aav integration into a progenitor population of nasal airway cells in mice. We isolated nasal septa from mice 12 months after being transduced with piggybac/aav and Ad-iPB7. Using a similar protocol as described for our in vitro integration library generation, we performed genomic DNA isolation followed by LAM- PCR. Using shot-gun cloning and Sanger sequencing, we mapped 47 piggybac genomic integrations (Figure 9, red arrows). Interestingly, we observed a much lower frequency of recovered vector (1 out of 48). This change in ratios may point to an in vitro sequencing artifact that may not be relevant in vivo. DNA transposition from a piggybac/ad vector into the genome To determine if an Ad5-based viral vector is a suitable delivery vehicle for piggybac transposon, we inserted the transposon sequence into the E1 region of a firstgeneration Ad5 vector. Similar to the AAV vector, the transposon expressed mcherry and puromycin resistance genes separated by a T2A element (Figure 10A). Transposasemediated genomic integration was determined by colony formation assays (Figure 10B). Unlike piggybac/aav, very few puromycin-resistant colonies were observed for piggybac/ad in the absence of ipb7. Interestingly, a piggybac/ad MOI of 10 was optimal for each increasing dose of Ad-iPB7. Addition of Ad-iPB7 (MOI = 100) resulted 42

61 in ~150-fold increase in colony count as compared to piggybac/ad without Ad-iPB7. The decreased colony formation counts in the high-dose groups may be the result, in part, of cellular toxicity associated with high MOI delivery of adenoviral vector. Transposase-mediated genomic integration in HeLa cells was again verified by LAM-PCR, shotgun cloning, and Sanger sequencing. Of the 65 informative sequences, 23.1% (15/65) were recovered vector and 76.9% (50/65) were confirmed genomic integrations. All but four of the genomic integrations occurred at canonical TTAA sites. The remaining four integrations occurred at TTAG, CTAA (two integrations), or TCAA sites. While the sample size is too small to draw conclusions concerning integration patterns, these data suggest that, like AAV, Ad is a functional delivery vehicle for piggybac transposon. To quantify gene transfer efficiencies, piggybac/ad expressing firefly luciferase was delivered to SCID mouse airways via nasal instillation. The bioluminescent signal was quantified using a CCD camera 5 minutes following intraperitoneal (i.p.) luciferin delivery. As shown (Figure 11A), in the presence of ipb7, expression stabilized by 8 weeks post-delivery at ~20 30% of the initial time point. In the absence of ipb7, expression continued to decline to ~2% of the initial time point. These results suggest that ipb7 confers persistent expression from a piggybac/ad vector in the airways of immunodeficient mice in vivo. Replicate studies in immunocompetent Balb/c mice resulted in loss of expression to naive levels, regardless of the presence or absence of transposase, by 4 weeks post-delivery (Figure 12). To discern the cell types transduced in vivo, similar experiments were repeated using a piggybac/ad vector expressing the visual reporter gene mcherry. As before, piggybac/ad was delivered via nasal instillation to SCID mice with either Ad-iPB7 or 43

62 Ad-Empty. Reporter gene expression in fixed frozen tissue was examined at 1 week and 21 weeks post-delivery. At 1 week post-delivery of piggybac/ad and Ad-iPB7, we observed abundant mcherry expression that was restricted to the surface epithelia of the conducting airways (Figure 11D G). At the 1 week time point, the pattern of mcherry expression was indistinguishable between the group receiving ipb7 and the group without ipb7 (data not shown). Furthermore, this pattern of expression is consistent with previous observations using Ad5 vector expressing β-galactosidase and formulated with methylcellulose (256). Consistent with the luciferase reporter gene results (Figure 11A), abundant expression was observed in the conducting airways of mice collected at the 21- week time point from the Ad-iPB7 cohort (Figure 11) but not the Ad-empty cohort (Figure 11I). As anticipated, no mcherry expression (Figure 11B) or evidence of inflammation (Figure 11C) was observed in control naive SCID mice. We isolated lungs from SCID mice 12 months after being transduced with piggybac/ad and Ad-iPB7. Importantly, abundant mcherry expression was still observed in large airways (Figure 11J, K) at levels similar to 21 weeks post-delivery (Figure 11H). We performed genomic DNA isolation followed by LAM-PCR, using a protocol similar to discussed for in vitro transduced HeLa cells. Using shot-gun cloning and Sanger sequencing, we mapped 49 piggybac genomic integrations (Figure 9, blue arrows). Similar to the piggybac/aav mice, but unlike the in vitro shot-gun cloning, we observed a low frequency of recovered vector (2 out of 51). These data strongly support the notion that the observed in vivo persistence of transgene expression in the absence of selection is the result of transposase mediated piggybac transposition from the Ad genome into the host genome. 44

63 piggybac/ad-mediated CFTR correction in CF primary airway cells Potential advantages of the piggybac/ad vector include a large packaging capacity and preparations with high titers. Because the CFTR cassette flanked by piggybac terminal repeats exceeded the packaging limit of AAV, we focused on piggybac/ad as a delivery vehicle. Specifically, we examined if a piggybac/ad-delivered CFTR cdna can persistently rescue the anion transport defect in CF airway epithelial cells. Well-differentiated human primary CF tracheal epithelia were cultured at air-liquid interface (235) and co-transduced with basolateral application of piggybac/ad-cftr (MOI = 50) with Ad-iPB7 or Ad-Empty. At progressive time points, chloride currents were measured in modified Ussing chambers as previously reported (236). Upon addition of forskolin and 3-isobutyl-1-methylxanthine (IBMX), a significant increase in campstimulated chloride current was observed in epithelia treated with piggybac/ad-cftr (Figure 13A). Changes in short circuit current are not observed in naive CF cultures (Figure 13A). Chloride current (I) increased by ~4 μa/cm2 in these cultures. This increase in camp-stimulated chloride current was significant compared to the Ad-empty controltreated cultures (Figure 13B; P< 0.001). Further, these levels of correction are consistent with our previous experience using Ad5-CFTR (256). This observation indicates that sufficient expression was achieved to functionally correct CF airway epithelia. Discussion The choice of a vehicle to deliver therapeutic genes to specific target tissues is a vital consideration. Important features for a CFTR gene transfer vector for airway delivery include a large packaging capacity, efficient transduction, persistent expression, 45

64 and the capacity to be concentrated and purified. There are multiple viral-based vectors for delivering genes to the airways, but none possess all of these attributes. Here, we codelivered piggybac/ad or piggybac/aav with Ad-transposase to cells in vitro and mouse airways in vivo and demonstrated efficient transduction, transposase-mediated integration, and persistent expression. Hybrid piggybac/ad and piggybac/aav vectors are valuable new tools for in vivo gene transfer. Our results indicate the piggybac/ad and piggybac/aav vectors transpose and express their transgenes in the host genome in vitro and in vivo. We observed transposase-dependent transposition activity in HeLa cells and in the airways of mice. These data suggest that these viral vectors can support piggybac transposition. This important observation contrasts with previous reports using the Sleeping Beauty based DNA transposon and Ad vectors (204). Yant et al. observed that the Ad genome required Flp-mediated recombination that resulted in circularization before releasing the Sleeping Beauty transposon. It is unclear why piggybac does not require this circularization requirement. The unmethylated Ad and AAV genomes may be more suitable for piggybac transposition as compared to Sleeping Beauty (258). The ability of the piggybac-based viral vectors to integrate without an intermediate recombination step greatly simplifies in vivo delivery and potentially increases integration efficiency. A prerequisite to life-long expression from a gene therapy vector is genomic integration into progenitor cells; therefore, integrating vector systems may have the greatest potential for treating genetic diseases. There is inherent risk when introducing a transgene with integrating vectors. Insertional mutagenesis may disrupt normal cell functions by inactivating an essential host gene or inappropriately causing expression of 46

65 an undesirable gene. However, the risk will vary depending on the vector used, the therapeutic gene, and the cell type targeted. In many cases, enhancer effects pose the greatest danger. Using deep sequencing, we mapped viral vector-delivered piggybac integrations in HeLa cells. Previous reports show that piggybac preferentially integrates near transcription start sites at a frequency of 16 20%, in a manner similar to murine moloney leukemia virus (248, ). However, we did not observe this pattern of integration near transcription start sites in our studies (Figure 7). In fact, we observed a slightly disfavored integration pattern at transcription start sites as compared to a matched random control consistent with our previously published mapping data using a plasmidbased delivery system (255). The reason for this discrepancy is unclear but may result from the transposon delivery method, the cell types transduced, or the library generation method. The integration profile of a high capacity Ad-delivered Sleeping Beauty vector was also recently evaluated using adapter mediated-pcr and a near random pattern was observed (261). This finding was consistent with a lentiviral-sleeping Beauty hybrid vector system or plasmid-based delivery ( ). Several features of adenoviral vectors make them attractive vehicles for delivering therapeutic genes such as CFTR, including their large carrying capacity, efficient gene transfer capabilities, ability to transduce nondividing cells, the ability to be grown to high titer, and ease of purification. Indeed, Ad-based viral vectors were the first to be used for gene therapy trials in CF patients (50). Therapeutic levels of CFTR mrna were achieved in the airway epithelium of CF patients, but expression quickly waned and subsequent administrations were limited by humoral immunity against the vector (72, 265). Thus, the two greatest impediments to adenoviral 47

66 vectors are robust immune responses and transient expression resulting from episomal expression. As demonstrated, the use of piggybac/ad can overcome the limitation of episomal expression; however, immune-deficient SCID mice were required to observe the effect. Consistent with previous observations using first-generation Ad based vectors, we observed that piggybac/ad-mediated transgene expression was transient in immunocompetent mice (183, 214, 266). This short-lived transgene expression is attributed to the induction of a cytotoxic T lymphocyte immune response against adenoviral antigens (266). Immunogenic responses to adenoviral vectors preclude the potential for readministration. In addition, there are increased levels of inflammation and cytotoxicity associated with adenoviral infections (267). In these studies, when both AdiPB7 and piggybac/ad were co-delivered, the total viral load delivered was pfu. This amount was chosen based on the following three considerations: (i) the optimal ratios determined in vitro, (ii) the titer of the vector preparations, and (iii) a maximal dose that could be delivered in a 25 μl volume. We chose Ad to deliver ipb7 because it efficiently and transiently produces a transgene product. Interestingly, in our Ad-iPB7 and piggybac/aav experiments, we only delivered a total dose of pfu of Ad-based vector and observed persistent transposon expression in immunocompetent Balb/c mice. As before, this amount was chosen based on the same three considerations. A potential explanation for persistent expression in immunocompetent mice from the Ad/AAV combination and not the Ad/Ad combination could be a threshold of Ad (~10 8 pfu) that triggers a cytotoxic T lymphocyte immune response when delivered intranasally. Perhaps, de-escalation studies would reveal a dose that results in stable expression of the Ad/Ad combination in immunocompetent mice. Pulmonary immune responses in mice 48

67 may not be predictive of humans. Our data serve as a proof of principle that piggybacmediated transgene delivery can lead to long-term transgene persistence in vivo but further testing in a large animal model is necessary. As we turn our attention to the future, there are at least four strategies to consider when addressing the immune response. (i) Helper dependent (HD)-Ad vectors lack any viral encoded genes and have precedence for long-term expression in vivo (197, 268, 269). Thus, helper-dependent (HD) piggybac/ad and HD-Ad-transposase vectors would likely attenuate a vector-mediated immune response. (ii) Transient immune suppression at the time of delivery could be considered. (iii) The vector load could be reduced by delivering both piggybac and a self-inactivating transposase from the same Ad vector. Alternatively, the transposase could be supplied as RNA or with an AAV-based vector. (iv) Alternate vector delivery systems could also be considered. Indeed, integrasedeficient lentiviral vectors have been successfully used to deliver Sleeping Beauty (262, 264). Our goal for CF gene therapy is to provide a life-long gene replacement strategy for the airways that would be efficacious regardless of the CFTR disease-causing mutation. Primary cultures of airway epithelia derived from humans with CF manifest defective CFTR-dependent anion transport. We determined that the hybrid piggybac/ad vector conferred CFTR expression in transduced cells and functionally corrected primary cultures of CF epithelial cells in vitro. We measured the persistence of camp-stimulated Cl- dependent short circuit current across the epithelia for 17 weeks (120, 129). The demonstration of functional correction of CFTR in primary CF epithelia is a notable benchmark that provides an important foundation for future in vivo studies. 49

