NEW STRATEGIES IN CANINE CORNEAL WOUND HEALING FOR THE VETERINARY OPHTHALMIC PATIENT _______________________________________ A Thesis presented to the

Faculty of the Graduate School at the University of Missouri-Columbia _______________________________________________________ In Partial Fulfillment of the Requirements for the Degree Master of Science _____________________________________________________ by ANN P. BOSIACK Dr. Elizabeth A. Giuliano, Thesis Supervisor DECEMBER 2012

The undersigned, appointed by the dean of the Graduate School, have examined the thesis entitled NEW STRATEGIES IN CANINE CORNEAL WOUND HEALING FOR THE VETERINARY OPHTHALMIC PATIENT presented by Ann Bosiack, a candidate for the degree of master of science, and hereby certify that, in their opinion, it is worthy of acceptance.

Professor Elizabeth A. Giuliano

Professor Jacqueline W. Pearce

Professor Cecil P. Moore

Professor Rajiv R. Mohan

ACKNOWLEDGEMENTS I want to express my deepest gratitude to Dr. Elizabeth Giuliano, my masters advisor as well as residency mentor. Her assistance and support have been instrumental to the completion of this project. She contributed an enormous amount of her time to helping me prepare and revise manuscripts and presentations. She is a patient and supportive mentor who has provided me with exceptional guidance and instruction throughout my residency and graduate career. I would like to sincerely thank Dr. Rajiv Mohan for making this project possible. He generously provided the laboratory, materials, and support needed to complete this project. His collaboration with the ophthalmology service at the veterinary school has been an invaluable aspect of my residency program. He helped extensively with the experimental design and execution as well as with the manuscript revision and submission. He has contributed an enormous amount of research in the corneal field of medicine and it has been an honor to work with him. I would also like to thank my other graduate committee members Dr. Jackie Pearce and Dr. Cecil Moore. Their guidance and support throughout my residency made it possible for me to accomplish a concurrent masters degree. As residency mentors I am, undeniably, ever grateful for all that they have taught me. Lastly, I would like to acknowledge and thank Rangan Gupta and all of the other members of Dr. Mohans laboratory for their support and assistance in the lab.

ii

TABLE OF CONTENTS ACKNOWLEDGEMENTS ................................................................................................ ii LIST OF ILLUSTRATIONS ............................................................................................. v LIST OF TABLES ............................................................................................................. vi Chapter 1. INTRODUCTION ....................................................................................................1 Canine Corneal Disease ......................................................................................1 Corneal Gene Therapy In Veterinary Medicine ...................................................2 Adeno-Associated Virus ......................................................................................4 Gene Transcription Modulators ...........................................................................5 2. GENE THERAPY IN COMPARATIVE OPHTHALMOLOGY ............................6 Vectors .................................................................................................................6 Potential Applications In Veterinary Medicine ....................................................9 Conclusions ........................................................................................................18 3. CANINE CORNEAL FIBROBLAST AND MYOFIBROBLAST TRANSDUCTION WITH AAV5 ...........................................................................20 Experimental Purpose ........................................................................................20 Materials And Methods ......................................................................................20 Results ................................................................................................................26 Discussion ..........................................................................................................28 4. EFFICACY AND SAFETY OF SUBEROYLANILIDE HYDROXAMIC ACID (VORINOSTAT) IN THE TREATMENT OF CANINE CORNEAL FIBROSIS................................................................................................................32 Experimental Purpose ........................................................................................32 Materials And Methods ......................................................................................32
iii

Results ................................................................................................................37 Discussion ..........................................................................................................39 APPENDIX ........................................................................................................................43 1. FIGURES ................................................................................................................43 BIBLIOGRAPHY ..............................................................................................................58

iv

LIST OF ILLUSTRATIONS Figure Page

1. Representative immunocytochemistry images demonstrating phalloidin staining for F-actin (shown in red) in canine corneal fibroblasts (a) and myofibroblasts (b). ......................................................................................................44 2. Representative images demonstrating delivered GFP staining in CCFs. .......................45 3. Representative images demonstrating delivered GFP staining in CCMs ........................46 4. Quantification of GFP positive cells showing the amount of gene delivery by AAV5 into CCFs as well as CCMs. .....................................................................47 5. Quantitative PCR measurements of AAV5-delivered GFP in CCFs and CCMs. ........................................................................................................................48 6. Phase-contrast microscopic images of CCF (a) and CCM (b) cultures taken 48 h after AAV transduction. ...........................................................................49 7. Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay demonstrating apoptotic cells (pink staining) in CCF cultures transduced with AAV5-GFP. Trypan Blue exclusion data (c) shows the percent viability of CCFs 24 h and 48 h following AAV transduction compared to controls. .................................................................................................50 8. A bar graph showing the cellular viability as determined by trypan blue assay in order to demonstrate the dose dependent effect of SAHA on CCFs. ..........................................................................................................................51 9. A composite image showing phase-contrast microscopy images of CCFs. ....................52 10. Representative images of TUNEL assay results detecting apoptosis of CCFs in the untreated control group (a) and in CCFs treated with 0.06% SAHA for 24 h (b). .........................................................................................53 11. Immunocytochemical staining using SMA antibody. ..................................................54 12. Bar graphs showing SMA immunostaining quantification. ..........................................55 13. Bar graphs showing quantitative real-time PCR results demonstrating the relative expression of SMA. ....................................................................................56 14. Western blot analysis of SMA protein. ........................................................................57
v

LIST OF TABLES Table Page

1. Characteristics of viral vectors used in corneal gene therapy..43

vi

CHAPTER 1: INTRODUCTION

1. CANINE CORNEAL DISEASE Canine corneal fibrosis is a significant clinical problem in veterinary ophthalmology. The cornea provides two thirds of the refractive power of the eye. Corneal transparency and clarity rely on the special organization of the cornea's constituent components.1 Corneal injury and/or infection often results in fibrosis2, which may cause significant visual impairment.3 Corneal wound healing following injury or infection often results in decreased corneal tissue transparency due to corneal fibrosis. The transformation of quiescent keratocytes to corneal fibroblasts and myofibroblasts has been identified as the primary event in corneal wound healing that is responsible for fibrosis. While the production of myofibroblasts is a normal part of the complex process of corneal wound healing, the overproduction of myofibroblasts can lead to vision loss due to corneal scarring.4 Canine ulcerative keratitis is a significant clinical problem in veterinary ophthalmology.3 Corneal fibrosis is a common outcome following complicated corneal ulceration and wound healing. Treatment of deep corneal ulcers typically demands both medical and surgical therapeutic strategies.5 Surgical procedures utilized for the repair of deep corneal ulceration include conjunctival grafts, corneal grafts, corneo-conjunctival transposition, natural material transplants, and application of tissue adhesives.6-9 These procedures and their associated inflammatory response typically results in moderate to severe corneal scarring. Currently, no

pharmacologic agents specifically aimed at reducing corneal fibrosis are routinely used in veterinary ophthalmology.

2. GENE THERAPY IN COMPARATIVE OPHTHALMOLOGY Almost all advances in ocular and vision research require the testing of new therapeutic strategies in animal models prior to pilot use in clinical trials. In vitro studies are commonly used prior to animal models in an effort to establish proof of principle and in some cases initial drug doses. Animal models are then needed to determine efficacy and establish toxicity levels associated with novel therapies. Although animal research is essential for furthering our understanding of vision and the treatment and prevention of ophthalmic disease in people, scientific advancement brought about through the use of animal testing is often slow to be adopted for use in the veterinary clinical patient compared to its use in physician ophthalmology. Leber Congenital Amaurosis (LCA) is an inherited degenerative retinal disease that causes blindness in children. A majority of LCA cases are caused by mutations in the RPE65 gene. A naturally occurring RPE65 mutation in Briard dogs was found, similar to LCA in people.10 A gene-knockout mouse model was produced, which in turn, greatly facilitated the development of gene-replacement strategies. Pioneer efforts in gene therapy for congenital LCA has demonstrated significant improvement of visual impairment in affected Briard dogs and is being used to help restore vision in children.11 This well-known scientific advancement serves to demonstrate the importance of animal models of ocular disease and the essential role they have played in the discovery of underlying genetic defects, mechanism of disease, and development of therapeutic strategies. While rodents are commonly used, use of larger animal models (i.e. dog,
2

cat, pig) has become increasingly attractive to assess efficacy and safety of a variety of treatment modalities that are being considered for clinical trials in people. The successful establishment of a canine animal model for retinal diseases such as LCA is likely one important reason why retinal gene therapy has advanced faster than corneal gene therapy. Comparative ophthalmic disease shared among many different species has long been recognized; however, recent international programmatical developments such as the One Health Initiative have fostered improved collaborations among health science professions, especially physicians and veterinarians. Similarities between the genomics and pathogenesis of many ocular diseases shared between people and animals has encouraged a convergence of knowledge between physician and veterinary ophthalmologists with significant potential to benefit the prevention, treatment, and diagnosis of many ophthalmic diseases. The cornea represents one of the most evolutionary conserved anatomic components across mammals with many shared pathological processes in various species. Corneal gene therapy is therefore an ideal target for collaborative biomedical research. The potential therapeutic applications of gene therapy are vast. A major advantage of gene therapy over the use of conventional drugs is the prospect of curing disease rather than providing transient relief by suppression of disease symptoms. Replacing defective genes with functional genes through the use of gene therapy offers the prospect for long-term therapeutic benefits without repeat drug application. Successful vision restoration through the use of gene therapy in patients affected with LCA has demonstrated the ability of gene therapy to potentially cure ocular disease and prevent long-term blindness.12, 13 Further research in gene therapy is needed to develop treatments for other vision-impairing ocular diseases.
3

The cornea is an ideal target for gene therapy due to its accessibility, immune-privileged nature, and ability to perform frequent non-invasive assessment.1, 14 Gene therapy has been studied in animal models of ocular corneal graft rejection, neovascularization, wound healing, fibrosis, macular degeneration, optic neuropathy, and retinal degeneration.1, 15-20 The success of corneal gene therapy depends on the type of vector, the extent of therapeutic gene expression, adverse immune responses, and the stage of intervention.21

3. Adeno-Associated Virus Several viral as well as non-viral vectors have been developed to transport genetic material into cells, each of which has advantages and limitations.1 Our previous studies found Adeno-Associated Virus (AAV) to be a safe and efficient viral vector for transducing corneal cells and delivering therapeutic genes in the cornea of a variety of species including rabbit, mouse, human and horse.1, 14, 22-25 To the best of our knowledge, no study has been conducted thus far to test the potential use of AAV for canine corneal gene therapy. Serotypes 19 of AAV are most commonly used for gene therapy.16 The AAV vector possesses several important properties that make it a highly successful vector for corneal gene therapy, including its ability to transduce dividing and non-dividing cells, low pathogenicity, low immunogenicity, and longterm transgene expression.1 Various studies have shown unique tropism and transduction efficacy of different AAV serotypes for specific tissues.16 The efficacy of AAV2 and AAV5 to deliver genes into the corneal stroma of rabbits and rodent in vivo, as well as the ability of AAV6, AAV8, and AAV9 to deliver genes in mouse cornea in vivo and donor human cornea ex vivo has been established.1, 16 The finding of variable transduction efficiency and tissue tropism
4

has led to comparative studies quantifying transduction efficiency of various AAV serotypes in corneal tissue. In a previous study, AAV5 was shown to have enhanced gene delivery when compared to AAV2 in the rabbit cornea.14

