OPTIMIZING OCULAR GENE THERAPY FOR LONG TERM INTRAOCULAR PRESSURE CONTROL IN ADAMTS10-OPEN ANGLE GLAUCOMA By Kristin Leigh Koehl A THESIS Submitted to Michigan State University in partial fulfillment of the requirement for the degree of Comparative Medicine and Integrative Biology – Master of Science 2018 ABSTRACT OPTIMIZING OCULAR GENE THERAPY FOR LONG TERM INTRAOCULAR PRESSURE CONTROL IN ADAMTS10-OPEN ANGLE GLAUCOMA By Kristin Leigh Koehl Glaucoma is an optic neuropathy that causes irreversible damage to retinal ganglion cell axons. Most forms of open angle glaucoma (OAG) are associated with an increased intraocular pressure (IOP) due to obstruction of the aqueous humor outflow pathways (AHOP). For years our team has worked to preserve vision by targeting these outflow pathways by delivering a novel capsid mutated adeno-associated virus (AAV) mediated gene replacement therapy in a clinically relevant and well-established OAG dog model caused by a mutation in ADAMTS10. Past studies with ssAAV2-(Y444F)-hGFP (targeting) and ssAAV2-(Y444F)-hADAMTS10 (therapy) at vector genome (vg) dose of 1-3x1011 vg administered intracamerally (IC) in a 50µL volume delivered safe and long-term transgene expression to the AHOPs; however, an IOP reducing therapeutic effect was not achieved. Our goal was to increase transduction efficiency to the AHOPs by delivering the same vector constructs at a higher dose. In Aim 1, we showed in 5 dogs that 2.78-5.56x1012vg provided increased and clinically observable transgene expression within the iridocorneal angle (ICA). In Aim 2, 2 of 3 dogs IC injected with ssAAV2-(Y444F)- hADAMTS10 at 1.25-1.43x1012vg experienced a significant reduction in IOP by 19-20 days post- injection with a steady and well-maintained reduction for at least 105 weeks. ssAAV2-(Y444F)- hADAMTS10 was well-tolerated and resulted in an increased outflow facility by 3-75%. These studies provide proof-of-concept that modification of gene expression within the AHOP is possible and can result in long-term IOP control. ACKNOWLEDGEMENTS The amazing work that encompasses this thesis is due to a great team of people and all require acknowledgement and thanks for their hard work and dedication for great science. My mentor deserves great acknowledgment and more thanks than I could ever return. Dr. András Komáromy is by far the kindest and most genuine human I have ever met and truly, one to model yourself after. Thank you for teaching me science and grace, the opportunities to advance my career, and your patience for my learning and questions. Thank you for the mentorship and the ability to always call you a friend. To Dr. Vilma Yuzbasiyan-Gurkan, thank you for the opportunity to advance my education. Your kindness, enthusiasm, and pride in all your students within the program is unparalleled. To my committee, Drs. Simon Petersen-Jones, Shannon Boye, Sayoko Moroi and Arthur Weber, your guidance over the years to completion of the thesis is appreciated; and, thank you for your collaborations, expertise and enlightenment during my Master’s academic career. To the Boye Laboratory at the University of Florida, thank you for your vast and exceptional knowledge of vector technology. To Dr. Juan Pedro Steibel at Michigan State University, thank you for meeting with me and sharing your knowledge in biostatistics. To Dr. Chris Pirie at Michigan State University, thank you for your tremendous contribution to our kinetics study and for being a witty and fun person to chat up science. To Dr. Dodd Sledge at iii Michigan State University’s Veterinary Diagnostic Laboratory, for your wealth and expertise in ocular pathology. To Christine Harman of the Komáromy Laboratory, thank you for everything! You are a dear friend that has imparted much wisdom in science and in friendship. To all the undergraduate students of the Komáromy Laboratory, past and present, thank you for all your hard work and impressive drive. You all are truly inspirational at your young ages. To all the people that care for the dogs. The animals deserve great lives and each of you make that happen. Thank you, Kim Williamson, Ramona Stambaugh, and Danae Walters for all your care and affection for our great animals. The financial support that make all things possible: NIH grants, R01EY025752, R01- EY024280, P30EY021721; The Glaucoma Research Foundation; Research to Prevent Blindness; Foundation Fighting Blindness; and Michigan State University Discretionary Funding Initiative. Thank you to the Edward Sheppard family for the wonderful unrestricted gift. For my friends and family outside of science. To my good friend Katie Meyer, you are an amazing woman who always inspires me with your generosity. To my parents, Terry and Margaret and my soon to be husband, David, thank you for your amazing patience and support through my life endeavors. Thank you for living all the smiles and tears with me. Of all the acknowledgements, most importantly, ALL the dogs. Thank you to each and every one of you! Your time and lives will never be forgotten. May God always be good to you. Until we meet again. iv TABLE OF CONTENTS LIST OF TABLES ..................................................................................................................................... vii LIST OF FIGURES ................................................................................................................................... viii KEY TO ABBREVIATIONS AND SYMBOLS ................................................................................................. x CHAPTER 1 - Introduction ....................................................................................................................... 1 1.1 Glaucoma Classification ................................................................................................................ 1 1.2 Aqueous Humor Dynamics ............................................................................................................ 5 1.2.1 Aqueous Humor Production .................................................................................................... 6 1.2.2 Aqueous Humor Drainage ....................................................................................................... 7 1.3 Trabecular Meshwork ................................................................................................................... 9 1.3.1 Uveal and Corneal Scleral Meshwork ...................................................................................... 9 1.3.2 Juxtacanalicular Tissue ............................................................................................................ 9 1.4 Trabecular Meshwork Extracellular Matrix (TM-ECM) ................................................................ 10 1.4.1 Mechanisms that Can Change the ECM Profile ...................................................................... 11 1.5 Schlemm’s Canal/Angular Aqueous Plexus ................................................................................. 12 1.6 Genetics of OAG .......................................................................................................................... 12 1.7 Animal Models of OAG ................................................................................................................ 16 1.7.1 A Disintegrin and Metalloproteinase with Thrombospondin Motifs (ADAMTS) – Canine OAG Models .......................................................................................................................................... 18 1.8 Preservation of Vision by Therapeutic Lowering of IOP .............................................................. 21 1.8.1 Current Therapies ................................................................................................................. 22 1.9 Adeno-Associated Virus (AAV) for Targeted Gene Therapy ......................................................... 25 1.10 Purpose ..................................................................................................................................... 29 CHAPTER 2 – Materials and Methods ................................................................................................... 31 2.1 Study Design ............................................................................................................................... 31 2.1.1 Aim 1 – Targeting the AH outflow pathway – hGFP ............................................................... 31 2.1.2 Aim 2 – Therapeutic delivery - ssAAV2-(Y444F)-hADAMTS10 ................................................. 31 2.2 Animals ....................................................................................................................................... 32 2.3 Anesthesia Protocols ................................................................................................................... 33 2.4 Clinical Examination .................................................................................................................... 35 2.4.1 Diurnal IOP – primary outcome measure ............................................................................... 35 2.4.2 Pachymetry ........................................................................................................................... 36 2.4.3 Gonioscopy ........................................................................................................................... 36 2.4.4 Ultrasound biomicroscopy (UBM) ......................................................................................... 37 2.5 AAV Production ........................................................................................................................... 37 2.6 Intraocular injections – IC and IVit .............................................................................................. 38 2.6.1 ssAAV2-(Y444F)-hGFP and ssAAV2-(Y444F)-hADAMTS10 ....................................................... 38 2.6.2 Optimizing Injection Technique ............................................................................................. 44 2.7 Aqueous Humor Outflow Facility - Pneumotonography .............................................................. 48 2.8 Aqueous Humor Flow - Fluorophotometry .................................................................................. 49 v 2.9 Optical Coherence Tomography (OCT) and A-scan Biometry ...................................................... 49 2.10 Electroretinography (ERG) ......................................................................................................... 50 2.11 Aqueous Paracentesis and Aqueous Humor Analyses ............................................................... 51 2.12 Tissue Collection for Histopathology and Immunohistochemistry ............................................ 52 CHAPTER 3 - Results .............................................................................................................................. 55 3.1 Aim 1 - Effect of Dosing and Injection Route on TM Transgene Expression ................................. 55 3.1.1 Increasing the AAV Dose ....................................................................................................... 55 3.1.2 Comparing IC vs IVit injections .............................................................................................. 57 3.1.3 Effect on Pupillary Aperture .................................................................................................. 58 3.1.4 Other Observations following IC and IVit ssAAV2-(Y444F)-hGFP ............................................ 60 3.1.5 Comparison of hGFP Expression Between Quadrants ............................................................ 63 3.1.6 Fluorescein Kinetics of the AC ............................................................................................... 64 3.1.7 Summary of Aim 1 – Targeting the AH outflow pathway – ssAAV2-(Y444F)-hGFP .................. 67 3.2 Aim 2 – Treatment of Canine ADAMTS10-OAG – ssAAV2-(Y444F)-hADAMTS10 .......................... 68 3.2.1 Long-term IOP control following IC injection of ssAAV2-(Y444F)-hADAMTS10 ....................... 68 3.2.2 Increased Outflow Facility in Successfully Treated Eyes ......................................................... 71 3.2.3 Aqueous Humor Flow Remained Unaffected by Gene Therapy .............................................. 73 3.2.4 ICA Morphology .................................................................................................................... 74 3.2.5 Optic Nerve Head (ONH) Morphology and Retinal Function .................................................. 76 3.2.6 ssAAV2-(Y444F)-hADAMTS10 Safety Measures – Clinical Examination, Pachymetry and A-scan Biometry........................................................................................................................................ 84 3.3 Adverse Events – Ocular Inflammation ....................................................................................... 86 3.3.1 Aim 1 - Severe Inflammation ................................................................................................. 86 3.3.2 Neutralizing Antibody (NAb) Assays ...................................................................................... 88 3.3.3 Aim 2 – Protein Concentration and Cell Count of AH ............................................................. 89 Chapter 4 – Discussion .......................................................................................................................... 91 REFERENCES ........................................................................................................................................ 100 vi LIST OF TABLES Table 1.1. Glaucoma Phenotypes .................................................................................................. 3 Table 1.2. Genes and Susceptible Loci ......................................................................................... 13 Table 1.3. Animal Models of OAG ................................................................................................ 17 Table 1.4. ADAMTS10 Mutation Comparisons ............................................................................ 20 Table 1.5. Traditional Topical Medication Synopsis .................................................................... 23 Table 1.6. Capsid Mutations Investigated by Bogner et al .......................................................... 30 Table 2.1. Aim 1 and 2 Study Participants ................................................................................... 33 Table 2.2. hGFP and hADAMTS10 Dosing Assignments ............................................................... 40 Table 2.3. Fluorescein Kinetics of the Anterior Chamber Assignments ...................................... 47 Table 2.4. Photopic ERG Protocol ................................................................................................ 51 Table 3.1. Summary of Target Tissues Identified in Aim 1 .......................................................... 68 Table 3.2. Dog 1: ERG a-, b-, and PhNR Wave Amplitudes and Peak Time Comparisons ........... 78 Table 3.3. Dog 2: ERG a-, b-, and PhNR Wave Amplitudes and Peak Time Comparisons ........... 80 Table 3.4. Dog 3: ERG a-, b-, and PhNR Wave Amplitudes and Peak Time Comparisons ........... 82 Table 3.5. Percent Variation of Pachymetry in ssAAV2-(Y444F)-hADAMTS10 Treated and Control Eyes .................................................................................................................................. 84 Table 3.6. A-scan Biometry Comparing ssAAV2-(Y444F)-hADAMTS10 Treated and Control Eyes ............................................................................................................................................... 85 Table 3.7. Aim 2 – NAb Titers for T Cell Response to scAAV2-(Y444F)-smCBA-mCherry ............ 89 Table 3.8. Protein Concentration and Cell Count of AH of Aim 2 Dogs ....................................... 90 vii LIST OF FIGURES Figure 1.1. Schematic drawing of aqueous humor drainage pathways. ..................................... 8 Figure 2.1. Anterior segment embedding and sectioning map. ................................................ 54 Figure 3.1. Effective TM targeting with ssAAV2-(Y444F)-hGFP in dogs. ................................... 56 Figure 3.2. Relative % area of IHC hGFP labeling within the ICA when comparing doses and routes of administration. ............................................................................................................. 58 Figure 3.3. Effect on relative % area of IHC hGFP labeling within the ICA when comparing pupil size when utilizing the 2.78E12vg dose. ............................................................................. 60 Figure 3.4. hGFP expression in the TM and other intraocular tissues when comparing IC injections with variable pupil sizes and IVit injection. ............................................................... 62 Figure 3.5. Comparison of % IHC hGFP expression between four quadrants following IC delivery of ssAAV2-(Y444F)-hGFP.. .............................................................................................. 64 Figure 3.6. AC fluorescein kinetics as a function of temperature and vehicle.......................... 66 Figure 3.7. Lens capsule fluorescence shown with blue slit beam. ........................................... 66 Figure 3.8. Anterior diffusion of AH through the pupillary aperture followed by convection- like movement within the AC. ..................................................................................................... 67 Figure 3.9. Diurnal IOP for ssAAV2-(Y444F)-hADAMTS10 treated and untreated control eyes. .............................................................................................................................................. 69 Figure 3.10. Linear regression fitted trendline with CI for treated and untreated eyes. ......... 70 Figure 3.11. Linear regression fitted trendline with CI for difference between treated and untreated eyes. ............................................................................................................................ 71 Figure 3.12. Outflow facility measured by pneumotonography. .............................................. 72 Figure 3.13. Aqueous humor flow measured by fluorophotometry. ........................................ 73 Figure 3.14. ICA morphology of treated and untreated fellow eyes compared by gonioscopic examination and UBM. ................................................................................................................ 75 viii Figure 3.15. Morphologic comparison of ONH and comparison of retinal function. ............... 77 Figure 3.16. Photomicrographs depicting inflammation induced by injection of ssAAV2- (Y444F)-hGFP. ............................................................................................................................... 88 ix KEY TO ABBREVIATIONS AND SYMBOLS Anterior Chamber Aqueous Humor Axial Length AAALAC Association for the Assessment and Accreditation of Laboratory Animal Care International Angular Aqueous Plexus Adeno Associated Virus Angle Closure Glaucoma Anterior Chamber Depth Aqueous Humor Outflow Pathway Anterior Chamber-associated Immune Deviation AAP AAV AC ACAID ACD ACG ADAMTS10 A Disintegrin and Metalloproteinase with Thrombospondin Motifs 10 AH AHOP ARVO AVMA AXL CCT cd CI CLANs CSTM CYP1B1 Association for Research in Vision and Ophthalmology American Veterinary Medical Association Corneoscleral Trabecular Meshwork Cross-linked Actin Networks Central Corneal Thickness Cytochrome P450 1B1 Candela Confidence Interval x Food and Drug Administration Green Fluorescent Protein Humanized Green Fluorescent Protein Heparin Sulfate Proteoglycans Hematoxylin and Eosin Stain Institutional Animal Care and Use Committee Extracellular Matrix Electroretinography Fibrilin 1 Glucocorticoid Phenylalanine Hydrochloride ECM ERG F FBN1 FDA GC GFP hADAMTS10 Human ADAMTS10 HCl hGFP HSPGs HSS Hz H&E IACUC IC ICA IHC IM IOP IV IVit Iridocorneal Angle High Salt Solution Intramuscular Intracameral Intravenous Intravitreal Hertz Intra-ocular Pressure Immunohistochemistry xi JCT LA LCA LT LOX m mg/dL MIGS mm MMP ms MSU MTP MYOC N/A n/a NAb NC NE nm NO NTF4 Juxtacanalicular Tissue Light Adapted Leber Congenital Amaurosis Lens Thickness Lysyl Oxidase Meter Milligrams per Deciliter Micro-invasive Glaucoma Surgery Millimeters Matrix Metalloproteinase Milliseconds Michigan State University Micro Total Protein Myocilin Not Available Not Applicable Neutralizing AAV Antibody Not Collected Not Examined Nanometer Nitric Oxide Neurotrophin 4 xii OAG OCT ONH OPTN PI RBC RGC RNFL ROCK s SC scAAV smCBA ss TBK1 TM TM-ECM tx UBM un-tx UTM vg Open Angle Glaucoma Optical Coherence Tomography Optic Nerve Head Optineurin Post Injection Red Blood Cells Retinal Ganglion Cells Retinal Nerve Fiber Layer Rho Kinase Second Schlemm’s Canal Self-complimentary AAV Small Chicken β-actin Single-stranded Tank-binding Kinase 1 Trabecular Meshwork Trabecular Meshwork Extracellular Matrix Treated Ultrasound Biomicroscopy Untreated Uveoscleral Trabecular Meshwork Vector Genomes xiii WDR36 WMS Y µV µL ™ ® WD Repeat Domain 36 Weill-Marchesani Syndrome Tyrosine Microvolts Microliters Trademark Registered xiv CHAPTER 1 - Introduction 1.1 Glaucoma Classification Glaucoma is a leading cause of irreversible vision loss. This collective of optic neuropathies, globally, affects more than 70 million people. By 2020, it is estimated to surpass 76 million people. And, between 2013 and 2040, our global populous afflicted by glaucoma is expected to increase by 74%(1, 2). These optic neuropathies result in vision loss by the progressive death of retinal ganglion cells (RGCs)(2). RGCs are located in the inner retina, and their axons relay visual information in the form of action potentials via optic nerve head and optic nerve to the brain(3). Their death is multifactorial, encompassing heterogenous genetic and environmental variables(4) which include excitotoxicity caused by amino acid release of glutamate and aspartate(5, 6); neurotrophin deprivation from inhibition of retrograde axonal transport(7-10); excessive intracellular calcium(11); blood flow impedance to the ONH and retina(12-18); oxidative stress(19, 20); inflammation and autoimmunity against retinal and optic nerve antigens(21-23); and reactive gliosis(24-27). IOP is a major risk factor of glaucoma and the main focus of this thesis. It is regulated by the careful balance of aqueous humor (AH) production and drainage in the anterior segment of the eye. This normal regulation not only maintains the necessary constant IOP needed for functional shape and vision but is also responsible for exchange of crucial metabolites for the avascular tissues of the globe, such as cornea and lens(2, 28, 29). In hypertensive forms of 1 glaucoma an increased AH outflow resistance through the ICA results in accumulation of AH and increased IOP. Depending on the presence or absence of an identifiable underlying etiology, glaucomas are categorized as primary or secondary, the latter being caused by a detectable disease process that results in impairment of AH drainage. Characterization of the phenotypes is what categorizes the various types of primary and secondary glaucomas (Table 1.1). The two main forms of primary glaucoma are OAG, the topic of this thesis, and angle-closure glaucoma (ACG). Clinically, ACG is characterized by the obstruction of the AHOP by the iris against the ICA(2). 