68 A gene transfer vector with the capacity to persistently express CFTR in airways in vivo may prevent or significantly slow CF disease progression. A high priority will be to demonstrate persistence and stable restoration of CFTR function in large-animal models, such as the CF pigs or ferrets (147, 184, 186, 270). Long-term expression following a single vector dose will require vector delivery to airway cells with progenitor capacity without causing toxicity. Access to potential airway progenitor cells, such as keratin 5-positive basal cells, will be assessed by screening in vivo vector delivery techniques. Transient disruption of tight junctions may help assure vector access to the appropriate cellular compartments. Attaining long-term expression of CFTR and prevention of the progression of changes associated with lung disease in vivo would provide a powerful proof-of-principle for translational gene therapy. Materials and Methods Constructs The piggybac transposon constructs in these studies expressed either an mcherry-t2apuromycin N-acetyl-transferase (Puror) cassette or firefly luciferase for in vitro and in vivo studies, respectively. The mcherry-t2a-puror cassette flanked by the piggybac terminal repeats (TRs) was designed in silico and in vitro synthesized (GenScript, Piscataway, NJ). Gene cassettes within the transposons were driven by an RSV promoter and cloned from puc57 into paav2 vector to create piggybac/aav. For these studies the RSV promoter was chosen because in our experience the RSV promoter is sufficient to drive life-long expression in mice. The AAV2 vector was pseudotyped with the AAV5 serotype capsid. The transposon inserts were separately cloned into the AAV vector using 50

69 the restriction enzymes XhoI and NotI. The piggybac/ad construct was cloned with the same restriction enzymes and ligation protocols from puc57 into the Ad5 vector. The Ad-iPB7 transposase was cloned by cutting the hyperactive insect piggybac transposase15 from pcdna3.1/myc-hisa (Invitrogen, Grand Island, NY) to the Ad5 vector by restriction enzymes EcoRI and NotI. All clones were sequence confirmed. Adenoviral and Adenoviral-Associated Vector production was performed as a fee for service at the University of Iowa Viral Vector Core ( Colony formation assay HeLa cells were transduced in a 24-well plate ( cells/well) with piggybac/aav alone, piggybac/ad alone or in combination with Ad-iPB7 at the indicated MOIs in quadruplicate for 4 hours. At 24 hours post-transduction, each well was trypsinized and expanded into a 100 mm plate and placed under puromycin selection (0.5 µg/ml). The cells were cultured under puromycin selection for two weeks and the selection media was changed three times per week. Following selection, puromycin-resistant colonies were fixed with 4% paraformaldehyde, stained with methylene blue and counted. Each assay was repeated at least three independent times. Integration site recovery for Illumina HiSeq2000 sequencing. Integration sites were recovered as described previously (193). Briefly, HeLa cells ( ) were transduced with piggybac/aav (MOI = 10,000; vector genomes (vg)) or piggybac/ad (MOI = 10; plaque forming units (pfu)) in combination with Ad-iPB7 (MOI = 10) for 4 hours. Integrants were selected with puromycin (0.5 µg/ml) for 3 weeks. Genomic DNA from three separate transfections was extracted from the integration library using the DNeasy tissue kit (Qiagen, Valencia, CA). Pooled DNA (2 51

70 µg) was digested overnight with ApoI or BstYI at 50 and 60 C, respectively; DNA was purified with the QIAquick PCR purification kit (Qiagen) and ligated to ApoI and BstYI linkers overnight at 16 C. Nested PCR was carried out under stringent conditions using transposon end-specific primers AAACCTCGATATACAGACCGATAAAACACATGCGTCAATTTTACGC (primary) and AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGC TCTTCCGATCTXXXXCGTACGTCACAATATGATTATCTTTC (secondary; XXXX denotes bar code, underlined sequence indicates Illumina cluster-generation sequence) and linker-specific primers CGTAGGGAGCAAGCAGAAGACGG (primary) and CAAGCAGAAGACGGCATACGAGCTCTTCCGATCT (secondary). DNA barcodes were included in the second-round PCR primers in order to track sample origin. The PCR products were gel purified, pooled, and sequenced using the Illumina HiSeq2000 sequencing platform. Ethics statement All animal procedures were previously approved (Animal Protocol # ) by the Institutional Animal Care and Use Committee (IACUC) and in accordance with National Institutes of Health guidelines. In vivo delivery and bioluminescence All mice for this study were housed at the University of Iowa Animal Care Facilities. Female mice (6 8-week-old) were transduced intranasally with piggybac/ad or piggybac/aav vector plus Ad-iPB7 or Ad-empty formulated 1:1 with 2% methylcellulose (50 µl total volume) as previously described (152). We observed no signs of visible toxicity or mortality as a result of viral delivery. Animals were imaged at 52

71 indicated time points using the in vivo imaging system (Caliper Life Sciences, Hopkinton, MA). Luciferin substrate (200 µl, 15mg/ml; Caliper Life Sciences) was administered via intraperitoneal injection and the mice were imaged for 5 minutes. Data were analyzed using Living Image software (Caliper Life Sciences). Polidocanol treatments were performed in mice that received piggybac/aav-luciferase. At 3 months postdelivery, 25 µl of 2% polidocanol (Sigma, St Louis, MO) was administered to mice intranasally. A second dose was delivered 48 hours later. Luciferase expression was quantified by bioluminescent imaging at indicated time points before and after polidocanol treatment (256). Primary epithelial cultures and electrophysiology studies Tracheal epithelial cells from human CF lungs were isolated by enzymatic digestion, seeded onto permeable filters, and grown at air-liquid interface as previously described (235). Cystic Fibrosis Transmembrane Conductance Regulator (CFTR)-null porcine tracheal epithelia cultures were studied in modified Ussing chambers as previously described (236). Briefly, epithelia were bathed on both surfaces with solution containing: 135 mmol/l NaCl, 2.4 mmol/l K2HPO4, 0.6 mmol/l KH2PO4, 1.2 mmol/l CaCl2, 1.2 mmol/l MgCl2, 10 mmol/l dextrose, 5 mmol/l HEPES (ph = 7.4) at 37 C and gassed with compressed air. Baseline transepithelial currents were measured. After apical addition of 100 µmol/l amiloride (Amil) and 100 µmol/l 4,4 -diisothiocyanoto-stilbene- 2,2 -disulfonic acid (DIDS), currents were allowed to stabilize and the apical solution was replaced with a 4.8 mmol/l Cl solution containing 135 mmol/l D-Gluconic Acid, 2.4 mmol/l K2HPO4, 0.6 mmol/l KH2PO4, 1.2 mmol/l CaCl2, 1.2 mmol/l MgCl2, 10 mmol/l dextrose, 5 mmol/l HEPES (ph = 7.4) at 37 C and gassed with compressed air. camp- 53

72 dependent Cl current was stimulated by apical addition of 10 µmol/l forskolin and 100 µmol/l 3-isobutyl-1-methylxanthine (IBMX), and CFTR-specific Cl transport was inhibited with 100 µmol/l GlyH-101. Transepithelial voltage (Vt) was maintained at 0 mv to measure transepithelial current (I). Transepithelial electrical conductance (Gt) was measured by intermittently clamping Vt to +5 and/or 5 mv. Spontaneous values of Vt were measured by transiently removing the voltage clamp. Immunohistochemistry Approximately pfu of piggybac/ad plus Ad-iPB7 or Ad-empty vector in a 50 µl volume with 1% methylcellulose (1:1) was delivered via nasal instillation to the airways of 3-month-old male SCID/NCr mice (NCI, MD, Balb/c background, 01S11) mice under ketamine/xylazine ( mg/kg) anesthesia. Fluorescence analyses of mcherry expression in mice lungs were examined as described in ref. (147). Briefly, mice lungs were harvested from a subset of mice and fixed in 4% paraformaldehyde in PBS at 4 C overnight. Prior to fixation, lungs were gently inflated with 15% sucrose via the trachea to maintain lung architecture. After fixation, lungs were submerged in 15% sucrose for 8 hours and then in 30% sucrose overnight. All steps were performed at 4 C. Lungs were embedded in OCT media, frozen in liquid nitrogen, and 10 µm sections were obtained using a microtome at 26 C. The serial sections were then stained with hematoxylin and eosin using standard techniques. Images were captured using an Olympus BX60 fluorescence microscope (Leeds Precision Instrument, Minneapolis, MN). 54

73 Statistics All numerical data are represented as mean ± standard error. Standard one and two sample t-tests were performed using the statistical computer program R ( with the lme4 package. (Originally published in Molecular Therapy (194).) 55

74 Figure 6: piggybac/aav transduction and colony formation assays. (A) Schematic representation of the piggybac transposon expressing mcherry and the puromycin resistance gene, puromycin N-acetyl-transferase (Puro r ), driven by an RSV promoter and delivered by AAV2/5. Hyperactive piggybac transposase (ipb7) is delivered in trans by an Ad5 vector. ITR, inverted terminal repeat of AAV; TR, piggybac terminal repeat. (B) HeLa cells were cotransduced with increasing vector multiplicity of infections as indicated. Cells were selected for days with 0.5 µg/ml puromycin, stained with methylene blue, and the drug resistant colonies were counted. 56

75 Figure 7: The heat map summarizes piggybac distributions to genomic features. Deep sequencing analysis done by Thomas Bair. 57

76 Figure 8: AAV delivered piggybac transposon mediates persistent transgene expression in mouse airways. (A) piggybac/aav2/5 expressing firefly luciferase was co-delivered to the nasal airways of Balb/c mice with Ad-iPB7 or Ad-Empty formulated with 2% methylcellulose. At the indicated time points, mice were given luciferin via i.p. injection and luciferase expression was quantified by bioluminescent imaging. At 12 weeks post-delivery, mice received two doses of polidocanol via nasal delivery, as indicated by arrows and described in Materials and Methods. Luciferase expression was measured as photons/sec/cm 2 and is reported as a percentage of levels from week one. *p = 0.002, n = 10. (B) Representative in vivo imaging system images at selected time points are shown. 58

77 Figure 9: Mapping piggybac/aav or piggybac/ad integrations in mice. Mouse genomic DNA was isolated from mice 12 months following delivery of piggybac/aav or piggybac/ad. Red arrows indicated piggybac/aav integrations and blue arrows indicate piggybac/ad integrations. 59

78 Figure 10: piggybac/ad transduction and colony formation assays. (A) Schematic representation of the piggybac transposon expressing mcherry and the puromycin resistance gene, puromycin N-acetyl-transferase (Puro r ), driven by an RSV promoter and delivered by Ad5 vector. ipb7 is delivered in trans by an Ad5 vector. (B) HeLa cells were co-transduced with increasing multiplicity of infections of vector as indicated. Cells were selected for days with 0.5 µg/ml puromycin, stained with methylene blue, and the drug resistant colonies were counted 60

79 Figure 11: Ad delivered piggybac transposon mediated persistent transgene expression in mouse airways. (A) piggybac/ad expressing firefly luciferase was co-delivered to the nasal airways of SCID mice with Ad-iPB7 or Ad-Empty formulated with 2% methylcellulose. At the indicated time points, mice were given luciferin via i.p. injection and luciferase expression was quantified by bioluminescent imaging. Luciferase expression was measured as photons/sec/cm 2 and is reported as a percentage of levels from week one. *P < 0.001, n = 10. (B I) Histological analysis of the airways was performed on SCID mice that received nasal delivery of piggybac/ad expressing mcherry. (B) Fluorescent expression and (c) hematoxylin and eosin (H&E) images of naïve mouse lung sections of age matched control mice. (D, F) Representative fluorescent expression and (E, G) H&E images of mice 1 week post-delivery of piggybac/ad-mcherry and Ad-iPB7. (F, G) Higher power images of boxed areas in D and E, respectively. (H) Mouse airways 21 weeks post-delivery of piggybac/ad-mcherry and Ad-iPB7. (I) Mouse airways 21 weeks post-delivery of piggybac/ad-mcherry and Ad-Empty. (J) Fluorescent expression and (K) hematoxylin and eosin (H&E) images of airways of mice 1 year post-delivery of piggybac/ad-mcherry and Ad-iPB7. Arrow indicates rare mcherry positive cell. Scale bar = 500 µm. Asterisks indicate large airways. A DAPI stain was used to label nuclei. Representative images from three mice/condition. 61

80 Figure 12: Luciferase expression in immunocompetent mice. PiggyBac/Ad expressing firefly luciferase was co-delivered to the nasal airways with AdiPB7 or Ad-Empty formulated with 2% methylcellulose. Mice were imaged using IVIS Live Imaging at 1 and 4 weeks post-delivery. 62

81 Figure 13: piggybac/ad-cftr restores Cl - transport in CF tracheal epithelia. (A) Ad-CFTR or piggybac/ad-cftr were delivered to well-differentiated primary cultures of airway epithelia from CF human donors. Anion channel correction was measured 1 week post-transduction by Cl current responses to Forskolin/IBMX and GlyH-101 in Ussing chambers. (B) Well-differentiated tracheal epithelia cultures were co-transduced with piggybac/ad-cftr (multiplicity of infection (MOI) = 50) with AdiPB7 (MOI = 50) or Ad-Empty (MOI = 50). At the indicated time points, transepithelial Cl currents were measured in Ussing chambers. Bars represent means of change in Cl current upon addition of Forskolin (10 µmol/l) and IBMX (100 µmol/l). Error bars represent SEM. n = 3 epithelial cultures/treatment. *p < Ussing chambers run by Samantha Osterhaus. 63