4. GENE TRANSCRIPTION MODULATORS Of the various cytokines that contribute to the development of corneal fibrosis following injury, TGF1 has been found to play a critical role by facilitating the transdifferentiation of corneal keratocytes to fibroblasts and myofibroblasts.2 It has been shown that inhibition of corneal myofibroblast formation reduces corneal fibrosis and increases corneal transparency.26-29 Much research has been focused on pharmacologic agents that modulate TGF function and the subsequent development of corneal fibrosis. Recently, our lab demonstrated that Trichostatin A (TSA) effectively reduces haze and scarring in vivo in the rabbit cornea without causing significant ocular side effects.30, 31 TSA is a reversible inhibitor of HDAC (HDACi). Histone deacetylases (HDAC) increase the binding affinity between histones and DNA and produces condensed chromatin, which decreases the transcriptional activity of genes.32 Therefore treatment of mammalian cells with inhibitors of HDAC activity, such as trichostatin A results in increased expression of a variety of genes.33 HDACs have been shown to be involved in fibrogenesis in a variety of organs. However, TSA is still under clinical trial and is not routinely used in the clinical setting to treat patients. A derivative of TSA, Vorinostat or suberoylanilide hydroxamic acid (SAHA) is a broad spectrum HDACi. It is approved by the US Food and Drug Administration (FDA) for clinical use as a chemotherapeutic agent for the treatment of cutaneous T-cell lymphoma.34
5

CHAPTER 2: CORNEAL GENE THERAPY IN VETERINARY MEDICINE: A REVIEW

1. VECTORS: Conventional vectors: Several viral as well as non-viral vectors have been developed to transport genetic material into cells, each of which has advantages and limitations (table 1).1 A detailed discussion of all currently used viral and non-viral vectors in ocular gene therapy is beyond the scope of this review. Rather, this section of the chapter will serve to review the most common vectors utilized for corneal gene therapy. Several studies have assessed the efficacy of Adenovirus (AV) vectors for corneal gene therapy.1 AV vectors have been shown to effectively deliver genes to the rodent cornea; however, they provide short-term transgene expression and stimulate moderate to severe inflammation.35, 36 Retroviral vectors can successfully deliver genes into the cornea but are not capable of transducing non-dividing cells.37 Like adenovirus, retroviral vectors have also been shown to stimulate mild-to-severe inflammatory responses and have oncogenic potential due to their integration near proto-oncogenes, making them undesirable.1, 38 Lentiviral vectors appear to be safer than retroviral vectors due to their lower risk of tumor oncogenesis.38 The origin of lentiviral vectors are from equine infectious anemia virus and the human immunodeficiency virus (HIV), thus safety remains a major concern. Attempts to improve safety of lentiviral vectors include creating vectors that lack the entire viral genome.39 Disabled lentivirus vectors
6

provide long-term transgene expression and are able to transduce slow/non-dividing cells, making them better options for the transduction of corneal keratocytes and endothelial cells. Adeno-Associated Virus (AAV) possesses several important properties that make it a highly successful vector for corneal gene therapy including its ability to transduce dividing and non-dividing cells, low pathogenicity, low immunogenicity, and long-term transgene expression.1 Our previous studies found AAV to be a safe and efficient viral vector for transducing corneal cells and delivering therapeutic genes into the cornea of rabbits, mice, horses, and people.1, 14, 22-25 Serotypes 19 of AAV are most commonly used for gene therapy.16 Various studies have shown unique tropism and transduction efficacy of different AAV serotypes for specific tissues.16 The finding of variable transduction efficiency and tissue tropism has led to comparative studies quantifying transduction efficiency of various AAV serotypes in corneal tissue. The efficacy of AAV2 and AAV5 to deliver genes into the corneal stroma of rabbits and rodents in vivo, as well as the ability of AAV6, AAV8, and AAV9 to deliver genes in mouse cornea in vivo and donor human cornea ex vivo has been established.1, 16 In a previous study, AAV5 was shown to have enhanced gene delivery when compared to AAV2 in the rabbit cornea.14 Non-viral gene therapy is the introduction of therapeutic genes via plasmid DNA into target cells without the use of a virus. In general these techniques demonstrate low toxicity, immunogenicity, and pathogenicity, rendering these techniques safer than viral methods. However, they demonstrate low transduction efficiencies, which make them less desirable. These techniques include microinjection, electroporation, sonoporation, gene gun, controlled
7

corneal dehydration, laser, and chemical delivery methods. The reader is referred to a thorough review of these techniques in a recent publication by Mohan et al.39 Contemporary vectors: Tyrosine-mutant AAV The AAV viral capsid plays a fundamental role in cellular tropism (i.e. binding and uptake) thus contributing to the efficiency of transgene expression. When the AAV vector is incorporated into the target cell it is vulnerable to normal cellular degradation pathways. By directly mutating the tyrosine residues on the viral capsid, the degradation mechanisms of the target cell are avoided, which leads to increased transduction.40 When injected subretinally or intravitreally these tyrosine-mutant AAV vectors demonstrate several fold higher transgene delivery in target retinal cells than non-mutant AAV vectors.41 The dose of tyrosine-mutant AAV vector required for transduction is also significantly lower than that of AAV which lessens the immune response.41 Nanoparticles Nanoparticles (NPs) offer an alternative to the use of viral vectors in gene therapy. NPs are particles that are 1-100 nm in size. There are several different forms of nanoparticles and they typically contain a segment of DNA or RNA that is compacted with a polycationic polymer.42 Due to their small size, NPs can readily interact with molecules on the cell surface or inside cells. Unlike their viral vector counterparts, NPs do not introduce exogenous genes into target cells. They tend to be less immunogenic and cytotoxic than viral vectors.43 Finally, NPs are able to incorporate numerous ligands such as DNA, antibodies, peptides, and probes and
8

therefore present an array of therapeutic modalities. Pathways of cellular internalization of NPs include phagocytosis, macropinocytosis, clathrin- or caveolae-mediated endocytosis, and other clathrin- and caveolae-independent endocytic pathways.42 There has been successful vector gene transfer with the use of NPs in phase I/II clinical trials designed to treat cystic fibrosis.44 Limited studies have investigated the use of NPs for ocular gene delivery.42, 45 Mohan et al. recently evaluated the safety and toxicity of gene transfer technique using a 2-kDa polyethylenimine conjugated to gold nanoparticles (PEI2-GNPs) in the human cornea in vitro and in rabbit cornea in vivo.45 PEI2-GNPs exhibited significant transgene delivery in human corneal fibroblasts and did not appear to alter cellular viability or phenotype. Rabbit in vivo studies revealed substantial gold uptake in the rabbit cornea with a slow clearance of GNPs over time. Our results to date suggest that PEI2-GNPs appear safe for use in the cornea and may have potential use for corneal gene therapy; however, further research examining the clearance of GNPs and their effects on the optical properties of the eye is needed.45

2. POTENTIAL APPLICATIONS IN VETERINARY MEDICINE Corneal disease is a commonly encountered clinical problem in veterinary medicine. The cornea provides two thirds of the refractive power of the mammalian eye. Therefore corneal disease that results in disruption of corneal health (e.g. clarity or refractive properties) can lead to significant visual impairment. Unique properties of the cornea that enable this living tissue to be essentially optically clear render it susceptible to limited pathologic responses to injury and often opacification results. Some of the common corneal diseases encountered in veterinary medicine include keratitis (i.e. ulcerative, immune-mediated, keratoconjunctivitis sicca (KCS)), neoplasia,
9

corneal dystrophy, dermoids, chemical burns, lacerations, and corneal degeneration, among others. We now present an overview of the potential applications of gene therapy for treatment of several common corneal diseases in veterinary medicine. Corneal Transplantation In physician ophthalmology, by virtue of well-maintained and highly regulated eyebanks, viable full-thickness corneal transplants are commonly used. In veterinary medicine the use of fresh corneal transplants is limited for a number of reasons. First, harvesting and preservation requirements of donor corneal tissue is not feasible in the majority of veterinary practices.46 Second, tissue typing and donor-recipient rejection concerns are not well understood. Third, medical-legal implication/permission of harvesting fresh, healthy donor cornea from a euthanized dog has not been established. Presently, veterinary ophthalmologists use frozen corneal transplants to provide tectonic support rather than optical clarity, as corneal endothelial cells do not survive the freezing process.46-48 Common ocular conditions that may be treated with a previously frozen corneal transplant include descemetoceles, stromal abscesses, inflammatory keratitis with or without malacia, corneal perforation/iris prolapse, and feline corneal sequestra. 46, 48, 49 Corneal graft rejection involves the recognition of foreign antigens from the donor graft, which leads to an immune response by the host. Graft rejection is manifested by a loss of graft transparency after an initial time period of graft clarity (approximately 2 weeks in people and several days in horses).47, 50 Corneal vascularization, inflammation, and/or infection in the host species prior to transplantation significantly increases the chance of corneal graft rejection or
10

failure.50 Human patients with these pre-existing corneal conditions are considered high-risk patients. Most veterinary patients that receive corneal transplants would therefore be considered high risk as almost all conditions in which corneal transplantation is indicated in veterinary medicine are characterized by these pre-existing conditions. The most important risk factor for graft rejection is host cornea vascularization.50 In stark contrast to physician ophthalmology, corneal vascularization is usually considered a desired healing response since vascularization substantially improves the patients ability to resist corneal infection. Thus, while the vast majority of the corneal transplants performed in veterinary ophthalmology undergo rejection by physician standards, vascularization of the donor cornea is considered an optimal progression in wound healing in the majority of cases.46, 47 Prevention of corneal graft rejection currently is achieved by suppressing the host immune response. In humans, this is done with liberal use of systemic and topical corticosteroids as well as cyclosporine. In veterinary medicine the use of corticosteroids is contraindicated in the face of corneal infection, which is present in many of the patients that undergo corneal transplantation.51 Therefore the development of an alternative therapy to corticosteroids, such as gene therapy, to prevent corneal graft rejection and the resultant longterm corneal fibrosis would be advantageous in veterinary medicine. The unique setting afforded by having an established tissue bank for corneal transplants is ideal for gene therapy, as donor cornea can easily be manipulated, ex vivo, prior to transplantation. Several gene therapy methods have been examined in an effort to improve graft survival by the delivery of therapeutic genes that modulate cellular apoptosis, angiogenesis, and
11

wound healing.52 One study investigated the delivery of an intracellular enzyme, indoleamine 2,3- dioxygenase (IDO), using a lentivirus vector in murine corneal endothelium.53 Once activated by cytokines, IDO leads to the immunoregulatory catabolism of tryptophan. The reduction of tryptophan is thought to halt activated T cells in the cell cycle, which encourages immune tolerance of the transplanted cornea. This same research group also evaluated the effect of IDO transduction on donor corneas ex vivo prior to transplantation and found significant prolongation of corneal allograft survival compared to controls.54 Another group found that the transfer of an anti-apoptotic gene, bcl-xL, using a lentivirus vector was effective at inhibiting apoptosis in the corneal endothelium in vitro in a rodent model, as well as significantly enhancing corneal graft survival in vivo.55 Another application of gene therapy use for corneal transplantation involves the preservation of corneal endothelial cell density in corneal tissue that is stored at eye banks. One study demonstrated that the overexpression of transcription factor E2F2, using recombinant adenovirus ex vivo produced an increased endothelial cell count in rabbit and human corneas.56, 57 The E2F2 over-expression induced cell-cycle progression of corneal endothelial cells, which are normally amitotic, in the first week after transduction and without inducing significant apoptosis. This therapy has great potential for use in veterinary ophthalmology, since it is more practical to use corneas that have been preserved long-term. This type of gene therapy may prove especially beneficial in the treatment of corneal endotheliopathies. Corneal Fibrosis/ Modulation of Corneal Wound Healing