2 Table 1.1. Glaucoma Phenotypes Type Open angle glaucoma IOP; Drainage Angle Elevated; open Significant Clinical Presentation ONH cupping Angle closure glaucoma Elevated; closed or partially closed Regional or full obstruction of ICA; ONH cupping Risk Factors Age, family history, African heritage, use of corticosteroids, elevated IOPs(2) Age, female gender, first degree relatives with disease, Asian heritage, eyes with short axial length(2) Age, female gender, Asian descent(30) Sporadic but mostly inherited and varies by ethnicity(31) Pigment dispersion syndrome(32) Exfoliation syndrome(33) Inducing ocular injury and any secondary complications in the closed globe or inducing ocular injury and any complications from the open-globe and its subsequent repair which may include use of corticosteroids(34, 35) Various retinal vessel occlusions, diabetic retinopathy, various ocular tumors, various systemic diseases and retinopathies(36) ONH cupping; various corneal changes; buphthalmia ONH cupping; illuminated iris defects; Krukenberg’s spindle; uniformly heavy pigmentation of trabecular meshwork (TM) ONH cupping; classic pattern deposits of exfoliation on anterior lens Closed: decreased baseline vision, hyphema, angle reduction, traumatic lens incidences, iris afflictions Open: damage to globe and globe anatomy, peripheral anterior iris synechia, vitreous hemorrhage ONH cupping ONH cupping; neovascularization of ICA and iris; vascular retinal changes Normal; open ONH cupping Normal-tension glaucoma Congenital glaucoma Pigmentary glaucoma Pseudoexfoliative glaucoma Traumatic glaucoma Typically, elevated; delayed/failed development Elevated; pigmented deposits Elevated; deposits Elevated; varying degrees of recession or exacerbating damage Neovascular glaucoma Elevated; development of vasculature 3 Table 1.1. (cont’d) Type Iridocorneal endothelial syndrome IOP; Drainage Angle Elevated; appearance of open but obstructed by clear endothelium Uveitic glaucoma Very elevated; initial appearance is open however chronic inflammation can create collapse or obstruction Significant Clinical Presentation ONH cupping; changes in corneal endothelium and cornea; iris atrophy; peripheral anterior iris synechia Risk Factors Chandler syndrome, progressive iris atrophy, Cogan-Reese syndrome(37) ONH cupping; proteinaceous and cellular particulates present in anterior chamber; anterior and posterior iris synechia No known propensities; however, some systemic conditions can result in chronic uveitis which can produce glaucomatous damage(38, 39) Abbreviations: ONH, optic nerve head; ICA, iridocorneal angle; TM, trabecular meshwork 4 Unlike ACG, OAG is clinically identified with an open and fully accessible ICA(2, 40). The iris remains stationed in its respective location and does not contribute to the primary impedance of AH outflow. OAG-affected individuals may live long lengths of their lives unaware of their disease until there are noticeable changes in their visual fields. The pathogenesis of increased AH outflow resistance in OAG is not fully understood but appears to be associated with plaque formation(41) within the extracellular matrix (ECM) within the ICA. The following risk factors have been identified for OAG: older age, family history in first degree relatives, African descent, Hispanic descent, elevated IOP, thin cornea, low perfusion pressure, and use of corticosteroids(2, 28, 29). 1.2 Aqueous Humor Dynamics Aqueous humor is a clear fluid medium that fills the anterior and posterior chambers of the eye. It is produced by the ciliary body and released into the posterior chamber. It then moves through the pupillary aperture into the anterior chamber, and then exits the eye through two pathways: either the ICA/TM (conventional pathway) or uveoscleral pathway (unconventional pathway)(28, 29). Due to temperature gradients, the AH circulates in a convection current within the anterior chamber, washing all interfacing structures before exiting through the ICA(42, 43). In order to remain optically clear, the protein concentration is lower in the AH compared to the blood plasma. It also contains electrolytes/ions, ascorbic acid, and a multitude of other components that include growth factors, enzymes, prostaglandins, cyclic adenosine monophosphate, catecholamines, steroid hormones and hyaluronic acid(43, 44). As AH moves in its normal capacity, its composition will change as it passes interfacing structures: vitreous, 5 lens, iris and cornea. The AH is responsible for the maintenance of the avascular cornea and lens and their transparency by providing them with nutrition and oxygen, removal of waste from all interfacing surfaces, and aiding in the ocular immune response (28, 43, 44). 1.2.1 Aqueous Humor Production AH is produced by the ciliary body at the site of its species specific, uniquely shaped processes(45). Each of the processes has stroma that is double layered by epithelium, containing a rich supply of fenestrated capillary beds that are necessary for the generation of oncotic pressure. The tight junctions that affix the inner, non-pigmented epithelium to the outer, pigmented epithelium create part of the blood-aqueous barrier of the eye. AH is formed by (1) active secretion, (2) ultrafiltration and (3) simple diffusion(43, 44, 46). Active Secretion: The majority, about 80-90%, of AH is secreted by active transport. This energy demanding movement of fluid across the bilayer requires changes in the electrochemical gradient of electrolytes and the use of two valuable ion transporters: Na+:K+:2Cl- cotransporter and the Na+/K+ pump. Movement of anions and cations with these transporters creates an osmotic pressure gap across the epithelial bilayer that is then supportive of hydrostatic and oncotic movement of fluid from the ciliary body stroma to posterior chamber via aquaporins (cell membrane water channels). Additionally, fluid is also influenced to move due to changes in the sodium gradient that are determined by bicarbonate formation that is subsequently regulated by the reverse hydration of carbon dioxide by carbonic anhydrase(43, 46, 47). 6 Ultrafiltration: Hydrostatic pressure or osmotic gradients created by the differences between the capillary beds, the current IOP of the eye, and difference in solute will allow for natural and passive movement of fluid and solutes down their gradient(44, 48). Simple Diffusion: Some ions and pooled fluid move passively, or leak, across the ciliary epithelial membrane in the direction of their electrical or concentration gradients(44, 48). 1.2.2 Aqueous Humor Drainage Trabecular Meshwork/Conventional Pathway: AH predominantly exits the anterior chamber via the conventional pathway (Figure 1.1). Varying by species, between 50% and 75% of AH will filter through the three layers of the TM prior to valvular-like entry into a species- specific structure such as Schlemm’s Canal (SC), in mice and primates, or the angular aqueous plexus (AAP) in canines and other mammals. Distal drainage from these trabecular vessels occurs through the adjacent aqueous veins and episcleral vessels(28, 45, 49). As the site that regulates volume changes within the eye, a measure of its resistance (or compliance) can help to characterize the change between a healthy or diseased state. Compliance can be estimated by calculating outflow facility which is the inverse of the resistance of the conventional outflow pathway(50, 51). Uveoscleral/Unconventional Pathway: The alternative flow of AH can include the corneal, iridial and retinal route but is more commonly considered as egress through the posterior uveal meshwork (Figure 1.1). The uveal meshwork will move a bulk of this fraction of the AH while the alternative routes move negligible volumes. The estimated fraction of AH that departs the anterior chamber via the uveoscleral pathway varies greatly among species; e.g. 3- 7 82% in humans (52) and 15% in canines(53). The AH reaches the orbital vasculature through the interstitial spaces of the ciliary muscle, suprachoroidal space, sclera and episclera(52). Comparatively between humans and canines, structural differences of the ICA correlate to the accommodation of the species and not to the physiology of movement and drainage of AH(54). In carnivores, the ciliary body musculature is more developed and in canines the wide and deep ciliary cleft requires additional support for appropriate anchorage of the iris(55). Unlike humans, this additional support of the iris is compensated for by a circumferential series of fibrous strands called the pectinate ligament(55). Figure 1.1. Schematic drawing of aqueous humor drainage pathways. 8 1.3 Trabecular Meshwork The TM is a multifunctional, complex network of connective tissue that bridges the full 360° circumference of the ICA to its’ distal drainage anatomy; it is the main focus of this thesis. Functionally, the TM is a filtration network for the AH and works intimately with SC or the AAP to regulate the resistance of AH outflow for healthy maintenance of IOP. The TM consists of three regions: (1) uveal meshwork, (2) corneal scleral meshwork and (3) juxtacanalicular tissue (JCT). 1.3.1 Uveal and Corneal Scleral Meshwork The uveal meshwork is the most proximal layer of the TM and is structured as a cobweb of collagen and elastin lamellae beams sheathed in a thin continuous monolayer of TM cells and oriented radially with large open spaces, also known as Fontana’s spaces(48). The middle layer, or corneal scleral meshwork, is comprised of similar components, collagen and elastin, but arranged more tightly in a plate-like fashion with holes(48), and sheathed in TM cells of same continuous fashion. Both proximal layers of the meshwork act as the filtration unit of the eye, removing biological components such as large cellular debris and reactive oxygen species that have been washed away as waste components in the AH. The phagocytic activity of these two layers is critical to the final and most distal layer’s, known as the JCT, regulating function. Additionally, these first two layers of the TM also function as a site of antigen presentation and production of histocompatible proteins(29). 1.3.2 Juxtacanalicular Tissue The most distal and perhaps most OAG influencing layer of the TM is a loose layer of connective tissue that bridges the tighter plate-like network of the corneal scleral layer to the 9 bordering inner wall of SC or AAP. This loose layer contains TM cells that are sheathed in ECM. This loose bridging tissue and the inner wall of SC/AAP are what regulate the resistance of AH outflow. Like the corneal scleral meshwork, the JCT also has contractility features of smooth muscle actin and myosin. These elastic extensions of smooth muscle reach and terminate at the inner wall of SC/AAP. This feature gives the JCT and SC/AAP the ability to regulate AH outflow resistance by countering tension of the ciliary muscle. As ciliary muscle contracts, it will open up the TM’s network thus increasing outflow; when the TM contracts this will counter regulate the outflow by increasing resistance with smaller openings(29). 1.4 Trabecular Meshwork Extracellular Matrix (TM-ECM) The TM-ECM consists of a multi-protein, loose collagen and elastin fiber filled ground material that embeds the collagen and elastin beams of the TM offering structural support and attachment of TM cells to each other and to SC/AAP endothelial cells. The ECM is responsible for a complex and diverse array of functions and cellular communications(56, 57); and, its molecules have a signature profile that changes to meet the needs of the embedded cells based on their external challenges. These external challenges on the TM-ECM include mechanical forces (e.g., eye rubbing, ocular movements, changes in body position) as well as metabolic, phagocytic and UV radiation stress(29, 57). The ECM of the JCT and its intimate association to the channel forming inner wall of SC/AAP, is where most of the resistance to AH outflow occurs, and it is thought to be where dysregulation and stiffening develop with age and OAG(48, 58). This interface between the JCT and inner wall of SC/AAP is thought to be the site of increased AH outflow resistance in OAG, 10 and it is caused by increased deposition, stiffening, or thickening of sheath-derived plaques of the ECM, resulting in an increase in IOP(2, 43, 58, 59). 1.4.1 Mechanisms that Can Change the ECM Profile The dynamically evolving maintenance profile of the TM-ECM is necessary for preservation of normal outflow facility(57) of AH; however, some biological factors can alter the rate at which ECM regulates deposition and remodeling of itself. Hormones and cytokines: Transforming growth factor beta-2 (TGFβ-2) belongs to the superfamily of transforming growth factors. It is a profibrotic cytokine that can arise from the depletion of TM cells and may contribute to OAG pathogenesis by upregulating expression of elastin, collagen and fibronectin of the TM-ECM(29, 59, 60). TGFβ-2 and one of its isoforms, TGFβ-1, can also signal connective tissue growth factor (CTGF), which in turn will activate production of ECM proteins - fibronectin and various collagens – modifying the profile to become less elastic and permeable. Additionally, TGFβ-2 can change signaling of ECM crosslinking enzymes and genes associated to them such as all five lysyl oxidase (LOX) genes(29, 57). Non-endogenous glucocorticoids (GC), typically received by an individual in the form of a topical steroid therapy, are known to alter the metabolism and cytoskeletal ultrastructure of the ECM as well as gene expression of the cell. Most notably, Myocilin (MYOC ) is modified by GC, resulting in formation of cross-linked actin networks (CLANs) within the TM cells, thereby decreasing their elasticity and increasing their rigidity(29, 58, 61, 62). 11 1.5 Schlemm’s Canal/Angular Aqueous Plexus Conventional outflow of AH will cycle fluid back to the primary blood supply by species specific drainage vessels, SC or AAP. Dynamic movement from JCT to either of these trabecular veins can be either paracellular or more commonly transcellular(43) through vacuole-like structures, pores or pinocytotic vesicles, of the inner endothelium. From the SC/AAP, the AH flows via radial collector channels on to the intrascleral venous plexus back to systemic blood supply (45, 48). 1.6 Genetics of OAG In humans, genome wide association studies have found numerous susceptible loci that contribute to the development of OAG (Table 1.2). 12 Table 1.2. Genes and Susceptible Loci (40, 63, 64) Gene/Susceptible Loci Associated Biological Process Associated Human Population 8q22 ABCA1 AFAP1 ARHGEF12 ATXN2 Non-coding enhancer of both the ciliary body and choroid plexus Lipid membrane and cholesterol metabolism ECM modulator Lipid membrane and cholesterol metabolism Neurodegenerative functions CAV1/CAV2 CDKN2B-AS1 Development of caveolae invaginations Cell division regulation United States - European Chinese and Australian Australian European United States - European Iceland Australian and Japanese Multiple populations Multiple populations CYP1B1 FNDC3B FOXC1 GAS7 GMDS NTF4 Monooxygenase involved in phase 1 metabolism Cytokine signaling for function of ECM Coding for ocular development United States - Cell division regulation First step in the coalescence of GDP-fucose to GDP-mannose Survival and differentiation within the nervous system European United States - European Australian Not specified 13 Systemic Therapy that Suggestively Benefits Glaucoma Statin Therapy Statin Therapy Expressed Outside of the TM Yes Yes Not specified Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Table 1.2. (cont’d) Gene/Susceptible Loci Associated Biological Process Associated Human Population Japanese Cellular function trafficking and signaling regulation Marshals the reaction of mannose 6-phosphate to mannose 1-phosphate Coding for ocular development United States - Chinese OPTN PMM2 SIX6 TBK1 TGFBR3 TMCO1 TXNRD2 Phosphorylation of upstream signals for gene transcription TGFβ-3 cytokine signaling ECM remodeling Transmembrane calcium channel of the endoplasmic reticulum Thioredoxin reductase 2 which is responsible for the breakdown of the mitochondrial by product reactive oxygen species T-cell activation European Not specified Multiple populations Australian United States - European Not specified WDR36 Abbreviations: ECM, extracellular matrix Systemic Therapy that Suggestively Benefits Glaucoma Expressed Outside of the TM Yes Not specified Yes Yes Yes Yes Yes Yes 14 The genes that have a causative role in OAG in humans are MYOC, optineurin (OPTN), cytochrome P450 1B1 (CYP1B1), neurotrophin 4 (NTF4), Tank-binding kinase 1 (TBK1) and WD repeat domain 36 (WDR36)(2, 63, 65, 66). MYOC mutations were discovered in 1997 and account for 4% of adult onset OAG(67, 68). The gene’s normal function is elusive but the thought on how the mutations initiate and advance OAG is by accumulation of myocilin within the cell. Myocilin should naturally be secreted; however, the mutations related to the gene seem to keep the protein sequestered in the peroxisomes allowing for accruement of myocilin deposits that are thought to be cytotoxic. Cytotoxicity will lead to a depletion of TM cells and as they become less, outflow facility of AH(68) decreases. Additional studies with transgenic mice expressing the same MYOC mutations will facilitate the evaluation of the sequestered protein and the effects on healthy TM cells and ECM(68). OPTN is a gene that is thought to interact with other pathways that arbitrate several cellular functions such as apoptosis, inflammation and morphogenesis(69). It is still unknown what functional role OPTN plays in the initiation and development of OAG but it has been associated normal tension glaucoma(2, 69). A possible change in OPTN expression leading to OAG could be attributed to the aging eye. People with CYP1B1 mutations typically present with congenital or juvenile OAG. Like OPTN, the role of CYP1B1 is still ambiguous but is thought to be a functional factor during ocular development in enzyme metabolism of compounds. The polymorphisms associated with 15 this gene result in a loss-of-function and decrease in overall numbers of enzymes, subsequently leading to OAG(70). NTF4 belongs to a family of neurotrophin proteins and is the upstream regulator of tyrosine kinase receptor B (TrkB) phosphorylation. Normal maintenance of this signaling pathway is thought to have neuroprotective qualities, and its disruption caused by NTF4 mutations negatively affect neuron survival, resulting in OAG(71). Copy number variants that span the TBK1 gene are what generate a downstream dysregulation of intermediate kinases that activate protein complex NF-κB (nuclear factor kappa-light-chain-enhancer of activated B cells). This complex is responsible for numerous cell functions and cytokine responses, and disruption of its expression is believed to be another possible etiology of normal-tensive OAG(72). Multiple sources seem to concur that genetic multiple variants of WDR36 are associated with OAG. Still elusive in its overall contribution to OAG, many believe that these variants act as modifying components that contribute to RGC death by apoptosis(66, 73). Many susceptibility loci and genes can contribute to an OAG phenotype in humans. Animal models of OAG, offer the ability to study disease etiology and contributing pathophysiology while also facilitating the development of novel therapies that can prevent or offset the risk factors that lead to RGC death. 1.7 Animal Models of OAG Table 1.3, summarizes a wide variety of animal models that have helped in the study of disease mechanism and therapeutic advancements in OAG. 16 Table 1.3. Animal Models of OAG (74-77) Species Mechanistic Mode Non- human primate Dog Cat Mouse Rat Natural occurring inheritance TM scarring by laser photocoagulation IC injection of microbeads IC injection of autologous erythrocytes Autosomal recessive inheritance (natural/spontaneous model) Autosomal recessive inheritance (natural/spontaneous model) Transgenic (many) Corticosteroid induced IC injection of microbeads Limbus scarring by laser photocoagulation Corticosteroid induced IC injection of microbeads Sclerosing scleral vessels TM scarring by laser photocoagulation Cauterization of episcleral veins Rabbit Zebrafish Genetic mutation Transgenic (many) Corticosteroid induced TM blockage – by rupturing zonular fibers with an intravitreal injection of α- chymotrypsin(80) IC injection of microbeads TM scarring by laser photocoagulation Steroid induced Steroid induced Light induced Sheep Cow Avian Abbreviations: IC, intracameral; TM, trabecular meshwork Gene or Protein or Colony Associated Cayo Santiago colony, based in Puerto Rico ADAMTS10 mutations(65) LTBP2 mutation(78) Human MYOC point mutation Mutations in collagen type I OPTN overexpression Nonsense mutation lrp2-/-(79) WDR36 17 Animal models are an indispensable tool in biomedical research. While all the models listed (Table 1.3) have their own relevance, the dogs with ADAMTS10-OAG offers a predictive and spontaneous model, and are the focus of this thesis. These dogs develop characteristic ECM changes in the AH outflow pathways that are similar to the changes observed in human OAG (65), therefore representing a relevant large animal model. 1.7.1 A Disintegrin and Metalloproteinase with Thrombospondin Motifs (ADAMTS) – Canine OAG Models ADAMTS genes represent a super family of metalloproteinases that have varied roles across the body and are heavily involved in the structure, function and integrity of ECM supramolecular networks(81, 82). These secreted proteinases’ genetic framework contains a pro-domain, catalytic domain, and an ancillary domain that are involved in regulating the enzyme’s activity associated with ECM composition. Notably, ADAMTS10 and ADAMTS17 that are homonymous to respective genes, inhibit appropriate interactions with fibrillin-1 (FBN1) and produce abnormal microfibrils when they are mutated. In humans, ADAMTS10 mutations result in Weill-Marchesani syndrome (WMS), an autosomal recessive disease. The WMS phenotype consists of short-stature, stunted digits, stiff joints, narrowing of cardiac valves, various lens anomalies including luxation, and glaucoma(82-84). ADAMTS17 mutations also result in a lack of proper microfibril formation and result in WMS-like syndrome. This WMS-like phenotype shares short-stature and ocular irregularities, like lens luxation and glaucoma with WMS but lacks the cardiac features, stunted digits and joint stiffness. The shared features are suggestive of mechanistic overlap(81, 85). 