82 CHAPTER 4: CFTR DELIVERY BY A HYBRID PIGGYBAC/AAV VECTOR CORRECTS THE AIRWAY DEFECT IN CYSTIC FIBROSIS PIGS IN VIVO Introduction Cystic fibrosis (CF) is an autosomal recessive genetic disease affecting multiple organ systems, however the leading cause of morbidity and mortality in CF patients is chronic lung disease. CF is caused by a mutant cystic fibrosis transmembrane conductance regulator (CFTR) gene which leads to dysregulated anion channel function (18, 271). Without adequate anion channel exchange, viscous mucus accumulates on the airway surface and becomes a breeding ground for bacteria ( ). Chronically established bacterial colonization and mucus accumulation lead to plugged airways and ultimately organ failure (276, 277). Currently, a lung transplant is the only way to remediate a lung damaged from the consequences of CF (278). CF is a single gene disease with over 2000 documented CFTR mutations ( Although the small molecule potentiator ivacaftor has provided relief for some patients with CF, this drug is only effective for 2-3% of the CF population (7, 279, 280). There is a need to treat all CFTR mutations, ideally using a single dose reagent to prevent onset of a lifelong disease. A major hurdle in the field of CF gene therapy has been the lack of an animal model that recapitulates CF human lung disease. The pig was selected as a CF model due to its size, large litters, and the similarity of its lung anatomy to humans (184). The CF pig is free from infection and inflammation at birth but develops lung disease within a few months (217). From this large animal model, we learned much about the development of CF lung disease. In the absence of CFTR, the ph of airway surface liquid 64

83 (ASL) is reduced and the lung exhibits a reduced bacterial killing ability (190). Through earlier studies with CF pigs, we established novel metrics for quantifying anion channel current, ASL ph, and bacterial killing ability in CF pigs. Additionally, both lentiviral (191) (Chapter 2) and adeno-associated viral (AAV) (192) vectors delivered to CF pigs partially corrected the phenotypic defect. As such, the CF pig is a relevant model for complementing CFTR to the airways and quantitatively measuring phenotypic correction. AAV is a promising gene therapy vector for multiple diseases, most notably neuromuscular and retinal diseases. Examples of AAV vectors in neuromuscular clinical trials include spinal muscular atrophy type 1 (SMN), intramuscular delivery of limbgirdle muscular dystrophy, type 2D (LGMD2D) (281) and Becker muscular dystrophy (BMD) (282, 283). Encouragingly, LGMD2D and BMD delivered by AAV demonstrated persistent correction in muscle function for at least 6 months in trial participants. Ocular gene therapy studies using AAV include delivery of RPE65 for treatment of Leber congenital amaurosis 2 (LCA2), Mer proto-oncogene tyrosine kinase (MERTK), and Rabescort protein 1 (REP1). These studies reported a safe vector delivery and gene correction which led to improved vision of participants (reviewed in (284)). AAV is a good vector choice for these models because muscle cells have low turnover (285) and retinal cells are relatively immunoprivileged (286) and highly permissive to AAV (287, 288). AAV has been tested in six Phase I and II CF clinical trials, but results from these trials reported no significant improvement in lung function (105, 106). The vector design in these trials were not optimal for several reasons: 1) The AAV vector relied on the weak AAV-ITR promoter to drive CFTR expression. Due to the size constraints of AAV, the 4.5kb CFTR cdna encompassed nearly the entire carrying capacity, leaving no room 65

84 for a strong promoter. To overcome this limitation, a shortened CFTR was created by removing a portion of the R-domain of CFTR, which retained functional anion channel activity (CFTRDR) (91). Additionally, a shortened synthetic promoter and poly adenylation signal were engineered to improve vector design (173). 2) The low prevalence of receptor and co-receptors for AAV2 on the apical surface of airway epithelial cells resulted in low gene transfer efficiency (289, 290). Since the AAV clinical trials, new AAV serotypes have been derived using directed evolution with enhanced tropism for airway epithelial cells (169, 192). 3) AAV is typically shuttled to the proteasome for degradation (291). Co-delivery with the proteasome inhibitor doxorubicin increases AAV expression through proteasome modulation (167). 4) Preclinical studies in the rabbit lung suggest that AAV does not persist in epithelial cells and will likely require readministration (87). Here, we describe an integrating AAV vector that delivers a transgene that integrates into the host genome to overcome the readministration requirement. In a recent publication by Steines et al., delivery of AAV-CFTRDR with tropism for pig airways (H22 capsid) to CF pigs led to increased anion channel activity, decreased ASL ph, and improved bacterial killing ability. Additionally, AAVH22 transduced ciliated and non-ciliated cells and expressed CFTR at the apical surface (173). In total, AAVH22-CFTRDR restores the phenotypic defect in CF pigs in vivo. AAV-mediated transgene expression persists for months in the lung (81). However, the ultimate goal is to express CFTR for the lifetime of an individual with CF. AAV episomal expression only persists for the life of the cell (80), but an integrating vector transducing a progenitor cell population would lead to continuous repopulation of 66

85 CFTR-positive cells at the airway surface. The nonviral piggybac DNA transposon is a two-part system composed of terminal repeats (TRs) flanking a gene of interest and a transposase that catalyzes integration of the gene of interest into the host genome (248). Using a viral vector as a delivery vehicle for the nonviral transposon system improves the delivery efficiency of the otherwise highly inefficient plasmid-based piggybac transposon to somatic cells. Previously, we reported that transduction by a hybrid piggybac/aav leads to persistent expression in mouse airways in vivo ((194), Chapter 3). In this study, we employ a piggybac with shortened TRs (292) to accommodate the size constraints of AAV to create an integrating hybrid piggybac/aav vector system for gene delivery. Using the gut-corrected CF pig model (186), we describe delivery of a shortened CFTR cdna using piggybac/aav pseudotyped with an H22 capsid. With this approach, we observed that a shortened CFTR cdna corrects the phenotypic defect in a CF pig model. Results piggybac/aav transposition in vitro We previously reported that AAV successfully delivers a piggybac transposon to the genome of HeLa cells (194). In that experiment, the piggybac transposon contained full length terminal repeats (TRs) (300 bp total) was delivered by a single stranded recombinant AAV and the transposase was supplied using an adenoviral vector (AdiPB7). In this study, we compared transposition activity by a transposon with minimal piggybac TRs (99 bp total) (292) delivered by single-stranded recombinant AAV delivered (Figure 14A). We performed a colony formation assay as previously described 67

86 (194). Briefly, HeLa cells were co-transduced with piggybac/aav-puro and Ad-GFP or Ad-transposase (ipb7) and maintained in puromycin-containing media for two weeks. Cells were then fixed, stained with methylene blue, and counted. Colony formation allows for indirect quantification of transposase-mediated integration. We observed colony formation in a dose-dependent manner only in the presence of the transposase (Figure 15A). To develop an alternative option to Ad for transposase delivery, we asked if transposase (194) delivered by an AAV vector would still confer high transposition efficiency. We delivered the transposase by a separate AAV vector and compared transposition to transposase delivered by an Ad vector (194). Additionally, we hypothesized that using a self-complimentary AAV (scaav) rather than a singlestranded AAV (ssaav) (Figure 14B) would lead to increased transposition activity since scaav bypasses second-strand synthesis and expresses earlier than a transgene delivered by ssaav. Unexpectedly, transposase delivered by ssaav led to equal or greater transposition activity as compared to scaav (Figure 15B). We conclude that the ssaavipb7 confers dose-dependent transposase mediated integration of the piggybac/aav- Puro transposon into the genome. Phenotypic correction in CF pigs by piggybac/aav-cftrdr In a short-term study to validate the piggybac/aav transposon in vivo, we aerosolized piggybac/aav-cftrdr (without transposase) formulated with 250 µm doxorubicin into the trachea and lungs of newborn gut-corrected CF pigs. In addition, a bolus dose was delivered to the nose of the same animals. Two weeks after delivery, we used a variety of CF metrics to measure the effects of CFTR complementation. Freshly excised trachea and bronchus tissues were mounted in Ussing chambers to assess their 68

87 bioelectric properties. The CF pigs that received piggybac/aav-cftrdr showed change in transepithelial current significantly greater than the untreated CF pigs upon treatment with the camp agonists forskolin and IBMX (F&I) and the CFTR-inhibitor GlyH-101 (GlyH) (Figure 16A). This was the case for both trachea and bronchus tissue. To determine if complementing CFTR in airway cells modified the tracheal ASL ph, we used a ph optode to measure the ASL ph of the tracheal surface. The CF pigs treated with piggybac/aav-cftrdr showed trends of an increased tracheal ASL ph and similar non-cf pigs, although not statistically significant (Figure 16B). Furthermore, we observed bacterial killing ability in treated animals that mirrored non-cf levels (Figure 16C). Based on the multiple metrics, complementing CFTR by a piggybac/aav vector corrects the phenotypic defect in CF pigs. Measuring ASL ph and viscosity in primary cells cultured from treated CF pigs We next examined ASL ph and viscosity in primary cultures of airway cells derived from CF pigs that received piggybac/aav-cftrdr compared to control primary cultures from naïve CF pigs and grew them at an air liquid interface until fully differentiated. Consistent with the trends of in vivo measurements, ASL ph was slightly elevated in cultures as compared to untreated controls, although not statistically significant (Figure 17A). In addition to decreased ASL ph, our working model suggests that another consequence of dysregulated ion exchange in CF is dehydrated and viscous mucus. We next assayed the ASL viscosity by fluorescence recovery after photobleaching (FRAP) (293). Here, we observed a reduced viscosity in cells cultured from CF pigs that received piggybac/aav-cftrdr relative to cultures from untreated CF pigs (Figure 17B). In CF cultures, the viscous mucus is partially immobilized after 69

88 being photobleached whereas CF+AAV cultures recover to near saline levels (Figure 17C). Together, these data indicate that the piggybac/aav-cftrdr vector can rescue the CF phenotypic defect as measured by 1) CFTR channel activity, 2) ASL ph, and 3) ASL viscosity. Discussion In this study, we report transposition by minimal piggybac TRs in a hybrid piggybac/aav vector system in vitro. Delivering the transposase by a separate AAV vector leads to transposase-mediated integrations in a dose-dependent manner. In vivo, we confirmed piggybac/aav-cftrdr function two weeks after aerosol delivery. In this experiment transposase was not included because the goal was to functionally validate CFTR expression from the transposon vector, not to achieve genomic integration. Future studies will be performed to validate persistence in vivo in pigs using a reporter gene. Phenotypic correction was observed by measuring a change in current, ASL ph, and bacterial killing ability. Additionally, we further assayed for ASL ph and viscosity changes in primary airway cells harvested from CF pigs after piggybac/aav-cftrdr delivery. Here, we validate a hybrid integrating viral vector system in vitro and correct the phenotypic defect in a CF pig model. A hybrid piggybac/aav is a nonviral transposon integrating vector system delivered by a viral vector. AAV is considered a low immunogenic vector and has been used to deliver CFTR and correct the anion channel defect in a large animal CF pig model (192). In a persistence study in mice, piggybac/aav-luciferase expression persisted for 6 70

89 months (duration of experiment) in the presence of transposase (Chapter 3, (194)). Polidocanol treatment was used to accelerate cell turnover at a 3-month time point by denuding surface epithelial cells. Luciferase expression following the polidocanol treatment drastically declined in the absence of the transposase, but stabilized in the mice that received transposase (194). In the CF pig, AAVH22 carrying a shortened CFTR transduces ciliated and non-ciliated airway epithelial cells and rescues the phenotypic defect in vivo (192). As such, these results are encouraging that AAV-delivered piggybac carrying a shortened CFTR could be a promising therapeutic to prevent onset of CFassociated lung disease. For a single dose, life-long gene addition strategy to be successful, genomic integration into progenitor cells is likely a necessity. As such, the risk of insertional mutagenesis must be weighed against the therapeutic potential. Deep sequencing is the current method of identifying where integrations occur, their proximity to oncogenes, and their likelihood of unintentionally activating potentially dangerous genes (294, 295). Studies in tumor prone mice are an established tool for determining if integration patterns lead to inadvertent oncogene activation (296, 297). In future studies, we intend to codeliver the piggybac/aav transposon with a ssaav carrying the transposase and map transposase-mediated integrations in wild-type pigs. The ultimate goal for CF gene therapy is to develop a single-dose reagent for treating CF. This could be accomplished by delivery of an integrating vector that transduces the appropriate cell type, restores the anion channel defect, raises the ASL ph, and repairs the bacterial killing defect. In CF individuals with the G551D-CFTR mutation who were treated with the drug ivacaftor, lung function, sputum bacterial content, 71