12

Corneal transparency and clarity rely on the special organization of the cornea's constituent components including corneal collagens and the extracellular matrix.1 Corneal wound healing following injury and/or infection often results in fibrosis/scarring2, which may cause significant visual impairment.58 Much of the focus on the human side of corneal gene therapy for treatment of corneal fibrosis is geared towards reduction of photorefractive keratectomy-induced laser injury to the cornea. In veterinary ophthalmology corneal fibrosis (scar) is most commonly an outcome following complicated corneal ulceration and wound healing. As previously discussed, because vascularization is a desired component of the healing response in many veterinary cases that would benefit from corneal transplantation, the role of gene therapy might better serve to address the fibrosis that results from graft rejection. Ulcerative keratitis is a significant clinical problem in veterinary ophthalmology.58 Treatment of deep corneal ulcers typically demands both medical and surgical therapeutic strategies.5 Surgical procedures utilized for the repair of deep corneal ulceration include conjunctival grafts, corneal grafts, corneo-conjunctival transposition, natural material transplants, and application of tissue adhesives.6-9 These procedures and their associated inflammatory response typically result in moderate to severe corneal scarring. Currently, pharmacologic agents that are specifically aimed at reducing corneal fibrosis are not routinely used in veterinary ophthalmology. The transformation of quiescent keratocytes to corneal fibroblasts and myofibroblasts has been identified as the primary event in corneal wound healing that is responsible for fibrosis. While the production of myofibroblasts is a normal part of the complex process of corneal

13

wound healing, the overproduction of myofibroblasts can lead to vision loss due to corneal scarring.4 Our lab investigated the safety and efficacy of AAV5 for delivering gene therapy in canine corneal fibroblasts (CCFs) as a model of the pre-fibrotic state as well as in canine corneal myofibroblasts (CCMs), as a model of corneal fibrosis. Our results demonstrated that AAV5 is an effective vector for gene therapy in canine corneal fibroblasts and myofibroblasts in vitro. The transduction rate in CCFs was 42.8% CCFs and 28.0% in CCMs. The tested AAV5 vector showed significantly less transduction of CCMs compared to CCFs. This difference in transduction efficiency of AAV5 between the two cell types may suggest that higher doses of vector are required for delivering therapeutic genes into fibrotic tissue and it may signify the importance of therapeutic strategies aimed at initial prevention of corneal fibrosis versus treatment of established fibrosis. However, other potential causes of this noted transduction efficiency difference may be due to the specific cell tropism of the viral vector. Future studies are underway to investigate different AAV serotypes in order to determine if a particular serotype demonstrates specific tropism for canine corneal myofibroblasts. Of the various cytokines that contribute to the development of corneal fibrosis following injury, TGF1 has been found to play a critical role by facilitating the transdifferentiation of corneal keratocytes to fibroblasts and myofibroblasts.2 It has been shown that inhibition of corneal myofibroblast formation reduces corneal fibrosis and increases corneal transparency.26-29 Therefore, it is thought that inhibiting TGF and/or its signal transduction may be a useful strategy to regulate uncontrolled corneal healing and to prevent or treat corneal scarring. Much

14

research has been focused on gene therapy approaches to modulate TGF function and the subsequent development of corneal fibrosis. Decorin, a small proteoglycan, which is a natural inhibitor of TGF has been successfully delivered to the cornea by the use of gene therapy. Mohan et al. recently showed that decorin gene transfection inhibits TGF-driven elevated expression of profibrogenic genes and transformation of corneal fibroblasts to myofibroblasts in human corneal fibroblast cultures.59 Our laboratory also demonstrated a significant inhibition of corneal scarring in vivo in the rabbit with the use of AAV5-mediated decorin gene therapy and did not detect any adverse effects at a time point of one-month following transduction.60 This was the first study to establish the therapeutic potential of decorin gene therapy for treating corneal diseases. Currently, long-term studies are underway to assess whether the over-expression of decorin will cause detectable ocular complications or changes in corneal function or transparency.39 The ability of gene therapy to prevent corneal scarring in experimental animals has been demonstrated in several previous studies as well. Seitz et al. evaluated the use of retroviral vector-mediated genetic transduction of rabbit keratocytes in vivo modulating keratocyte proliferation and decreasing corneal scarring.61 They employed the use of a suicide gene and pro-drug combination of the herpes simplex virus thymidine kinase (HSVtk) gene with ganciclovir. HSVtk triggers cell death in proliferating keratocytes by converting ganciclovir to a toxic metabolite. This combination gene therapy reduced corneal scarring 10 and 21 days post administration. Galiacy et al. utilized an AAV vector expressing a fibril collagenase, matrix metalloproteinase 14 (MMP14), in a murine model of corneal scarring.62 They observed that MMP14 gene delivery reduced corneal opacity and expression of several major genes involved
15

in corneal scarring. These findings are encouraging with regards to new developments for the prevention and treatment of corneal fibrosis with possible direct application to the veterinary ophthalmic patient. Corneal Dystrophies Corneal stromal dystrophy refers to a group of inherited corneal diseases that are typically bilateral, symmetric and slowly progressive, with absent inflammation.63 They are caused by an accumulation of material such as lipid or cholesterol crystals in the cornea. There are a several forms of corneal dystrophy, which vary in their expression, location within the cornea, and degree of vision loss. The dog is the most common nonhuman species that is affected by corneal dystrophy and the condition is rare in other species. This condition manifests as bilateral crystalline or white opacities in the central or paracentral cornea.64 Specific breeds that are described in the literature to be affected by stromal dystrophies include the Shetland Sheepdog, Beagle, Siberian Husky, Cavalier King Charles Spaniel, Airedale Terrier and Rough Collie.65 In people the resulting corneal opacity causes visual impairment. In dogs, stromal dystrophies infrequently lead to significantly appreciable functional vision loss. Nevertheless, complete vision loss secondary to corneal dystrophy has been reported in Airedale Terriers and Siberian Huskies.63 Boston Terriers, Chihuahuas and Dachshunds are affected by corneal endothelial dystrophy, and the American Cocker Spaniel has been reported to develop a posterior polymorphous dystrophy.65 Endothelial dystrophies (except the posterior polymorphous dystrophy in the American Cocker Spaniel) are characterized by progressive corneal edema, which can lead to vision loss, bullous keratopathy, and corneal ulcers.63

16

Several genetic mutations in people with corneal dystrophies have been identified.66, 67 The pathogenesis and localization of many additional genes likely responsible for these corneal diseases have yet to be identified. A paucity of research in the realm of gene therapy for the treatment of corneal dystrophies exists for several reasons. There is a lack of animal models for most of the corneal dystrophies described in people. The few mouse lines that have been created to express corneal dystrophy genetic mutations do not exhibit similar clinical manifestations of these diseases as observed in people.68 Additionally, the common use of corneal transplantation as a relatively effective treatment in people may imply that corneal dystrophies are less visually debilitating than other ocular diseases for which no treatment exists and as such, limited research funding may be available to study corneal dystrophies at this time. Specific gene localization and targeting could be vital for the development of animal models of this category of corneal disease and for gene therapy treatment advances. As is the case in physician ophthalmology, in veterinary ophthalmology there is a lack of research focused on treatment of corneal dystrophies. Corneal dystrophies generally do not respond to medical treatment; therefore, unless they cause an obvious loss of vision or have complications (e.g. corneal ulceration) they are often treated with benign neglect. Treatment of progressive corneal edema with endothelial dystrophy is palliative. When dogs develop recurrent ulcers as a complication of dystrophies or persistent bullous keratopathy secondary to endothelial dystrophy they may benefit from thermokeratopasty or keratoleptynsis.65, 69 Definitive treatment includes a living (i.e. one with functional donor corneal endothelial cells) corneal transplant as previously described but is associated with relatively high failure rates in dogs. Presently, there are no treatment strategies utilizing gene therapy for corneal dystrophies
17

in veterinary ophthalmology. The corneal tissue-selective gene therapy approaches that are currently under investigation may have potential for the treatment of corneal dystrophies in dogs. Other Mucopolysaccharidosis (MPS) is a group of lysosomal storage diseases characterized by the defective metabolism of glycosaminoglycans (GAG). Five types of MPS have been described in dogs (MPS I, II, III, VI, VII). While MPS is relatively rare in veterinary medicine, the use of dogs in research is not uncommon because they serve as a large animal model for human disease. Corneal pathology typically involves opacification of the cornea due to lysosomal storage in keratocytes.52 Intravenous gene therapy in neonatal MPS dogs using a retroviral vector has been shown to prevent GAG accrual and resultant corneal clouding.70, 71 Corneal clouding in untreated MPS I dogs resulted in a blurry view of the fundus as well as a granular appearance of the cornea due to the lysosomal storage. A marked reduction in corneal clouding was shown with systemic gene therapy using a retroviral vector in neonatal MPS I dogs.70

3. CONCLUSIONS: Animal models represent the corner stone of gene-therapy research. Many of the advancements made to date have been possible through the appropriate use of rodents, rabbits, dogs, cats, pigs, and non-human primates. Given the obvious comparative ophthalmic implications to many of the recent scientific discoveries in vision science, the potential for translational research to directly impact our veterinary ophthalmic patients is exciting. Our
18

laboratorys recent research in canine and equine corneas in vitro as well as significant progress and advancement in the field of corneal gene therapy in the last decade holds promise for establishing important models and for the improvement of long-term vision in our veterinary patients and thus, improvement of human and animal welfare.

19

CHAPTER 3: CANINE CORNEAL FIBROBLAST AND MYOFIBROBLAST TRANSDUCTION WITH AAV5

1. EXPERIMENTAL PURPOSE The purposes of this study were to determine the efficacy of AAV5 for delivering therapeutic genes in both normal and fibrotic canine keratocytes and to study the safety of AAV5 for the canine corneal gene therapy using an in vitro model of canine corneal fibroblasts (CCFs) and myofibroblasts (CCMs).

2. MATERIALS AND METHODS Canine corneal fibroblast and myofibroblast cultures All experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Missouri. Ten full-thickness 6 mm axial corneal buttons were aseptically harvested immediately post-mortem from five healthy research dogs undergoing humane euthanasia for reasons unrelated to this study. Slit-lamp biomicroscopy was performed by an experienced veterinary ophthalmologist (EAG) prior to euthanasia to ensure that all samples were harvested from dogs free of anterior segment disease. The corneal biopsy samples were washed with minimal essential medium (MEM, Gibco; Grand Island, NY, USA) and the epithelium and endothelium were removed by careful manual scraping using a # 10 Bard Parker scalpel blade (BD, Franklin lakes, NJ, USA). The corneal stroma was sectioned into four smaller explants and placed in 100 X 20 mm tissue culture plates (BD BioSciences, Durham,
20

NC, USA) containing MEM supplemented with 10% fetal bovine serum (Gibco, NY, USA), then incubated in a humidified CO2 incubator (Thermo Scientific, Asheville, NC, USA) at 37oC in order to obtain CCF cultures. In approximately 3-10 days primary fibroblasts began migrating from the stromal sub-sections. On attaining ~90% confluence, the explants were manually removed using rat-toothed forceps, and discarded. The confluent cells were trypsinized and the cellular concentrations were calculated using Countess automated cell counter (Invitrogen) and replated in 100 mm tissue culture dishes. Passages 1-3 of CCFs were used for this experiment and all experiments were repeated in triplicate. Serum-free MEM medium containing 1ng/mL TGF1 (PeproTech Inc, Rocky Hill, NJ, USA) was used to generate CCM cultures from fibroblasts.