18 Several breeds of dogs also have mutations within the ADAMTS10 and ADAMTS17 genes, but only a partial phenotypic profile is shared with humans(65), (Table 1.4). And while it is unknown if the dogs feature other sub-clinical attributes (such as body size, joint stiffness or narrowing of cardiac valves) that are clinically expressed in their human counterparts, they typically present within normal limits of health. However, the microfibrillopathies that induce similar clinical ocular anomalies of OAG (ADAMTS10 mutations in Beagles and Norwegian Elkhound and ADAMTS17 mutations in some breeds, such as PBGV) and lens luxation (ADAMTS17 in many breeds, including small terriers) across the many breeds are well documented(65, 86-88). 19 Table 1.4. ADAMTS10 Mutation Comparisons(88-92) Species Inheritance Mutation Domain Location Phenotype Human Autosomal Recessive Dog Autosomal Recessive Mouse Induced model Null mutations: non-sense (R237X); splice mutation (1190+1G→A); splice mutation (810+1G→A) Loss of function: missense mutation (73G>A); nonsense mutation (952C>T) Missense mutation: G661R (Beagle) Missense mutation: 1441G>A (Norwegian Elkhound) Null allele: Frame-shift mutation in exon 5 Null mutations: metalloprotease domain Loss of function mutations: 73G>A, peptidase cleavage site; 952C>T, catalytic domain Beagle mutation: cysteine -rich domain Norwegian Elkhound mutation: metalloprotease domain Null allele: catalytic domain Abbreviations: ECM, extracellular matrix; OAG, open-angle glaucoma 20 Short stature; short digits; stiff joints; microspherophakia; ectopia lentis; myopia; glaucoma; occasional cardiac anomalies Expression Outside of the Eye Chondrocytes (accounting for short stature and digits); heart (attributing to cardiac defects); skin fibroblast with abnormal extrafibrillar space and increase bundle size of microfilaments (suggests ECM impairment) Requires continued analysis Beagle mutation: primary OAG Norwegian Elkhound mutation: secondary lens luxation and primary OAG Decreased body weight compared to wild-type mice of the same litters; decrease in anterior chamber depth after 2 years of life; subtle increase in alveolarization of alveolar ducts; stiff skin Limbs: cartilage; bone; skeletal muscle; tendons and skin Thoracic and abdominal parenchyma: lungs and vessels of the heart and liver ADAMTS10-OAG in Beagles is caused by a missense mutation at amino acid 661 of ADAMTS10, where the glycine residue has been substituted for an arginine (G661R)(88). This missense mutation results in sheath derived plaques in the JCT of the TM where it interfaces the AAP(65, 88, 93). These plaque formations create an increase in resistance to AH outflow that subsequently generates an increase in IOP, contributing to the development of glaucoma(93). In Beagles with this mutation, seemingly only OAG is a shared phenotype with humans. It is yet to be known if the other phenotypes listed for humans (Table 1.4) are shared by the dogs on a subclinical level. A well-established colony of Beagles with this deleterious mutation, rendering them a clinically relevant form of OAG is housed at Michigan State University (MSU). This ADAMTS10-mutant dog model is the focus species of this thesis as it shares the same molecular obstruction of AHOPs as humans. The advantageous connection of a highly expressed gene in the TM of both dogs and humans(88), poises the canine eye as a significantly relevant large animal OAG model to test the concept of TM-targeted gene therapy. Advantages of canine ocular disease models: Dogs share many anatomical features of the eye with humans. Outside of a few evolutionary accommodations between the two species, both have eyes that are very similar in size, anatomy, and physiology. In recent years, the canine eye has been used many times as a translatable model for application of drug therapies and diagnostic tools for lateral use in humans(94-100), including AAV-mediated ocular gene therapy, thus supporting the use of this dog model in our study. 1.8 Preservation of Vision by Therapeutic Lowering of IOP Preservation of vision and the overall comfort and quality of life are the goals in glaucoma therapeutics. Current therapies for all species are limited to lowering IOP, a major 21 risk factor of glaucoma, and comprise of topical medicines, surgical procedures, or a combination of the two. IOP is lowered by either increasing outflow of AH or by decreasing its production(101). 1.8.1 Current Therapies Topical medications are a predominant method for both humans and canines in reduction of IOP (Table 1.5). The most effective IOP reducing drug class are prostaglandin analogs, and while the full mechanism of action is elusive, they are known to induce remodeling of the ECM by upregulation of matrix metalloproteinases (MMPs) resulting in an increase in uveoscleral outflow of AH (101). Second-line topical drug classes are β-blockers, alpha2- agonists, and carbonic anhydrase inhibitors. Beta blockers are thought to cause vasoconstriction in the ciliary body, reducing blood supply and therefore decreasing AH production. Alpha2-agonists work similarly in decreasing AH production; however, they additionally can modulate uveoscleral outflow by altering expression of MMPs. Carbonic anhydrase inhibitors also decrease AH production by inhibiting formation of bicarbonate and reducing active secretion. Third-line topicals are parasympathomimetic drugs and are utilized less commonly due to the frequency of needed application within a 24-hour period. These topicals induce miosis via iris sphincter muscle contraction thus allowing for increased conventional outflow(101, 102). 22 Table 1.5. Traditional Topical Medication Synopsis(101-103) Drug Class First Line Second Line Third Line Mechanism of Action Prostaglandin analogs: MMP remodeling of ECM; relaxation of ciliary muscle Β-blockers: Adrenoceptors β1 and β2 are blocked; ciliary body also experiences vasoconstriction Carbonic anhydrase inhibitors: Hinders formation of bicarbonate Alpha2-agonists: (rare in use) Alpha2 adrenergic receptors are blocked while prostaglandin expression is increased Parasympatho- mimetics: Inducing miosis through contraction of iris sphincter muscle Target Side Effects Drug Example Outflow Conjunctival hyperemia; increase in eyelash length and pigmentation; increased darkening of iris; increased pigmentation of periocular skin; Latanoprost; travoprost; bimatoprost; tafluprost; unoprostone Production Bradycardia; exacerbation of heart block with possible cardiac rhythm effects; hypotension; increased intolerance to exercise; depression of central nervous system Timolol; carteolol; betaxolol; levobunolol; metipranolol Production Blurred vision; fatigue/weakness; bitter/metallic taste and possible associated nausea and vomiting; punctate keratopathy Dorzolamide; brinzolamide; acetazolamide; methazolamide Production and outflow Outflow Apraclonidine; brimonidine Pilocarpine; echothiophate iodide Generalized ocular irritation which can include aching, photopsia/blurring of vision, pruritis and eyelid edema; ocular ischemia and blanching of conjunctiva; vasovagal syncope and hypotension; fatigue; depression and anxiety; headache, dizziness and dry mouth Posterior iris synechiae; cataract development; increased risk for angle closure glaucoma; increased retinal sensitivity along with increased risk for retinal tears and detachments; possible changes to color vision; punctal stenosis and epiphora; increased gastric secretions with abdominal cramps Abbreviations: MMP, matrix metalloproteinases; ECM, extracellular matrix 23 Recently emerged topicals on the market include nitric oxide (NO) donors and rho kinase (ROCK) inhibitors. NO donors are one of the first topicals that can have vast effects on the conventional outflow pathway by relaxing and shrinking the cells of the TM(102). The moiety they offer to smooth muscle regulators can also help improve the permeability and dilation of SC(104, 105). ROCK inhibitors also target the conventional outflow pathway by decreasing the rigidity and relaxing TM contraction, thereby opening Fontana’s spaces. ROCK inhibitors also reduce ECM synthesis and increase SC permeability resulting in a greater frequency of giant vacuoles as well as widening of the vessel itself(101, 102, 106). Traditional and new classes of topicals can be used in combination to achieve a more effective, multi- parameter approach to IOP reduction. A multitude of surgical procedures and implants are being used separately or in combination with medical therapies to lower IOP. Just like topical medicine approaches, the ultimate goal of the listed procedures is to alter the production or outflow of AH(43, 107). These procedures include the placement of AH drainage devices, trabeculectomy, laser trabeculoplasty, lens extraction and cyclophotocoagulation (the partial destruction of the ciliary epithelium for reduction of AH production). Recent advances in glaucoma therapy for more effective IOP control include micro-invasive glaucoma surgery (MIGS) and continuous, sustained drug delivery by ocular implants. Despite all these advances, vision loss continues to progress in many glaucoma patients. Non-compliance or nonadherence to prescribed therapies and surgical complications can exacerbate the rate of vision loss(108). A more effective IOP control could be achieved by newer treatments that specifically target underlying disease mechanisms. 24 1.9 Adeno-Associated Virus (AAV) for Targeted Gene Therapy Gene therapy is a delivery tool that offers much to the idea of preventative or potentially curative approach to ocular defects. It is a technique that utilizes genetic code to prevent and potentially treat heritable diseases. This therapeutic approach offers the ability to replace or inactivate mutated genes or even introduce a new gene that can combat disease(109). Targeting of the TM with gene therapy that can replace genetic defects causing elevated IOPs could be considered for long-term control of IOP by maintaining open and patent AHOPs. In the early infancy of TM targeted gene therapy there were several investigations carried-out to determine the appropriate viral vector tool approach that could safely and efficaciously express transgene long-term. Early attempts in targeting the TM, with adenoviruses were successful in species such as rodents, canines and non-human primates; however, their ability to maintain expression was limited to just a few weeks or months(110) which is not sufficient for the duration of effect for a chronic disease like glaucoma. Additionally, lentiviruses (FIV and HIV-1) had been used and successfully expressed in explanted and perfused human trabecular meshwork for 16 days(111) and as long as 10 months when delivered in vivo to cats (lentivirus, FIV)(112). While adenovirus has had some short-term success and lentivirus proving good long- term success both can come with adverse effects. Adenovirus is associated with generating a robust innate immune response(113) and lentivirus integrates into the genome which risks changes in expression of adjacent genetic code(114). AAV is a nonenveloped parvovirus with low immunogenicity and when stripped of its rep-binding genetic code, loses its ability to 25 integrate into the host genome. Twelve serotypes and several hybrid combinations of these serotypes have been identified over the decades of research, all allowing for unique benefits in cellular tropism and immune response(115). By the mid-90’s, AAV had breakthrough discoveries in its ability to mediate long-term transgene expression(116) making it a viable option for long-term targeted TM gene therapy. This non-pathogenic, safe and efficient delivery tool targets many cell types including in the eye, and even quiescent, post-mitotic cells(117-119). AAV notably, has had a lot of success in safety and efficacy in both pre-clinical and clinical research targeting the retinal disease, RPE65-Leber congenital amaurosis (LCA)(94, 120-124). In October of 2017, the Food and Drug Administration’s (FDA), Tissue and Gene Therapy Advisory Committee had a unanimous vote to recommend approval of the single-dose treatment of Luxturna, an AAV2 mediated gene therapy targeting RPE65-LCA(125). On December 19th, 2017, the FDA released to the press its groundbreaking approval of Luxturna, which is the first ever gene therapy approved to target and treat an inherited genetic mutation(126). While there has been some grand success in retinal approaches with AAV, the initial attempts of using this vector tool to target the TM showed no success in transducing the target tissue across multiple species and human tissue culture(110). Interestingly, pioneering attempts in 2010 by Buie et al., challenged the use of AAV again by approaching with a strategy to transduce living TM cells by utilizing self-complimentary AAV (scAAV). Buie et al. hypothesized that the reason for AAV’s failure in past attempts to transduce the TM was because of a rate limiting step at the point of double stranded DNA synthesis. This limitation was thought to be the confounding factor for successful transgene expression. When utilizing scAAV in rats and monkeys this team was able to feature the ability to target multiple 26 structures in the front of the eye including the TM(127). Additionally, this team also used scAAV to deliver MMP1, a protein thought to be downregulated in steroid responsive individuals, in a steroid induced sheep model of glaucoma and successfully reduced IOP for at least 3 weeks(128). The ability of AAV capsids to transduce cells depends on their ability to attach to glycans in the vicinity. Serotype AAV2 can bind to heparin sulfate proteoglycans (HSPGs) which are receptors that are very plentiful and integral in the ECM of the TM(129, 130). This serotype was used in both experiments aforementioned by Buie et al and Borras et al. for successful targeting of the TM(127, 128). The limiting factor of utilizing scAAVs is the reduced cargo capacity of the cassette (131), thus decreasing the size of the transgene within the cassette by half. Recognizing this limitation, Wang et al. compared the ability of scAAV2 to target both cornea and TM in vitro and in vivo (mice) with single-stranded (ss) AAV2, AAV8, and the synthetic, capsid mutated AAV Anc80L65. While traditional AAVs were still unsuccessful in targeting the TM the ss synthetic AAV proved capable of overcoming the rate limiting step of second stranded synthesis(92), previously hypothesized by other teams. Point mutations of specific amino acids on the icosahedral structure of the AAV capsid can alter the affinity of AAV to a specific cell’s external profile, allow for evasive maneuvering for avoidance of the proteasome while in the cytoplasm, and adjust capsid stability for appropriate breakdown once in the nucleus(132, 133). Bogner et al. used these capsid mutations in varying degrees to improve transduction efficiency of traditional AAVs to the TM(134). Given AAV2’s natural affinity to HSPGs and previous success in transduction of the retina and TM, it along with AAV8 (an AAV that lacks the receptor affinity to HSPG but has 27 targeting capability to other anterior eye structures where HSPGs are present) were used as parent serotypes to test different mutant capsid structures. These mutations were acquired by modifying surface exposed amino acid tyrosine (Y) to phenylalanine (F). Utilizing rodent cohorts, these scAAVs and varying mutations (Table 1.6) were equipped with reporter gene, humanized green fluorescent protein (hGFP) and injected IC. Cohorts that utilized scAAV2- (Y444F)-hGFP and scAAV2-(Y444+500+730F)-hGFP had the most success in transducing the AHOPs(134). These two successful vector constructs from the Bogner et al. team, were then moved forward into higher order species (canine) studies with wild type and mutant ADAMTS10-OAG Beagles(135). Given the size of the ADAMTS10 gene, the full cargo capacity of AAV is required to deliver the entire sequence of the replacement gene. Therefore, ssAAV was utilized to deliver both reporter gene (hGFP) and therapeutic trials of human ADAMTS10 (hADAMTS10). Of the vector constructs trialed for TM targeting capabilities, only one construct was successful. IC injections of ssAAV2-(Y444F)-hGFP in both wild type and ADAMTS10-OAG Beagles (at doses 108-1010 vector genomes (vg)) successfully targeted transgene expression to the TM (135). However, during therapeutic trials of this single mutated capsid construct, ssAAV2-(Y444F)- hADAMTS10 with doses similar to the ssAAV2-(Y444F)-hGFP trials (1010 vg delivered IC) did not result in a therapeutic IOP lowering effect(136). It was hypothesized that a therapeutic response could potentially be achieved with a dose escalation of the single mutant capsid construct. 28 1.10 Purpose The purpose of our study was to further develop the single capsid mutant gene therapy vector ssAAV2-(Y444F) for treatment of OAG by (Aim 1) optimizing the targeting of transgene expression to the canine AHOPs, including the TM, and (Aim 2) modifying gene expression by gene replacement therapy within the AHOPs in order to achieve a therapeutic effect in a clinically relevant and well-established canine model of ADAMTS10-OAG. 29 Table 1.6. Capsid Mutations Investigated by Bogner et al.(134) scAAV Number of Mutated Site of Amino Acid Gene Full Construct 2 Surface Exposed Amino Acids 1 3 7 1 444 444, 500, 730 252, 272, 444, 500, 700, 704, 730 733 hGFP scAAV2-(Y444F)-hGFP scAAV2-(Y444+500+730F)-hGFP scAAV2-(Y252+272+444+500+700+704+730F)-hGFP scAAV8-(Y733F)-hGFP 8 Abbreviations: hGFP, human green fluorescent protein 30 CHAPTER 2 – Materials and Methods 2.1 Study Design 2.1.1 Aim 1 – Targeting the AH outflow pathway – hGFP Because of our previously failed attempts to achieve an IOP-lowering, therapeutic effect(136), the goal of Aim 1 (targeting) was to improve transgene expression within the ICA by increasing vector dose and optimizing administration techniques, most importantly intracameral (IC) vs. intravitreal (IVit) injection and variation of pupil size with IC injection. Utilizing ssAAV2-(Y444F)-hGFP, we injected either IC or IVit, 14 eyes across 9 normal dogs (Table 2.1). These dogs were then clinically followed for 5-20 weeks. The following techniques were used to closely monitor safety: slit lamp biomicroscopy, indirect ophthalmoscopy and gonioscopy. Immediately after the 5-20 week time-frame, eyes were collected and evaluated by immunohistochemistry (IHC) for hGFP reporter gene expression. Additionally, to optimize our IC injection techniques, we injected 10 eyes across 5 normal, healthy dogs with fluorescein solutions and immediately monitored distribution within the anterior chamber. 2.1.2 Aim 2 – Therapeutic delivery - ssAAV2-(Y444F)-hADAMTS10 Applying the optimal injection technique developed under Aim1, three ADAMTS10- mutant Beagles with early stages of ADAMTS10-OAG were treated unilaterally by IC injections of ssAAV2-(Y444F)-hADAMTS10, leaving the fellow eye as an untreated control. The dogs were clinically followed for 62-105 weeks. The main outcome measured was weekly diurnal IOP by rebound tonometry. The following additional techniques were used to closely monitor efficacy and safety: slit lamp biomicroscopy, indirect ophthalmoscopy, gonioscopy, pneumotonography (AH outflow facility), fluorophotometry (AH flow), high-resolution imaging of the anterior and 31 posterior segments by optical coherence tomography (OCT) and ultrasound biomicroscopy (UBM), A-scan ultrasonography (globe size), pachymetry (central corneal thickness, CCT), and electroretinography (ERG; retinal function). AH and serum samples were collected throughout the post-injection period to measure protein concentration, cell counts and neutralizing AAV antibody (NAb) titers. 2.2 Animals All dogs (Table 2.1) were group-housed in the MSU College of Veterinary Medicine’s Vivarium with a 12-hour light-dark cycle. This facility is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care International (AAALAC). All performed procedures were approved by the MSU Institutional Animal Care and Use Committee (IACUC) and done in accordance with the Association for Research in Vision and Ophthalmology (ARVO) statement on the use for Animals in Ophthalmic and Vision Research. Purpose-bred, normal dogs (n=14) aged 4.9-33 months (Marshall BioResources, North Rose, NY, USA) and ADAMTS10-mutant beagle dogs (n=3) aged 21-25.5 months old were used in this study. All MSU-bred dogs from the ADAMTS10-mutant line were genotyped to determine the genotype regarding the G661R ADAMTS10, autosomal recessive, missense mutation(137). 32 Table 2.1. Aim 1 and 2 Study Participants Dog ID Gender Status Age at injection (months) 5.1 4.9 4.9 5.1 6.5 6.5 6.5 6.5 6.5 11.5 22 33 33 16 21 25.5 25.5 Vector/Study Assignment Eye treated ssAAV2-(Y444F)-hGFP 100mg/mL Fluorescein 100mg/mL Fluorescein 0.2mg/mL Fluorescein 0.2mg/mL Fluorescein 0.2mg/mL Fluorescein ssAAV2-(Y444F)-hADAMTS10 OD OD OS OS OU OU OU OU OU OU OU OU OU OU OS OS OD Normal M Ja55 M Ja76 M Es12 M St80 M Na25 F Me66 F Wi43 F Cl24 F Su01 M Ba84 M Ma32 F Tu63 F La98 Ha57 M Ep00/Dog 1 M Em40/Dog 2 M El42/Dog 3 M Abbreviations: OD, right eye. OS, left eye. OU, both eyes. ADAMTS10- mutant 2.3 Anesthesia Protocols Most clinical evaluation techniques, such as routine ophthalmic examination, tonometry, pachymetry, gonioscopy and UBM required no anesthesia. Pneumotonography and fluorophotometry required only light sedation with a single intravenous (IV) injection of 0.3mg/kg of butorphanol tartrate (Henry Schein Animal Health, Dublin, OH, USA) and 0.3mg/kg of midazolam HCl (Hospira, Inc., Lake Forest, IL, USA). Dogs received either an oral dose of 4mg of ondansetron (Heritage Pharmaceuticals Inc., Eatontown, NJ, USA) up to one hour prior to delivery of sedation or an IV dose of 0.25mg/kg ondansetron within 5 minutes of sedation for relief of any nausea that may have been associated with sedation. If additional sedation was 33 needed for either procedure, butorphanol tartrate was re-administered at a 0.1mg/kg dose and never exceeding a total amount of 0.5mg/kg. General anesthesia was utilized to immobilize the dogs for longer periods of time and the need to control eye positioning. For ERG and OCT, dogs were premedicated with acepromazine maleate IV at 0.02mg/kg (Butler Schein Animal Health, Dublin, OH, USA), followed by aseptic placement of a peripheral IV catheter (either cephalic or saphenous vein), IV induction with propofol (PropoFlo™28, Abbott Laboratories, North Chicago, IL, USA) at 4mg/kg or to effect and maintenance with inhalant isoflurane (Akorn Inc., Lake Forest, IL, USA) in O2. A-scan measures were performed at the end of OCT under general anesthesia. For longer procedures like ERG and OCT, dogs were given IV replacement fluids (Normosol-R, Hospira Inc., Lake Forest, IL, USA) at 5-10mls/kg/hour as needed based on preliminary physical examination findings (hydration status was assessed by capillary refill time, tackiness of gums, skin turgor and other general overall appearances such as a dry nose) and intra-procedure blood pressure measurements. For intraocular injections or aqueous paracentesis, dogs were premedicated with buprenorphine hydrochloride (HCl) (Par Pharmaceutical, Chestnut Ridge, NY, USA) at 0.015mg/kg and acepromazine maleate at 0.02mg/kg either IV or intramuscularly (IM), followed by aseptic peripheral catheter placement, IV induction and maintenance with propofol. Dogs that participated in a short study to optimize IC injection technique with fluorescein were premedicated as previously noted for IC injections, followed by aseptic placement of an IV catheter and then delivered induction and maintenance anesthesia as described for ERG and OCT. 34 2.4 Clinical Examination For safety assessment and measure of any adverse effect, all animals received baseline and regularly scheduled ocular examination by a boarded veterinary ophthalmologist. Immediately post IC injections, examinations were performed daily for up to one week, followed by weekly for up to one year, then followed monthly for the remaining duration of follow-up examinations. Each exam consisted of slit-lamp biomicroscopy, indirect ophthalmoscopy and rebound tonometry (TONOVET, iCare). CCT was measured monthly by pachymetry (PachPen, Accutome Inc.). Gonioscopy (RetCam II, Clarity) was performed monthly. Lastly, UBM (UBM Plus ®, 48MHz, Accutome Inc.) was performed at least once over the duration of the study for assessment of safety. 