90 inflammation, and chest computed tomography (CT) scans were measured before and after one year of treatment. Using this small molecule potentiator resulted in decreased sputum P. aeruginosa density, inflammation, and mucus plugging with modest improvements in lung disease as measured by CT (298). Restoring CFTR function had a global effect on lung function even after onset of lung disease. However, in these studies, P. aeruginosa infections still persisted. Perhaps using a gene therapy approach to restore CFTR function prior to bacterial colonization in the lung could prevent onset of lung disease entirely. Steines et al. used directed evolution to modify AAV tropism for improved lung delivery in pigs. Following generation of an AAV vector that transduces the apical surface of pig airways in vivo, they delivered AAV-CFTRDR to gut-corrected CF pigs and rescued the phenotypic defect in these animals (173, 192). In our studies, we also delivered an AAV-CFTRDR (91) vector to gut-corrected CF pigs, however our vector included minimal terminal repeats of the piggybac transposon. While our study evaluated correction at an early time point in the absence of the transposase, we showed proof-ofprinciple evidence that our vector undergoes transposition in vitro. Future studies will include measuring persistence of piggybac/aav in the presence and absence of the transposase in wild-type pigs. We evaluated delivering the transposase by an Ad, ssaav, and scaav vector. While an efficient way to deliver the transposase is by a viral vector, other delivery options may be evaluated. The transposase is required for transposition into the host genome, but is not needed long term and transient expression may be preferred. Transposase delivery by mrna is a promising way to improve transposition efficiency 72

91 and quality of integration (299, 300). Another method of transposase delivery is through incorporation of the transposase protein into integrase-deficient lentiviral particles (301). Expressing the transposase short term could prevent any undesired effects of persistent transposase expression. Efforts to inactivate transposase following transposition include co-delivery with Herpes Simples Virus thymidine kindase (HSV-tk) for inactivation following ganciclovir treatment (302). Overall, there are several options for transient delivery of the transposase that confers persistent expression from an integrated transposon. In these studies, we show dose-dependent transposition activity with minimal transposon terminal repeats. We also show functional correction in CF pigs following aerosolized delivery of a shortened CFTR by a hybrid piggybac/aav vector as measured by anion channel activity, tracheal ASL ph, and bacterial killing. In the future, we plan to compare persistence of piggybac/aav in the presence and absence of the transposase. We predict that following lung cell turnover, a population of cells containing the integrated transposon will repopulate the airway epithelium and persistently express the gene of interest. The hybrid piggybac/aav transposon system is a promising vector system to efficiently and persistently complement CFTR in airway cells and prevent the onset of CF related lung disease. Materials and Methods Constructs Single-stranded recombinant (ss)aav carrying the piggybac transposon with minimal terminal repeats (TRs) carrying an F5Tg83 promoter driving mcherry-t2a-puromycin 73

92 was designed in silico and synthesized by Genscript (Piscataway, NJ). Human CFTRDR cdna was a kind gift from Lynda Ostedgaard (91) and was cloned in to the piggybac/aav vector with minimal TRs with restriction enzymes SpeI and SalI. The Ad-CMV-iPB7 transposase was previously described(194). ssaav and scaav vectors with F5Tg83 driving ipb7 expression was cloned using SpeI and KpnI restriction enzyme sites. AAV vectors were generated using AAV2 genome and pseudotyped with the AAV5 capsid for in vitro studies and AAVH22 capsid for in vivo pig studies. Ad and AAV vectors were produced as a fee for service at the University of Iowa Viral vector core (medicine.uiowa.edu/vectorcore/). Colony formation assay HeLa cells were transduced in a 24-well plate (5x10 4 cells/well) with piggybac/ssaav- F5Tg83-mCherry-T2A-Puro R with or without Ad or AAV-delivered ipb7 as indicated. AAV MOIs ranged from and Ad-GFP or Ad-iPB7 was delivered at an MOI of 10. Twenty-four hours post-transduction, cells were trypsinized and expanded into 100 mm plates in puromycin-containing media. Media was changed 3 times per week. Two weeks after expansion, cells were fixed in 2% paraformaldehyde, stained with methylene blue, and colonies were counted. CF pigs All animal procedures were reviewed and approved by the University of Iowa IACUC, Iowa City, Iowa in accordance with the United States Department of Agriculture and National Institutes of Health guidelines. CF pigs were generated by homologous recombination in fibroblasts as previously described(185). Gut-corrected pigs were created by somatic cell nuclear transfer cloning(186). All pigs were housed at the 74

93 University of Iowa Animal Care for the duration of the study. Animals were humanely euthanized by i.v. euthasol (90 mg/kg). In vivo viral vector administration Newborn (<1 week old) gut-corrected CF pigs were sedated with inhaled 2% isoflurane for viral delivery while monitoring oxygen levels, heart rate, and respiratory rate. Viral vector was aerosolized into the trachea by passing a Penn Century Microsprayer through a 2.0 endotracheal (ET) tube. Approximately 1x10 12 vector genomes (vg) of piggybac/aav-hcftrdr formulated with 200 µm doxorubicin (Sigma Aldrich, St. Louis, MO) was delivered to each pig. Vector was delivered to the nasal turbinates by a bolus dose through a 34G Teflon catheter. Two weeks later, pigs were analyzed for phenotypic correction. Ussing chamber studies CFTR anion channel correction was measured by Ussing chamber analysis. Freshly excised tissue or well-differentiated airway epithelial cultures were mounted in Ussing chambers and apical and basolateral chambers were maintained under a symmetrical Ringers solution (135 mm NaCl, 5 mm HEPES, 0.6 mm KH 2 PO 4, 0.4 M K 2 HPO 4, 1.2 mm MgCl 2, 1.2 mm CaCl 2, 5 mm Dextrose). Transepithelial current was measured as previously described (236). Baseline currents were measured and the following drugs were added to inhibit ion channels: Amiloride (100 µm) to inhibit Na + channels (Sigma Aldrich), and 4,4 -Dilsothiocya-no-2,2 -stilbenedifulonic acid (DIDS) (Sigma Aldrich) to inhibit Cl - (100 µm). After establishing a low Cl - gradient in the apical chamber, camp agonists Forskolin (10 µm) (Cayman Chemical) and 3-isobutyl-1-meth-ylxanthine (IBMX) (100 µm) (Sigma Aldrich) were added. After the current stabilized, GlyH

94 was added to block CFTR-mediated Cl - current. Transepithelial currents (I T ) relative to baseline measurements are reported. Tracheal airway surface liquid (ASL) ph and bacterial killing Pigs were anesthetized with ketamine (20 mg/kg) and xylazine (2 mg/kg) and sedation was maintained with propofol (1 mg/kg). After opening a tracheal window, a ph sensitive foil was placed on the tracheal surface as previously described ( , 219). Bacterial killing assays were performed as previously described ( ). Briefly, S. aureus isolate SA43 was conjugated to electron microscopy grids via biotin and streptavidin interaction and placed on the airway surface of the sedated pigs for 1 minute. Live/dead bacteria were imaged by confocal microscopy following SYTO and propidium iodide staining (Invitrogen, Carlsbad, CA). Live/dead bacteria was quantified using ImageJ. ph ASL measurements The airway-surface liquid ph (ph ASL ) was assayed using the ratiometric ph-sensitive dye SNARF-1. The SNARF-1 was conjugated to 70 kda dextran (Molecular Probes a division of Fisher Scientific) to trap the dye on the apical surface of the airway cultures. Two hours before assaying the cultures, a minimal amount of SNARF-1 dextran was applied to the cell cultures using a cell strainer. Fluorescence was measured with a Zeiss LSM510 inverted confocal microscope using an apochromatic 40x water objective (Zeiss) with the pinhole set to produce a 2 µm plane. SNARF-1 was excited with the 510nm argon laser and emissions were collected at and through an acousto optical tunable filter. Calibrations were performed in vitro using ringer solutions titrated the day of calibration at and maintained in a chamber during the fluorescence 76

95 measurement. Consistent with other studies ( ), our empirical SNARF-1 pk a of 7.39 is acid-shifted to the documented pk a of ~7.5 at room temperature. Each culture was placed within a humidified 5% CO 2 chamber containing a HEPES ringer (equilibrated to ph 7.4 with 5% CO 2 at prior to the experiment), and images were acquired. All incubations, measurements, and calibrations were performed at 37 C. For each cell culture, 4 images over 1 minute were acquired from at least 3 separate regions. Raw intensities from each region acquired by NES elements software were background subtracted in Microsoft Excel. Each region was averaged for each culture and then averaged per pig. Mean pig ratios were converted to ph values in GraphPad Prism by interpolating the calibration curve. Data are represented as mean values among donors. Viscosity and dye immobilization As described in (293), FITC conjugated Dextran was applied to the surface of airway cultures. Viscosity was measured by fluorescence recovery after photobleaching (FRAP), specifically the duration of time it took for the FITC-Dextran to recover in the photobleached regions of the culture. In the instances where full recovery did not occur, the proportion of the net dye loss was presented as a percentage of immobilized dye. 77

96 Figure 14: Schematics of hybrid vectors. (A) piggybac/aav transposon as delivered by a single stranded (ss) AAV. The transposon terminal repeats (TRs) flank thef5tg83 promoter driving expression of mcherry-t2a-puromycin or hcftrdr followed by a shortened synthetic polya(173). (B) The piggybac transposase (ipb7) driven by the F5Tg83 promoter followed by the short polya represented schematically in either the ssaav or self-complementary (sc) AAV vectors. 78

97 Figure 15: piggybac/aav forms puromycin-resistant colonies in the presence of the transposase. (A) HeLa cells were left untransduced or transduced with piggybac/aav-puro R at MOIs of 10 3, 10 4, or 10 5 in the absence of Ad, with Ad-GFP (MOI=10), or with Ad-iPB7 (MOI=10) for 4 hours. Twenty-four hours after transduction, cells were expanded from a 24-well plate into 100 mm tissue culture dishes and grown in puromycin-containing media for 2 weeks. HeLa cells were then fixed, stained with methylene blue, and colonies were counted. (B) ssaav-ipb7 and scaav-ipb7 form puromycin resistant colonies. HeLa cells were transduced as in (A) with indicated viral vectors. *p<0.05, n=4. 79

98 Figure 16: piggybac/aav2/h22-cftrdr corrects the anion channel defect in vivo. 3 newborn CF pigs received ~1x10 12 vg of H22 pseudotyped piggybac/aav-cftrdr via intratracheal aerosolization and nasal instillation. AAV-iPB7 was not included in this experiment. Two weeks after delivery, pigs were assayed for phenotypic correction. (A) Freshly excised tracheal and bronchus tissues from untreated CF pigs or CF pigs treated with piggybac/aav-cftrdr were mounted in Ussing chambers to measure their bioelectric properties. Changes in transepithelial current was measured in response to F&I and GlyH. (B) Tracheal ASL ph was obtained upon opening a tracheal window following sedation. A ph sensitive foil was placed on the tracheal surface and ph was measured by a ph optode. (C) Bacterial killing was determined by placing a S. aureuscoated electron microscopy grid on the tracheal surface for 1 minute. Grids are stained with a live/dead stain, imaged by confocal microscopy, and quantified using ImageJ. *p<0.05, **p<0.005, ***p< n=3. 80

99 Figure 17: Primary airway epithelia cultured from piggybac/aav-cftrdr treated pigs exhibits increased ASL ph and decreased viscosity compared to untreated CF pigs. Trachea/bronchus and turbinate primary airway epithelial cells were cultures from untreated CF pigs or CF pigs treated with piggybac/aav-cftrdr. (A) Airway epithelial cultures were treated with SNARF-1 and ASL ph measurements were obtained by confocal microscopy. (B) Viscosity was measured by coating primary cells with FITC-conjugated dextran and measuring tau of ASL and saline by fluorescence recovery after photobleaching (FRAP). (C) Viscosity of the cultures presented as the percentage of immobilized dextran. Threshold dotted line indicates saline background. *p<0.05, n=3. 81

100 CHAPTER 5: WIDESPREAD AIRWAY DISTRIBUTION AND PHENOTYPIC CORRECTION OF CYSTIC FIBROSIS PIGS FOLLOWING AEROSOL DELIVERY OF PIGGYBAC/ADENOVIRUS Introduction Cystic fibrosis (CF) is a common autosomal recessive genetic disease affecting more than 70,000 people worldwide, with approximately 1,000 new cases of CF each year (Cystic Fibrosis Foundation, CF affects multiple organ systems; however, lung disease is the leading cause of morbidity and mortality in CF patients. CF is caused by a mutant cystic fibrosis transmembrane conductance regulator (CFTR) gene which results in dysregulated anion exchange at the airway surface. Reduced levels of Cl - and HCO - 3 in airway surface liquid (ASL) leads to a relatively acidic ASL ph and decreased bacterial killing ability. There are greater than 2000 CFTR mutations ( associated with CF. Small molecule potentiators and correctors provide relief to a small percentage of CF patients with specific mutations (8, 307, 308), however there is still a great need to develop a universal treatment for all people with CF. Gene therapy is a promising candidate for CF regardless of the diseasecausing mutation. Importantly, CFTR delivery prior to the onset of lung disease could prevent CF-related lung complications. In 1992, three separate Phase 1 gene therapy protocols for delivering an adenovirus (Ad)-based vector to CF patients were approved (reviewed in(265)). The vector was used to deliver CFTR to the nasal or intrapulmonary airways of small cohorts of CF patients. Evidence for correction occurred in all cases but the effect was transient and immune responses were observed. Since that time, we have learned much about 82