AAV vector production Production of AAV5 expressing Enhanced Green Fluorescent Protein (GFP) under control of hybrid cytomegalovirus (CMV) + chicken (-actin (CBA) promoters (AAV5-gfp) was performed by the gene therapy core, University of Florida, Gainesville, Florida and was procured from Drs. Gregory S. Schultz and Vince A. Chodo. The AAV2 plasmid pTRUF11 expressing green fluorescent protein gene under control of a hybrid promoter (cytomegalovirus enhancer and chicken -actin) and simian virus 40 polyadenylation site was packaged into AAV5. In brief, AAV titer expressing GFP was produced by the 2-plasmid, co-transfection method 29. One Cell Stack (Corning Inc., Corning, NY, USA) with approximately 1 X 109 HEK 293 cells was cultured in Dulbeccos Modified Eagles Medium (Hyclone Laboratories, Inc. Logan UT, USA), supplemented with 5% fetal bovine serum and antibiotics. A calcium phosphate
21

transfection precipitation was set up by mixing a 1:1 molar ratio of AAV2 plasmid DNA and AAV5 repcap helper plasmid DNA. This precipitate was applied to the cell monolayer and the transfection was incubated at 37C, at 7% CO2 for 60 h. The cells were then harvested and lysed by freeze/thaw cycles and subjected to discontinuous iodixanol gradient-ultra centrifugation at 350,000g for 1 hour (h). This iodixanol fraction was further purified and concentrated by column chromatography on a 5-ml HiTrap Q Sepharose column using a Pharmacia AKTA FPLC system (Amersham Biosciences, Piscataway, NJ, USA). The vector was eluted from the column using 215 mM NaCl buffer, pH 8.0, and the rAAV peak collected. The AAV5-GFP vectorcontaining fraction was then concentrated and buffer exchanged in Alcon BSS with 0.014% Tween 20, using a Biomax 100K concentrator (Millipore, Billerica, MA, USA). The vector was titered for DNAse-resistant vector genomes by Real-Time PCR relative to a standard.

AAV5 transduction Seventy-percent confluent CCF or CCM cultures in 12-well plates were used for all experiments. Cultures were serum deprived for 2 h prior to viral transduction (viral titer of 6.5x1012 genomic copies/mL). AAV5 expressing GFP marker gene (2 L for each well of a 12 well plates) was applied in serum-free MEM medium. After 6 h, the medium containing viral vector was replaced with fresh medium supplemented with 10% serum and the cells were incubated for 48 h prior to microscopic evaluation of fluorescence.

Cytotoxicity Studies

22

The toxicity and safety of AAV5 for canine corneal gene therapy was evaluated by comparing phenotype, cell death, and cellular viability in AAV5-treated and untreated CCFs and CCMs. Phase-contrast light microscopy images were obtained 24 h and 48 h after application of AAV5 to observe any phenotypic changes in CCF and CCM cultures. Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay was used to detect cellular apoptosis due to AAV vector transduction in CCFs and CCMs. Cells were fixed with 4% paraformaldehyde at room temperature for 15-20 min and then washed with phosphate buffered saline. A fluorescence-based TUNEL assay was used according to the manufacturers instructions using ApopTag apoptosis detection kit (Chemicon international, Temecula, CA, USA). Cellular viability was assessed using trypan blue (Sigma-Aldrich, St. Louis, MO, USA) exclusion 24 h and 48 h after transduction. Cell count was performed using an automated cell counter.

Immunocytochemistry Immunocytochemical analysis was performed in order to determine transduction efficiency. Direct fluorescence microscopy at 100X magnification was used to count GFP transduced CCFs and DAPI-stained nuclei in ten randomly selected, non-overlapping areas for each treatment. Analysis was performed on a PC computer using the public domain NIH Image JAVA program (developed at the U.S. National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/nih-image). Percent transduction efficiency was calculated by dividing the total number GFP-positive cells by the total number of 4',6-diamidino-2-phenylindole (DAPI) stained nuclei.
23

Immunofluorescent staining for SMA, a marker for myofibroblasts, was performed using mouse monoclonal antibody for and SMA. At study end point, cultures were washed twice with PBS and incubated at room temperature with mouse monoclonal antibody for SMA (DAKO, Carpinteria, CA, USA) at a 1:200 dilution in 1X PBS for 90 min. Cultures were then washed twice with PBS and the incubated at room temperature with secondary antibody Alexa 488 goat anti-mouse IgG (Invitrogen, Carlsbad, CA, USA) at a dilution of 1:500 for 1 h. Immunofluorescent staining for F-actin, which stains cytoskeletal cellular structures, was performed using phalloidin (Invitrogen, Calsbad, CA, USA). At study end point, cultures were washed twice with PBS and incubated at room temperature with phalloidin (Invitrogen, Calsbad, CA, USA) at a 1:200 dilution in 1X PBS for 90 min. Cells were mounted with Vectashield containing 4-6-Diamidino-2-phenylindole (DAPI) (Vector laboratories, Inc., Burlingame, CA, USA) to allow visualization of nuclei. Irrelevant isotype-matched primary antibody, secondary antibody alone, and tissue sections from nave eyes were used as negative controls for each immunocytochemistry experiment. Immunostained cultures were examined and photographed with Leica fluorescent microscope (Leica, Wetzlar, Germany) equipped with a CCD digital camera (SpotCam RT KE, Diagnostic Instruments Inc., Sterling Heights, MI, USA). The immunocytochemistry data quantification was performed by counting GFP, TUNEL, and DAPI-positive colored cells in 10 randomly selected non-overlapping areas of fluorescent cultures at 100X magnification using a Leica Fluorescent microscope (Leica).

RNA extraction, cDNA synthesis, and quantitative real-time PCR

24

The total RNA from CCF and CCM cultures transduced with AAV5-GFP was extracted using RNeasy kit (Qiagen Inc., Valencia, CA, USA) and reverse-transcribed to cDNA following manufacturers instructions (Promega, Madison, WI, USA). Real-time PCR was performed using iQ5 real-time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA). A 20 L reaction mixture containing 2 L cDNA, 2 L forward (200nM), 2 L reverse primer (200nM), and 10 L 2X iQ SYBR green super mix (Bio-Rad Laboratories) was run at universal cycle (95oC 3 min, 40 cycles of 95oC 30 sec followed by 60oC 60 sec) following manufacturers instructions. For GFP, forward primer sequence GCGTGCAGTGCTTTTCCAGA and reverse primer sequence CTTGACTTCAGCGCGGGTCT were used. The house-keeping gene beta actin forward primer sequence was CGGCTACAGCTTCACCACCA and reverse primer sequence was CGGGCAGCTCGTAGCTCTTC. The threshold cycle (CT) was used to detect the increase in the signal associated with an exponential growth of PCR product during the log-linear phase. The relative gene expression was calculated using the following formula: 2-CT. The CT is a PCR cycle number at which the fluorescence meets the threshold in the amplification plot and CT is a subtraction product of target and housekeeping genes CT values. The 2-CT is a method for determining relative target mRNA quantity in samples. The amplification efficiency for all used templates was validated by ensuring that the difference between linear slopes for all templates <0.1. Three independent reactions were performed and the average ( standard error of the mean or SEM) results were calculated.

Quantification and statistical analysis

25

Results are expressed as mean standard error of the mean (SEM). Statistical analysis was performed using analysis of variance (ANOVA) followed by Bonferroni test for comparing data between control and AAV vector. Data was tested for normality using Kolmogorov-Smirnov Test and probability plots. Microsoft excel and Graph Pad Prism statistical software was used.

3. RESULTS AAV Transduction Efficiency Figure 1 shows establishment of canine corneal fibroblast (representing early healing state) and myofibroblast (representing late healing state) monolayer cultures using phalloidin (Factin) immunostaining. The cells grown in 10% serum supplemented medium reveal classic spindle shaped fibroblast morphology (Figure 1a); cells grown in serum-free medium containing TGF1 (1ng/ml) show large and wide morphology typical of the myofibroblast phenotype (Figure 1b). The growth kinetics of these cells was similar to previously reported findings of equine corneal cells except canine corneal cells grew at a rate of about 2.5-3 fold faster.72 Successful transduction of GFP with AAV5 vector into CCFs and CCMs was determined via fluorescent microscopy (Figures 2 and 3, respectively). The dual immunostaining experiments were performed to confirm transgene delivery into the two cell types. Myofibroblast marker, SMA, is shown in red. Detection of ~ 3-5% myofibroblast cells in corneal fibroblast cultures (Fig 2a) is normal and unavoidable.72 CCFs treated with AAV5-GFP (Figure 2b) demonstrated significant transduction as evident from GFP-positive cells (green fluorescence showing spindle shaped morphology) when compared to the untreated control cells
26

(Figure 4a) indicating successful and efficient GFP gene transduction. Likewise, myofibroblasts treated with AAV5-GFP (Figure 3b) demonstrated a significantly high of number of GFPpositive cells compared to the untreated control cells (Figure 3a). Quantification of GFP-positive CCFs and CCMs was completed at 100X magnification following previously described methods (Figure 4). The AAV5-GFP vector transduced 42.8% CCFs and 28.0% CCMs. The tested AAV5 vector showed significantly less (15%; p 0.01) CCM transduction compared to CCFs.

Quantification of real-time PCR The relative delivery of the GFP gene in AAV5-treated CCFs and CCMs and their respective controls was quantified using real-time PCR (Figure 5). In both cell populations there is significant transduction of AAV5-GFP compared to the untreated controls. Percentage of GFP-mRNA was significantly lower in CCM compared to CCF cultures.

AAV Safety No observable changes in phenotype were detected in phase-contrast light microscopy images between untreated control cells and AAV5-treated cells at 24 h and 48 h after the application of AAV5. Figure 6 shows CCFs and CCMs 48 h after the application of AAV5. These images demonstrate the classic morphology of CCFs and CCMs as well as the successful progression to confluence. Similar results were seen at 24 h (data not shown). TUNEL assay was used to detect CCF and CCM apoptosis, and to a lesser extent necrosis. No detectable difference in cellular apoptosis was observed between AAV5-GFP
27

treated CCFs and untreated controls (Figure 7a and 7b). Similar results were seen for CCMs (results not shown). Trypan blue exclusion assay demonstrated that cellular viability of the AAV5-GFP treated CCFs was greater than 95% at both 24 and 48 h (Figure 7c). Cellular viability of the AAV5-GFP treated CCMs was also greater than 95% at both 24 and 48 h (results not shown).