2.4.1 Diurnal IOP – primary outcome measure All ssAAV2-(Y444F)-hADAMTS10 treated dogs had diurnal IOPs measured by rebound tonometry once a week. Consistently on the same day each week, at the same morning (8am), noon (12pm) and early evening (4pm) times, we evaluated both treated and untreated eyes for variable ranges and up to 105 weeks. Each eye was measured for a total of 6 single values at each of the three time points. Single values were calculated by the iCare TONOVET following 6 rebound measures; the device excludes the 2 most extreme outliers and averages the remaining 4 measures. Daily averages for each eye were calculated from the 18 single values. These daily values were compared between treated and non-treated eyes in a generalized linear model and estimating equation. At completion, statistical analysis included a fitted trendline of these average daily IOPs over time. For this, values were obtained using the non- parametric LOESS curve fitting method(138) and corresponding confidence intervals (CI) 35 determined with Bonferroni correction of multiple comparisons, set at 1-α=0.99. Testing for significant changes in IOP was done by calculating the difference in daily average IOPs between the treated and the untreated eye. From the difference, a fitted trendline was again obtained with corresponding CIs. Calculated differences that were considered significant, when α≤0.01, were for values that had CIs not containing the number zero (values would have to be less or greater than zero). 2.4.2 Pachymetry Once monthly CCT was measured by pachymetry. Each eye was locally anesthetized with proparacaine HCl 0.5% ophthalmic solution (Akorn Inc., Lake Forest, IL, USA), and the pachymeter placed perpendicularly on the central cornea. The instrument measured 9 individual readings, discarding the most extreme outlier and averaging the remaining 8 for a single value. Percent variation of CCT was calculated for the duration of the study as ((highest reading across study timeline – lowest reading across study timeline)/lowest reading across study timeline)x100(139, 140). 2.4.3 Gonioscopy Monthly gonioscopy was performed(141) to ensure that the ICA remained open and does not show any signs of AAV-associated adverse effects. Proparacaine HCl 0.5% ophthalmic solution was applied for ocular surface anesthesia, followed by ultrasound gel (Sonotech Clear Image, Medium Viscosity, A Division of Illinois Tool Works Inc., Bellingham, WA, USA) for optimal contact between the RetCam lens and the cornea for viewing of the ICA. Inferior, temporal, superior and nasal quadrants were assessed separately. 36 2.4.4 Ultrasound biomicroscopy (UBM) Assessment of the ciliary cleft and surrounding structures was done by high-resolution, cross-sectional UBM(142). Dogs rarely needed sedation with 0.02mg/kg of acepromazine maleate IV. Ocular surface anesthesia was applied with topical proparacaine HCl 0.5% ophthalmic solution prior to application of ultrasound gel and covered probe tip emerged in sterile water. 2.5 AAV Production The two recombinant AAV2 vectors, ssAAV2-(Y444F)-GFP and ssAAV2-(Y444F)- hADAMTS10, were designed and manufactured as previously described by Dr. Shannon Boye’s laboratory at the University of Florida(143, 144). The surface-exposed capsid amino acid 444 was modified from a tyrosine (Y) to a phenylalanine (F) on the AAV helper plasmid containing AAV2 “Cap”. Because no TM-specific promoter was available to drive gene expression, vectors were assembled with a ubiquitous small chicken β-actin promoter (smCBA). The small size of the promoter permitted fitting of the larger hADAMTS10, cDNA. Generation of vectors was done by plasmid co-transfection in HEK293T cells. Further purification and concentration of nuclear and cytoplasmic fractions was accomplished with iodixanol (Sigma-Aldrich, St. Louis, MO, USA) gradient centrifugation and ion exchange column chromatography (HiTrap Sp Hp 5 mL, GE Healthcare Bio-Sciences, Piscataway, NJ, USA). Final vector titers were established by real-time PCR and purity was met by use of silver-stained sodium dodecyl sulfate- polyacrylamide gel electrophoresis. Final aliquots of ssAAV2-(Y444F)-hGFP were resuspended in sterile balanced salt solution (BSS, Alcon Laboratories, Fort Worth, TX, USA) containing 0.014% Tween 20; ssAAV2-(Y444F)-hADAMTS10 was resuspended in sterile BSS/0.014% Tween 37 supplemented with NaCl, we termed high salt solution (HSS), to increase the ionic strength of the suspension so to reduce self-aggregation of AAV2(145). BSS/0.014% Tween 20 suspension has a concentration established at ~300mOsm, when supplemented with NaCl the concentration will increase to ~600mOsm. The ~3.3 kb hADAMTS10 amino acid code for the therapeutic vector, was synthesized according to Genbank accession number NM_030957, with an addition of a consensus ‘Kozak’ sequence as well as a silent G to A change at nucleotide 3150. This resulted in a full mRNA of about 4.3 kb, including 5’ and 3’ untranslated regions. Due to the functionally efficient cargo capacity of AAV (4.1-4.5 kb) (119), untranslated regions were dropped to achieve a more efficacious fit. The substitution of G to A resulted in a loss to internal restriction enzyme Not I. Following a Not I/Sal I digest, the synthesized cDNA sequence was finally cloned into the AAV vector plasmid containing the smCBA promoter. 2.6 Intraocular injections – IC and IVit 2.6.1 ssAAV2-(Y444F)-hGFP and ssAAV2-(Y444F)-hADAMTS10 The AAV vectors varied in genetic contents (hGFP vs. hADAMTS10 transgene), total virus particles and volume delivered (Table 2.2). As described under study design, treatments with AAV were staged over two aims 1) targeting and 2) therapeutics. Because of our previously failed attempts to achieve an IOP-lowering, therapeutic effect(136), the goal of Aim 1 (targeting) was to improve transgene expression within the ICA by increasing vector dose and optimizing administration techniques, most importantly IC vs. IVit injection and variation of pupil size with IC injection. Normal dogs that were used for Aim 1 had either one randomly selected eye or both eyes injected (Table 2.2) with ssAAV2-(Y444F)-hGFP. Additionally, to 38 optimize our IC injection techniques, we injected, in vivo, 10 eyes of 5 normal, healthy dogs with fluorescein solutions and immediately monitored distribution within the anterior chamber (see 2.6.2 Optimizing Injection Technique). Applying the optimal injection technique developed under Aim1, 3, ADAMTS10-mutant Beagles were selected for Aim 2 (therapeutics) with 1 eye per dog randomly chosen for ssAAV2- (Y444F)-hADAMTS10, leaving the fellow eye as an untreated control (Table 2.2). 39 Table 2.2. hGFP and hADAMTS10 Dosing Assignments Dog ID Ja76 M Eye treated Dose - Vector Genomes (vg) 1x109vg Gender Vector/Study Assignment ssAAV2-(Y444F)-hGFP OD St80 M ssAAV2-(Y444F)-hGFP OS Ja55 M ssAAV2-(Y444F)-hGFP OD Es12 M ssAAV2-(Y444F)-hGFP OS 1x1010vg 1x1011vg 1x1011vg Na25 M ssAAV2-(Y444F)-hGFP OD 2.78x1012vg ssAAV2-(Y444F)-hGFP OS 2.78x1012vg Me66 F ssAAV2-(Y444F)-hGFP OD 2.78x1012vg ssAAV2-(Y444F)-hGFP OS 2.78x1012vg Wi43 F ssAAV2-(Y444F)-hGFP OD 2.78x1012vg ssAAV2-(Y444F)-hGFP OS 2.78x1012vg Cl24 F ssAAV2-(Y444F)-hGFP OD 2.78x1012vg ssAAV2-(Y444F)-hGFP OS 2.78x1012vg 40 Vehicle BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween BSS/0.014% Tween Volume 50µL Route Pupil Size IC 50µL 50µL 50µL 50µL 50µL 50µL 50µL IC IC IC IC IC IC IC mydriatic miotic 50µL IVit 50µL IVit 50µL 50µL IC IC mydriatic miotic Gender Vector/Study Table 2.2. (cont’d) Dog ID Su01 F Assignment ssAAV2-(Y444F)-hGFP OD Eye treated Dose - Vector Genomes (vg) 5.56x1012vg ssAAV2-(Y444F)-hGFP OS 5.56x1012vg Ep00 M Em40 M El42 M ssAAV2-(Y444F)- hADAMTS10 ssAAV2-(Y444F)- hADAMTS10 ssAAV2-(Y444F)- hADAMTS10 OS OS OD 1.25x1012vg 1.43x1012vg 1.43x1012vg Vehicle BSS/0.014% Tween BSS/0.014% Tween HSS HSS HSS Volume 100µL Route Pupil Size IC 100µL IC 75µL 75µL 75µL IC IC IC Abbreviations: OD, right eye; OS, left eye; BSS, balanced salt solution; HSS, BSS high salt solution 41 Pre- and post-operative standard of care protocols with ophthalmic medications were given bilaterally to all dogs, so not to bias IOP readings. In order to prevent post-injection ocular hypertension following the IC or IVit injection of reagent without the need of an additional aqueous paracentesis, the eyes were pretreated with the IOP-lowering glaucoma medication dorzolamide hydrochloride-timolol maleate (Bausch & Lomb Inc., Tampa, FL, USA). Pre-injection treatment also included the topical non-steroidal anti-inflammatory drug flurbiprofen sodium 0.03% (Bausch & Lomb Inc., Tampa, FL, USA), the topical steroid prednisone acetate 1% (Pacific Pharma, Irvine, CA, USA) and the systemic steroid prednisolone administered orally and tapered over 3 weeks post-injection (Henry Schein Animal Health, LLOYD Inc., Shenandoah, IA, USA; 1 mg/kg, twice daily for 7 days, then once daily for 8 days, then once every other day for 6 days). For eyes that enrolled in Aim 1 requiring alteration of pupillary aperture, were either pre-treated with topical latanoprost 0.005% ophthalmic solution (Bausch & Lomb Inc., Tampa, FL, USA) to induce miosis or topical tropicamide 1% ophthalmic solution (Akorn, Inc., Lake Forest, IL, USA) to induce mydriasis. Treatment with the systemic antibiotic amoxicillin/clavulanic acid (Clavamox®, GlaxoSmithKline, Zoetis Inc., Kalamazoo, MI, USA; orally, at 14-22mg/kg, twice daily) was started before the injection and continued for 7 days to minimize the risk of bacterial infections. Anesthesia protocols were followed as previously described (see 2.3 Anesthesia Protocols). Once a surgical plane of anesthesia was achieved, dogs were placed in sternal recumbency. While procedures were very short, body temperature was thermo-regulated by a warm water circulating blanket (T/Pump®, Gaymar Industries Inc., Orchard Park, NY, USA). Patient parameters (heart rate, oxygen saturation, end-tidal CO2, respiratory rate and blood 42 pressure) were assessed by hand and a patient monitor (PM-9000Vet, Shenzhen Mindray Bio- Medical Electronics Co., Ltd., Nanshan, Shenzhen, China). Ocular surface and periocular skin of the treatment eye was prepped aseptically with 0.2% dilute povidone-iodine solution (Poviderm™ Solution, Henry Schein Animal Health, Dublin, OH, USA) then topically anesthetized with proparacaine HCl 0.5% ophthalmic solution. Conjunctival stay sutures (4-0 silk; Ethicon Inc., San Lorenzo, Puerto Rico, USA) were placed near the superior and inferior limbus at the 12 and 6 o’clock positions to assist with stability and gaze of the eye. Prepared vector was delivered via insulin syringe (30 gauge x 1/2 inch needle; BD Insulin Syringes with BD Ultra- Fine™ needle, BD, Franklin Lakes, NJ, USA) by IC injection at the superior-temporal limbus. The scleral-limbal margin was visualized with use of a binocular loupe (EyeMag Pro, Carl Zeiss Inc., Oberkochen, Germany) with illumination. The anterior chamber was entered in an obliqued angle to the iris surface by tunneling the needle over several millimeters through the conjunctiva, sclera, and cornea, starting posterior to the limbus. Once the 50-75 µL vector solution was delivered, the needle remained in place for up to 10 minutes to reduce leakage of any AH containing virus particles on exit. Upon removal of the needle, the surrounding conjunctiva was held closed by 0.12mm Castroviejo forceps for 2 minutes in order to further prevent leakage. IVit injections followed a similar protocol; however, the injection entry was premeasured to 6 mm posterior of the dorsolateral limbus at the location of the pars plana of the ciliary body. Post-operatively, a 4-mg subconjunctival injection of triamcinolone acetonide suspension was administered (Kenalog®-40, Bristol-Myers Squibb Company, Italy) followed by neomycin-polymyxin B sulfates--dexamethasone 0.1% ophthalmic ointment (Bausch & Lomb Inc., Tampa, FL, USA). Post injection treatment was continued with twice daily topical 43 application of the latter ointment for 21 days and then tapered to once daily for 5 days. For dogs participating in Aim 2, an initial attempt to discontinue the topical steroid dexamethasone resulted in minor anterior uveitis in the AAV-injected eyes; hence, topical steroid therapy was continued in both eyes with twice-daily administration of 0.05% difluprednate ophthalmic emulsion (Durezol®; Alcon Laboratories Inc., Fort Worth, TX, USA) or neomycin-polymyxin B sulfates-dexamethasone 0.1% ophthalmic ointment (Bausch & Lomb Inc., Tampa, FL, USA) for the duration of study. In order to minimize the development of steroid keratopathy/corneal degeneration, the twice daily topical steroid treatment was combined with 0.2% topical cyclosporine ointment (Optimmune®, Intervet International B.V., Merk Animal Health, France). 2.6.2 Optimizing Injection Technique Between Aims 1 and 2 a short fluorescein kinetics of the anterior chamber study was performed to assess the full circumferential reach of the ICA following a one-time IC injection. This short study was performed in two stages to assess the following parameters: body and globe position, temperature of reagent, and vehicle suspending reagent (Table 2.3). All normal dogs that participated followed the previously described anesthetic protocol for the short fluorescein kinetics study (see 2.3 Anesthesia Protocols) as well as the pre-operative ophthalmic drop protocols as described for dogs participating in Aims 1 and 2 (see subsection 2.6.1 ssAAV2- (Y444F)-hGFP and ssAAV2-(Y444F)-hADAMTS10). The post-operative care protocol for this study was less intense than following AAV injections since no protein or virus was administered. The dogs received a one-time subcutaneous 4.4-mg/kg injection of non-steroidal anti- inflammatory drug carprofen (Rimadyl®, Zoetis LLC., Lincoln, NE, USA)) and a one-time topical application of neomycin-polymyxin B sulfates-dexamethasone 0.1% ophthalmic ointment. The 44 fluorescein kinetics study procedures mimicked the following parameters of Aims 1 and 2: aseptic preparation of the ocular surface and periocular skin, visualization and delivery approach, IC delivery of reagent, volume delivered and needle track closure. All chamber distributions were monitored immediately upon injection and followed every two minutes for 10 minutes. Image acquisition at each injection and consecutive time points was by use of a modified (full spectrum) dSLR camera, dSLR camera adaptor, and camera lens. The camera for digital acquisition used was a Canon 7D (Canon, Tokyo, Japan) with lens (Canon EF 85 mm f/1.4L USM lens). For conducting angiography, excitation and barrier filters were inserted within the illumination and optical pathways of the adaptor, respectively. Camera settings used included a shutter speed of 1/100 second, effective aperture of f/8, and a sensitivity (International Standards Organization) setting of 200. The excitation filter was MF479/40 nm (Thorlabs, Newton, NJ, USA) with peak transmission centered at 479 nm with a width of 40nm and barrier MF525/39 nm (Thorlabs, Newton, NJ, USA) with peak transmission centered at 525 nm and width of 39 nm. Two dogs participated in STAGE ONE, which focused on body position and globe stability. Both eyes of these dog were IC injected with 50 µL of 10% fluorescein (Fluorescite; Alcon Inc., Fort Worth, TX). The first dog (Ma32, Table 2.3) in our standard IC injection technique where the head is placed in an up-right position, similar to Aim 1 and 2. The second dog (Ba84, Table 2.3) was first placed in dorsal recumbency with head positioning so that the iris surface was in a horizontal plane; this is consistent with the positioning for cataract surgery/phacoemulsification. The first eye was maintained in forward gaze position by conjunctival stay sutures (5-0 Ethilon, Ethicon Inc., San Lorenzo, Puerto Rico, USA). Once the 45 injection and imaging of the first eye was completed, the dog was positioned in lateral recumbency and the second eye was stabilized by a retrobulbar saline injection (0.9% Sodium Chloride, Hospira Inc., Lake Forest, IL, USA). Immediately after injection of the second eye, the dog was rotated from left to right lateral positions every 1-2 minutes for the duration of 10 minutes. Three dogs participated in STAGE TWO, focusing on vehicle and temperature of the preparation. Each dog was randomly assigned to a preparation vehicle that was either BSS, BSS containing 0.014% Tween 20, or HSS. Vehicles were used to dilute fluorescein to a 0.2mg/mL concentration and each vehicle dilution had a 37°C (body temperature) and cold temperature (~4°C on ice, standard temperature of AAV in vehicle due to survivability of vector through transport and storage) preparation. Standard protocol for IC injections described under Aims 1 and 2 were followed. The right eyes of all dogs received a 37°C, 50-µL IC injection of their assigned vehicle with fluorescein dilution while all fellow left eyes received a 50-µL IC injection of the cold preparation of that same vehicle dilution (Table 2.3). Images taken in this study were assessed qualitatively as a function of time to determine how well/uniform the injected reagent distributes within the anterior chamber. 46 Table 2.3. Fluorescein Kinetics of the Anterior Chamber Assignments Stage Dog Eye Fluorescein ID Ma32 OD 100mg/mL Concentration 1 2 Vehicle; Volume n/a; 50µL °C Body Position 37°C Standard IC injection technique OS 100mg/mL n/a; 50µL 37°C Standard IC injection technique Ba84 OD 100mg/mL n/a; 50µL 37°C Left lateral recumbency, rotating OS 100mg/mL n/a; 50µL 37°C Dorsal recumbency; head position, perpendicular from body lateral recumbency every 1-2 minutes for 10 minutes Tu63 OD 0.2mg/mL BSS; 50µL 37°C Standard IC injection technique OS 0.2mg/mL BSS; 50µL 4°C Standard IC injection technique La98 OD 0.2mg/mL OS 0.2mg/mL Ha57 OD 0.2mg/mL BSS/0.014% Tween; 50µL BSS/0.014% Tween; 50µL HSS; 50µL 37°C Standard IC injection technique 4°C Standard IC injection technique 37°C Standard IC injection technique OS 0.2mg/mL HSS; 50µL 4°C Standard IC injection technique Globe Stability 5-O Ethilon stay sutures 5-O Ethilon stay sutures Retrobulbar injection of saline 5-O Ethilon stay sutures; eye gaze to ceiling 4-O Silk stay sutures 4-O Silk stay sutures 4-O Silk stay sutures 4-O Silk stay sutures 4-O Silk stay sutures 4-O Silk stay sutures Pupil Size Atropine mydriasis Atropine mydriasis Atropine mydriasis Atropine mydriasis Table 2.3. 37°C, room temperature. 4°C, cold preparation. Atropine sulfate 1% ophthalmic solution (Akorn, Inc., Lake Forest, IL, USA) 1 drop OU was given post-injection for one-time preventative treatment of uveitis. Abbreviations: n/a, not applicable 47 2.7 Aqueous Humor Outflow Facility - Pneumotonography In order to determine if any IOP-lowering therapeutic effect was based on an increase in conventional AH outflow, outflow facility was measured in the three dogs of Aim 2 by use of pneumotonography (Reichert Technologies, Reichert Inc., Depew, NY, USA)(146). Dogs were sedated as previously described (see 2.3 Anesthesia Protocols) followed by application of one drop of 0.5% proparacaine HCl ophthalmic solution to each eye for ocular surface anesthesia. With use of a calibrated pneumotonograph, a seated and supine IOP was measured for one eye. The dog would remain in a supine position by gentle restraint, while a 10-gram weighted sensor head placed on the cornea recorded the IOP continuously for a duration of 2 or 4 minutes. The fellow eyes followed the same order of events and were measured immediately after the first eye. The ocular surface moisture/hydration was maintained by repeated topical administration of BSS as necessary. The print outs of the tonography recordings (change in IOP over 2 or 4 minutes) were scanned into CurveSnap software for windows (Softonic, Softonic International S.A., Edificio MediaTIC, Barcelona, Spain) for analysis of the digitized IOP scans by Excel (Microsoft® Office Excel, Windows 10, Redmond, WA, USA). Starting (time 0) and ending IOPs (time 2 or 4 minutes) read from the digitized plot, along with the seated and supine IOPs captured at the time of the recording, were then entered into the pneumotonograph to calculate outflow facility. Because the pneumotonograph was not available for baseline measurements, AH outflow facility values were compared between treated and untreated eyes in each dog. 48 2.8 Aqueous Humor Flow - Fluorophotometry AH flow was measured by fluorophotometry (Fluorotron Master II, Ocumetrics, Inc., Mountain View, CA) in Aim 2 in order to determine the effect of AAV gene therapy on AH turnover. Performance of scans were done as previously described by others (147-150). A 20µL drop of 10% fluorescein (AK-FLUOR® 10%, Akorn, Inc., Lake Forest, IL, USA) was applied to the corneal surface of each eye followed 5 minutes later by a thorough flush with sterile eye wash (Ophthalmic solution eyewash, Akorn, Inc., Lake Forest, IL, USA). Four hours elapsed to allow establishment of equilibrium of fluorescein concentrations within the cornea and anterior chamber. Dogs were sedated as previously described (see 2.3 Anesthesia Protocols) and placed into sternal recumbency. They were then gently held in place in front of the optic head of the Fluorotron instrument for fluorophotometric scans. Duplicate scans were performed and averaged at four time points in 45-minute intervals for each eye. Because the fluorophotometry was not available for baseline measurements, AH flow was compared between treated and untreated eyes in each dog. 2.9 Optical Coherence Tomography (OCT) and A-scan Biometry High-resolution imaging by OCT (Heidelberg Spectralis, Heidelberg Engineering Inc., Heidelberg, Germany) of cornea, retina, and ONH was performed to assess any adverse events and natural progression of glaucoma in dogs of Aim 2. IOPs were measured by rebound tonometry followed by application of one drop of tropicamide 1% ophthalmic solution for pupil dilation in order to optimize visualization of the ocular fundus. Dogs were anesthetized as previously described (see 2.3 Anesthesia Protocols) and placed in sternal recumbency with their heads stabilized on an adjustable platform (headrest) to a custom-designed examination table. 