101 adenovirus biology. For example, the receptor was assumed to be apically localized on airway cells simply because Ad is a respiratory virus. However, in 1997, the basolaterally localized Coxsackievirus and Adenovirus receptor (CAR) (309) was determined to be the receptor (124). Considerable research has identified multiple strategies to facilitate receptor access. Vector co-administration with lysophosphatidylcholine (LPC), a natural airway surfactant, drastically increases viral transduction of Ad in the lung (159). Ad has many attributes, such as the ability to grow to high titers, a large carrying capacity, and wide tropism. To achieve persistent gene delivery to airway cells, we created an integrating hybrid nonviral/viral vector termed piggybac/ad. The nonviral DNA transposon piggybac is comprised of a two-part system including a transposon (i.e., terminal repeats flanking a gene of interest) and a transposase which provides the catalytic transposition activity to mediate genomic integration. Using an Ad vector to deliver this transposon system, we observed transgene expression for at least one year in SCID mice ((194), Chapter 3). Additionally, we mapped transposase-mediated integrations in HeLa cells in vitro and to the mouse genome in vivo. Delivery of piggybac/ad carrying CFTR to the basolateral surface of primary human CF airway epithelia persistently corrected the anion channel defect in vitro for at least 17 weeks (194). Thus, we validated this hybrid vector system in vitro and in a small animal model. Moving forward to a large animal CF model, we selected the pig because of its similarities to humans in lung anatomy, size, and physiology (217). CF pigs develop lung disease similar to humans with CF (216, 310). Previous studies indicate that by 6 weeks of age all CF pigs have airway inflammation and remodeling. While the airways of the 83

102 newborn CF pigs are free of inflammation and infection, they exhibit a reduced ASL ph and bacterial eradication defect (216, 311). Investigation of these disease features and their physiologic and biologic underpinnings has allowed for novel assay development and validation, which influenced our study design and the selection of endpoints. In this short-term study, we delivered piggybac/ad expressing GFP and Gaussia luciferase to the airways of wild-type pigs to determine viral distribution in the airways by quantifying the number of cells transduced and to identify cell types transduced. Using gut-corrected CF pigs (186), we delivered piggybac/ad-cftr to the airways and measured phenotypic correction by anion channel activity, ASL ph, and bacterial killing ability. We also measured the ASL ph and viscosity in primary cells cultured from these pigs. Here, we report quantification of whole lung viral distribution of piggybac/ad vector and phenotypic correction in a large animal CF model. Results Distribution of piggybac/ad in pig pulmonary airways We performed a thorough analysis of the number and types of cells transduced in pig airways after transduction with an integrating hybrid vector, termed piggybac/ad. This vector confers persistent gene expression in vitro and in vivo (194). The nonviral DNA transposon piggybac is made up of terminal repeats (TRs) that flank a transgene of interest. In our studies, we used a construct carrying the TRs flanking a promoter driving GFP-T2A-Gaussia luciferase (gluc); this cassette was then cloned into the E1 region of Ad (Figure 18A). In the absence of the transposase, piggybac/ad is functionally indistinguishable from first generation Ad, but in the presence of the transposase, the 84

103 gene expression cassette is integrated into the genome. In these studies, we aerosolized piggybac/ad carrying GFP and gluc with Ad-transposase (ipb7) or Ad-mCherry formulated with 0.1% LPC to the trachea of 6 newborn non-cf pigs (Figure 18A). The Ad-mCherry served as a negative control for the transposase as well as way to measure how frequently two Ad vectors transduced the same cells. Five days post-transduction, serum, broncheoalveolar lavage (BAL), and lungs were collected. Each of the six lung lobes were divided into two to four sections for whole-lung analysis, in addition to proximal and distal portions of the trachea (Figure 16B). We next quantified the number of GFP positive cells in large and small conducting airways. In nearly all cases, we consistently observed greater than 30% transduction in airways ranging from 50 µm 2 to >1001 µm 2 (Figure 18C). GFP positive cells were identified in airways throughout the entire lung. In order for transposition to occur, both parts of the integrating piggybac system must be in the same cell. We co-delivered piggybac/ad-gfp-t2a-gluc with an AdmCherry vector. To measure the percentage of dual positive transduction in vivo, GFP and mcherry positive cells were quantified by counting. We observed 55% of fluorescent cells to be dual positive (Figure 19). These data suggest that if a single airway cell is permissive to Ad transduction, a second vector is likely to transduce the same cell. To determine relative distribution of GFP throughout pig airways, we performed quantitative real-time PCR on genomic DNA (gdna) or quantitative reversetranscriptase PCR on mrna isolated from twenty different regions of the lung from each pig (as shown in Figure 18B). As a general trend, we observed GFP gdna and mrna in all regions of each lung lobe (Figure 20A, 20B). These results, in combination with the 85

104 fluorescence studies, suggest that piggybac/ad aerosolized to the airways of newborn pigs will distribute to all regions of the lung. Ad transduces all major cell types in the airways including ciliated and nonciliated epithelial cells, goblet cells, basal cells, and submucosal glands (SMGs) (196). Here we used immunofluorescence to visualize cell types transduced by piggybac/ad. Strikingly, we observed GFP positive cells in the surface epithelium of the distal trachea as well as high levels of transduction in the SMGs along the trachea (Figure 21A). At a higher magnification, we found that both ciliated (c) and nonciliated (nc) cells were GFP positive (Figure 21B-C). Additionally, basal cells were GFP positive (Figure 21D-F). As expected, piggybac/ad has wide cell tropism in the airways; however, the levels of SMG transduction were particularly remarkable. Phenotypic correction via piggybac/ad-cftr Functional correction in a large animal model is vital to validating piggybac/ad as a gene therapy vector. Here, piggybac/ad-cftr was aerosolized to four newborn CF pigs. Three days post-transduction, we measured phenotypic correction by a variety of metrics. In freshly excised tracheal tissues from pigs that received piggybac/ad-cftr, we observed increased current in response to the camp agonists Forskolin and IMBX (F&I) and decreased transepithelial current in response to the CFTR inhibitor GlyH-101 (GlyH). In contrast, these responses are nearly absent in tissues from untreated CF pigs (Figure 22A). In these same animals, we also observed a restoration of ASL ph (Figure 22B) and bacterial killing ability (Figure 22C) to wild-type levels. Lastly, CFTR colocalizes with GFP positive cells in vivo (Figure 23A-D). Based on these end-point 86

105 metrics, delivery of CFTR by a piggybac/ad hybrid vector corrects the CF phenotype in vivo. To determine if the ASL mucus viscosity had been modified, primary airway epithelial cells from untreated, treated, or noncf (wild-type) pigs were seeded onto clear membranes and grown at an air liquid interface for two weeks (Figure 24A). ASL ph and viscosity were measured as previously described (293, 312). ASL ph was measured using SNARF-1 and imaged by confocal microscopy. As shown in Figure 24B, the ASL ph from the primary airway epithelial cultures from piggybac/ad-cftr treated CF pigs appear to have an ASL ph similar to noncf pigs. These cultures also tend to have a decreased viscosity trending towards noncf levels (Figure 24C). In addition to presenting the viscosity data as a ratio of t ASL /t saline, viscosity can also be expressed as a percentage of the immobilized FITC-Dextran dye after fluorescence recover after photobleaching (FRAP). In Figure 24D, the primary cultures from piggybac/ad treated pigs have less immobilized dye, indicating that the dye can freely refill the photobleached region of the cultures under less viscid conditions. In total, complementing CFTR using piggybac/ad in CF pigs rescues the phenotypic defect as measured by Cl - current, ASL ph, bacterial killing, and ASL viscosity. Discussion Here we present two methods of assessing gene transfer efficacy in the airways of a large animal model in vivo: quantifying reporter gene distribution and measuring phenotypic correction following delivery of a therapeutic gene. To thoroughly investigate the deposition of piggybac/ad particles following aerosolization to wild-type pigs in 87

106 vivo, we systematically divided pig lungs into twenty regions and assessed GFP expression in each sample. We found high levels of transduction by immunofluorescence and whole lung distribution at the DNA and RNA level. We show Ad-mediated transduction of surface airway epithelia and submucosal glands. These data systematically detail the distribution pattern within the lung. In CF pigs, we delivered piggybac/ad-cftr and assayed for functional correction by quantifying the change in transepithelial current in response to F&I and GlyH in freshly excised tracheal tissue. We also measured ASL ph and bacterial killing in the trachea. Additionally, we cultured primary airway epithelia from untreated, CF pigs, piggybac/ad treated CF pigs, and non-cf pigs. Here, we measured the ASL ph and viscosity of the cultures and found that the primary airway cells from piggybac/ad- CFTR treated pigs exhibited a trend toward non-cf levels of ASL ph and viscosity. In total, we quantified viral vector distribution in pig airways and show phenotypic correction of a large animal CF model by a hybrid piggybac/ad vector carrying CFTR. A major hurdle for Ad as a gene transfer vector for CF gene therapy is the immune response which results in transient transgene expression in the airways ( ). Cytotoxic T cells (CTLs) target and eliminate transduced airway cells (63) and neutralizing antibodies prevent vector readministration (318, 319). Adenoviral genes activate multiple signaling pathways which induce expression of several proinflammatory cytokines, including TNF-a, IL-1b, IL-6, IFN-g (158), IL-2, IL-10 (73), and IL-8 (320). Transgene expression begins to decrease within 24 hours post-delivery (314) and cytokine levels are substantially decreased within 10 days of delivery (321). In our studies, while we did not collect samples at an early time point, we performed ELISAs at 88

107 5 days post-delivery on the BAL and serum. By 5 days, the cytokine levels were relatively low (data not shown) compared to early cytokine levels previously reported (322). Ad hexon antibody staining on pig airway sections 5 days post-delivery did not detect any hexon protein (data not shown). As a proof-of-principle for using gluc as a persistence reporter gene in vivo, we quantified the amount of secreted luciferase in the BAL and serum. In both cases, we observed drastic increases in relative light units (RLUs) compared to samples from untreated pigs (Figure 25). Although this is a shortterm study, we validated a secreted reporter gene as a potential marker for persistence studies. In these studies, we used piggybac/ad as a tool to quantify viral vector distribution throughout pig airways. Moving forward, we aim to generate a hybrid piggybac/helper-dependent Ad (HDAd) vector carrying CFTR to validate persistence in a large animal model in vivo. Our goal is to develop a vector in which a single delivery to the airways could prevent onset of CF lung disease. HDAd is an attractive gene therapy vector because it retains a high transduction efficiency in the lung but lacks all viral coding genes, yielding a lower immunogenicity (323). Multiple studies highlight the tropism for HDAd, particularly in a large animal pig model (324). Hu and colleagues have evaluated the ability to readminister HDAd vectors in the lung (176, 181, 197, 324) and reported transgene expression in the submucosal glands (196). Transducing a progenitor cell population such as basal cells or submucosal glands are important to persistently express CFTR and repopulate the surface airway epithelium following cell turnover. The submucosal gland is an important target because those cells express high levels of CFTR in noncf human tissues (325). In a pilot study, we delivered HDAd 89

108 expressing GFP to the airways of pigs and observed high levels of GFP expression in the trachea (Figure 26A) and conducting airways (Figure 26B). In these studies, we show high levels of transduction by an integrating hybrid vector system in key progenitor populations including cytokeratin 5 positive basal cells and submucosal glands. We previously validated the hybrid piggybac/ad system for expression and persistence in mice and showed that expression persisted in SCID mice for 1 year when co-delivered with the transposase. Another hybrid nonviral/viral vector system, the transposon Sleeping beauty (SB) employing a Flp-recombinase system was delivered to the liver of mice by helper-dependent (HDAd) and persisted for at least 22 weeks in the presence of the transposase following a partial hepatectomy (204). Other studies using HDAd vectors to deliver the SB system also reported persistent expression in vivo (252, 326). An advantage to using the piggybac transposon in a hybrid vector is that it does not require a Flp-recombinase to integrate. Additionally, piggybac has a large carrying capacity and, unlike SB, does not leave a footprint following excision (327). Life-long gene expression from a single dose of a gene delivery vector likely requires genomic integration. Integrating vectors show promise for treating genetic diseases; however, there is inherent risk when introducing foreign DNA into a cell. Insertional mutagenesis may disrupt normal cell functions by inactivating an essential host gene or inappropriately causing expression of an undesirable gene. To date, malignant cell transformation after vector-mediated insertional mutagenesis has only been observed in three clinical entities: X-linked severe combined immunodeficiency (SCID-X1), chronic granulomatous disease (CGD), and Wiskott-Aldrich syndrome (WAS), all of which occurred in conjunction with the use of first-generation gamma- 90