4. DISCUSSION The results of this study demonstrate that AAV5 is a safe and effective vector for gene therapy in CCFs and CCMs in vitro. To our knowledge this is the first study to test the efficacy of an AAV5 vector for canine corneal gene therapy. The immunogenicity of AAV5 in the canine cornea is currently not known. This in vitro study demonstrated no significant cell death or phenotypic change associated with AAV5 gene therapy on CCFs and CCMs. Our previous studies have demonstrated AAV to be nonpathogenic and safe for corneal gene therapy in vitro in equine corneal fibroblasts, as well as in vivo in both the rabbit and mouse without causing significant immune reactions.1, 14, 21, 25 However, future canine studies are warranted in order to establish the safety and immunogenicity of AAV5 in vivo. Investigation of the effects of varying viral concentrations, administration time, and gene delivery systems are warranted. Use of an in vitro model allows for experimentation within a controlled environment without risk to a clinical patient until safety parameters have been established. The in vitro transduction efficiency of AAV5 determined for various cell types including lung epithelium, skeletal muscle, gastrointestinal cells, and corneal fibroblasts is similar to reported in vivo transduction efficiencies.73-75 Recently our lab demonstrated a higher level of gene transduction
28

in vivo than previously published transgene expression in vitro with the topical application of serotypes AAV9 and AAV8.16, 21 Therefore, the transduction efficiency percentage obtained from this in vitro study concerning the efficacy of AAV5 as a vector for canine corneal gene therapy may, in fact, underestimate the potential transgene expression that we can expect in vivo. The discrepancies reported in the literature among transgene expression in vitro and in vivo highlights the importance of validating in vitro findings with in vivo studies.16 The main cell type that plays a crucial role in fibrosis and scarring is the myofibroblast which is derived from activated fibroblasts. The present study aimed to investigate the potential use of gene therapy in CCFs as a model of the pre-fibrotic state as well as in CCMs, as a model of corneal fibrosis. We found a significant decrease in transduction efficiency in myofibroblasts compared to corneal fibroblasts. This difference in transduction efficiency of AAV5 between the two cell types may suggest that higher doses of vector are required for delivering therapeutic genes into fibrotic tissue. However, it may also be due to specific cell tropism of the viral vector. Different AAV serotypes have varying shell proteins and therefore differ in their capability of binding to and thus transfecting different host cells.21 Future studies are underway to investigate different AAV serotypes in order to determine if a particular serotype demonstrates specific tropism for canine myofibroblasts. This noted difference in transduction efficiency may also signify the importance of therapeutic strategies aimed at initial prevention of corneal fibrosis versus treatment of established fibrosis. Furthering our understanding of the cascade of events that occurs in canine corneal wound healing is critical to developing medical therapies that aim to prevent or reverse corneal fibrosis. Corneal gene therapy makes use of a vector system to transfer a gene into the cornea so
29

that a transgenic protein will be expressed to modulate congenital or acquired disease. The protein may be structural (e.g. collagen), or functionally active (e.g. an enzyme, cytokine or growth factor).52 Interfering with the activation of signaling transduction molecules using gene transfer could be an important strategy to prevent or treat the undesirable outcomes of tissue injury resulting in vision loss.15 Fibrotic diseases are characterized by the presence of myofibroblasts, excessive accumulation of disorganized extracellular matrix, and subsequent tissue contraction, all of which are modulated by a multitude of cytokines. Of those cytokines TGF has been found to play a critical role in the development of corneal fibrosis following injury by facilitating the trans- differentiation of keratocytes to myofibroblasts.15 Decorin, a small proteoglycan, which is a natural inhibitor of TGF has been successfully delivered to the cornea by the use of gene therapy. Mohan et al. recently showed that decorin gene transfection inhibits TGF-driven elevated expression of profibrogenic genes and transformation of corneal fibroblasts to myofibroblasts in human corneal fibroblast cultures.59 The ability of gene therapy to prevent corneal scarring in experimental animals has been demonstrated in several other recent studies as well. Seitz et al. evaluated the use of retroviral vector-mediated genetic transduction of keratocytes in vivo modulating keratocyte proliferation and decreasing corneal scarring.61 They employed the use of a suicide gene and pro-drug combination of the herpes simplex virus thymidine kinase (HSVtk) gene with ganciclovir. HSVtk triggers cell death in proliferating keratocytes by converting ganciclovir to a toxic metabolite. This combination gene therapy reduced corneal scarring 10 and 21 days post administration. Galiacy et al. utilized an AAV vector expressing a fibril collagenase, matrix metalloproteinase 14 (MMP14), in a murine model of corneal scarring.76 They observed that MMP14 gene delivery reduced corneal opacity
30

and expression of several major genes involved in corneal scarring. These findings are encouraging with regards to new developments for the prevention and treatment of corneal fibrosis. To our knowledge, this study is the first to investigate the effect of gene transfer in the canine cornea. Our results demonstrate that AAV5 is a safe and effective vector for canine corneal gene therapy in this in vitro model. In vivo studies are warranted.

31

CHAPTER 4: EFFICACY AND SAFETY OF SUBEROYLANILIDE HYDROXAMIC ACID (VORINOSTAT) IN THE TREATMENT OF CANINE CORNEAL FIBROSIS

1. EXPERIMENTAL PURPOSE The purpose of this project is to examine the safety and efficacy of suberoylanilide hydroxamic acid (SAHA) to treat canine corneal fibrosis using an in-vitro model.

2. MATERIALS AND METHODS Canine corneal fibroblast and myofibroblast cultures Corneal buttons were aseptically harvested from purpose-bred healthy research dogs receiving no topical or systemic medications, euthanized for reasons unrelated to the present study. Slit-lamp biomicroscopy was performed by an experienced veterinary ophthalmologist (EAG) prior to euthanasia to ensure that all samples were harvested from dogs free of anterior segment disease. Corneal biopsy samples were washed with minimal essential medium (MEM, Gibco; Grand Island, NY, USA) and the epithelium and endothelium were removed by careful manual scraping using a # 10 Bard Parker scalpel blade (BD, Franklin lakes, NJ, USA). Corneal stroma was sectioned into four smaller explants and placed in 100 X 20 mm tissue culture plates (BD BioSciences, Durham, NC, USA) containing MEM supplemented with 10% fetal bovine serum (Gibco, NY, USA) then incubated in a humidified CO2 incubator (Thermo Scientific, Asheville, NC, USA) at 37oC to produce canine corneal fibroblast (CCF) cultures. In
32

approximately 3-10 days, primary fibroblasts began migrating from the stromal sub-sections. On attaining ~90% confluence the explants were manually removed using rat-toothed forceps and discarded. Confluent cells were trypsinized and cellular concentrations were calculated using Countess automated cell counter (Invitrogen, Carlsbad, CA, USA) and replated in 100 mm tissue culture dishes. Passages 1-3 of canine corneal fibroblasts (CCFs) were used for this experiment. Seventy-percent confluent CCF cultures in 12-well plates were used for all experiments. To find the optimal dose of SAHA, dose-dependent studies were first performed. Briefly, SAHA at concentrations of 0.01%, 0.02%, 0.06%, and 0.12% (2 L for 12 well plates) was applied in serum-free MEM medium for 5 min. After 5 min, the cells were washed and re-plated with fresh medium, and cellular viability was assessed (see cytotixicity studies). Following preliminary dose finding studies, a concentration of 0.06% (1 L of 25nM) SAHA was used for all subsequent experiments. At a time point of 4 h after treatment with TGF-1 CCFs were treated with 0.06% SAHA in serum-free MEM medium with TGF-1 for 5 minutes (group 1) and for 24 h (group 2). TGF-1 was used to transform CCFs into myofibroblasts as an in-vitro model of canine corneal fibrosis. For myofibroblast generation, CCF cultures were exposed to 1 ng/ml TGF-1 (PeproTech Inc, Rocky Hill, NJ, USA) under serum free conditions. Qualitatively the formation of myofibroblasts was confirmed using immunohistochemical staining for -smooth muscle actin (SMA), which is specific to myofibroblasts. Quantitative assessment of alterations in myofibroblast formation due to drug treatment was evaluated using real-time PCR and immunoblotting.

33

Cytotoxicity Studies Cellular viability was assessed using trypan blue assay (Sigma-Aldrich, St. Louis, MO, USA). Cells were trypsinized 48 hours after initiation of treatment and mixed with 0.4% trypanblue solution (Invitrogen, Carlsbad, CA, USA). Cell counts were performed using an automated cell counter (Invitrogen, Carlsbad, CA, USA) and cellular viability was calculated as a percent. Light microscopy images were obtained to observe any phenotypic changes in cell cultures after the application of SAHA. CCF cultures treated with SAHA were examined with a Leica fluorescent microscope (Leica, Wetzlar, Germany) and photographed at 8 h, 24 h, and 48 h after treatment. Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay was used to detect cellular apoptosis due to SAHA treatment. Cells were fixed with 4% paraformaldehyde at room temperature for 15-20 min and then washed with phosphate buffered saline. A fluorescence-based TUNEL assay was used according to the manufacturer's instructions using ApopTag apoptosis detection kit (Chemicon international, Temecula, CA, USA). Photographs were obtained using a fluorescent microscope (Leica, Wetzlar, Germany)

equipped with a digital camera (SpotCam RT KE, Diagnostic Instruments Inc., Sterling Heights, MI, USA).

Immunocytochemistry Immunocytochemical analysis was performed to confirm cell type. Immunofluorescent staining for F-actin, which stains cytoskeletal cellular structures, and for SMA, a marker for myofibroblasts, was performed using mouse monoclonal antibody for phalloidin and SMA,
34

respectively. At study end point cultures were washed twice with PBS and incubated at room temperature with mouse monoclonal antibody for phalloidin (Invitrogen, Calsbad, CA, USA) or SMA (DAKO, Carpinteria, CA, USA) at a 1:500 dilution in 1X PBS for 90 minutes. After incubation with the primary antibody, cells were washed twice with PBS, then incubated with secondary antibody Alexa 488 or 594 goat anti-mouse IgG (Invitrogen, Carlsbad, CA, USA) at a dilution of 1:500 for 1 h. Cells were mounted with Vectashield containing 4'-6'-Diamidino-2phenylindole (DAPI) (Vector laboratories, Inc., Burlingame, CA, USA) to allow visualization of nuclei. Irrelevant isotype-matched primary antibody, secondary antibody alone, and tissue sections from nave eyes were used as negative controls for each immunocytochemistry experiment. Immunostained cultures were examined and photographed with Leica fluorescent microscope (Leica, Wetzlar, Germany) equipped with a CCD digital camera (SpotCam RT KE, Diagnostic Instruments Inc., Sterling Heights, MI, USA).

Immunoblotting Cells were washed with ice-cold PBS and lysed directly on plates using RIPA protein lysis buffer containing protease inhibitor cocktail (Roche Applied Sciences, Indianapolis, IN, USA). Samples were suspended in Laemmli's denaturing sample buffer (30l) containing mercaptoethanol, vortexed for 1 min, centrifuged for 5 min at 10,000 x g, and then boiled for 70oC for 10 min. Protein samples were resolved by 4%-10% SDS-PAGE and transferred to a 0.45-m pore sized PVDF membrane (Invitrogen, San Diego, CA). SMA monoclonal antibody was purchased from DAKO (Carpinteria, CA, USA) and GAPDH antibodies from Santa Cruz

35

Biotechnology Inc (Santa Cruz, CA, USA). Anti-mouse or goat-AP antibodies were obtained from Santa Cruz Biotech (Santa Cruz, CA, USA).