49 Rebound tonometry of each eye was repeated to verify that mydriasis did not result in any egregious IOP spikes. The ocular surface of one eye was then anesthetized with 0.5% proparacaine HCl ophthalmic solution, and positioned with conjunctival stay sutures as previously described with 4-0 Silk, while the fellow eye was held closed by medical tape to prevent drying of the cornea. The cornea of the imaged eyes were kept moist by repeated application of BSS sterile irrigating solution (Alcon ®, Alcon Laboratories, Fort Worth, TX, USA). Once imaging was complete, A-scan biometry (A-scan; Scanmate, DGH Technology) was performed for evaluation of total axial length, anterior chamber depth, lens thickness, and vitreous chamber depth. Subsequently, the fellow eye was examined and once complete, both eyes were treated with latanoprost 0.005% ophthalmic solution to reverse the pupil dilation and any potentially associated adverse IOP spikes. ONH cross sections were compared subjectively between treated and contralateral untreated eyes. 2.10 Electroretinography (ERG) Photopic full-field ERGs (RETI-port, Roland Consult) were recorded for assessment of retinal function loss, specifically photopic negative responses (PhNR) representing RGC function in the 3 dogs of Aim 2 (151, 152). As described for OCT, all dogs had rebound tonometry performed before and after application of tropicamide 1% ophthalmic solution for pupil dilation. All dogs were anesthetized, placed in sternal recumbency with their head stabilized and eyes held into a forward gaze with conjunctival stay sutures as previously described (see 2.9 Optical Coherence Tomography (OCT) and A-scan Biometry). ERG-Jet® corneal contact lens electrodes (Fabrinal SA, Switzerland) were filled with hypromellose 2.5% ophthalmic demulcent solution (Goniotaire, ALTAIRE Pharmaceuticals, Inc., Aquebogue, NY, USA) and placed on the 50 cornea as active electrodes. Subdermal platinum needle electrodes (GRASS Technologies, West Warwick, RI, USA) were used as reference and ground electrodes (subdermally placed 3-5 cm posterior to the lateral canthus and top of the head, respectively). Eyes were light stimulated and recorded simultaneously (Table 2.4). Light adaptation to background luminance was 5 minutes for each wavelength. Once testing was completed, mydriasis was reversed with 0.005% latanoprost ophthalmic solution. The following ERG amplitudes were compared between treated and untreated eyes in each dog: a- and b-wave, and PhNR(153, 154). Table 2.4. Photopic ERG Protocol Stimulus of Light Luminance Over Time (cd s m-2) White: (1.0) White: (3.0) Red: (1.0); (625 nm) Red: (3.0); (625 nm) Blue: (1.0); (470 nm) Background Luminance (cd m-2) White: (25) Blue: (25); (470 nm) Amber: (25); (590 nm) Amplification and Filtration Bandpass (Hz) 1-300 Abbreviations: ERG, electroretinography; cd, candela; s, second; m, meter; Hz, Hertz; nm, nanometer 2.11 Aqueous Paracentesis and Aqueous Humor Analyses AH samples, totaling a volume of 100-200µL per eye, were collected for protein concentration, cell counts, and Nab titers. Repeated aqueous paracentesis was performed in the dogs of Aim 2 while under general anesthesia following ERG or OCT recordings. The preparation of the eyes and paracentesis technique were the same as for the IC injections (see 2.6 Intraocular Injections – IC and IVit) and were also performed with an insulin syringe (30 gauge x 1/2 inch needle). Post procedure care included a 4.4 mg/kg subcutaneous injection of carprofen (Rimadyl®, Zoetis Inc., Kalamazoo, MI, USA) and a one-time application of neomycin- polymyxin B sulfates-dexamethasone 0.1% ophthalmic ointment. Protein concentration and 51 cell counts were obtained by the MSU Veterinary Diagnostic Laboratory, and AAV NAb titers by the Boye Laboratory at the University of Florida. Along with AH collection, whole blood was collected via venipuncture of the jugular or cephalic veins, and aliquots of the serum were measured for AAV NAb titers at the Boye Laboratory, University of Florida. Samples were tested at 1:10, 1:40, 1:160, 1:640, and 1:2,560 dilutions using scAAV2-(Y444F)-smCBA-mCherry and ARPE19 cells as previously described(155). 2.12 Tissue Collection for Histopathology and Immunohistochemistry All dogs from Aim 1 were humanely euthanized in a method consistent with the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals. They were pre-medicated with 0.03mg/kg of acepromazine maleate IV, followed by a peripheral placement of an IV catheter and IV overdose administration of sodium pentobarbital (Fatal Plus, Vortech Pharmaceuticals, Dearborn, MI, USA). Eyes were enucleated and fixed in 4% paraformaldehyde solution for 3 hours at 4°C. Anterior and posterior segments were then separated at the pars plana of the ciliary body and fixed in 2% paraformaldehyde for 21 hours. Tissues were incubated in 15% and 30% sucrose each for 24 hours, respectively. The bisected anterior and posterior segments were cut into four quadrants and embedded in OCT media (Tissue-Tek OCT, Sakura Finetek USA Inc., Torrence, CA, USA). The frozen blocks were stored at -80°C until further processing. Ten µm sections were made by Leica CM3050S cryostat (Leica Microsystems, Buffalo Grove, IL, USA) at the vertical and horizontal sagittal planes (Figure 2.1) and mounted on plus-charge slides (Brain Research Laboratories, Newton, MA, USA). After slides air dried they were then stored at -20°C 52 until stained routinely with hematoxylin and eosin (H&E)(156). The slides were reviewed, and the degree of inflammation graded by a boarded veterinary pathologist as follows: 0=no inflammation, 1=mild inflammation, and 2=marked inflammation. The population of inflammatory cells was characterized. Sections of all areas were also incubated with hGFP antibody (rabbit FITC conjugated, 1:1000, Invitrogen Corp., Carlsbad, CA, USA) for IHC as previously described(157). Each slide was cover-slipped with an anti-fading reagent with DAPI for nuclear stain (DNA binding fluorescent stain (4′,6-diamidino-2-phenylindole, ProLong® Gold, Life Technologies)). Slide imaging was performed with a fluorescence microscope (Nikon Eclipse 80i, Nikon instruments Inc., Melville, NY, USA) at 4x, 10x, 40x and 100x magnification and images captured with software (MetaVue, Molecular Devices, Sunnydale, CA, USA). Sections of hGFP IHC were qualitatively assessed for their labeling intensity. One representative low-mid grade slide image was taken through a 4x objective from each anterior segment quadrant and area of %hGFP green fluorescence quantitatively assessed using ImageJ (National Institute of Health, USA)(158). For statistical comparison of quadrants across varying dosages, the quantitative area of fluorescence was normalized by calculating each quadrant’s % of the sum of the 4 quadrants within an eye. Fluorescence was compared between quadrants by the use of the function Imp from the package ImPerm in R(159). Permutation based ANOVA calculated from this program included the effects of: quadrant, eye, and the rearrangement of quadrants within the eyes. 53 Figure 2.1. Anterior segment embedding and sectioning map. Vertical and horizontal sagittal planes are the borders establishing the 4 quadrants of each eye. Cryo-sectioning started at vertical and horizontal sagittal planes. Red arrows indicate the cutting direction for the 10µm sections into each quadrant of the eye. On this map, corneas are facing down. 54 CHAPTER 3 - Results 3.1 Aim 1 - Effect of Dosing and Injection Route on TM Transgene Expression In previous studies, we have shown the ability to successfully target gene expression to the AHOPs of normal and ADAMTS10-mutant dogs with ssAAV2-(Y444F)-hGFP(135). Subsequently, we treated ADAMTS10-mutant dogs with early stages of hypertensive OAG with the corresponding therapeutic vector ssAAV2-(Y444F)-hADAMTS10 at a dose of 1-3x1011vg. We showed that the treatment was safe and resulted in long-term hADAMTS10 transgene expression for at least 35 months; however, no therapeutic IOP lowering effect could be achieved(136). In our attempt to improve transgene expression and to achieve a therapeutic effect, we pursued the following strategies under Aim 1: 1) increase the AAV delivery dose by one log unit; 2) compare IC vs. IVit AAV administration; and 3) test the effect of pupillary aperture on AAV vector distribution with an IC injection. 3.1.1 Increasing the AAV Dose When assessing the ICA hGFP transgene expression across a range of 4 log units, 1x109- 2.78 and 5.56x1012 vg, following IC administration of ssAAV2-(Y444F)-hGFP, the increase from 1x1011 to 2.78-5.56x1012 vg resulted in an observable increase at 5 weeks post injection, which could be observed both clinically, by fluorescent gonioscopy, and by IHC (Figure 3.1). Because of the small sample sizes, no statistical analyses were performed. Despite the non-specific smCBA promoter, we confirmed our previous results of strongest transgene expression along the AHOPs, with some off-target transgene expression on the anterior iris surface and the iris epithelium on the posterior surface. Because there is no TM-specific IHC marker, TM cells were identified by location and shape. We observed that within the ICA, hGFP expression was not 55 limited to the TM, but was also strong along the uveoscleral outflow pathway, suggesting that a comparable therapeutic vector construct will likely treat both conventional and non- conventional pathways. It appeared that non-TM cells along the AH outflow pathways were also targeted. No hGFP expression was observed within the cornea, including the corneal endothelium, which minimizes concerns about potential corneal toxicity. Figure 3.1. Effective TM targeting with ssAAV2-(Y444F)-hGFP in dogs. A1 and A2 are gonioscopic images of the same canine ICA (RetCam II). Note in A1, with use of 488nm blue light, fluorescence within the ciliary cleft just beyond the landmark pectinate ligaments. A2, depicting the white light image of the same landmark pectinate ligaments. B, hGFP expression of the corneoscleral TM (CSTM), uveoscleral TM (UTM) and continuing posteriorly along the uveoscleral outflow pathway (arrow). Bar=50µm. 56 3.1.2 Comparing IC vs IVit injections In an attempt to slow the passage of ssAAV2-(Y444F)-hGFP through the ICA and maximize the exposure of bioavailable virus particles to the TM, we injected the vector IVit, trialing the vitreous as a slow release reservoir. When comparing IC vs. IVit injection of ssAAV2- (Y444F)-hGFP at 2.78-5.56x1012vg/50µL, we observed comparable transgene expression (hGFP fluorescence) within the ICA (Figure 3.2). Because of this observation, we continued with the IC injection approach in all subsequent treatments. Not surprisingly, the IVit administration of ssAAV2-(Y444F)-hGFP not only resulted in hGFP fluorescence along the AHOPs, but also the anterior and posterior iris surface, the ciliary body epithelium, and the retina (see subsection 3.1.4 Other Observations following IC and IVit ssAAV2-(Y444F)-hGFP; Figure 3.4). 57 Figure 3.2. Relative % area of IHC hGFP labeling within the ICA when comparing doses and routes of administration. (Graph modified with permission(158)) Overall a one log-unit increase by IC injection from 1E11 to 2.78E12 vg resulted in a clear increase of hGFP expression. IC and IVit injection of the same vector dose (2.78E12 vg) resulted in comparable hGFP expression within the ICA. Each data point/blue mark represents area of hGFP fluorescence within one quadrant of the ICA. For IC injections: 1E9, N=1 eye (4 quadrants); 1E10, N=1 eye (4 quadrants); 1E11, N=2 eyes (8 quadrants); 2.78E12, N=2 eyes (8 quadrants); 5.56E12, N=2 eyes (8 quadrants). For IVit injections: 2.78E12, N=2 eyes (8 quadrants). 3.1.3 Effect on Pupillary Aperture To determine the effects of pupil size on transgene expression, we compared relative % area of IHC hGFP labeling results from IC injections of ssAAV2-(Y444F)-hGFP at 2.78x1012vg in 58 50µL with un-altered, miotic, and mydriatic pupils. We hypothesized that a small pupil could increase vector transduction within the ICA by (1) minimizing posterior loss of vector through the pupil, and (2) maximizing TM exposure to vector within a more open angle that is under tension from the miotic pupil. On the other hand, a dilated pupil and associated crowed ICA may slow AH drainage and increase contact time between vector and target cells. Miosis was induced with one drop of 0.005% latanoprost ophthalmic solution and mydriasis with one drop of 1% atropine sulfate ophthalmic solution. While median area of IHC hGFP fluorescence may suggest that expression was higher in miotic pupils, there was no obvious difference in hGFP expression over all quadrants evaluated when comparing the three pupil sizes (Figure 3.3). 59 Figure 3.3. Effect on relative % area of IHC hGFP labeling within the ICA when comparing pupil size when utilizing the 2.78E12vg dose. A (graph modified with permission(158)), hGFP labeling with an overall similar % representation across 3 different pupil statuses: mydriatic, miotic and un-altered. Each data point/blue mark represents area of hGFP fluorescence within one quadrant of the ICA. B, hGFP IHC with green fluorescence shows robust hGFP labeling of the TM (arrows), regardless of pupil sizes. For pupil aperture status: mydriatic, N=2 eye (8 quadrants); miotic, N=2 eye (8 quadrants); un-altered, N=2 eyes (8 quadrants). Blue DAPI, 4’6- diamidino-2-phenylondole (DAPI), utilized for labeling of nuclear DNA. 3.1.4 Other Observations following IC and IVit ssAAV2-(Y444F)-hGFP While the ICA was our primary target for transgene expression, hGFP was expressed also in other ocular cells (not graded or scored, only observed findings reported). Following vector injection into the anterior vitreous (via an IVit injection), hGFP expression was found anteriorly in the ciliary body epithelium, iris epithelium and anterior iris surface as well as posteriorly scattered across the retina, spanning portions of the nerve fiber layer and reaching down into cells of the retinal ganglion cell and inner nuclear layers (Figure 3.4). IC injections resulted in 60 stronger hGFP expression in cells surrounding the posterior chamber with increasing pupil size. In contrast, a miotic pupil resulted in hGFP expression of the anterior iris surface and to a lesser extent, pupillary margin and iris epithelium following IC vector administration. Unaltered pupils during IC injection, resulted in hGFP transgene expression of the of the iris surface, pupillary margin and iris epithelium. Mydriatic pupils during IC injection resulted in targeting of similar structures/cells as with an unaltered pupil, but with further posterior diffusion, more robust hGFP expression within the iris and ciliary body epithelium (Figure 3.4). Collectively, all IC injections, regardless of pupil size resulted in no transgene expression posterior to the ciliary body. 61 Figure 3.4. hGFP expression in the TM and other intraocular tissues when comparing IC injections with variable pupil sizes and IVit injection. All routes of administration resulted in transgene expression within the TM; IC routes more robustly than IVit. Pupil size and route of administration seemed to correlate with the ability to target other cell populations. Yellow arrows point to the iris epithelium. The fluorescence of the iris epithelium covering the posterior iris surface seemed more robust, more distantly from the pupillary margin with increased pupil size. Iris surface expression (white arrowheads) with an IVit approach seemed paltry. The ciliary body epithelium was reached by IVit vector administration or IC administration with mydriasis (yellow double asterisks). IVit vector administration also targeted the retina (white double asterisks) spanning the nerve fiber layer and even a few cells that reach into the inner nuclear layer. Blue DAPI was used to stain nuclear DNA. ONL, outer nuclear layer. INL, inner nuclear layer. 62 3.1.5 Comparison of hGFP Expression Between Quadrants To determine if the ICA is targeted uniformly through the whole circumference, we compared areas of hGFP expression between the four quadrants following IC injection of ssAAV2-(Y444F)-hGFP. Dosing differences were corrected as described in Methods (see 2.12 Tissue Collection for Histopathology and Immunohistochemistry). In general, hGFP was uniformly expressed when comparing the four quadrants (Figure 3.5). Dorsolateral quadrants seemingly trended toward higher hGFP expression while ventrolateral quadrants trended towards less; however, with use of permutation ANOVA, there was no significant differences in expression among the quadrants. 63 Figure 3.5. Comparison of % IHC hGFP expression between four quadrants following IC delivery of ssAAV2-(Y444F)-hGFP. Data sets were normalized so that total percentage of hGFP expression added up to 100%. This was done to account for dosing differences (1x109 to 2.78x1012vg). Each of the 14 eyes, of 9 dogs, is shown in a different color. 3.1.6 Fluorescein Kinetics of the AC To confirm that our IC injection technique of a small volume (50μL) results in even filling of the AC, a prerequisite for even circumferential targeting of the ICA, we injected 10 eyes of 5 normal dogs with 0.2mg/ml and 100mg/ml fluorescein in the same vehicles that were used for AAV administration. We explored different variables that could possibly effect AC filling in order to identify the best technique for future vector administration. Head and body position, globe stability, composition and temperature of vehicle were evaluated (Table 2.3 in Methods). 64 When injecting full 100mg/ml concentration of fluorescein, which was not diluted with vehicle, the reagent seemed buoyant in the AH and would pool within the AC. For the dog positioned upright, this was the dorsal aspect of the chamber and for the dog in dorsal recumbency, pooling occurred behind the central cornea. When a dog was placed in an initial position of lateral recumbency, flipping from side to side (left lateral to right lateral recumbency), pooling occurred at the temporal aspect of the AC that was the opposite side of the finishing recumbency that the dog was placed. All other protocols resulted in a uniform distribution within 1-253 seconds (Figure 3.6), with reagent diluted in HSS distributing the fastest (<1 second) and reagent diluted in BSS/0.014% Tween being the slowest (253 seconds). All eyes exhibited swirling of the reagent within the AH following the initial motion of injection. In 4 eyes/2 dogs, topical atropine was applied following fluorescein injection for treatment of the mild anterior uveitis associated with any IC injection. In these eyes we observed fluorescence of the anterior and posterior lens capsule 1 day post-injection; this fluorescence gradually faded and migrated into the lens cortex over several weeks (Figure 3.7). This observation confirmed our findings with ssAAV2-(Y444F)-hGFP that a mydriatic pupil permits diffusion of IC injected reagent posteriorly and allowing contact with the cells and tissues bordering the posterior chamber, including the lens (Figure 3.7). When pupils were not pharmacologically dilated, we were able to observe the slow anterior diffusion of clear AH from the posterior to the anterior chamber through the pupil in a convection-like manner (Figure 3.8), suggesting that the reagent is not diffusing posteriorly. 65 Figure 3.6. AC fluorescein kinetics as a function of temperature and vehicle. 6 eyes of 3 dogs, positioned in upright recumbency, were IC injected with 50μL of three different vehicles (BSS, BSS/Tween, and High Salt Solution (HSS)) at two different temperatures (37°C and 4°C). White arrows point to the site of injection. Figure 3.7. Lens capsule fluorescence shown with blue slit beam. Lens capsule exposure to fluorescein reagent due to a dilated pupil. Over an 8-week observation, fluorescein faded and diffused into lens cortex and nucleus. 66 Figure 3.8. Anterior diffusion of AH through the pupillary aperture followed by convection- like movement within the AC. Clear AH can be seen diffusing through the pupil (white bracket) into the AC (yellow arrows). A, 37°C, BSS vehicle. B, 37°C, BSS/Tween vehicle. 3.1.7 Summary of Aim 1 – Targeting the AH outflow pathway – ssAAV2-(Y444F)-hGFP For robust transgene expression in the ICA dosing matters: escalation of the dose by an additional log unit, from 1011 to 1012 vg, resulted in greater transgene expression of the AHOPs which became detectable, clinically, by gonioscopy. IVit injection did not result in stronger ICA transgene expression when compared to IC injection, but resulted in additional transgene expression in the retina, ciliary body and iris epithelium. Pupil size also did not appear to affect amount of ICA transgene expression within the ICA; however, a mydriatic pupil resulted in ‘loss’ of vector posteriorly through the pupil with targeting of cells and tissues bordering the posterior chamber. These findings are summarized in Table 3.1. 67 Table 3.1. Summary of Target Tissues Identified in Aim 1 ICA Anterior Iris + + + + Injection site/pupil IC/miotic IC/normal IC/mydriatic IVit/normal Abbreviations: NE, not examined. +, positive target. -, negative target. (+), light or sparingly positive target. Ciliary Body epithelium - - + + surface + + + (+) Lens (capsule) NE - + NE Iris epithelium (+) + + (+) Retina - - - + In conclusion of Aim 1 and the fluorescein kinetics of the AC, we found that in the adult canine eye, the IC injection of 50-75µL reagent in either BSS/0.014% Tween or HSS at either 4°C or 37°C resulted in even filling of the anterior chamber with a subsequent uniformly targeting of the ICA following delivery if 1012 vg. 3.2 Aim 2 – Treatment of Canine ADAMTS10-OAG – ssAAV2-(Y444F)-hADAMTS10 Aim 1 studies have allowed us to optimize our ability to successfully target gene expression to the canine ICA. In Aim 2, we treated 3, 21-25.5-month-old ADAMTS10-mutant dogs unilaterally with 1.25-1.43x1012 vg of ssAAV2-(Y444F)-hADAMTS10 (Table 2.2, Methods) and followed them for 62-105 weeks. At the age of injection, IOP was already beginning to increase as part of the natural disease process. 