109 retroviral vectors harboring LTRs with strong enhancer/promoter sequences ( ). These studies were conducted in immunocompromised patients where gene transfer conferred a selective advantage to corrected cells. Considerable effort has been put toward mapping integration patterns and determining the functional consequences of lentiviruses (LV), DNA transposons, and retroviral vectors (reviewed in (333, 334)). Montini and colleagues demonstrated that LV integrations, even at high vector titer loads, did not accelerate tumorigenesis in tumor prone mice. In contrast, gamma-retroviral vector transduction triggered a dose-dependent acceleration of tumor onset (296). PiggyBac/Ad vector is breaking new ground in terms of integration mechanisms and delivery efficiencies. We are currently mapping the integration patterns of piggybac/ad co-delivered with the transposase and look forward to analyzing the data to determine the integration profile of the piggybac system in these experiments. Overall, these studies contribute a significant advancement for developing an integrating vector for long-term preventative treatment of CF airway disease. Materials and Methods Pigs CF pigs were generated by homologous recombination as previously described (185) and the gut-corrected pigs were generated by somatic cell nuclear transfer cloning (186). Wild-type pigs were acquired from Exemplar. Pigs were anesthetized using inhaled 2% isoflurane for viral delivery and i.m. ketamine (20 mg/kg) and xylazine (2 mg/kg) (both Akorn Animal Health) and maintained with i.v. propofol (Fresenius Kabi USA ;1 mg/kg) for the bacterial killing assays. During anesthesia, oxygen levels, heart rate, pulse and 91

110 respiratory rate were monitored. Pigs were euthanized by i.v. Euthasol (90 mg/kg). All pigs were housed at the University of Iowa during the duration of the studies. All animal procedures were reviewed and approved by the University of Iowa IACUC, Iowa City, Iowa in accordance with the United States Department of Agriculture and National Institutes of Health guidelines. Constructs and vector production The hybrid piggybac/ad vector was designed with the piggybac transposon terminal repeats (TRs) flanking a CMV-eGFP-T2A-Gaussia Luciferase (gluc) cassette. The piggybac/ad construct was modified from the hybrid vector in ref.(194). Briefly, the CMV-eGFP-T2A-gLuc cassette was cloned into the piggybac/ad vector using HindIII and SpeI. PiggyBac/Ad-CFTR with an F5Tg83 promoter was also cloned using HindIII and Spe restriction enzyme sites. Ad-iPB7 was created as in ref (194). Adenoviral vectors were produced by the University of Iowa Viral Vector Core (medicine.uiowa.edu/vectorcore). In vivo viral vector delivery Ad vector titers ranging from 2x10 10 to 2x10 11 pfu/ml were mixed with lysophosphatidylcholine (LPC) for a final concentration of 0.1% LPC in a total volume of 2ml. The viral vectors were aerosolized into pig airways by intratracheal instillation using the PennCentury Microsprayer. Ussing chamber studies Freshly excised tracheal explant tissues or well-differentiated primary airway epithelial cultures were mounted in Ussing chambers to measure their bioelectric properties as referenced (191). The apical and basolateral chambers were bathed in symmetrical 92

111 Ringers solution (135 mm NaCl, 5 mm HEPES, 0.6 mm KH2PO4, 2.4 mm K2HPO4, 1.2 mm MgCl2, 1.2 mm CaCl2, 5 mm Dextrose) and CFTR Cl- current was measured as previously described (236). Bacterial killing assay The bacterial killing assay was developed by Pezzulo et al. (190). Briefly, S. aureus (strain SA43) was immobilized on gold electron microscopy grids through a streptavidin and biotin interaction. Bacterial-coated grids were placed on the tracheal surface of a sedated pig for 1 minute and queried for cell death by a propidium iodide stain (Live/Dead Bacterial Viability Assay, Invitrogen) to determine cell viability. Grids were imaged by confocal microscopy and analyzed using ImageJ. Immunofluorescence Frozen sections were probed with anti-acetyl tubulin (Cell Signaling Technology, D20G3, K40, Danvers, MA), anti-cytokeratin 5 (CK5) polyclonal antibody (ThermoFisher, PA , Waltham, MA) or anti-cftr. Secondary antibody was Donkey Anti-rabbit Alexa 546. Slides were counterstained and mounted with Vectashield containing DAPI (Burlingame, CA). Quantitative Real-Time PCR Genomic DNA was harvested using the DNeasy Genomic Isolation Kit (Qiagen, Germantown, MD). RNA was harvesting using the RNeasy Lipid Tissue Mini Kit (Qiagen, Germantown, MD) with RNase-Free DNase Set (Qiagen, Germantown, MD) and cdna libraries were generated using Superscript IV Vilo Master Mix (ThermoFisher, Waltham, MA). GFP gdna and mrna abundance was normalized to 50 ng of each sample and quantified by qpcr using SYBR green (Applied Biosystems, 93

112 Foster City, CA). Primer sequences for GFP were: Forward 5 - GGCGACGGCCCCGTGCTGCTGC 3 ; Reverse 5 CACGAACTCCAGCAGGACCATG 3. For gdna, Pig RPL4 was used as a housekeeping gene (Forward: 5 - AGCGCTGGTCATGTCTAAAG 3 ; Reverse: 5 TTCCAGGCCTTAAGCTTATTA 3. b-actin was used as a housekeeping gene for mrna: Forward: 5 - CTGCGGCATCCACGAAAC - 3 Reverse: 5 - GTGATCTCCTCCTGCATCCTGTC - 3. Statistics All statistically significant differences were calculated using one-way ANOVA in Graphpad Prism. All data are presented as a mean ±SEM. P<0.5 was marked as statistically significant. 94

113 Figure 18: Experimental design and quantification of transduced cells in conducting airways. (A) Ad5 carrying a piggybac transposon with the terminal repeats (TRs) flanking a reporter gene cassette. Vector was co-delivered with AdiPB7 or Ad-mCherry and 0.1% LPC via intratracheal instillation to newborn wild-type pigs. 5 days post-delivery, lungs were harvested for analysis. TR=piggyBac terminal repeats, pa=poly adenylation signal. Schematic of pig lung diagram exemplifies how tissues were portioned for analysis. (B) Schematic of lung regions for whole lung analysis. Prox trach=proximal trachea, dist trac=distal trachea. Each lobe was divided into two to four regions. (C) Quantification of GFP positive cells in conducting airways of sizes ranging from 50 to <1001 µm 2. Sections were counterstained with a DAPI nuclear stain. Airway sizes were calculated using ImageJ. 95

114 Figure 19: Dual positive GFP and mcherry cells in the airways. Individual channels of (A) GFP, (B) mcherry, or (C) merged image of transduced airways. (D) Representation of percentage of GFP and mcherry dual positive cells among transduced airways. Scale bar=500 µm. 96

115 Figure 20: Distribution of GFP in pig lungs. Pig lungs were divided into twenty regions as depicted in Figure 18B and sampled for whole lung distribution. (A) Quantitative real-time PCR of relative GFP levels in genomic DNA from each region of the pig lung 5 days post-delivery of piggybac/ad- GFP. GFP levels are normalized to housekeeping gene RPL4 and fold changes are normalized relative to untreated animals (n=3). (B) GFP mrna expression levels from each region of the pig lung. Gene expression is normalized to the housekeeping gene pig b-actin, and fold changes are expressed relative to levels in untreated animals (n=3). 97

116 Figure 21: piggybac/ad-gfp co-localizes with major cell types. (A) piggybac/ad-gfp transduces submucosal glands throughout the trachea. Each region of the pig lung went through a sucrose gradient, was embedded in OCT, and frozen. Tissues were sectioned at 10 µm, counterstained with DAPI, and mounted for imaging by immunofluorescence. (B and C) Tissues stained with acetyl-tubulin (red), counterstained with DAPI, and imaged as in (A) show that piggybac/ad-gfp transduces ciliated (c) and non-ciliated (nc) cells at the airway surface. (D, E and F) piggybac/ad-gfp co-localizes with CK5 positive basal cells (red) within in submucosal glands (n=6). Scale bars= 500 µm. 98

117 Figure 22: piggybac/ad-cftr corrects the CF pig phenotype in vivo. (A) Schematic of experimental design: piggybac/ad-cftr was co-delivered with either Ad-transposase or Ad-GFP. (B) Anion channel current was measured in freshly excised tracheal tissues from CF pigs, CF pigs that received piggybac/ad-cftr (3 days postdelivery), or non-cf pigs. Change in current was measured in response to F&I and GlyH following establishment of a low Cl - gradient. (C) Tracheal ASL ph was measured in vivo and using the noninvasive dual lifetime referencing planar optode. (D) Bacterial killing was quantified by immobilizing S. aureus (pig isolated SA43) on electron microscopy grids through streptavidin and biotin conjugation. Grids were placed on the tracheal surface of a sedated pig for 1 minute, stained with propidium iodide, and imaged by confocal microscopy. Live/dead bacteria were quantified using Image J (n=4). Ussing data collected by Laura Marquez-Loza, bacterial killing performed with assistance from Brajesh Singh. 99

118 Figure 23: piggybac/ad-cftr co-localizes with Ad-GFP in vivo. PiggyBac/Ad-CFTR was co-delivered with Ad-GFP to newborn CF pigs. 5 days postdelivery, tissues were fixed in 4% paraformaldehyde, subjected to a sucrose gradient, and embedded in OCT. Frozen tissues were sectioned and immunostained for CFTR using an Alexa 647 secondary. (A) Merged GFP, CFTR, DAPI, and DIC image. (B) GFP and CFTR co-localization. (C) GFP only. (D) CFTR immunostaining only. Scale bar=10 µm. Immunofluorescence performed by Lynda Ostedgaard. 100

119 Figure 24: Airway epithelia cultured from piggybac/ad-cftr treated CF pigs retains correction in ASL ph and viscosity. Primary airway epithelial cells from untreated and treated CF as well as noncf pigs were isolated into a single cell suspension and grown at an air liquid interface for two weeks. (A) ASL ph of the primary cultures was measured by confocal microscopy following treatment with SNARF-1. (B) Viscosity of primary cultures was measured by fluorescence recover after photobleaching (FRAP) using a FITC-Dextran and is expressed as a ratio of t ASL to t saline. (C) Recovery of FITC-Dextran after FRAP was quantified as a percentage immobilized dye for untreated CF, treated CF, or noncf pig cultures (n=4). Data collected by Ian Thornell. 101

120 Figure 25: Gaussia luciferase expression in BAL and serum. Newborn noncf pigs received piggybac/ad expressing GFP-T2A-Gaussia luciferase (gluc) and tissues were collected 5 days after delivery. Upon tissue harvest, broncheoalveolar lavage (BAL) and serum were collected to measure gluc expression. Data presented as relative light units (RLUs) over untreated noncf pigs. n=3. 102

121 Figure 26: HDAd transduces trachea and conducting airways in vivo. HDAd-CMV-eGFP was aerosolized into the airway of a newborn pig. 5 days postdelivery, tissues were collected and imaged. (A) En face image of freshly excised trachea by an immunofluorescence dissecting microscope. (B) Tissues were fixed, embedded, and sectioned. Sections were collected and slides were counterstained with DAPI. n=1. 103

122 CHAPTER 6: DISCUSSION Progress for CF gene therapy has been partially hindered by the lack of an animal model that develops lung disease. In these studies, we report functional correction of CF pigs by three integrating vector systems including lentivirus (Chapter 2), piggybac/aav (Chapter 4), and piggybac/ad (Chapter 5). We measure phenotypic correction based on anion channel activity, tracheal ASL ph, and bacterial killing. The remainder of the discussion focuses on the fundamental properties of CF gene therapy and which CF metrics have been established in the field. Topics will include correcting appropriate cell types, options for integrating vectors, animal models, and outcome measures. This chapter will end with discussing the challenges associated with CF gene therapy and the potential to overcome these hurdles. Correcting the appropriate cells Because pulmonary disease is generally the most life-limiting complication of CF, gene therapy strategies focus on lung delivery of CFTR. Regardless of the gene delivery tool, an important consideration for CF gene therapy is the target cell. In the proximal airways, CFTR is normally most abundant in surface epithelial cells including ciliated cells, surface columnar cells, and submucosal gland epithelia (SMGs) (335); in distal airways only superficial epithelia express CFTR. With this information in mind, we face two important questions. 1) What cell types need to be transduced to attain lasting expression? 2) What percentage of cells needs to be transduced to correct CF lung disease? A goal of gene transfer to the pulmonary epithelium with integrating vectors is to correct the CFTR defect in a population of cells that could pass the corrected gene to 104