Quantification using real-time PCR Total RNA from CCF cultures treated with SAHA was extracted using RNeasy kit (Qiagen Inc., Valencia, CA, USA) and reverse-transcribed to cDNA following manufacturer's instructions (Promega, Madison, WI, USA). Real-time PCR (RT PCR)was performed using iQ5 real-time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA). A 50 L reaction mixture containing 2 L cDNA (250 ng), 2 L forward (200nM), 2 L reverse primer (200nM), and 25 L 2X iQ SYBR green super mix (Bio-Rad Laboratories) was run at universal cycle (95oC for 3 min, 40 cycles at 95oC for 30 sec, and 60oC for 60 sec) following manufacturer's instructions. For SMA, forward primer sequence TGGGTGACGAAGCACAGAGC and reverse primer sequence CTTCAGGGGCAACACGAAGC were used. House keeping gene actin forward primer sequence was CGGCTACAGCTTCACCACCA and reverse primer sequence was CGGGCAGCTCGTAGCTCTTC. Cycle threshold (CT) was used to detect increases in signal associated with an exponential growth of PCR product during the log-linear phase. Relative gene expression was calculated using the following formula: 2-CT, where CT is a PCR cycle number at which the fluorescence meets the threshold in the amplification plot and CT is a subtraction product of target and housekeeping genes CT values. The 2-CT is a method for determining relative target mRNA quantity in samples. Amplification efficiency for all used templates was validated by insuring that the difference between linear slopes for all

36

templates <0.1. Three independent reactions were performed and the average ( standard error of the mean or SEM) results were calculated.

Quantification and statistical analysis Immunocytochemistry data quantification was performed by counting SMA and DAPIpositive cells in 10 randomly selected non-overlapping areas of fluorescent cultures at 100X magnification field under Leica Fluorescent microscope (Leica, Wetzler, Germany). Results are expressed as mean standard error of the mean (SEM). Statistical analysis was performed using two-way analysis of variance (ANOVA) followed by Bonferroni test for comparisons test for cell toxicity assay. The RT PCR results were analyzed using one-way ANOVA followed by Tukey's multiple comparison tests. A P-value of less than 0.05 was considered significant. Immunoblotting data were analyzed using J 1.38 X image analysis software (NIH, Bethesda, MD, USA)

3. RESULTS SAHA safety Cellular viability of CCFs after SAHA treatment was determined by trypan blue exclusion assay (Fig. 8). A dose of 0.06% SAHA treatment was the highest concentration tested that did not significantly alter cellular viability. Toxicity and safety of SAHA treatment was evaluated by comparing phenotype and cell death in SAHA-treated and untreated CCFs. Light microscopy images of untreated and 0.06% SAHA-treated CCFs were obtained to observe any
37

phenotypic cellular changes. SAHA 0.06% treatment for 5 min (group 1) or 24 h (group 2) 4 h after TGF1 treatment did not alter CCF phenotype (Fig. 9) compared to corresponding controls. Similar results were seen 48 h after treatment in both groups (data not shown). The TUNEL assay, which predominantly detects apoptosis and to a lesser extent necrosis, demonstrated the safety of SAHA in this in vitro model. No detectable difference in cellular apoptosis was observed between 0.06% SAHA treated (group 1 or 2) cells and untreated controls (Fig. 10).

Immunocytochemistry and SMA quantification Results of Immunocytochemistry staining of myofibroblast transformation is shown in Figure 11. Approximately 3-5% myofibroblast cells in corneal fibroblast cultures (Fig. 11a) is normal and unavoidable.77 Treatment of CCFs with TGF1 stimulated transdifferentiation of the majority of fibroblasts into myofibroblasts (Fig. 11b), as depicted by a significant increase in SMA staining. Both group 1 (5 min) and group 2 (24 hr) 0.06% SAHA treatments showed a significant decrease in SMA immunostaining when compared to the TGF1 treated control groups (Fig 11c and 11d). Quantification of SMA immunostaining is shown in Figure 12. TGF1 treatment of CCFs produced a significantly higher level of SMA staining (89 2.2%; p<0.01) when compared to the untreated control group (8 2%). Treatment with 0.06% SAHA significantly decreased TGF1-induced myofibroblast production in both group 1 (26 7.4%, p<0.01) and

38

group 2 (23 1.6%, p<0.01). The comparison of SMA immuno-detection data between SAHA treatment groups 1 and 2 showed no significant difference in SMA levels.

SMA real-time PCR and western blot The relative expression of SMA in the untreated control group, the TGF1 treatment group, and 5 min (group 1) and 24 h (group 2) SAHA treatment groups was quantified using real-time PCR (Fig. 13). In both SAHA treatment groups 1 and 2 a significant decrease in SMA expression was detected when compared to the TGF1 treatment group. The percentage of SMA mRNA level was significantly lower in both SAHA treatment groups 1 and 2 compared to control. SMA mRNA levels were not significantly different between SAHA treatment groups 1 and 2. Western blot analysis of SMA protein (Fig. 14) was consistent with our quantitative mRNA and immunocytochemistry findings. Treatment of CCFs with 0.06% SAHA effectively suppressed SMA expression in both treatment groups 1 and 2.

4. DISCUSSION Corneal wound healing involves numerous cellular mechanisms including corneal epithelial and keratocyte migration, proliferation, apoptosis, matrix remodeling, and transdifferentiation of keratocytes to fibroblasts and myofibroblasts.26, 78, 79 This process is regulated by several factors including various growth factors, cytokines, and chemokines.26, 80
39
79,

81

Of the various cytokines involved in the formation of corneal fibrosis, TGF has been found

to play a pivotal role in initiating this transdifferentiation. The development of corneal fibrosis is greatly dependent on the generation of myofibroblasts.4, 82, 83 Myofibroblasts are highly contractile cells with decreased intracellular crystallin production resulting in reduced corneal transparency. Currently, topical mitomycin C (MMC), an alkylating chemotherapeutic agent, is commonly used in physician ophthalmology to prevent laser-induced corneal haze. MMC blocks keratocyte proliferation. Recently our lab reported that MMC effectively reduced the formation of myofibroblasts in the equine77 and canine cornea84 in vitro and in rabbit cornea in vivo.85 However, MMC application is associated with several complications including corneal endothelial damage, abnormal wound healing, limbal and scleral necrosis, loss of keratocytes, and the inhibition of keratocyte repopulation.85, 86 As a class of drugs, HDACi alter gene expression profiles at the epigenetic level and change protein function by inhibiting the activity of HDAC. HDACi were originally developed for their anti-proliferative properties in transformed cells.87, 88 They are currently being extensively investigated in cancer research for their anti-angiogenic properties and their ability to induce apoptosis and cell cycle arrest in malignant cells.89 Additionally, HDACi also prevent pro-inflammatory cytokine production and have demonstrated anti-inflammatory activities in animal models of inflammatory bowel diseases, multiple sclerosis and SLE.90-93 The therapeutic potential of HDACi as antifibrotic agents has been recently reported in lung and cardiac fibrosis.94, 95 In this study we demonstrated that 0.06% SAHA, an HDACi, significantly

40

decreased the amount of myofibroblast formation without significant cellular toxicity using an established canine corneal in vitro model. The precise mechanism through which SAHA inhibits TGF1 stimulated myofibroblast formation remains unclear. It is reported that the inhibition of HDAC4, HDAC6, and HDAC8 impair TGF1 induced SMA expression.96 The inhibition of HDAC4 has been found to prevent morphological changes associated with TGF1 stimulation.96 Down-regulation of HDAC4 stimulated the expression two endogenous repressors of the TGF signaling pathway but did not stimulate inhibitory Smad7.96 Therefore HDAC4 may be a potential specific target for treating fibrotic disease. Studies specifically aimed at examining the use of SAHA in fibrotic disease are limited. A recent report investigated the effects of SAHA on human lung fibroblast cultures as a therapeutic potential for pulmonary fibrosis. That study showed an inhibition of TGF1 induced transdifferentiation of fibroblasts to myofibroblasts, demonstrated by the suppression of SMA expression. Furthermore, SAHA treatment reduced TGF1 expression and collagen I deposition in the tested cell type without causing apoptosis.94 A significant reduction in SMA expression due to SAHA treatment was consistent with our findings reported here. Inflammation typically precedes fibrosis in fibro-proliferative diseases, thus the antiinflammatory properties of SAHA may help to prevent fibrosis. Known anti-rheumatic activities of SAHA include growth arrest of rheumatoid arthritis synovial fibroblasts, suppression of proinflammatory cytokines and nitrous oxide (NO) release, and down-regulation of angiogenesis and MMPs.97 SAHA has also been reported to reduce circulating levels of multiple pro41

inflammatory cytokines in a rat model of hypertensive cardiomyopathy.95 Although we did not specifically evaluate fibroblast proliferation in this study, future studies may involve the use of an MTT ((3-4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay to assess fibroblast proliferation as well as investigation of underlying molecular mechanisms of SAHAmediated anti-fibrosis in the canine cornea. To the authors knowledge, this is the first study to evaluate the use of an HDACi, specifically SAHA, to prevent corneal fibrosis in a canine in vitro model. Our results suggest that SAHA may be a potent agent to treat fibrosis in the canine cornea. In vivo studies are warranted.

42

APPENDIX
Transduction of Dividing (D) or Non-dividing (ND) cells D D & ND D & ND D & ND D

Vector AV AAV Lentivirus Disabled Lentivirus Retrovirus

Immunogenicity Moderate-severe No-mild Mild-moderate Mild-moderate Mild-severe

Long-term expression No Yes No Yes No

Table 1: Characteristics of viral vectors used in corneal gene therapy.

43

Figure 1. Representative immunocytochemistry images demonstrating phalloidin staining for Factin (shown in red) in canine corneal fibroblasts (a) and myofibroblasts (b). Cellular nuclei are stained blue with DAPI. Myofibroblasts (b) show significantly higher amount of F-actin staining compared to and fibroblasts (a) revealing the levels of fibrosis.

44

Figure 2. Representative images demonstrating delivered GFP staining in CCFs. AAV5transduced CCFs (b) show significantly higher number of GFP-positive cells compared to the untreated control cell population (a). Dual staining was performed to detect SMA (shown in red), a marker for myofibroblasts.

45

Figure 3. Representative images demonstrating delivered GFP staining in CCMs (b). AAV5transduced CCMs (b) show significantly higher number of GFP-positive cells compared to the untreated control cell population (a). Dual staining was performed to detect SMA (shown in red), a marker for myofibroblasts.

46

Figure 4. Quantification of GFP positive cells showing the amount of gene delivery by AAV5 into CCFs as well as CCMs. In both cell populations there is significant transduction of AAV5GFP (grey bar) compared to the untreated controls (black bar). However, the percentage of GFP-positive cells is significantly lower in CCMs compared to CCFs. Error bars indicate standard error. * p <0.01 versus control; p<0.01 versus AAV5-GFP CCFs.

47

Figure 5. Quantitative PCR measurements of AAV5-delivered GFP in CCFs and CCMs. Both cell populations showed significant transduction of AAV5-GFP (crossed bar) compared to the untreated controls (black solid bar). However, the levels of GFP-mRNA were significantly lower in CCMs compared to CCFs. Error bars indicate standard error. * p <0.01 versus control; p<0.01 versus AAV5-GFP CCFs.

48

Figure 6. Phase-contrast microscopic images of CCF (a) and CCM (b) cultures taken 48 h after AAV transduction showed that AAV5 viral vector application did not alter the morphology of the fibroblast and myofibroblast culture population. Similar results were seen at 24 h.