3.2.1 Long-term IOP control following IC injection of ssAAV2-(Y444F)-hADAMTS10 We observed a robust lowering and long-term steady maintenance of weekly diurnal IOPs in 2/3 ADAMTS10-mutant eyes treated for the entire observation period of 62-105 weeks (Figures 3.9). In comparison, weekly diurnal IOPs of the fellow untreated eyes and both the treated and untreated eyes of the third dog continued to increase and vary with natural disease 68 progression (Figure 3.9). With use of local regression models, Loess, a trendline of average daily IOPs over time was obtained with corresponding CIs accounting for Bonferroni correction of 1-α=0.99 (Figure 3.10). To test the significance between the paired data of the treated and fellow untreated eyes, we calculated the difference between the two, and again obtained a fitted trendline with CIs (Figure 3.11). Any CI not containing zero in its value was considered significant with a Bonferroni adjustment of multiple comparisons (α≤0.01). For Dog 1 and 2, this clear pattern of a lowered and maintained IOP was first determined to be significant at day 20 and 19, respectively. Figure 3.9. Diurnal IOP for ssAAV2-(Y444F)-hADAMTS10 treated and untreated control eyes. Dogs 1 and 2 showed a clear lowering of IOP and maintained this pattern for the duration of the study. Both eyes of Dog 3 and fellow untreated eyes of Dogs 1 and 2 exhibited the natural disease progression of increasing IOP. Shaded in grey is the target IOP range of 10-20mmHg. 69 Figure 3.10. Linear regression fitted trendline with CI for treated and untreated eyes. Linear regression model was used to create a smooth and fitted trendline across multiple weekly IOPs. CIs (shaded in grey) were obtained for a Bonferroni correction level of 1-α=0.99. 70 Figure 3.11. Linear regression fitted trendline with CI for difference between treated and untreated eyes. A smoothed and fitted trendline with corresponding CIs (shaded in grey) was established for the difference between treated and untreated eyes. Significance is determined when the CI of the difference does not contain the value 0, when α≤0.01. 3.2.2 Increased Outflow Facility in Successfully Treated Eyes Pneumotonography was utilized to measure conventional outflow facility. Facility is calculated as the inverse measure of trabecular resistance to AH outflow(51): less TM 71 resistance results in increased conventional outflow facility . Since pneumotonography was not available to us until after administration of ssAAV2-(Y444F)-hADAMTS10, treated eye outflow facility measurements were compared to the untreated fellow eye rather than baseline measurements in the same eye. Overall, the measurements were quite variable over the observation period, but when comparing the treated to untreated eyes, outflow facility was higher in the successfully treated eyes on 3 out of 4 (Dog 1) and 3 out of 5 (Dog 2) time points with a 3-75% increase (Figure 3.12). Figure 3.12. Outflow facility measured by pneumotonography. Shaded in grey is the normal range established in our laboratory (2 dogs/4 eyes). Values below the black line (0.2μl/min/mmHg) were previously considered indicative of OAG in ADAMTS10-mutant Beagles (146). Assuming that the variability in outflow facility was not solely based on technical artifacts, these measurements are not fully consistent with the stable IOP lowering observed in the successfully treated eyes. Therefore, we assume that the IOP decrease was a result of both increase in conventional and uveoscleral outflow as suggested by our hGFP expression data under Aim 1. The in vivo estimation of uveoscleral outflow is challenging and has not yet been established for dogs. 72 3.2.3 Aqueous Humor Flow Remained Unaffected by Gene Therapy Fluorophotometry was newly added to the lab after administration of ssAAV2-(Y444F)- hADAMTS10. Because no baseline data was collected, the AH flow measurements were compared between treated and untreated fellow eyes in each dog. Our studies represent the first measurements of AH flow in ADAMTS10-mutant dogs, and they show that the G661R missense mutation in ADAMTS10 resulted in a significant decrease in AH flow (1.01- 5.79µL/minute) compared to normal control eyes (6.2-13.47µL/minute), regardless of age/OAG-disease state. While the number of measurements need to be expanded in the future, our initial observations show that AH flow is not affected by our gene therapy. When assessing disappearance rate of fluorescein by fluorophotometry, both treated and untreated eyes seem to maintain a low AH flow of 1.01-5.79μL/minute (Figure 3.13). Figure 3.13. Aqueous humor flow measured by fluorophotometry. Shaded in grey is the normal range measured by us (2 dogs/4 eyes) and others(148, 149, 160, 161). AH flow in all three ADAMTS10-mutant Beagles was lower regardless of treatment. 73 3.2.4 ICA Morphology All 3 AAV-treated dogs had gonioscopic and UBM examinations of the ICA performed. No changes were observed over time. ICA morphology was essentially normal with open angles and ciliary cleft (Figure 3.14) 74 Figure 3.14. ICA morphology of treated and untreated fellow eyes compared by gonioscopic examination and UBM. Gonioscopic images (RetCam II) and high-resolution ultrasound image (48-MHz probe) exhibit an open ICA and ciliary cleft in all AAV treated and untreated fellow eyes. Cornea, lens, iris and ciliary body pointed out by white arrows. Natural recession of posterior ciliary cleft depicted by black arrowheads. 75 3.2.5 Optic Nerve Head (ONH) Morphology and Retinal Function All dogs were monitored for morphologic changes of the ONH by OCT imaging and for retinal functional changes by ERG, respectively. For the 2/3 dogs that responded with long-term and steady reductions in IOPs, the ONHs of the treated eyes were protected from glaucomatous changes compared to the untreated fellow eyes that showed the characteristic ONH cupping. Both eyes of dog 3 showed ONH cupping, confirming non-response to therapy. While untreated and non-responding eyes all seemed to exhibit morphological changes classic to glaucoma, all 6 eyes of the 3 dogs retained retinal cell function that is comparable to normal dogs (Figure 3.15 and Table 3.2-3.4), demonstrating that clinically ONH atrophy can be detected earlier in the disease process than loss of retinal function in canine ADAMTS10-OAG. 76 Figure 3.15. Morphologic comparison of ONH and comparison of retinal function. OCT (Heidelberg; Spectralis) images of the ONH. 2/3 eyes responded to AAV treatment exhibited by preservation of the optic disc morphology, while all untreated and the non-responding eye of dog 3 exhibited classic optic disc cupping (white arrow). ERGs (RETI-port, Roland Consult), 3.0 cd.sec/m2 (0.5-Hz, minimum of 25 averages) of all eyes tested exhibited comparable functionality to a normal, adult Beagle. Wave analysis: a-wave, photoreceptor (rods and cones) response; b-wave, bipolar cell response; photopic negative response (PhNR), retinal ganglion cell response. White measure bars for OCT images = 600µm; ERG, vertical measure bars = 10µV, horizontal bar = 100 milliseconds. 77 Table 3.2. Dog 1: ERG a-, b-, and PhNR Wave Amplitudes and Peak Time Comparisons White 1.0 cd s m-2 on White 25 cd m-2 a-wave Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) b-wave PhNR 9.2 10.8 9.8 16.0 13.8 22.7 24.3 24.3 23.5 33.7 7.0 7.8 14.9 12.5 15.7 11.1 10.0 13.6 21.2 13.3 23.5 25.8 25.8 25.8 33.7 12.8 10.2 16.4 5.7 16.3 10.5 17.0 9.9 11.8 17.2 Timepoint months PI AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx Baseline 7 13 18 24 White 3.0 cd s m-2 on White 25 cd m-2 Baseline 7 13 18 24 Red 1.0 cd s m-2 on Blue 25 cd m-2 Baseline 7 13 18 24 20.1 29.8 23.1 27.0 25.6 23.2 34.1 30.8 33.8 25.9 26.6 29.0 29.0 28.2 29.0 29.0 29.7 29.0 28.2 29.0 5.3 3.7 5.6 6.0 6.3 3.5 1.3 0.8 1.1 x 11.7 12.5 12.5 11.0 14.1 2.0 2.5 2.5 2.4 4.8 5.9 3.6 6.7 5.6 2.9 1.9 1.4 1.8 3.7 0.1 27.6 23.2 36.2 46.9 23.6 11.5 8.1 13.1 15.4 10.6 32.2 28.3 43.0 48.7 25.0 6.7 7.2 9.7 16.7 7.6 12.5 10.2 14.1 13.3 14.1 13.3 16.4 11.0 15.7 x 40.7 43.8 41.5 41.5 42.3 37.6 29.7 x 39.1 x 39.9 41.5 43.1 40.7 43.1 39.9 43.1 42.3 39.9 41.5 37.6 39.9 40.7 39.1 45.4 5.4 4.4 x 12.9 x 9.0 9.5 6.2 9.8 x 21.1 22.7 22.7 21.9 22.7 21.9 23.5 23.5 21.1 x 45.4 45.4 45.4 42.3 46.2 2.8 2.1 0.1 2.1 0.0 11.7 13.3 13.3 12.5 13.3 14.1 12.5 13.3 16.4 14.1 78 Table 3.2. (cont'd) Red 3.0 cd s m-2 on Blue 25 cd m-2 a-wave Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) b-wave PhNR 13.3 14.1 13.3 13.3 16.4 19.0 14.4 16.9 25.5 19.6 Timepoint months PI AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx 4.8 Baseline 1.5 7 1.8 13 5.4 18 6.6 24 Blue 1.0 cd s m-2 on Amber 25 cd m-2 19.8 Baseline 7 11.8 9.5 13 21.3 18 24 11.8 Abbreviations: ERG, electroretinography; PhNR, photopic negative response; PI, post injection; µV, microvolts; ms, milliseconds; tx, treated; Un-tx, untreated. Any ‘x’ that is within the table, represents a wave that could not be measured. 8.8 21.3 17.4 27.2 17.0 24.3 27.4 27.4 25.8 27.4 38.4 39.9 39.9 37.6 47.7 24.3 26.6 27.4 25.8 27.4 18.9 17.5 27.4 21.9 14.5 60.1 42.4 52.5 59.8 54.4 19.3 16.2 27.2 33.6 16.5 26.9 9.8 30.9 49.8 18.2 12.5 15.7 12.5 12.5 12.5 17.2 15.7 17.2 15.7 16.4 33.7 46.2 40.7 40.7 47.0 47.0 46.2 45.4 42.3 44.6 3.6 2.3 3.9 0.3 2.4 17.3 9.9 12.0 11.8 12.1 63.4 64.4 41.6 68.6 51.0 43.8 46.2 43.8 42.3 44.6 14.9 15.7 15.7 14.1 16.4 26.5 21.8 23.9 43.9 21.1 29.7 32.1 31.3 29.7 32.9 30.5 32.9 31.3 29.7 32.1 79 Table 3.3. Dog 2: ERG a-, b-, and PhNR Wave Amplitudes and Peak Time Comparisons White 1.0 cd s m-2 on White 25 cd m-2 a-wave Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) b-wave PhNR 13.3 11.0 10.2 12.5 12.5 11.0 11.0 14.1 13.3 12.5 13.3 11.7 12.5 12.5 15.7 12.5 11.0 12.5 13.3 14.1 23.5 25.0 25.0 24.3 25.8 1.8 2.0 4.3 3.0 3.3 3.4 3.3 4.4 4.5 3.6 13.7 15.7 13.2 9.2 6.6 14.6 22.3 13.5 8.8 9.7 Timepoint months PI AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx Baseline 3 9 15 20 White 3.0 cd s m-2 on White 25 cd m-2 Baseline 3 9 15 20 Red 1.0 cd s m-2 on Blue 25 cd m-2 Baseline 3 9 15 20 25.3 32.6 21.7 18.7 20.0 33.1 32.4 30.2 22.7 21.2 2.0 5.6 2.8 1.8 1.2 2.4 1.9 3.4 5.0 5.8 0.4 2.0 2.2 2.3 0.2 11.5 14.3 16.5 14.8 x 28.0 24.5 25.7 23.5 18.0 25.0 25.0 25.0 24.3 25.8 29.7 28.2 29.0 27.4 29.7 12.2 10.2 13.7 13.2 8.5 31.4 31.9 37.4 30.3 19.0 44.6 46.2 43.8 43.1 47.7 44.6 44.6 44.6 42.3 45.4 2.6 14.7 x 7.6 3.3 47.0 41.5 4.5 41.5 x 47.0 43.8 41.5 42.3 44.6 x 43.8 x 42.3 31.3 29.7 28.2 28.2 27.4 29.7 23.5 21.9 22.7 22.7 23.5 32.9 44.6 x 42.3 32.1 7.8 11.0 9.4 12.5 14.1 10.2 12.5 10.2 12.5 13.3 8.1 6.6 7.9 6.9 4.7 0.7 1.7 1.3 0.1 1.1 9.0 9.5 5.7 7.1 5.2 23.5 21.9 20.4 21.1 23.5 x 10.8 x 9.9 4.6 80 b-wave PhNR Table 3.3. (cont'd) Red 3.0 cd s m-2 on Blue 25 cd m-2 a-wave Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) Timepoint months PI AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx Baseline 3 9 15 20 Blue 1.0 cd s m-2 on Amber 25 cd m-2 Baseline 3 9 15 20 Abbreviations: ERG, electroretinography; PhNR, photopic negative response; PI, post injection; µV, microvolts; ms, milliseconds; tx, treated; Un-tx, untreated. Any ‘x’ that is within the table, represents a wave that could not be measured. 17.3 25.5 14.1 18.2 18.6 58.6 65.3 36.4 34.7 37.6 16.7 23.3 18.0 15.5 12.1 28.6 35.2 37.5 39.1 22.2 44.6 43.1 42.3 41.5 45.4 45.4 43.8 44.6 43.8 47.7 26.6 25.8 26.6 25.0 26.6 12.5 13.3 13.3 13.3 11.7 16.0 16.8 22.9 14.5 11.0 27.4 25.8 25.8 25.0 27.4 9.4 14.1 14.1 14.1 15.7 14.8 20.9 22.0 17.7 14.2 51.8 51.0 45.6 36.3 36.2 0.6 2.5 2.3 0.7 1.4 11.6 11.4 10.6 9.0 10.8 0.7 4.3 1.3 3.3 2.5 13.6 13.5 10.0 12.0 11.3 14.9 14.9 15.7 14.1 16.4 27.2 30.7 38.9 41.7 24.0 29.0 32.1 31.3 30.5 32.9 29.0 32.1 30.5 31.3 32.1 14.9 15.7 13.3 14.9 16.4 50.1 47.7 46.2 47.0 49.3 50.1 47.0 45.4 45.4 51.7 81 Table 3.4. Dog 3: ERG a-, b-, and PhNR Wave Amplitudes and Peak Time Comparisons White 1.0 cd s m-2 on White 25 cd m-2 a-wave Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) b-wave PhNR N/A 10.2 14.9 12.5 13.3 N/A 4.7 12.5 12.5 11.7 N/A 12.5 12.5 11.0 11.7 N/A 12.5 11.0 12.5 12.5 N/A 2.8 2.7 2.4 0.8 N/A 2.5 4.5 2.6 2.2 N/A 12.8 15.8 15.6 10.5 N/A 9.6 8.5 12.8 8.8 Timepoint months PI AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx Baseline 3 9 15 20 White 3.0 cd s m-2 on White 25 cd m-2 Baseline 3 9 15 20 Red 1.0 cd s m-2 on Blue 25 cd m-2 Baseline 3 9 15 20 N/A 19.4 17.8 25.2 19.9 N/A 30.2 33.0 37.2 25.3 N/A 38.4 39.1 40.7 42.3 N/A 0.8 1.7 3.5 2.2 N/A 2.5 4.0 3.4 4.0 N/A 1.4 1.7 1.8 0.3 N/A x 10.0 9.7 4.2 N/A 16.9 19.1 18.8 9.8 N/A 10.5 15.2 9.7 5.7 N/A 21.8 35.7 27.1 14.5 N/A 23.5 25.0 25.0 25.0 N/A 27.4 28.2 27.4 29.0 N/A 24.3 25.0 24.3 25.8 N/A 27.4 28.2 27.4 29.0 N/A 9.2 9.9 10.0 6.6 N/A x 39.9 43.8 43.8 N/A 44.6 42.3 41.5 43.1 N/A 31.3 38.4 32.9 46.2 N/A 41.5 42.3 40.7 45.4 N/A 39.9 37.6 29.9 39.9 N/A 21.1 21.9 21.1 22.7 N/A 3.6 6.9 5.5 5.9 N/A 13.3 13.3 12.5 12.5 N/A 6.8 8.5 7.9 4.8 N/A 9.0 10.1 10.2 7.1 N/A 21.9 22.7 20.4 21.9 N/A 1.2 1.1 3.0 0.5 N/A 14.1 12.5 12.5 12.5 82 Table 3.4. (cont'd) Red 3.0 cd s m-2 on Blue 25 cd m-2 a-wave Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) Amplitude (µV) Peak Time (ms) b-wave PhNR N/A 20.5 28.1 26.3 19.1 Timepoint months PI AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx AAV tx Un-tx Baseline 3 9 15 20 Blue 1.0 cd s m-2 on Amber 25 cd m-2 Baseline N/A 3 46.2 9 45.4 15 43.8 20 49.3 Abbreviations: ERG, electroretinography; PhNR, photopic negative response; PI, post injection; N/A, not available; µV, microvolts; ms, milliseconds; tx, treated; Un-tx, untreated. Any ‘x’ that is within the table, represents a wave that could not be measured N/A 17.2 18.3 22.7 16.6 N/A 49.5 57.7 53.7 46.4 N/A 21.0 29.6 46.6 27.0 N/A 25.0 25.8 25.0 26.6 N/A 35.2 43.8 41.5 44.6 N/A 12.5 10.2 13.3 13.3 N/A 14.9 14.1 14.1 14.9 N/A 4.8 13.4 13.9 11.1 N/A 42.3 49.7 52.2 37.5 N/A 11.7 10.2 14.9 14.1 N/A 14.9 15.7 14.9 16.4 N/A 25.0 25.8 25.8 26.6 N/A 29.0 29.7 29.0 30.5 N/A 16.5 21.0 17.9 11.6 N/A 21.8 28.3 39.9 27.7 N/A 0.9 0.9 0.5 2.5 N/A 11.6 12.9 17.8 15.0 N/A 1.9 1.9 3.3 3.6 N/A 10.7 13.3 16.3 13.8 N/A 29.7 29.7 29.0 29.7 N/A 49.3 46.2 44.6 47.7 N/A 39.9 41.5 39.9 42.3 83 3.2.6 ssAAV2-(Y444F)-hADAMTS10 Safety Measures – Clinical Examination, Pachymetry and A- scan Biometry Overall ssAAV2-(Y444F)-hADAMTS10 was well tolerated in Aim 2. Only mild uveitis was observed on clinical examination and easily controlled by twice daily topical corticosteroid administration, which subsequently induced a steroid keratopathy in both eyes of all 3 dogs. Pachymetry was collected at a similar frequency as clinical eye examination and % variation (Table 3.5) was calculated over the duration of the study as reported in Methods (see 2.4.2 Pachymetry). Percent variation in all eyes is similar to reported diurnal ranges (2.7-18.7%) in normal Beagles(139). At the finish of each OCT, A-scan biometry was performed to assess axial length of globe, anterior chamber depth (Table 3.6) and depth of vitreous (not calculated). Overall, all eyes show fluctuations in length over the duration of study; however, when looking at globe axial length only, treated eyes of Dogs 1 and 2 tended to be shorter in length than the untreated fellow. Dog 3, whose treated eye experienced elevated IOP compared to its untreated fellow, also tended to exhibit shorter globe lengths in the eye that had lower IOPs, suggesting that an elevation in IOP can alter globe length. Table 3.5. Percent Variation of Pachymetry in ssAAV2-(Y444F)-hADAMTS10 Treated and Control Eyes Dog 1 2 3 Treated Eye 11.1% 9.27% 13.3% Control Eye 13.4% 18.6% 11% 84 Table 3.6. A-scan Biometry Comparing ssAAV2-(Y444F)-hADAMTS10 Treated to Control Eyes Dog Eye Length Measured 1 2 3 Treated AXL ACD LT Control AXL ACD LT Treated AXL ACD LT Control AXL ACD LT Treated AXL ACD LT Control AXL ACD LT Baseline (mm) 20.78 3.88 7.36 20.83 4.20 6.95 22.05 4.71 7.4 21.85 4.60 7.22 22.04 4.67 6.95 21.73 4.35 7.03 4 Months PI (mm) 21.78 5.22 7.00 21.36 4.18 7.12 22.38 4.43 8.01 22.03 4.51 7.28 21.58 4.11 7.78 21.48 3.98 7.17 10-12 Months PI (mm) * * * 21.93 4.05 7.62 21.68 4.16 7.44 23.59 5.98 7.49 22.97 4.91 7.48 21.96 4.28 7.33 16 Months PI (mm) 20.43 4.06 7.07 21.17 4.10 7.14 21.09 4.00 7.23 21.31 4.43 7.51 NE NE NE NE NE NE 18-20 Months PI (mm) 20.51 3.54 7.32 21.25 3.9 7.24 21.67 3.38 7.01 22.34 4.54 7.07 NE NE NE NE NE NE 22 Months PI (mm) 20.38 3.38 7.01 21.93 4.54 7.08 NE NE NE NE NE NE NE NE NE NE NE NE 24 Months PI (mm) 20.88 3.91 7.08 21.48 4.17 7.18 21.89 3.98 7.17 22.62 4.11 7.77 22.24 4.94 6.99 21.71 5.41 7.29 Abbreviations: AXL, axial length; ACD, anterior chamber depth; LT, lens thickness; NE, not examined; mm, millimeters. * indicates that testing was discontinued prematurely (likely due to anesthesia). 85 3.3 Adverse Events – Ocular Inflammation 3.3.1 Aim 1 - Severe Inflammation While ssAAV2-(Y444F)-hADAMTS10 was well tolerated in Aim 2, dogs in Aim 1 (7/9 dogs, 12/18 eyes) developed severe uveitis following ssAAV2-(Y444F)-hGFP administration. While 2/9 dogs, 2/18 eyes managed to recover with interventive therapies, the remaining cohort did not and required premature euthanasia and tissue collection. The most severe ocular inflammation occurred at the highest dose of 1012vg either IC or IVit with dramatic clinical changes within 33-37 days of injection. Observed changes in 2/5 dogs was a drastic decrease in IOP in both treated eyes while 1/5 dogs had a severe increase in IOP in both eyes (so severe, creating diffuse corneal edema that was difficult to visualize past); 3/5 dogs experienced mild to severe hyperemia; and 3/5 dogs had detectable flare and cells that ranged from few to +3 with the most severe obstructing visualization of fundus. The 1/5 dog that had a severe spike in IOP and corneal edema too difficult to assess beyond, also experienced the most severe hyperemia and ciliary flush, blepharospasm and corneal neovascularization. The severe uveitis required an aggressive regimen of interventive topical and oral therapies that included topical 1% prednisolone acetate ophthalmic solution in 4/5 dogs and oral prednisone in 3/5 dogs. Additional pain management was required in 1/5 dogs with injectable buprenorphine, as well as topical dorzolamide-timolol and latanoprost glaucoma medications had to be administered to control elevated IOPs in both eyes. Routine histopathology was performed on all the eyes. H&E sections were graded on a 0-2 scale (0=none, 1=mild and 2= marked) for level of inflammation. Anterior segments from 4/5 dogs were scored as a 2 (marked), while 1/5 dogs were so severe, both eyes scored 2+ (marked +). 86 Anterior segments on 5/5 dogs contained a lymphoplasma-histiocytic cell population that heavily infiltrated the ciliary cleft, ciliary body and iris stroma (Figure 3.16). For 2/5 of the dogs, inflammation included multifocal areas of retinal vasculitis with mild to moderate perivascular lymphoplasmacytic cell populations (Figure 3.16). The second most severe inflammation occurred in 2/4 dogs that had received varying low-doses (1x1010 and 1x1011vg) of ssAAV2-(Y444F)-hGFP. Vector administration in these 2 dogs occurred 3 months after a previous administration of AAV2 to the contralateral AC. For 1/2 of the dogs, fibrin appeared in the AC (of the newly treated ssAAV2-(Y444F)-hGFP eye) within 24 hours of injection and cleared by the next day with medical therapy. The second dog had severe inflammation within 24 hours of injection and required the addition of treatment with topical dorzolamide-timolol and latanoprost glaucoma medications to control the elevated IOPs of the newly injected eye. Unlike the previous 4/5 dogs, the just aforementioned second dog, had alleviation of his glaucoma within 24 hours and clearance of his inflammation within 6 days. Pathologic review of the eyes from these 2 dogs had no observable inflammatory cell populations in any of the sections of tissue, likely because of successful anti-inflammatory therapy (not pictured). 87 Figure 3.16. Photomicrographs depicting inflammation induced by injection of ssAAV2- (Y444F)-hGFP. Inflammatory scoring for the top and bottom rows of ICA, posterior cleft, and ciliary body: 2 (marked). The middle row of ICA, posterior cleft, and ciliary body: 2+ (marked+); in the center picture, black arrowheads point to inflammatory cells, that in abundance, have overrun the ciliary body and now reside in the posterior chamber. While no inflammatory score was assigned to the retina sections, both the highest IC dose and IVit treatment exhibited multifocal areas of retinal vasculitis with mild to moderate perivascular lymphoplasmacytic cell populations. All IC injections, that administered 2.78E12 vg or less, retained normal histologic appearance of the retina (top right). CB, Ciliary body. Hematoxylin-eosin stain. Scale bars, black 50μm (ICA and Posterior Cleft and Ciliary Body columns); white 100µm (Retina column). 3.3.2 Neutralizing Antibody (NAb) Assays Because of the severe uveitis observed in Aim 1 and the mild anterior uveitis observed following treatment with ssAAV2-(Y444F)-hADAMTS10, the presence of AAV neutralizing antibodies was measured in serum and AH samples over repeated time points that coincided 88 with either ERG or OCT. We found positive serum titers in all 3 dogs, 2 of them were tested before AAV administration and found negative (Table 3.7), which was consistent with other dogs tested in the same facility. In addition, AH titers were measured in dog 1 and found to be positive in the injected but negative in the non-injected eye, confirming the generation of a humoral response and breakdown of the immune-deviation mechanisms of the anterior chamber. This was likely the reason for the development of anterior uveitis. Table 3.7. Aim 2 – NAb Titers for T Cell Response to scAAV2-(Y444F)-smCBA-mCherry Collection Timepoints Serum Titers Dog 1 Dog 2 Dog 3 Baseline NAb 1 2 3 Post injection NAb Post injection NAb Post injection NAb NC 3 months >2560 20 months 2560 24 months 2560 <10 8 months 2560 16 months 640 <10 16 months 640 Untreated Aqueous Humor Titers Dog 1 AAV treated 24 months 640 24 months <5 Abbreviations: NAb, neutralizing antibody; NC, not collected. 