123 their progeny, thus eliminating the need for vector readministration. There appear to be several epithelial cell types in the lung that provide these functions, which has led to controversy regarding which cells to target for CF gene therapy. Arguments can be made in support of the necessity to correct basal cells (336, 337) and non-ciliated columnar cells of the airways ( ), SMGs ( ), club cells (344, 345), and alveolar type II cells (346, 347) in the distal lung. Compelling evidence from both in vitro and in vivo studies indicate that basal cells are multipotent proximal airway progenitor cells that repopulate pulmonary epithelia under normal conditions and during regeneration (reviewed in ( )). Cell-labeling experiments with transgenic mice show that basal cells give rise to labeled basal, ciliated, and club cells, thus fulfilling the definition of progenitor cells (351, 352). Several studies suggest that basal cells from human trachea or bronchi will repopulate denuded tracheal xenografts or differentiated epithelial cells in vitro ( ). Hematopoietic stem cells are an example showing that a single stem cell type can reconstitute a whole organ; however, there is no convincing evidence that a multipotent airway stem cell is capable of replenishing all regions of the intrapulmonary epithelium. The current literature supports that tracheal, bronchiolar, and alveolar epithelia are maintained by regionally distinct progenitor cell lineages. What percentage of cells needs to be transduced to functionally correct the CF phenotype in vivo? This is one of the most important questions in the field of CF gene therapy, but remains unanswered. At least five studies examined the relationship between percentage of cells expressing CFTR and transepithelial Cl- secretion (27, ). With relatively good agreement, they suggest that expressing CFTR in 5% 15% of cells 105

124 restores Cl- secretion to near wild-type levels. As such, the benchmark of correcting ~10% of the cells is often cited. However, there are many caveats to this number. Indeed, one limitation is that many of these studies were performed using in vitro models. In addition, as discussed below, other studies suggest that defective HCO - 3 transport through CFTR might be more relevant to early disease pathogenesis than Cl- secretion ( ). The relationships between HCO - 3 secretion, airway surface liquid ph, bacterial killing, mucociliary clearance, and mucus viscosity may be as important as Cl- secretion as metrics for disease correction. The short answer to the question posed above is we do not know ; however, given current animal models and improved vector technologies, the experiments necessary to address the question are feasible. Indeed, as we discuss below, existing vector technologies are being optimized for lung gene transfer and novel integrating vectors are being engineered. Options for integrating gene delivery Lentivirus Lentiviruses comprise a genus of the virus family Retroviridae. All retroviruses are defined by the ability to reverse transcribe their RNA genome and integrate proviral DNA into the genome of the host cell (202). Several features of lentiviral vectors (LVs) make them attractive vehicles for delivering therapeutic genes, including their large coding capacity, efficient gene transfer, persistent expression, directed tropism via pseudotyping, and lack of virus-encoded proteins that could elicit undesirable immune responses (140, 365, 366). Unlike gamma-retroviral vectors such as murine Moloney leukemia virus (MMLV)-based vectors, the pre-integration complex of lentiviruses can 106

125 transverse the nuclear envelope and integrate its cargo into the genomes of non-dividing cells (367). The first recombinant lentiviral vectors were based on human immunodeficiency virus type-1 (HIV-1) and remain the most widely used lentiviral vector for gene transfer applications (368). Feline immunodeficiency virus (FIV) is a non-primate LV with a less complex genome than HIV. Unlike HIV, wild-type FIV naturally lacks tat, vpr, vpu, and nef. In addition, FIV-based vector has vif deleted. LV delivery of CFTR is a promising option for CF gene therapy. In proof of principle experiments using reporter genes, HIV conferred gene transfer to both ciliated and basal cells of the mouse, sheep, marmoset, and ferret airways (149). Xenografts transduced with HIV expressing CFTR achieved functional correction as assessed by the measurement of transepithelial potential difference (128). In a CF mouse model, delivery of CFTR by HIV resulted in sustainable transgene expression for 18 months in ciliated, non-ciliated, and basal cells (140) and partially recovered the anion channel defect (225). Reporter gene studies demonstrate that FIV transduces cells in the conducting airways, bronchioles, and alveoli (147). FIV-mediated delivery of CFTR corrects the anion channel defect in airway epithelia (129). These findings support HIV as a vector candidate for CF gene therapy. Studies using FIV to deliver CFTR in a large animal CF model restores anion channel defect, rescues ph, and increases bacterial killing ability (Chapter 2). For LVs, the native envelope glycoprotein is deleted and a heterologous envelope is supplied. This strategy, termed pseudotyping, modifies vector tropism. Glycoproteins from a wide variety of enveloped viruses can be used to package LVs and, as will be discussed, multiple groups have identified envelopes that confer lung gene transfer. The 107

126 envelope glycoprotein from vesicular stomatitis virus (VSV-G) efficiently pseudotypes LVs, confers wide tropism, and is the most commonly used envelope glycoprotein. However, in well-differentiated airway epithelial cells, VSV-G pseudotyped LV (VSVG- LV) preferentially transduces the basolateral surface (133, 215). By using pretreatments or vehicles that transiently disrupt epithelial tight junctions, VSVG-LV accesses the basolateral surface of airway cells following luminal delivery (129). This strategy has also been shown to greatly improve VSVG-LV in vivo gene transfer efficiency in the lungs of mice (225). FIV pseudotyped with the baculovirus envelope glycoprotein (GP64) preferentially transduces polarized airway epithelia at the apical surface (215), results in persistent gene expression in mice [63], and supports gene transfer to airways of pigs (147). To prolong virus exposure to the airways immediately following delivery, formulating vector with a viscoelastic gel increases transduction efficiency in vivo (152, 369). Simian immunodeficiency virus (SIV)-based LV also successfully transduces airway epithelia. SIV carries little pathogenicity for its own host; therefore, modified strains of SIV could potentially be a safer alternative to HIV-based LVs (370). In studies using Sendai virus envelope proteins (F and HN) pseudotyped SIV (F/HN-SIV), a single dose persisted for the lifetime in the nasal epithelia of a mouse and achieved a dosedependent increase in reporter gene expression upon vector readministration (146). In vitro, F/HN-SIV carrying CFTR can generate functional chloride channels (145). Additionally, SIV transduction results in persistent transgene expression in both differentiated human airway cells and freshly excised human lung tissue (146). 108

127 Equine Infectious Anemia Virus (EIAV) is another non-primate lentivirus that has been investigated as a gene transfer vector. It has been studied extensively for neurological disease applications, such as Parkinson s disease (371). EIAV pseudotyped with the influenza HA envelope can transduce neonatal mouse airways, most notably, the nasal and lung epithelium. Readministration of HA-EIAV resulted in decreased gene transfer efficiency (227). DNA transposons Recombinant DNA transposons are integrating nonviral vectors that confer efficient and stable transgene expression in a variety of cell types. The DNA transposon Sleeping Beauty (SB) transposes its genetic cargo into host genomic loci using its catalytic transposase activity (243). Other transposons such as piggybac and Tol2 have also been used in gene transfer applications (reviewed in (372, 373)). DNA transposons are attractive tools for gene therapy because they have a large carrying capacity and integrate into the genome. Generally, transposon-based vectors are delivered as plasmids. This poses limitations for delivery to some somatic cell types. To improve delivery, several formulations have been investigated. Belur and colleagues described a protocol for complexing SB with polyethyleneimine (PEI) for delivery to the airways of mice. This non-viral vector approach overcomes biological barriers and allows for chromosomal integration; however, to date, this strategy has not been applied to delivering CFTR to the in vivo airways. Hybrid DNA transposon/viral vectors Delivery of a transposon system was first shown by Yant et al. Here, they showed that SB delivered by a helper-dependent Ad vector can integrate into the host 109

128 chromosome in a transposase-dependent mechanism (204, 243). Interestingly, in this system, the initial transposition of SB out of the Ad genome required the use of Flprecombinase to successfully deliver the transposon transgene (204). In our studies, the generation of a hybrid piggybac/ad and piggybac/aav facilitated delivery of a nonviral transposon. However, unlike SB, neither piggybac/ad nor piggybac/aav required an extra recombinase step for transposition to occur (194). In contrast to the hybrid SB, piggybac has not exhibited overexpression inhibition or limitations on size of the genetic cargo (374). PiggyBac/Ad hybrid vector successfully delivered a CFTR expression cassette to primary airway epithelial cultures in vitro that corrected the anion transport defect up to 4 months in culture. In reporter gene studies, transgene expression persisted for the 1-year duration of the experiment in mice (194). Animal models of cystic fibrosis Animal models serve an important testing ground for somatic cell gene transfer applications. Mice with null mutations (29, ), specific disease associated CFTR mutations ( ), and conditional CFTR null alleles (381) have contributed to the understanding of molecular mechanisms of CF. However, mice do not recapitulate several aspects of CF lung disease pathogenesis. As discussed above, many studies evaluating integrating gene transfer vector delivery to the lungs of mice have been conducted. For these reasons, we now discuss efforts to deliver integrating vector to new animal models CF. 110

129 CF rat Tuggle and colleagues used zinc finger nucleases to disrupt CFTR exon 3 in rats (382). CFTR -/- rats recapitulate many aspects of human disease including intestinal obstruction, obstruction of the vas deferens, and abnormalities in nasal mucus production. It is currently not clear if CFTR null rats develop lung disease. To date, no gene correction studies have been reported in this novel model. CF ferret CFTR null ferrets were developed using AAV-mediated gene targeting in somatic cells, nuclear transfer, and cloning (270). Unlike CF mice, CF ferrets develop early and reproducible lung infections that make it a promising platform for testing lung-directed CF therapies (270). There are several potential reasons for these species differences. First, Ca++ -activated Cl- channels in the mouse airway may compensate for campmediated CFTR Cl- transport (99, 383), a pathway that appears to be less active in humans or ferrets (270, ). Second, in humans and ferrets, goblet cells are the predominant secretory cell type of the cartilaginous airways (184, ), whereas in mice the analogous secretory cell type is the club cell (390, 391). Third, SMGs are virtually absent in murine cartilaginous airways, with only a handful in the most proximal regions of the trachea (391, 392). SMGs are important for airway innate immunity in the ferret (393) and humans (394, 395), and a potentially valuable site for CFTR expression(325, ). Lentiviral gene transfer to the wild-type neonatal ferrets using EIAV- and FIVbased vectors expressing fluorescent reporter genes was recently reported (399). The EIAV was pseudotyped with hemagglutinin (HA) from avian influenza A virus (227) and 111

130 the FIV vector was pseudotyped with GP64 (215). A liquid bolus of the vector was delivered to newborn ferrets via a tracheal incision. Significant transgene infection was noted in respiratory epithelia of all lobes in both the conducting and small airways with both vectors. Cmielewski and colleagues delivered VSVG-HIV expressing the LacZ reporter to the lungs of 7 8-week-old ferrets (400). Considerably less gene transfer was observed as compared to the HA or GP64 pseudotyped LV gene transfer in the neonatal ferrets. Currently, it is unclear if these differences are due to the vector pseudotype, age of the ferrets, or delivery protocol; however, these data suggest that ferrets may be useful pre-clinical models for lentiviral vector development. CF pig Pigs are an important model for many studies of human cardiovascular diseases, injury and repair, surfactants, inflammation, and pulmonary diseases (reviewed in (184)). Compared to rodents, the pig lung is anatomically and physiologically more similar to humans (401, 402) and has been studied extensively in xenotransplantation. The prenatal maturation of the pig lung is similar to humans and includes extensive alveolarization (403). Pig airway branching and cell composition is much more akin to human airways than to those of mice. The cell types comprising the conducting airway epithelium in pigs and humans are similar, and notably lack the high percentage of club cells typical of mice. The pig bronchial epithelium is pseudostratified and contains ciliated, basal, and goblet cells, and abundant SMGs (reviewed in (184)). Importantly, the distribution of SMGs in the conducting airways and the CFTR-dependent and -independent secretion of liquid and macromolecules is similar to humans (396, ). 112

131 Pigs with CFTR null and F508 knock-in alleles were generated by AAVmediated homologous recombination and somatic cell nuclear transfer (385). Breeding heterozygous male and females generated homozygous CFTR-/- pigs, and their striking neonatal phenotype was described (185, 385). Newborn CF pigs exhibit severe disease similar to humans including pancreatic insufficiency, meconium ileus with intestinal obstruction, absence of the vas deferens, and evidence of liver and gall bladder disease (185). Importantly, CFTR null and F508 pigs spontaneously develop lung disease with many features similar to humans with CF including bacterial infection, inflammation, abnormal mucociliary clearance, bronchiectasis, and remodeling. In a recent study, we compared HIV- and FIV-based lentiviral vectors in welldifferentiated human and pig airway epithelia (147). FIV transduced pig airway epithelia with greater efficacy than HIV, but both FIV and HIV transduced human airway epithelia with equal efficacy (147). We further screened a number of envelope glycoproteins and identified GP64 as one of the most efficient pseudotypes for transduction and persistent expression in both pig and human epithelial cells (147). A mcherry marker virus was delivered to wild-type pigs 4 weeks of age. A bolus dose of GP64-FIV vector was delivered to the ethmoid sinuses or to the tracheal lobe through a catheter threaded through the suction channel of a pediatric bronchoscope. We estimated the range of transduction efficiencies in the pig airways to be from <1 to 7%. In future studies, we will deliver CFTR expressing vector to CF pigs determine the preferred gene transfer targets and the level of CFTR correction required to prevent or slow disease progression. 113