49

Figure 7. Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay demonstrating apoptotic cells (pink staining) in CCF cultures transduced with AAV5-GFP. AAV5-GFP treated cells (b) demonstrated a comparable number of TUNEL-positive cells (white arrows) to untreated controls (a). Similar results were seen for CCMs. Nuclei are stained blue with DAPI. Trypan Blue exclusion data (c) shows the percent viability of CCFs 24 h and 48 h following AAV transduction compared to controls. Percent cellular viability slightly decreased (results not significant) in AAV5-GFP treated CCFs (black bars) and CCMs (results not shown) compared to the untreated controls (white bars). Error bars indicate standard error.

50

Figure 8: A bar graph showing the cellular viability as determined by trypan blue assay in order to demonstrate the dose dependent effect of SAHA on CCFs. Doses of 0.06% SAHA or less did not alter cellular viability. However, 0.12% SAHA demonstrated mild cellular toxicity.

51

Figure 9: A composite image showing phase-contrast microscopy images of CCFs at 5 minutes and 24 hours. No changes in CCF morphology were observed in group 1 (b) or group 2 (d) 0.06% SAHA treatment, when compared with their respective control groups (a & c). Bar = 10 m

52

Figure 10: Representative images of TUNEL assay results detecting apoptosis of CCFs in the untreated control group (a) and in CCFs treated with 0.06% SAHA for 24 h (b). No detectable differences in cellular apoptosis were observed between treated and control groups.

53

Control

TGF

Group 1 (5 min)

Group 2 (24 hrs)

Figure 11: Immunocytochemical staining using SMA antibody (green). Detection of ~ 3-5% myofibroblast cells in corneal fibroblast cultures (a) is normal and unavoidable.25 Treatment of CCFs with TGF1 showed significant increase in SMA staining (b). Both group 1 (c) and group 2 (d) 0.06% SAHA treatment groups showed a significant decrease in SMA immunostaining when compared to the TGF1 treated control group (b).

54

Figure 12: Bar graphs showing SMA immunostaining quantification. TGF1 treatment of CCFs produced a significantly higher level of SMA staining when compared to the untreated control group. Treatment with 0.06% SAHA significantly decreased TGF1-induced myofibroblast production in both the brief and extended treatment groups. # denotes p<0.01 (control vs TGF1) and * denotes p<0.01 (TGF1 vs SAHA treatment groups). No significant difference in SMA levels were found between the brief and extended SAHA treatment groups.

55

Figure 13: Bar graphs showing quantitative real-time PCR results demonstrating the relative expression of SMA . In both group 1 (5 min) and group 2 (24 h) SAHA treatment groups there was a significant decrease in SMA expression compared to the TGF1 treatment group. * denotes p<0.01 (control vs TGF1) and ** denotes p<0.01 (TGF1 vs SAHA treatment groups 1 & 2). SMA mRNA levels were not significantly different between the brief and extended SAHA treatment groups.

56

Figure 14: Western blot analysis of SMA protein. Treatment of CCFs with 0.06% SAHA effectively suppressed SMA expression in both treatment groups 1 (5 min)& 2 (24 h).

57

BIBLIOGRAPHY Mohan RR, Sharma A, Netto MV et al. Gene therapy in the cornea. Progress in Retinal 1. and Eye Research. 2005; 24: 537-559. 2. Tandon A, Tovey JC, Sharma A et al. Role of transforming growth factor Beta in corneal function, biology and pathology. Curr Mol Med. 2010; 10: 565-578. 3. Gilger BC. Diseases and Surgery of the Canine Cornea and Sclera. In: Veterinary Ophthalmology. (ed. Gelatt KN). 4th ed. Blackwell Publishing Ltd, Ames, Iowa, 2007; 690-752. 4. Jester JV, Huang J, Barry-Lane PA et al. Transforming growth factor(beta)-mediated corneal myofibroblast differentiation requires actin and fibronectin assembly. Investigative Ophthalmology and Visual Science. 1999; 40: 1959-1967. 5. Wichayacoop T, Briksawan P, Tuntivanich P et al. Anti-inflammatory effects of topical supernatant from human amniotic membrane cell culture on canine deep corneal ulcer after human amniotic membrane transplantation. Veterinary Ophthalmology. 2009; 12: 28-35. 6. Barros PS, Safatle AM, Godoy CA et al. Amniotic membrane transplantation for the reconstruction of the ocular surface in three cases. Veterinary Ophthalmology. 2005; 8: 189-192. 7. Brightman AH, McLaughlin SA, Brogdon JD. Autogenous lamellar corneal grafting in dogs. Journal of American Veterinary Medical Association. 1989; 195: 469-475. 8. Hansen PA, Guandalini A. A retrospective study of 30 cases of frozen lamellar corneal graft in dogs and cats. Veterinary Ophthalmology. 1999; 2: 233-241. Vanore M, Chahory S, Payen G et al. Surgical repair of deep melting ulcers with porcine 9. small intestinal submucosa (SIS) graft in dogs and cats. Veterinary Ophthalmology. 2007; 10: 93-99. 10. Aguirre GD, Baldwin V, Pearce-Kelling S et al. Congenital stationary night blindness in the dog: common mutation in the RPE65 gene indicates founder effect. Molecular Vision. 1998; 4: 23. 11. Acland GM, Aguirre GD, Ray J et al. Gene therapy restores vision in a canine model of childhood blindness. Nature Genetics. 2001; 28: 92-95. 12. Bainbridge JW, Ali RR. Success in sight: The eyes have it! Ocular gene therapy trials for LCA look promising. Gene Therapy. 2008; 15: 1191-1192. Maguire AM, Simonelli F, Pierce EA et al. Safety and efficacy of gene transfer for 13. Leber's congenital amaurosis. New England Journal of Medicine. 2008; 358: 2240-2248. 14. Mohan RR, Schultz GS, Hong JW et al. Gene transfer into rabbit keratocytes using AAV and lipid-mediated plasmid DNA vectors with a lamellar flap for stromal access. Experimental Eye Research. 2003; 76: 373-383. 15. Saika S, Yamanaka O, Sumioka T et al. Fibrotic disorders in the eye: targets of gene therapy. Progress in Retinal and Eye Research. 2008; 27: 177-196. 16. Sharma A, Ghosh A, Hansen ET et al. Transduction efficiency of AAV 2/6, 2/8 and 2/9 vectors for delivering genes in human corneal fibroblasts. Brain Research Bulletin. 2010; 81: 273-278. 17. Behrens A, Gordon EM, Li L et al. Retroviral gene therapy vectors for prevention of excimer laser-induced corneal haze. Investigative Ophthalmology and Visual Science. 2002; 43: 968-977.
58

Cheng HC, Yeh SI, Tsao YP et al. Subconjunctival injection of recombinant AAV18. angiostatin ameliorates alkali burn induced corneal angiogenesis. Molecular Vision. 2007; 13: 2344-2352. 19. Klausner EA, Peer D, Chapman RL et al. Corneal gene therapy. Journal of Controlled Release. 2007; 124: 107-133. 20. Parker DG, Brereton HM, Coster DJ et al. The potential of viral vector-mediated gene transfer to prolong corneal allograft survival. Current Gene Therapy. 2009; 9: 33-44. 21. Sharma A, Tovey JC, Ghosh A et al. AAV serotype influences gene transfer in corneal stroma in vivo. Experimental Eye Research. 2010. 22. Monahan PE, Samulski RJ. Adeno-associated virus vectors for gene therapy: more pros than cons? Molecular Medicine Today. 2000; 6: 433-440. 23. Monahan PE, Samulski RJ. AAV vectors: is clinical success on the horizon? Gene Therapy. 2000; 7: 24-30. 24. Royo NC, Vandenberghe LH, Ma JY et al. Specific AAV serotypes stably transduce primary hippocampal and cortical cultures with high efficiency and low toxicity. Brain Research. 2008; 1190: 15-22. 25. Buss DG, Giuliano EA, Sharma A et al. Gene delivery in the equine cornea: a novel therapeutic strategy. Veterinary Ophthalmology. 2010; 13: 301-306. 26. Netto MV, Mohan RR, Ambrosio R, Jr. et al. Wound healing in the cornea: a review of refractive surgery complications and new prospects for therapy. Cornea. 2005; 24: 509-522. Conrad GW, Funderburgh JL. Eye development and the appearance and maintenance of 27. corneal transparency. Transactions of the Kansas Academy of Science. 1992; 95: 34-38. 28. Boote C, Dennis S, Newton RH et al. Collagen fibrils appear more closely packed in the prepupillary cornea: optical and biomechanical implications. Investigative Ophthalmology and Visual Science. 2003; 44: 2941-2948. Maurice DM. The transparency of the corneal stroma. Vision Research. 1970; 10: 10729. 108. 30. Sharma A, Mehan MM, Sinha S et al. Trichostatin a inhibits corneal haze in vitro and in vivo. Investigative Ophthalmology and Visual Science. 2009; 50: 2695-2701. 31. Kitano A, Okada Y, Yamanka O et al. Therapeutic potential of trichostatin A to control inflammatory and fibrogenic disorders of the ocular surface. Mol Vis. 2010; 16: 2964-2973. 32. Struhl K. Histone acetylation and transcriptional regulatory mechanisms. Genes and Development. 1998; 12: 599-606. Yoshida M, Horinouchi S, Beppu T. Trichostatin A and trapoxin: novel chemical probes 33. for the role of histone acetylation in chromatin structure and function. Bioessays. 1995; 17: 423430. 34. Duvic M, Talpur R, Ni X et al. Phase 2 trial of oral vorinostat (suberoylanilide hydroxamic acid, SAHA) for refractory cutaneous T-cell lymphoma (CTCL). Blood. 2007; 109: 31-39. 35. Ritter T, Yang J, Dannowski H et al. Effects of interleukin-12p40 gene transfer on rat corneal allograft survival. Transplant Immunology. 2007; 18: 101-107. 36. Yu H, Wu J, Li H et al. Inhibition of corneal neovascularization by recombinant adenovirus-mediated sFlk-1 expression. Biochemical and Biophysical Research Communications. 2007; 361: 946-952.
59

Somia N, Verma IM. Gene therapy: trials and tribulations. Nature Reviews Genetics. 37. 2000; 1: 91-99. 38. Montini E, Cesana D, Schmidt M et al. Hematopoietic stem cell gene transfer in a tumorprone mouse model uncovers low genotoxicity of lentiviral vector integration. Nature Biotechnology. 2006; 24: 687-696. 39. Mohan RR, Tovey JC, Sharma A et al. Gene therapy in the Cornea: 2005-present. Progress in Retinal and Eye Research. 2011. 40. Zhong L, Li B, Jayandharan G et al. Tyrosine-phosphorylation of AAV2 vectors and its consequences on viral intracellular trafficking and transgene expression. Virology. 2008; 381: 194-202. 41. Petrs-Silva H, Dinculescu A, Li Q et al. High-efficiency transduction of the mouse retina by tyrosine-mutant AAV serotype vectors. Molecular Therapy. 2009; 17: 463-471. 42. Cai X, Conley S, Naash M. Nanoparticle applications in ocular gene therapy. Vision Research. 2008; 48: 319-324. 43. Farjo R, Skaggs J, Quiambao AB et al. Efficient non-viral ocular gene transfer with compacted DNA nanoparticles. PLoS One. 2006; 1: e38. 44. Konstan MW, Davis PB, Wagener JS et al. Compacted DNA nanoparticles administered to the nasal mucosa of cystic fibrosis subjects are safe and demonstrate partial to complete cystic fibrosis transmembrane regulator reconstitution. Human Gene Therapy. 2004; 15: 1255-1269. 45. Sharma A, Tandon A, Tovey JC et al. Polyethylenimine-conjugated gold nanoparticles: Gene transfer potential and low toxicity in the cornea. Nanomedicine : nanotechnology, biology, and medicine. 2011; 7: 505-513. 46. Hansen PA, Guandalini A. A retrospective study of 30 cases of frozen lamellar corneal graft in dogs and cats. Veterinary Ophthalmology. 1999; 2: 233-241. 47. Brooks DE, Plummer CE, Kallberg ME et al. Corneal transplantation for inflammatory keratopathies in the horse: visual outcome in 206 cases (1993-2007). Veterinary Ophthalmology. 2008; 11: 123-133. 48. Pena Gimenez MT, Farina IM. Lamellar keratoplasty for the treatment of feline corneal sequestrum. Veterinary Ophthalmology. 1998; 1: 163-161. 49. Brooks DE. Targeted lamellar keratoplasty in the horse: a paradigm shift in equine corneal transplantation. Equine Veterinary Journal. 2010; 42 Suppl 37: 24-30. 50. Panda A, Vanathi M, Kumar A et al. Corneal graft rejection. Survey of Ophthalmology. 2007; 52: 375-396. Denis HM. Equine corneal surgery and transplantation. Veterinary Clinics of North 51. America Equine Practice. 2004; 20: 361-380, vi-vii. 52. Williams KA, Coster DJ. Gene therapy for diseases of the cornea - a review. Clin Experiment Ophthalmol. 2010; 38: 93-103. 53. Beutelspacher SC, Tan PH, McClure MO et al. Expression of indoleamine 2,3dioxygenase (IDO) by endothelial cells: implications for the control of alloresponses. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons. 2006; 6: 1320-1330. 54. Beutelspacher SC, Pillai R, Watson MP et al. Function of indoleamine 2,3-dioxygenase in corneal allograft rejection and prolongation of allograft survival by over-expression. European Journal of Immunology. 2006; 36: 690-700.
60

Barcia RN, Dana MR, Kazlauskas A. Corneal graft rejection is accompanied by apoptosis 55. of the endothelium and is prevented by gene therapy with bcl-xL. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons. 2007; 7: 2082-2089. 56. Joyce NC, Harris DL, Mc Alister JC et al. Effect of overexpressing the transcription factor E2F2 on cell cycle progression in rabbit corneal endothelial cells. Investigative Ophthalmology and Visual Science. 2004; 45: 1340-1348. 57. McAlister JC, Joyce NC, Harris DL et al. Induction of replication in human corneal endothelial cells by E2F2 transcription factor cDNA transfer. Investigative Ophthalmology and Visual Science. 2005; 46: 3597-3603. 58. Wilkie DA, Whittaker C. Surgery of the cornea. Veterinary Clinics of North America Small Animal Practice. 1997; 27: 1067-1107. 59. Mohan RR, Gupta R, Mehan MK et al. Decorin transfection suppresses profibrogenic genes and myofibroblast formation in human corneal fibroblasts. Experimental Eye Research. 2010; 91: 238-245. Mohan RR, Tandon A, Sharma A et al. Significant inhibition of corneal scarring in vivo 60. with tissue-selective, targeted AAV5 decorin gene therapy. Investigative Ophthalmology and Visual Science. 2011; 52: 4833-4841. 61. Seitz B, Moreira L, Baktanian E et al. Retroviral vector-mediated gene transfer into keratocytes in vitro and in vivo. American Journal of Ophthalmology. 1998; 126: 630-639. Galiacy SD, Fournie P, Massoudi D et al. Matrix metalloproteinase 14 overexpression 62. reduces corneal scarring. Gene Therapy. 2010. 63. Cooley PL, Dice PF, 2nd. Corneal dystrophy in the dog and cat. Veterinary Clinics of North America Small Animal Practice. 1990; 20: 681-692. 64. Crispin SM. Crystalline corneal dystrophy in the dog. Histochemical and ultrastructural study. Cornea. 1988; 7: 149-161. 65. Gilger BC, Bentley E, Ollivier FJ. Diseases and Surgery of the Canine Cornea and Sclera. In: Veterinary Ophthalmology. (ed. Gelatt KN). 4th ed. Blackwell Publishing, Ames, Iowa, 2007; 690-752. 66. Kim JH, Ko JM, Lee I et al. A novel mutation of the decorin gene identified in a korean family with congenital hereditary stromal dystrophy. Cornea. 2011; 30: 1473-1477. 67. Aldave AJ. The genetics of the corneal dystrophies. Developments in Ophthalmology. 2011; 48: 51-66. Weiss JS. Corneal dystrophies: molecular genetics to therapeutic intervention--Fifth 68. ARVO/Pfizer Ophthalmics Research Institute Conference. Investigative Ophthalmology and Visual Science. 2010; 51: 5391-5402. 69. Maguire LJ, Shearer DR. A simple method of conjunctival dissection for Gunderson flaps. Archives of Ophthalmology. 1991; 109: 1168-1169. 70. Traas AM, Wang P, Ma X et al. Correction of clinical manifestations of canine mucopolysaccharidosis I with neonatal retroviral vector gene therapy. Molecular Therapy. 2007; 15: 1423-1431. 71. Wang B, O'Malley TM, Xu L et al. Expression in blood cells may contribute to biochemical and pathological improvements after neonatal intravenous gene therapy for mucopolysaccharidosis VII in dogs. Molecular Genetics and Metabolism. 2006; 87: 8-21.
61

Buss DG, Giuliano EA, Sharma A et al. Isolation and cultivation of equine corneal 72. keratocytes, fibroblasts and myofibroblasts. Veterinary Ophthalmology. 2010; 13: 37-42. 73. Ghosh A, Yue Y, Duan D. Viral serotype and the transgene sequence influence overlapping adeno-associated viral (AAV) vector-mediated gene transfer in skeletal muscle. Journal of Gene Medicine. 2006; 8: 298-305. 74. Limberis MP, Vandenberghe LH, Zhang L et al. Transduction efficiencies of novel AAV vectors in mouse airway epithelium in vivo and human ciliated airway epithelium in vitro. Molecular Therapy. 2009; 17: 294-301. 75. Polyak S, Mah C, Porvasnik S et al. Gene delivery to intestinal epithelial cells in vitro and in vivo with recombinant adeno-associated virus types 1, 2 and 5. Digestive Diseases and Sciences. 2008; 53: 1261-1270. 76. Galiacy SD, Fournie P, Massoudi D et al. Matrix metalloproteinase 14 overexpression reduces corneal scarring. Gene Therapy. 2011; 18: 462-468. 77. Buss DG, Sharma A, Giuliano EA et al. Efficacy and safety of mitomycin C as an agent to treat corneal scarring in horses using an in vitro model. Veterinary Ophthalmology. 2010; 13: 211-218. 78. Vij N, Sharma A, Thakkar M et al. PDGF-driven proliferation, migration, and IL8 chemokine secretion in human corneal fibroblasts involve JAK2-STAT3 signaling pathway. Molecular Vision. 2008; 14: 1020-1027. 79. Jester JV, Petroll WM, Cavanagh HD. Corneal stromal wound healing in refractive surgery: the role of myofibroblasts. Progress in Retinal and Eye Research. 1999; 18: 311-356. 80. Imanishi J, Kamiyama K, Iguchi I et al. Growth factors: importance in wound healing and maintenance of transparency of the cornea. Progress in Retinal and Eye Research. 2000; 19: 113-129. 81. Netto MV, Mohan RR, Sinha S et al. Stromal haze, myofibroblasts, and surface irregularity after PRK. Experimental Eye Research. 2006; 82: 788-797. 82. Maltseva O, Folger P, Zekaria D et al. Fibroblast growth factor reversal of the corneal myofibroblast phenotype. Investigative Ophthalmology and Visual Science. 2001; 42: 24902495. 83. Jester JV, Huang J, Petroll WM et al. TGFbeta induced myofibroblast differentiation of rabbit keratocytes requires synergistic TGFbeta, PDGF and integrin signaling. Experimental Eye Research. 2002; 75: 645-657. 84. Gupta R, Yarnell BW, Giuliano EG et al. Mitomycin C: a promising agent for the treatment of canine corneal scarring. Veterinary Ophthalmology. 2011: 1-9. 85. Netto MV, Mohan RR, Sinha S et al. Effect of prophylactic and therapeutic mitomycin C on corneal apoptosis, cellular proliferation, haze, and long-term keratocyte density in rabbits. Journal of Refractive Surgery. 2006; 22: 562-574. 86. Nassaralla BA, McLeod SD, Nassaralla JJ, Jr. Prophylactic mitomycin C to inhibit corneal haze after photorefractive keratectomy for residual myopia following radial keratotomy. Journal of Refractive Surgery. 2007; 23: 226-232. 87. Marks PA. Discovery and development of SAHA as an anticancer agent. Oncogene. 2007; 26: 1351-1356. 88. Marks PA, Breslow R. Dimethyl sulfoxide to vorinostat: development of this histone deacetylase inhibitor as an anticancer drug. Nature Biotechnology. 2007; 25: 84-90.
62

Bolden JE, Peart MJ, Johnstone RW. Anticancer activities of histone deacetylase 89. inhibitors. Nat Rev Drug Discov. 2006; 5: 769-784. 90. Tong X, Yin L, Giardina C. Butyrate suppresses Cox-2 activation in colon cancer cells through HDAC inhibition. Biochemical and Biophysical Research Communications. 2004; 317: 463-471. 91. Adcock IM. HDAC inhibitors as anti-inflammatory agents. British Journal of Pharmacology. 2007; 150: 829-831. 92. Choo QY, Ho PC, Lin HS. Histone deacetylase inhibitors: new hope for rheumatoid arthritis? Current Pharmaceutical Design. 2008; 14: 803-820. 93. Lin HS, Hu CY, Chan HY et al. Anti-rheumatic activities of histone deacetylase (HDAC) inhibitors in vivo in collagen-induced arthritis in rodents. British Journal of Pharmacology. 2007; 150: 862-872. 94. Wang Z, Chen C, Finger SN et al. Suberoylanilide hydroxamic acid: a potential epigenetic therapeutic agent for lung fibrosis? European Respiratory Journal. 2009; 34: 145155. Iyer A, Fenning A, Lim J et al. Antifibrotic activity of an inhibitor of histone 95. deacetylases in DOCA-salt hypertensive rats. British Journal of Pharmacology. 2010; 159: 1408-1417. 96. Glenisson W, Castronovo V, Waltregny D. Histone deacetylase 4 is required for TGFbeta1-induced myofibroblastic differentiation. Biochimica et Biophysica Acta. 2007; 1773: 1572-1582. 97. Choo QY, Ho PC, Tanaka Y et al. Histone deacetylase inhibitors MS-275 and SAHA induced growth arrest and suppressed lipopolysaccharide-stimulated NF-kappaB p65 nuclear accumulation in human rheumatoid arthritis synovial fibroblastic E11 cells. Rheumatology. 2010; 49: 1447-1460.

63