132 Outcome measures In pre-clinical studies of CF gene therapy, it is vital to define metrics of correction before the studies are initiated. It would be naïve to simply deliver vector to the airways and look and see if the airway disease is cured. CF is a complex disease with many phenotypic features and clearly defining the disease progression in an untreated CF animal model is vital. As experience is gained with new animal models, additional assays for correction will be established and refined. Importantly, these metrics may apply to multiple gene correction but may not be feasible in all animal models. Quantitative real-time PCR and CFTR protein expression Quantitative real-time RT-PCR is a sensitive assay for measuring vector expressed CFTR mrna (185). At progressive time-points post vector delivery, whole tissue or brushings of nasal or tracheobronchial epithelia can be obtained. Silent mutations can be engineered into the vector expressed CFTR cdna so that transgene expression can be differentiated from endogenous CFTR. Using a similar strategy, vector genome copy number can be estimated. Genomic DNA from a portion of the same tissue or epithelial brushings can be purified and copy number estimated by normalizing to endogenous DNA and a standard curve. Appropriate controls would include wild type and untreated affected littermates. To identify cells expressing the CFTR protein, immunohistochemistry and immunofluorescence protocols have been reported (185). Using these approaches, the percentage of cells expressing CFTR and the cell types expressing CFTR can be determined (407). 114

133 Functional correction Nasal potential difference (NPD) and Ussing chamber studies are established assays for demonstrating in vivo and in vitro, respectively. Correction of CFTRdependent Cl - transport ( ). For many of these studies, vector was delivered nasally and the NPD was used as evidence of CFTR complementation. Ideally, for integrating vectors, pre-delivery, early post-delivery, and late post-delivery timepoints in the intrapulmonary airways should be measured. As early as 1 week after gene transfer, the nasal voltage and Ussing chambers and their responses to amiloride, DIDS, low Cl -, camp agonists Forskolin and IBMX (F&I), and the CFTR inhibitor GlyH-101 (GlyH) (109, 185, 378). Animals could be followed with serial monthly nasal voltage measurements over a 12-month or longer period to document persistence of expression. Importantly, CFTR is an anion channel that conducts both Cl - and HCO - 3. As mentioned above, correcting ~10% of cells is often cited as a benchmark for restoring Cl - transport and correcting the clinical phenotype. However, other studies also suggest that defective HCO - 3 transport might be relevant to disease. Multiple studies of CF mouse cervical mucus (361, 362), CF mouse small intestinal mucus (363), and human CF nasal SMGs (364) support the importance of CFTR-dependent HCO3- transport in CF pathogenesis. Loss of CFTR-dependent HCO - 3 transport acidifies liquid produced by surface (190, 411) and secretions from SMGs (364). Thus, measurements of HCO - 3 transport may also be an important metric of functional correction. As a result of abnormal CFTR-dependent HCO - 3 secretion, airway surface liquid ph is acidified. The ASL of primary cultures of CF pig airway epithelia (190), newborn CF pig airways (190), and the nasal ph of newborn babies with CF (209) is acidic. In CF 115

134 pigs, the acidity has been shown to impair bacterial killing (190). In addition, there are new techniques to measure mucus viscosity, mucociliary clearance (MCT), and lung function in large animal models (184). Importantly, defects in airway MCT and SMG mucus detachment recently were identified in CF (210). The bacterial killing defect is a quantifiable characteristic of CF airways. Bacterial killing is impaired as a result of reduced bicarbonate anion secretion and increasing ASL ph rescues bacterial killing (190). Reduced infection and inflammation A goal of gene therapy for CF is to prevent the onset or reduce the progression of lung disease. Signs of reduced infection and inflammation in treated animals can be visually inspected in the airways. Bronchoscopy can be used to detect signs of inflammation, such as mucosal inflammation and excessive purulent secretions. Total cell counts, cell differentials, and cytokine levels are obtained from broncheoalveolar lavage (BAL) as standard assay for infection and inflammation. In addition, standard quantitative microbiologic techniques are used to identify and quantify BAL bacteria (412). Biopsies can also be obtained from larger animal models such as the CF pig. These samples can also be used for sequencing-based analyses. As we learn more about the disease progression in new animal models of CF, improved metrics of functional correction are being developed. High-resolution computerized tomography (HRCT) facilitates detailed structural analysis of the airways (184). HRCT scans can discern anatomic changes in the airways over time in control and treated animals (413). 116

135 Challenges to pulmonary gene transfer with integrating vectors Delivery The lung is an attractive target for gene therapy because, unlike most other tissues, the vector can be topically delivered. Vector delivery to the airways of mice, rats, and newborn ferrets is most easily accomplished by nasal or intratracheal bolus delivery of vector resuspended in a liquid vehicle such as buffered saline (134), LPC (149, 225), or a viscoelastic gel (152, 369). In small animals, bolus delivery using a relatively small volume of vector (25-50 µl) can achieve widespread gene expression throughout the airways. However, in large animal models such as pigs or sheep (and ultimately humans), aerosolization will likely be required to achieve a widespread pulmonary distribution. In general, devices for generating airborne vector fall into three categories: aerosolizing catheters, nebulizers, and atomizers. All of these devices convert liquids into particles small enough to be respired. Aerosolizing catheters convert liquids into particles at the point of expulsion. Typically, an aerosolizing catheter is first passed into the trachea and then the vector is instilled. The Microsprayer (PennCentury) and the AeroProbe catheter (Trudell Medical International) are examples of delivery systems for this application. According to the manufacturer, it is possible to generate particles with aerodynamic diameters of 4 8 µm. The Trudell AeroProbe was previously used to aerosolize helper-dependent adenovirus vectors to rabbit airways (181, 414) and to deliver Sendai virus vectors to sheep (415). Since Sendai virus and lentiviruses are both enveloped, it is likely that this approach is feasible with lentiviral vectors. In addition, aerosolized VSVG-LVs have been successfully delivered to the airways of mice (416). 117

136 Successful vector aerosolization has been reported in mice (179), rabbits (181, 414), pig (196) and sheep airways (417, 418). By 2 3 weeks of age, wild type pigs are large enough to be sedated and have a pediatric bronchoscope passed into the trachea. The AeroProbe catheter can be passed to the carina via the bronchoscope suction channel with the animal breathing spontaneously. In this way, an integrating vector can be aerosolized and targeted to specific bronchial segments. In the case of nebulizers, the liquid is first converted into mist and then passively inhaled. Using this strategy, a plasmid-based vector was delivered to the airways of CF patients in a phase IIB gene therapy trial (419). This approach could potentially be used to deliver DNA transposon or hybrid vectors. However, this strategy may not be feasible with enveloped viral vectors because this class of vectors may not be stable enough to withstand nebulization. In addition, nebulization requires a large volume of concentrated material; therefore, would be the least economic delivery strategy for LVs. Atomizers are a subclass of aerosolizing catheters that deliver larger sized particles. An atomizer, such as the MADgic TM (LMA) atomizer, delivers large droplets (~30 90 µm in diameter), which may vary in size depending on the force applied to the syringe plunger. This type of device is often used to topically deliver medications to the airways (420). Our group has observed that this type of atomizer is an effective delivery device for multiple viral vectors including FIV, particularly when formulated with a viscoelastic material such as methylcellulose. Insertional mutagenesis Since persistent gene expression from lentiviral vectors requires genomic integration, they show promise for treating life-long genetic diseases; however, there is 118

137 inherent risk when introducing a transgene with integrating vectors. Insertional mutagenesis may disrupt normal cell functions by inactivating an essential host gene or inappropriately causing expression of an undesirable gene. The risk will vary depending on the vector used, the transgene cassette, and the cell type targeted. In many cases, enhancer effects pose the greatest danger. So far, malignant cell transformation after vector-mediated insertional mutagenesis has only been observed in three clinical entities (X-linked severe combined immunodeficiency (SCID-X1), chronic granulomatous disease (CGD), and Wiskott Aldrich syndrome (WAS)), all of which occurred in conjunction with the use of first-generation gamma-retroviral vectors harboring LTRs with strong enhancer/promoter sequences ( ). These studies were conducted in immunocompromised patients where gene transfer conferred a selective advantage to corrected cells. The vector and disease settings likely influenced the risks for insertional mutagenesis and subsequent clonal expansion. Modern LVs are engineered to lack enhancer/promoter sequences within the LTRs and delivering CFTR to somatic cells has no known selective advantage. Considerable effort has been put toward mapping integration patterns and determining the functional consequences of LVs and retroviral vectors (reviewed in (333, 334)). LV integration analyses conducted on adrenoleukodystrophy clinical trial patients demonstrate that the genomic distribution maintain a polyclonal pattern (421). Montini and colleagues demonstrated that LV integrations, even at high vector titer loads, did not accelerate tumorigenesis in tumor prone mice. In contrast, gamma-retroviral vector transduction triggered a dose-dependent acceleration of tumor onset (296). 119

138 The burden of proof has fallen on LV researchers to demonstrate that LVs do not cause cancer via insertional mutagenesis. It is unlikely that this can ever be demonstrated with absolute certainty; however, the evidence to date suggests that current LVs are considerably safer than the gamma-retroviral vectors that were first brought to clinical trials. In fact, results from human clinical trials using LVs are encouraging and the feasibility of gene therapy for monogenetic diseases is now firmly established (422). Recent promising examples include Wiskott Aldrich Syndrome (423, 424), metachromatic leukodystrophy (425, 426), acute lymphoid leukemia (427), lymphoma (428, 429), and multiple primary immuno-deficiencies (430). Gene transfer and phenotypic correction using a porcine model In pilot studies, we aerosolized Ad-LacZ to the airways of wild-type pigs. Three days post transduction, we observed a variety of b-gal positive pig airway cells, such as ciliated cells, non-ciliated cells, basal cells (Figure 27A-B), and epithelial cells within sub-mucosal glands (Figure 27C). In the first gene therapy studies in CF pigs, Ad-CFTR was delivered to the airways of gut corrected CF pigs and assessed for functional CFTR expression. Forskolin and IBMX stimulated Cl - current and then GlyH-101 inhibited current, indicating CFTR function in freshly excised tracheal epithelia (Figure 27D). Further, we observed bacterial killing activity (Figure 27E) and CFTR expression in conducting airways (Figure 27F). Although unpublished, these data are the first evidence of phenotypic correction following CFTR complementation in CF pigs. 120

139 Conclusions Within a year of the discovery of CFTR, investigators validated the concept that gene replacement could reverse the ion transport defect in vitro, suggesting that gene therapy may be possible (21, 431). We and others have demonstrated that CFTR delivery by integrating vectors can correct the CF anion defect in vitro and in vivo, and although further pre-clinical trials are warranted, there is great potential for translating this strategy to the clinic. As discussed, estimates of the percent of CF epithelia requiring correction vary and there is debate about which cell types must be corrected to achieve phenotypic correction; however, interest in CF gene therapy remains strong as barriers to gene transfer are identified, outcome measures are established, CF animal models with lung disease are developed, and better delivery systems are engineered. Demonstration of corrective gene transfer to pristine newborn lungs in CF animal models is a vital first step before looking ahead to correcting more diseased lungs. Newborn screening for CF is now established in all 50 states, allowing early disease detection. This offers an opportunity to introduce an integrating therapeutic gene transfer vector to the airway epithelium prior to the onset of chronic infection and inflammation. This strategy is a potentially life-long curative therapy regardless of the disease-causing mutation. (Originally published in InTech Cystic Fibrosis in the Light of New Research (432).) 121

140 Figure 27: Adenovirus transduces airway epithelial cells, submucosal glands, and corrects the anion channel defect in vivo. Ad-b-gal vector was formulated with 0.1% LPC and aerosolized to noncf pigs. Five days after delivery, tissues were analyzed for b-gal expression. (A) Tracheal airway epithelia expressing b-gal positive cells at the airway surface. (B) High power magnification shows b-gal positivity in ciliated and non-ciliated cells, as indicated by black arrows. (C) b-gal positive cells in the submucosal glands (SMG). (D-F) Ad-CFTR with 0.1% LPC was aerosolized to CF pig airways in vivo and assayed for functional correction. (D) Tracing of tracheal explant Cl- current in response to Forskolin/IBMX and GlyH-101. (E) Bacterial killing assay quantifying percentage of S. aureus SA43 dead after 1 minute application to the tracheal surface. (F) Immunohistochemistry of CFTR, as indicated by black arrows. Scale bars: 500 µm. (A-C) n=3, (D-F) n=1. b-gal and CFTR staining performed by David Meyerholz, Ussing data collected by Sarah Ernst, bacterial killing data collected by Mahmoud Abou Alaiwa. 122