3.3.3 Aim 2 – Protein Concentration and Cell Count of AH Due to the dramatic uveitis associated with hGFP expression in Aim 1 and the mild but controllable uveitis of all 3 dogs in Aim 2, AH was collected 1-2 times, from each eye of each dog in Aim2. Analysis of protein concentrations and cell counts were performed by a boarded veterinary pathologist at MSU Veterinary Diagnostic Laboratory. AH submissions for all eyes contained low protein concentrations that were consistent with normal AH (Table 3.8). Four of 89 the 10 submissions did result in findings of nucleated cells however, they were few and were not indicative of an inflammatory response (Table 3.8). Table 3.8. Protein Concentration and Cell Count of AH of Aim 2 dogs Dog Procedure 18-22 Months post injection AAV treated Untreated ** 0 20-24 Months post injection AAV treated Untreated 18 17 <1 <1 unavailable Acellular fluid unavailable Acellular fluid No intact <1 MTP (mg/dL) ** Nucleated 0 cell/µL RBCs/µL Cytology Findings MTP (mg/dL) 23 Nucleated cell/µL RBCs/µL Cytology Findings 0 unavailable Acellular fluid consistent with normal AH 13 0 unavailable Acellular and low protein consistent with normal AH 1 2 3 nucleated cells 43 2 2 33 intact nucleated cells consisting of large mononuclear cells and small mononuclear cells; lack of granulocytes and no indication of an inflammatory response 29 <1 <1 No intact nucleated cells 28 1 1 4 intact and nucleated cells found but not able to identify 21 1 MTP (mg/dL) Nucleated cell/µL RBCs/µL Cytology Findings 0 5 intact mono nuclear cells Abbreviations: AH, aqueous humor; MTP, micro total protein; RBC, red blood cells; mg/dL, milligrams per deciliter; µL, microliter; **submission error with sample. 2 3 intact mono nuclear cells 90 Chapter 4 – Discussion Previously our team has demonstrated successful targeting and long-term transgene expression of the canine TM using a single mutant capsid ssAAV2-(Y444F) with either hGFP or hADAMTS10 cDNA; however, when applying the latter for a therapeutic effect, we failed to demonstrate a reduction in IOP in Beagles with ADAMTS10-OAG(135). The purpose of our current study was to further optimize the targeting of transgene expression with the same vector to the canine AHOPs, including the TM (Aim 1) and to reapply the optimized methods in a second therapeutic trial in ADAMTS10-mutant Beagles (Aim 2). In our study we evaluated the ability to increase the area of transgene expression of the TM by altering dosing, comparing IC vs. IVit injection techniques, assessing the effects of pupil diameter when paired with an IC injection, and anterior chamber filling capabilities of suspension vehicles. Alas the most dramatic improvement in transgene expression with ssAAV2-(Y444F)-hGFP, was achieved by a dose escalation from 1011 to 1012vg delivered as a single administration with our standard IC injection technique. This increase in hGFP transgene expression was so robust that we could clinically observe hGFP fluorescence by gonioscopic examination. This observed fluorescence emanating from the ciliary cleft was not witnessed in previous studies utilizing lower doses (109 to 1011) of ssAAV2-(Y444F)-hGFP. With this dose escalation, we are right at the limit of achievable vector concentrations with current AAV production methods. If further improvements in transgene expression were required, other avenues would have to be sought for improved transduction efficiency. 91 Perfusion of vector at large volumes distributed at a slow rate through the AC could be a possible approach to increase the number of vector particles delivered to the anterior chamber without increasing IOP. However, our goal is to have a minimally invasive approach that requires only one entry into the eye with a small gauge needle, 30-gauge, delivering a well- tolerated small volume of 50-75µL. Since the small volumes are well endured by the canine eye and do not require removal of AH to avoid severe IOP increases, we can avoid more inflammation naturally associated with any entry of the anterior chamber with a needle(162, 163). Additionally, having only one port for delivery reduces potential loss of vector. Other avenues that could also increase transduction efficiency include additional modifications to the viral capsid, use of a different serotype, use of a vehicle that can increase the bioavailability of the virus particles, and use of a more efficient or TM-specific promotor. While Aim 1 was very successful with increasing robust hGFP signal while suspended in a vehicle of BSS/0.014% Tween, the evolving science encompassing the improvement of vector administration with our University of Florida collaborators resulted in the use of HSS (with Tween) when administering hADAMTS10 in Aim 2(145). Using Tween (polyoxyethylenesorbitan monolaurate) as a vehicle prevents adherence of capsids to the wall of the storage tube and delivery device, thereby minimizing the loss of vector before injection. The addition of NaCl to create HSS further improves the bioavailability of capsids by reducing self-aggregation and increasing solubility of capsid particles(145). Previous studies have shown the ability to target the ICA by IVit injection(164). The vitreous has a slower turnover rate than AH which could lend the potential to act as a slow- release reservoir for AAV to the ICA allowing for increased contact time with the cells along the 92 AHOPs. When comparing an IVit administration approach to an IC approach, we were able to confirm that the canine ICA can be targeted by an IVit injection; however, hGFP expression levels were not increased. An IVit approach results in more off-target expression than the IC technique. Both injection techniques target the ICA, anterior iris surface, iris epithelium and, depending on pupil size, both can also target the ciliary body epithelium. IVit vector injection also targets the retina and is considered an attractive, less invasive administration technique compared to subretinal injection for retinal gene therapy(157, 165). Since we can minimize loss of vector posteriorly and off-target expression by keeping the pupil small, we opted to continue with IC injection as our preferred administration technique. Varying the pupil size gave us interesting insights on vector distribution following IC injection. Surprisingly, neither a mydriatic nor a miotic pupil resulted in obvious change in level of transgene expression along the AHOPs. While the ICA’s expression levels may not have been affected by the change in pupil size, posterior chamber structures had robust targeting with a mydriatic pupil, suggesting loss of vector into the posterior chamber. A dilated pupil allowed increased contact of IC injected reagents with the iris epithelium, lens capsule and ciliary body epithelium. While these are all off-target tissues for our current therapy attempts, these findings may be useful for future therapeutic trials with the intended targets being the iris, ciliary body epitheliums, or lens. A miotic pupil, while providing no detectable increase in TM transgene expression, prevented the posterior diffusion and loss of vector into the posterior chamber, thus maximizing the effective dose in the AC. Fluorescein reagents injected into the AC provided visual evidence that IC delivery of small volumes (50-75µL) resulted in uniform filling of the AC within seconds. The HSS vehicle 93 appeared to be the quickest to fill the AC, but ultimately all vehicles trialed revealed efficient uniform distribution. IHC analysis of hGFP expression within the ICA following small-volume IC injections also supports uniform circumferential distribution of vector to all four quadrants. In conclusion, Aim 1 insightfully exhibited that dose matters for our ability to robustly target the AHOPs. While we do not have a cell-specific promotor for gene expression of the TM, an IC dose of 1012 vg in a volume of 50-75µL resulted in robust transgene expression within the AHOPs. Despite minimal off-target transgene expression, the natural egress of AH through the ICA resulted in the desired effective targeting of the AHOPs. This strong transgene expression justified subsequent therapeutic trials of Aim 2. In Aim 2, we treated 3 ADAMTS10-mutant dogs with early stages of OAG unilaterally with ssAAV2-(Y444F)-hADAMTS10 suspended in HSS at a dose of 1.25-1.43x1012 vg as a one- time injection given IC. The dose was delivered at 21 to 25.5 months, at an age when the disease process is just starting to develop in the ADAMTS10-OAG Beagle model with IOPs around 20-24mmHg (54, 93, 166, 167). This age demographic is strategic in delivery, as injecting the same 1012vg dose in ADAMTS10-OAG Beagles aged 2-4 months (prior to disease progression) also resulted in no therapeutic effect (unpublished data). The natural disease progression for this OAG canine model is a steady increase in IOP starting at about 1.5-2 years of age and increased in range of pressure fluctuation(54). In 2/3 of the dogs that we treated with this gene therapy, a significant drop in IOP was achieved by 19-20 days post injection and maintained for at least 105weeks. This steady and maintained reduction in IOP resulted in effective protection of the ONH, the ultimate goal of any glaucoma therapy. 94 We observed low-grade anterior uveitis subsequently after therapy, but it was well controlled with topical administration of corticosteroids. While topical steroid therapy has been shown to result in an increase in IOP in these dogs with ADAMTS10-OAG(168) it did not mask the treatment effect of the gene therapy. Though the low-grade uveitis could contribute to an IOP decrease, it is unlikely that the uveitis alone is maintaining these low pressures as protein and cell analysis findings support that the ocular environment of these eyes are within normal limits. Despite variability in our tonography data, we were able to see a fairly consistent increase in conventional outflow facility of the treated eye versus the untreated eye in the 2 responders, supporting that the IOP decrease was based on an improved AH outflow. Conventional outflow facility as measured by pneumotonography has previously been shown to be reduced in ADAMTS10-mutant dogs(146). Because the therapeutic IOP effect is more robust and stable than the tonography data, we suspect that our treatment not only resulted in an increase in conventional outflow but also an increase in uveoscleral outflow. This notion is supported by our results in Aim 1, that demonstrated strong hGFP transgene expression along the conventional and uveoscleral outflow pathways. Currently, we do not have the ability to measure uveoscleral outflow in vivo, as the estimation via fluorophotometry still requires optimization for the canine eye(51). Uveoscleral outflow can be measured with terminal studies using tracers injected into the anterior chamber and measured in tissue samples(51), which were not feasible here because of our study design. With a reduction in IOP and an increase in outflow facility, we were expecting to find an increase in AH flow; but, to our surprise, we found that both the treated and untreated eyes 95 maintained the low AH flow as seen in pre-glaucomatous and glaucomatous ADAMTS10-mutant dogs. Our study is the first documentation of lower AH flow in ADAMTS10-mutant dogs compared to normal dogs (148, 149, 160, 161). While we currently do not have an explanation for the low AH flow, it has been shown that ADAMTS10-mutant dogs not only experience changes in the ECM of the TM, which contributes to the decrease in outflow facility, but they also have other ocular anomalies such as lens zonular dysplasia and disruptions of ciliary body epithelium basement membrane(169) which could contribute to the low AH production rate. While 2/3 dogs successfully responded to treatment with ssAAV2-(Y444F)-hADAMTS10, the third dog did not respond, suggesting that our current AAV administration protocol is at its threshold of effective therapy. We think a further increase in transduction efficiency may result in an outcome closer to 100% for therapeutic response. In general, the treatment with ssAAV2-(Y444F)-hADAMTS10 was safe. However, increasing our AAV administration dose, with and without HSS, not only resulted in increased hGFP transgene expression and a therapeutic response in 2/3 ADAMTS10-mutant dogs, it was also associated with anterior uveitis to a degree which we have not seen in previous intracameral injections. The most dramatic/severe inflammation was observed with delivery of ssAAV2-(Y444F)-hGFP at a 1012vg concentration, comparable to what we have previously reported with IVit administered hGFP vectors following vitrectomy, and likely based on a combination of immune response to AAV and hGFP(164) and a potential breakdown in the environmental or systemic suppressing pathways of the ACAID (Anterior Chamber Associated Immune Deviation) of the eye that lend the globe its immune-privileged properties. 96 While Aim 2 dogs also had anterior uveitis, it was less severe than the dogs from Aim 1, probably because in contrast to hGFP, which originates from jelly fish, hADAMTS10 is a mammalian protein. In all 3 dogs, uveitis was well controlled with topical corticosteroid therapy, associated with mild steroid keratopathy, which in dog 3 was reversed after cessation of drug application. The mildest inflammation of the 3, belonged to dog 3 who had no IOP reducing effect from treatment, and his inflammation was nicely controlled with topical dexamethasone, whereas the 2/3 responding dogs required a more potent topical steroid difluprednate. The fact that the 2 responding dogs needed a stronger topical steroid, supports a higher rate of transduction in the successfully treated eyes. Future experiments will include the measurement of hADAMTS10 protein expression within the treated eyes. We are hoping that this could even be achieved by evaluation of AH samples, which could be collected easily and repeatedly in non-terminal studies. The next steps in understanding our therapeutic results consists of a detailed histologic analysis, including evaluation of the ultra-structures of the TM to assess a potential reduction of plaque formation. Additionally, axon counts and/or RGC mapping will be necessary to confirm our OCT findings that the ONH is protected by our therapy. Our ERG and OCT findings have similar results as several human based studies, where OCT can detect structural changes to the ONH and retinal never fiber layer (RNFL) and circumpapillary RNFL prior to detectable ERG functional loss of the retina(170-173). Some suggest that there is redundancy or reserve of RGCs accommodating the structural loss(171). The aforementioned additional steps will be necessary to confirm cellular changes that match structural changes. 97 For safety assessment, a full body necropsy would be beneficial in reviewing the overall health status of other vital systems. Various tissues should be collected to evaluate systemic biodistribution of the viral vector. We suspect that following IC administration, some of the AAV vector ‘escaped’ through the AHOP into blood circulation, resulting in exposure to the systemic immune system and breakdown of the immune-deviation mechanisms within the anterior chamber as shown by the AAV NAb production in serum and treated eye of our dogs. Since the uveitis was more severe following the use of hGFP vector, we suspect that the immune response was also targeted against the hGFP as previously shown(164). Our goal is to reach a 100% success rate in therapeutic trials. To get there we need to understand the overall outcome of the successful therapy and determine what modifications we can make to deliver the safest and most effective therapy. An ideal therapy would be one that has the least concentrated vg with the highest transduction rate. Once optimized, our gene therapy can be translated into clinical trials for ADAMTS10-mutant human patients with WMS. While this is an orphan disease, it allows proof of concept that we can modify/alter gene expression in the AHOP, resulting in a therapeutic effect. More common forms of OAG are complex diseases with multiple genetic and environmental risk factors. However, if common molecular disease pathways can be identified, they could be targeted with our optimized AAV vector. We need to keep in mind that lowering IOP may not be sufficient to prevent progression of glaucoma, because RGCs may still continue to die for various reasons. Effective future glaucoma therapy will likely consist of a multi-strategic, comprehensive approach. Much effort is invested in the development of neuroprotective (and neuroregenerative) treatments to 98 achieve a true cure for glaucoma – some of these strategies also involve the use of AAV and are being evaluated in our laboratory. The future of glaucoma therapy is here, and it is very promising. 99 REFERENCES 100 REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Tham YC, et al. (2014) Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis. Ophthalmology 121(11):2081-2090. Weinreb RN, Aung T, & Medeiros FA (2014) The pathophysiology and treatment of glaucoma: a review. Jama 311(18):1901-1911. Dubal PM, Svider PF, Gupta A, Eloy JA, & Liu JK (2015) Chapter 30 - Injuries of the Cranial Nerves. Nerves and Nerve Injuries, eds Tubbs RS, Rizk E, Shoja MM, Loukas M, Barbaro N, & Spinner RJ (Academic Press, San Diego), pp 451-468. Iglesias AI, et al. (2015) Genes, pathways, and animal models in primary open-angle glaucoma. Eye (London, England) 29(10):1285-1298. Seki M & Lipton SA (2008) Targeting excitotoxic/free radical signaling pathways for therapeutic intervention in glaucoma. Progress in brain research 173:495-510. Brooks DE, Garcia GA, Dreyer EB, Zurakowski D, & Franco-Bourland RE (1997) Vitreous body glutamate concentration in dogs with glaucoma. American journal of veterinary research 58(8):864-867. Pease ME, McKinnon SJ, Quigley HA, Kerrigan-Baumrind LA, & Zack DJ (2000) Obstructed axonal transport of BDNF and its receptor TrkB in experimental glaucoma. Investigative ophthalmology & visual science 41(3):764-774. Knox DL, Eagle RC, Jr., & Green WR (2007) Optic nerve hydropic axonal degeneration and blocked retrograde axoplasmic transport: histopathologic features in human high-pressure secondary glaucoma. Archives of ophthalmology 125(3):347-353. Salinas-Navarro M, et al. (2010) Ocular hypertension impairs optic nerve axonal transport leading to progressive retinal ganglion cell degeneration. Experimental eye research 90(1):168- 183. Fahy ET, Chrysostomou V, & Crowston JG (2016) Mini-Review: Impaired Axonal Transport and Glaucoma. Current eye research 41(3):273-283. 11. Ward NJ, Ho KW, Lambert WS, Weitlauf C, & Calkins DJ (2014) Absence of transient receptor potential vanilloid-1 accelerates stress-induced axonopathy in the optic projection. The Journal of neuroscience : the official journal of the Society for Neuroscience 34(9):3161-3170. 12. 13. Agarwal R, Gupta SK, Agarwal P, Saxena R, & Agrawal SS (2009) Current concepts in the pathophysiology of glaucoma. Indian J Ophthalmol 57(4):257-266. Flammer J, Haefliger IO, Orgul S, & Resink T (1999) Vascular dysregulation: a principal risk factor for glaucomatous damage? Journal of glaucoma 8(3):212-219. 101 14. 15. 16. 17. 18. 19. 20. Michelson G, Langhans MJ, Harazny J, & Dichtl A (1998) Visual field defect and perfusion of the juxtapapillary retina and the neuroretinal rim area in primary open-angle glaucoma. Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie 236(2):80-85. Chung HS, Harris A, Kagemann L, & Martin B (1999) Peripapillary retinal blood flow in normal tension glaucoma. The British journal of ophthalmology 83(4):466-469. Gelatt KN, Miyabayashi T, Gelatt-Nicholson KJ, & MacKay EO (2003) Progressive changes in ophthalmic blood velocities in Beagles with primary open angle glaucoma. Veterinary ophthalmology 6(1):77-84. Gelatt-Nicholson KJ, Gelatt KN, MacKay EO, Brooks DE, & Newell SM (1999) Comparative Doppler imaging of the ophthalmic vasculature in normal Beagles and Beagles with inherited primary open-angle glaucoma. Veterinary ophthalmology 2(2):97-105. Brooks DE, Samuelson DA, & Gelatt KN (1989) Ultrastructural changes in laminar optic nerve capillaries of beagles with primary open-angle glaucoma. American journal of veterinary research 50(6):929-935. Mozaffarieh M, Grieshaber MC, Orgul S, & Flammer J (2008) The potential value of natural antioxidative treatment in glaucoma. Survey of ophthalmology 53(5):479-505. Liu Q, et al. (2007) Oxidative stress is an early event in hydrostatic pressure induced retinal ganglion cell damage. Investigative ophthalmology & visual science 48(10):4580-4589. 21. Wax MB & Tezel G (2009) Immunoregulation of retinal ganglion cell fate in glaucoma. Experimental eye research 88(4):825-830. 22. 23. 24. 25. 26. 27. Bell K, et al. (2013) Does autoimmunity play a part in the pathogenesis of glaucoma? Progress in retinal and eye research 36:199-216. Pumphrey SA, Pizzirani S, Pirie CG, Anwer MS, & Logvinenko T (2013) Western blot patterns of serum autoantibodies against optic nerve antigens in dogs with goniodysgenesis-related glaucoma. American journal of veterinary research 74(4):621-628. Bringmann A, et al. (2006) Muller cells in the healthy and diseased retina. Progress in retinal and eye research 25(4):397-424. Son JL, et al. (2010) Glaucomatous optic nerve injury involves early astrocyte reactivity and late oligodendrocyte loss. Glia 58(7):780-789. Inman DM & Horner PJ (2007) Reactive nonproliferative gliosis predominates in a chronic mouse model of glaucoma. Glia 55(9):942-953. Neufeld AH & Liu B (2003) Glaucomatous optic neuropathy: when glia misbehave. Neuroscientist 9(6):485-495. 102 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. Pattabiraman PP & Toris CB (2016) The exit strategy: Pharmacological modulation of extracellular matrix production and deposition for better aqueous humor drainage. European journal of pharmacology 787:32-42. Stamer WD & Clark AF (2017) The many faces of the trabecular meshwork cell. Experimental eye research 158:112-123. Mi XS, Yuan TF, & So KF (2014) The current research status of normal tension glaucoma. Clinical interventions in aging 9:1563-1571. Zagora SL, et al. (2015) Primary congenital glaucoma outcomes: lessons from 23 years of follow- up. American journal of ophthalmology 159(4):788-796. Lahola-Chomiak AA & Walter MA (2018) Molecular Genetics of Pigment Dispersion Syndrome and Pigmentary Glaucoma: New Insights into Mechanisms. Journal of ophthalmology 2018:5926906. Ritch R (2008) The management of exfoliative glaucoma. Progress in brain research 173:211- 224. Sihota R, et al. (2008) Early predictors of traumatic glaucoma after closed globe injury: trabecular pigmentation, widened angle recess, and higher baseline intraocular pressure. Archives of ophthalmology (Chicago, Ill. : 1960) 126(7):921-926. Bojikian KD, Stein AL, Slabaugh MA, & Chen PP (2015) Incidence and risk factors for traumatic intraocular pressure elevation and traumatic glaucoma after open-globe injury. Eye (London, England) 29(12):1579-1584. Rodrigues GB, et al. (2016) Neovascular glaucoma: a review. International journal of retina and vitreous 2:26. Sacchetti M, et al. (2015) Diagnosis and Management of Iridocorneal Endothelial Syndrome. BioMed research international 2015:763093. Sherman ER & Cafiero-Chin M (2018) Overcoming diagnostic and treatment challenges in uveitic glaucoma. Clinical & experimental optometry. Sng CC, Ang M, & Barton K (2015) Uveitis and glaucoma: new insights in the pathogenesis and treatment. Progress in brain research 221:243-269. 40. Weinreb RN, et al. (2016) Primary open-angle glaucoma. Nature reviews. Disease primers 2:16067. 41. Rohen JW & Witmer R (1972) Electrn microscopic studies on the trabecular meshwork in glaucoma simplex. Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie. Albrecht von Graefe's archive for clinical and experimental ophthalmology 183(4):251-266. 103 Avtar R & Srivastava R (2006) Modelling the flow of aqueous humor in anterior chamber of the eye. Applied Mathematics and Computation 181(2):1336-1348. Goel M, Picciani RG, Lee RK, & Bhattacharya SK (2010) Aqueous humor dynamics: a review. The open ophthalmology journal 4:52-59. To CH, Kong CW, Chan CY, Shahidullah M, & Do CW (2002) The mechanism of aqueous humour formation. Clinical & experimental optometry 85(6):335-349. Gelatt KN, Gilger BC, & Kern TJ (2013) Veterinary Ophthalmology : Two Volume Set (John Wiley & Sons, Incorporated, Hoboken, UNITED STATES). Kiel JW, Hollingsworth M, Rao R, Chen M, & Reitsamer HA (2011) Ciliary blood flow and aqueous humor production. Progress in retinal and eye research 30(1):1-17. Prasanna G, Dibas A, Hulet C, & Yorio T (2001) Inhibition of Na-ATPase by Endothelin-1 in Human Nonpigmented Ciliary Epithelial Cells. Journal of Pharmacology and Experimental Therapeutics 296(3):966. Pizzirani S & Gong H (2015) Functional Anatomy of the Outflow Facilities. The Veterinary clinics of North America. Small animal practice 45(6):1101-1126, v. McDougal DH & Gamlin PD (2015) Autonomic control of the eye. Comprehensive Physiology 5(1):439-473. Tian B, Hu Y, Gabelt BAT, & Kaufman PL (2006) Factors affecting outflow facility calculations. Experimental eye research 83(6):1515-1520. Toris C, Camras, C (2007) Measuring the Outflow of Aqueous Humor. Glaucoma Today (September/October):15-22. Johnson M, McLaren JW, & Overby DR (2017) Unconventional aqueous humor outflow: A review. Experimental eye research 158:94-111. Barrie KP, Gum GG, Samuelson DA, & Gelatt KN (1985) Quantitation of uveoscleral outflow in normotensive and glaucomatous Beagles by 3H-labeled dextran. American journal of veterinary research 46(1):84-88. Gelatt KN, Brooks DE, & Samuelson DA (1998) Comparative glaucomatology. I: The spontaneous glaucomas. Journal of glaucoma 7(3):187-201. Samuelson DA (1996) A Reevaluation of the Comparative Anatomy of the Eutherian Iridocorneal Angle and Associated Ciliary Body Musculature. Veterinary and Comparative Ophthalmology Volume 6(No. 3). Hynes RO (2009) The extracellular matrix: not just pretty fibrils. Science (New York, N.Y.) 326(5957):1216-1219. 104 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. Mouw JK, Ou G, & Weaver VM (2014) Extracellular matrix assembly: a multiscale deconstruction. Nature reviews. Molecular cell biology 15(12):771-785. Vranka JA, Kelley MJ, Acott TS, & Keller KE (2015) Extracellular matrix in the trabecular meshwork: intraocular pressure regulation and dysregulation in glaucoma. Experimental eye research 133:112-125. Pizzirani S (2015) Definition, Classification, and Pathophysiology of Canine Glaucoma. The Veterinary clinics of North America. Small animal practice 45(6):1127-1157, v. Daopin S, Piez KA, Ogawa Y, & Davies DR (1992) Crystal structure of transforming growth factor- beta 2: an unusual fold for the superfamily. Science (New York, N.Y.) 257(5068):369-373. 61. Wordinger RJ & Clark AF (1999) Effects of glucocorticoids on the trabecular meshwork: towards a better understanding of glaucoma. Progress in retinal and eye research 18(5):629-667. 62. Duffy L & O'Reilly S (2018) Functional Implications of Cross-Linked Actin Networks in Trabecular Meshwork Cells. Cellular physiology and biochemistry : international journal of experimental cellular physiology, biochemistry, and pharmacology 45(2):783-794. 63. Wiggs JL & Pasquale LR (2017) Genetics of glaucoma. Human molecular genetics 26(R1):R21-r27. 64. Online Mendelian Inheritance in Man, OMIM®. McKusick-Nathans Institute of Genetic Medicine, Johns Hopkins University, (Baltimore, MD), {12-01-2018}. World Wide Web URL: https://omim.org/ 65. 66. 67. 68. 69. 70. 71. 72. Komaromy AM & Petersen-Jones SM (2015) Genetics of Canine Primary Glaucomas. The Veterinary clinics of North America. Small animal practice 45(6):1159-1182, v. Khan AO (2011) Genetics of primary glaucoma. Current opinion in ophthalmology 22(5):347-355. Hewitt AW, Craig JE, & Mackey DA (2006) Complex genetics of complex traits: the case of primary open-angle glaucoma. Clinical & experimental ophthalmology 34(5):472-484. Kwon YH, Fingert JH, Kuehn MH, & Alward WL (2009) Primary open-angle glaucoma. The New England journal of medicine 360(11):1113-1124. Sarfarazi M & Rezaie T (2003) Optineurin in primary open angle glaucoma. Ophthalmology clinics of North America 16(4):529-541. Pasutto F, et al. (2010) Heterozygous loss-of-function variants in CYP1B1 predispose to primary open-angle glaucoma. Investigative ophthalmology & visual science 51(1):249-254. Pasutto F, et al. (2009) Heterozygous NTF4 mutations impairing neurotrophin-4 signaling in patients with primary open-angle glaucoma. American journal of human genetics 85(4):447-456. Fingert JH, et al. (2011) Copy number variations on chromosome 12q14 in patients with normal tension glaucoma. Human molecular genetics 20(12):2482-2494. 105 Footz TK, et al. (2009) Glaucoma-associated WDR36 variants encode functional defects in a yeast model system. Human molecular genetics 18(7):1276-1287. Bouhenni RA, Dunmire J, Sewell A, & Edward DP (2012) Animal models of glaucoma. Journal of biomedicine & biotechnology 2012:692609. Zeiss CJ (2013) Translational models of ocular disease. Veterinary ophthalmology 16 Suppl 1:15- 33. Morrison JC, Cepurna WO, & Johnson EC (2015) Modeling glaucoma in rats by sclerosing aqueous outflow pathways to elevate intraocular pressure. Experimental eye research 141:23- 32. Almasieh M & Levin LA (2017) Neuroprotection in Glaucoma: Animal Models and Clinical Trials. Annual review of vision science 3:91-120. Kuehn MH, et al. (2016) A Mutation in LTBP2 Causes Congenital Glaucoma in Domestic Cats (Felis catus). PloS one 11(5):e0154412. Sherpa T, Hunter SS, Frey RA, Robison BD, & Stenkamp DL (2011) Retinal proliferation response in the buphthalmic zebrafish, bugeye. Experimental eye research 93(4):424-436. Best M, Rabinovitz AZ, & Masket S (1975) Experimental alphachymotrypsin glaucoma. Annals of ophthalmology 7(6):803-810. Hubmacher D & Apte SS (2015) ADAMTS proteins as modulators of microfibril formation and function. Matrix biology : Journal of the international society for matrix biology 47:34-43. Kelwick R, Desanlis I, Wheeler GN, & Edwards DR (2015) The ADAMTS (A Disintegrin and Metalloproteinase with Thrombospondin motifs) family. Genome biology 16:113. Kutz WE, et al. (2011) ADAMTS10 protein interacts with fibrillin-1 and promotes its deposition in extracellular matrix of cultured fibroblasts. The Journal of biological chemistry 286(19):17156- 17167. Mularczyk EJ, et al. (2018) ADAMTS10-mediated tissue disruption in Weill-Marchesani Syndrome. Human molecular genetics. Hubmacher D, et al. (2017) Unusual life cycle and impact on microfibril assembly of ADAMTS17, a secreted metalloprotease mutated in genetic eye disease. Scientific reports 7:41871. Forman OP, Pettitt L, Komáromy AM, Bedford P, & Mellersh C (2015) A Novel Genome-Wide Association Study Approach Using Genotyping by Exome Sequencing Leads to the Identification of a Primary Open Angle Glaucoma Associated Inversion Disrupting ADAMTS17. PloS one 10(12):e0143546. 106 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. Oliver JA, Forman OP, Pettitt L, & Mellersh CS (2015) Two Independent Mutations in ADAMTS17 Are Associated with Primary Open Angle Glaucoma in the Basset Hound and Basset Fauve de Bretagne Breeds of Dog. PloS one 10(10):e0140436. Kuchtey J, et al. (2013) Screening ADAMTS10 in dog populations supports Gly661Arg as the glaucoma-causing variant in beagles. Investigative ophthalmology & visual science 54(3):1881- 1886. Ahonen SJ, et al. (2014) A novel missense mutation in ADAMTS10 in Norwegian Elkhound primary glaucoma. PloS one 9(11):e111941. Dagoneau N, et al. (2004) ADAMTS10 mutations in autosomal recessive Weill-Marchesani syndrome. American journal of human genetics 75(5):801-806. Kutz WE, et al. (2008) Functional analysis of an ADAMTS10 signal peptide mutation in Weill- Marchesani syndrome demonstrates a long-range effect on secretion of the full-length enzyme. Human mutation 29(12):1425-1434. 92. Wang L, Xiao R, Andres-Mateos E, & Vandenberghe LH (2017) Single stranded adeno-associated virus achieves efficient gene transfer to anterior segment in the mouse eye. PloS one 12(8):e0182473. 93. 94. 95. 96. 97. 98. 99. Samuelson DA, Gum GG, & Gelatt KN (1989) Ultrastructural changes in the aqueous outflow apparatus of beagles with inherited glaucoma. Investigative ophthalmology & visual science 30(3):550-561. Cideciyan AV, et al. (2013) Human retinal gene therapy for Leber congenital amaurosis shows advancing retinal degeneration despite enduring visual improvement. Proceedings of the National Academy of Sciences of the United States of America 110(6):E517-525. Tao W, et al. (2002) Encapsulated cell-based delivery of CNTF reduces photoreceptor degeneration in animal models of retinitis pigmentosa. Investigative ophthalmology & visual science 43(10):3292-3298. Beltran WA, et al. (2015) Successful arrest of photoreceptor and vision loss expands the therapeutic window of retinal gene therapy to later stages of disease. Proceedings of the National Academy of Sciences of the United States of America 112(43):E5844-5853. Beltran WA, et al. (2012) Gene therapy rescues photoreceptor blindness in dogs and paves the way for treating human X-linked retinitis pigmentosa. Proceedings of the National Academy of Sciences of the United States of America 109(6):2132-2137. Acland GM, et al. (2001) Gene therapy restores vision in a canine model of childhood blindness. Nature genetics 28(1):92-95. Komaromy AM, et al. (2010) Gene therapy rescues cone function in congenital achromatopsia. Human molecular genetics 19(13):2581-2593. 107 100. Gaub BM, et al. (2014) Restoration of visual function by expression of a light-gated mammalian ion channel in retinal ganglion cells or ON-bipolar cells. Proceedings of the National Academy of Sciences of the United States of America 111(51):E5574-5583. 101. O'Callaghan J, Cassidy PS, & Humphries P (2017) Open-angle glaucoma: therapeutically targeting the extracellular matrix of the conventional outflow pathway. Expert opinion on therapeutic targets 21(11):1037-1050. 102. 103. 104. 105. 106. Alario AF, Strong TD, & Pizzirani S (2015) Medical Treatment of Primary Canine Glaucoma. The Veterinary clinics of North America. Small animal practice 45(6):1235-1259, vi. Girkin CA, Bhorade AM, Crowston JG, Giaconi JA, Medeiros FA, Sit AJ, Tanna AP. (2018) Glaucoma. in 2018-2019 Basic and Clinical Science Course (American Academy of Ophthalmology). Aliancy J, Stamer WD, & Wirostko B (2017) A Review of Nitric Oxide for the Treatment of Glaucomatous Disease. Ophthalmology and therapy 6(2):221-232. Cavet ME, Vittitow JL, Impagnatiello F, Ongini E, & Bastia E (2014) Nitric oxide (NO): an emerging target for the treatment of glaucoma. Investigative ophthalmology & visual science 55(8):5005- 5015. Honjo M & Tanihara H (2018) Impact of the clinical use of ROCK inhibitor on the pathogenesis and treatment of glaucoma. Japanese journal of ophthalmology 62(2):109-126. 107. Maggio F & Bras D (2015) Surgical Treatment of Canine Glaucoma: Filtering and End-Stage Glaucoma Procedures. The Veterinary clinics of North America. Small animal practice 45(6):1261-1282, vi-vii. 108. Quigley HA (2018) 21st century glaucoma care. Eye (London, England). 109. Genetics Home Reference, U.S. National Library of Medicine, NIH, National Institutes of Health {12-1-2018}. World Wide Web URL: https://ghr.nlm.nih.gov/primer/therapy/genetherapy. 110. 111. 112. Borras T, Brandt CR, Nickells R, & Ritch R (2002) Gene therapy for glaucoma: treating a multifaceted, chronic disease. Investigative ophthalmology & visual science 43(8):2513-2518. Loewen N, et al. (2001) Genetic modification of human trabecular meshwork with lentiviral vectors. Human gene therapy 12(17):2109-2119. Loewen N, et al. (2004) Long-term, targeted genetic modification of the aqueous humor outflow tract coupled with noninvasive imaging of gene expression in vivo. Investigative ophthalmology & visual science 45(9):3091-3098. 113. Wold WSM & Toth K (2013) Adenovirus vectors for gene therapy, vaccination and cancer gene therapy. Current gene therapy 13(6):421-433. 114. Durand S & Cimarelli A (2011) The inside out of lentiviral vectors. Viruses 3(2):132-159. 108 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. Athanasopoulos T, Munye MM, & Yanez-Munoz RJ (2017) Nonintegrating Gene Therapy Vectors. Hematology/oncology clinics of North America 31(5):753-770. Samulski RJ & Muzyczka N (2014) AAV-Mediated Gene Therapy for Research and Therapeutic Purposes. Annual review of virology 1(1):427-451. Boye SE, Boye SL, Lewin AS, & Hauswirth WW (2013) A comprehensive review of retinal gene therapy. Molecular therapy : the journal of the american society of gene therapy 21(3):509-519. Schon C, Biel M, & Michalakis S (2015) Retinal gene delivery by adeno-associated virus (AAV) vectors: Strategies and applications. European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V 95(Pt B):343- 352. Dong JY, Fan PD, & Frizzell RA (1996) Quantitative analysis of the packaging capacity of recombinant adeno-associated virus. Human gene therapy 7(17):2101-2112. Cideciyan AV, et al. (2014) Pseudo-fovea formation after gene therapy for RPE65-LCA. Investigative ophthalmology & visual science 56(1):526-537. Jacobson SG, et al. (2012) Gene therapy for leber congenital amaurosis caused by RPE65 mutations: safety and efficacy in 15 children and adults followed up to 3 years. Archives of ophthalmology (Chicago, Ill. : 1960) 130(1):9-24. Pierce EA & Bennett J (2015) The Status of RPE65 Gene Therapy Trials: Safety and Efficacy. Cold Spring Harbor perspectives in medicine 5(9):a017285. Testa F, et al. (2013) Three-year follow-up after unilateral subretinal delivery of adeno- associated virus in patients with Leber congenital Amaurosis type 2. Ophthalmology 120(6):1283-1291. Bainbridge JW, et al. (2015) Long-term effect of gene therapy on Leber's congenital amaurosis. The New England journal of medicine 372(20):1887-1897. Griswold A (2017) Unanimous vote lights path to FDA approval for LCA gene therapy. (American Academy of Ophthalmology). 126. United States Food and Drug Administration, FDA (2017) FDA approves novel gene therapy to treat patients with a rare form of inherited vision loss, Luxturna is the first gene therapy approved in the U.S. to target a disease caused by mutations in a specific gene. December 19th, 2017 {12-1-2018}. World Wide Web: https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm589467.htm 127. Buie LK, et al. (2010) Self-complementary AAV Virus (scAAV) Safe and Long-term Gene Transfer in the Trabecular Meshwork of Living Rats and Monkeys. Investigative ophthalmology & visual science 51(1):236-248. 109 128. Borras T, Buie LK, & Spiga MG (2016) Inducible scAAV2.GRE.MMP1 lowers IOP long-term in a large animal model for steroid-induced glaucoma gene therapy. Gene therapy 23(5):438-449. 129. Mietzsch M, Broecker F, Reinhardt A, Seeberger PH, & Heilbronn R (2014) Differential adeno- associated virus serotype-specific interaction patterns with synthetic heparins and other glycans. Journal of virology 88(5):2991-3003. 130. Acott TS & Kelley MJ (2008) Extracellular matrix in the trabecular meshwork. Experimental eye research 86(4):543-561. 131. McCarty DM (2008) Self-complementary AAV vectors; advances and applications. Molecular therapy : the journal of the American Society of Gene Therapy 16(10):1648-1656. 132. 133. 134. Boye SL, et al. (2016) Impact of Heparan Sulfate Binding on Transduction of Retina by Recombinant Adeno-Associated Virus Vectors. Journal of virology 90(8):4215-4231. Zhong L, et al. (2008) Next generation of adeno-associated virus 2 vectors: point mutations in tyrosines lead to high-efficiency transduction at lower doses. Proceedings of the National Academy of Sciences of the United States of America 105(22):7827-7832. Bogner B, et al. (2015) Capsid Mutated Adeno-Associated Virus Delivered to the Anterior Chamber Results in Efficient Transduction of Trabecular Meshwork in Mouse and Rat. PloS one 10(6):e0128759. 135. Oh A, Harman CD, Koehl K, Chiodo V, Boye SL, Boye S, Hauswirth WW, Komaromy AM (2014) Targeting of gene expression to the wildtype and ADAMTS10-mutant canine trabecular meshwork by nonself-complementary AAV2. Annual Proceedings of the Association for Research in Vision and Ophthalmology Abstract: 5669 (Coeur d'Alene, ID). 136. Oh A (2014) Targeting of Gene Expression to the Trabecular Meshwork of Glaucomatous Beagles by Non-Self-Complimentary AAV2. Master of Science (Michigan State University, ProQuest). 137. 138. Palko JR, et al. (2016) Influence of Age on Ocular Biomechanical Properties in a Canine Glaucoma Model with ADAMTS10 Mutation. PloS one 11(6):e0156466. Cleveland WS & Devlin SJ (1988) Locally Weighted Regression: An Approach to Regression Analysis by Local Fitting. Journal of the American Statistical Association 83(403):596-610. 139. Martín-Suárez E, et al. (2014) Diurnal variations of central corneal thickness and intraocular pressure in dogs from 8:00 am to 8:00 pm. The Canadian veterinary journal = La revue veterinaire canadienne 55(4):361-365. 140. 141. Harper CL, et al. (1996) Diurnal variations in human corneal thickness. The British journal of ophthalmology 80(12):1068-1072. Gibson TE, Roberts SM, Severin GA, Steyn PF, & Wrigley RH (1998) Comparison of gonioscopy and ultrasound biomicroscopy for evaluating the iridocorneal angle in dogs. Journal of the American Veterinary Medical Association 213(5):635-638. 110 142. 143. 144. Bentley E, Miller PE, & Diehl KA (2003) Use of high-resolution ultrasound as a diagnostic tool in veterinary ophthalmology. Journal of the American Veterinary Medical Association 223(11):1617-1622, 1599. Jacobson SG, et al. (2006) Safety of recombinant adeno-associated virus type 2-RPE65 vector delivered by ocular subretinal injection. Molecular therapy : the journal of the American Society of Gene Therapy 13(6):1074-1084. Zolotukhin S, et al. (2002) Production and purification of serotype 1, 2, and 5 recombinant adeno-associated viral vectors. Methods (San Diego, Calif.) 28(2):158-167. 145. Wright JF, et al. (2005) Identification of factors that contribute to recombinant AAV2 particle aggregation and methods to prevent its occurrence during vector purification and formulation. Molecular therapy : the journal of the American Society of Gene Therapy 12(1):171-178. 146. 147. Gelatt KN, Gum GG, Mackay EO, & Gelatt KJ (1996) Estimations of aqueous humor outflow facility by pneumatonography in normal, genetic carrier and glaucomatous beagles. Veterinary and Comparative Ophthalmology 6(3):148-152. Fischer KM, Ward DA, & Hendrix DV (2014) Two time point versus 4 time point data acquisition in aqueous humor flow rate determinations by fluorophotometry. Journal of glaucoma 23(9):613-615. 148. Ward DA, Cawrse MA, & Hendrix DV (2001) Fluorophotometric determination of aqueous humor flow rate in clinically normal dogs. American journal of veterinary research 62(6):853- 858. 149. 150. Toris CB, Lane JT, Akagi Y, Blessing KA, & Kador PF (2006) Aqueous flow in galactose-fed dogs. Experimental eye research 83(4):865-870. Toris CB, Koepsell SA, Yablonski ME, & Camras CB (2002) Aqueous humor dynamics in ocular hypertensive patients. Journal of glaucoma 11(3):253-258. 151. Ofri R, Dawson WW, Foli K, & Gelatt KN (1993) Chronic ocular hypertension alters local retinal responsiveness. The British journal of ophthalmology 77(8):502-508. 152. 153. 154. 155. Kuze M, et al. (2017) Electrophysiological responses from intrinsically photosensitive retinal ganglion cells are diminished in glaucoma patients. Journal of optometry 10(4):226-232. Ekesten B, Komaromy AM, Ofri R, Petersen-Jones SM, & Narfstrom K (2013) Guidelines for clinical electroretinography in the dog: 2012 update. Doc Opthalmol 127(2):79-87. Takada S, Kinoshita J, Iwata N, Imaoka M, & Tani Y (2017) Response Characteristics and Retinal Origin of the Photopic Negative Response of the Electroretinogram in Dogs. Current eye research 42(9):1302-1307. Boye SE, et al. (2012) The human rhodopsin kinase promoter in an AAV5 vector confers rod- and cone-specific expression in the primate retina. Human gene therapy 23(10):1101-1115. 111 156. 157. 158. 159. 160. 161. Fischer AH, Jacobson KA, Rose J, & Zeller R (2008) Hematoxylin and Eosin Staining of Tissue and Cell Sections. Cold Spring Harbor Protocols 2008(5):pdb.prot4986. Boyd RF, et al. (2016) Photoreceptor-targeted gene delivery using intravitreally administered AAV vectors in dogs. Gene therapy 23(2):223-230. Harman CD, Koehl KL, Oh A, Chiodo VA, Boye SL, Hauswirth WW, Boye SE, Komaromy AM (2017) Optimizing of transgene expression in the trabecular meshwork (TM) by comparing capsid mutant AAV2 vectors. Annual Proceedings in the Association for Research and Vision Ophthalmology (Baltimore, MD). Team RC (2018) R: A language and environment for statistical computing. ed Computing RFfS (Vienna, Austria). Cawrse MA, Ward DA, & Hendrix DV (2001) Effects of topical application of a 2% solution of dorzolamide on intraocular pressure and aqueous humor flow rate in clinically normal dogs. American journal of veterinary research 62(6):859-863. Skorobohach BJ, Ward DA, & Hendrix DV (2003) Effects of oral administration of methazolamide on intraocular pressure and aqueous humor flow rate in clinically normal dogs. American journal of veterinary research 64(2):183-187. 162. Ward DA, Ferguson DC, Kaswan RL, Green K, & Bellhorn RW (1991) Fluorophotometric evaluation of experimental blood-aqueous barrier disruption in dogs. American journal of veterinary research 52(9):1433-1437. 163. 164. Allbaugh RA, Roush JK, Rankin AJ, & Davidson HJ (2011) Fluorophotometric and tonometric evaluation of ocular effects following aqueocentesis performed with needles of various sizes in dogs. American journal of veterinary research 72(4):556-561. Boyd RF, et al. (2016) Reduced retinal transduction and enhanced transgene-directed immunogenicity with intravitreal delivery of rAAV following posterior vitrectomy in dogs. Gene therapy 23(6):548-556. 165. Mowat FM, et al. (2014) Tyrosine capsid-mutant AAV vectors for gene delivery to the canine retina from a subretinal or intravitreal approach. Gene therapy 21(1):96-105. 166. 167. 168. Brooks DE, Samuelson DA, Gelatt KN, & Smith PJ (1989) Morphologic changes in the lamina cribrosa of beagles with primary open-angle glaucoma. American journal of veterinary research 50(6):936-941. Gelatt KN, et al. (1981) Primary open angle glaucoma: inherited primary open angle glaucoma in the beagle. The American journal of pathology 102(2):292-295. Gelatt KN & Mackay EO (1998) The ocular hypertensive effects of topical 0.1% dexamethasone in beagles with inherited glaucoma. Journal of ocular pharmacology and therapeutics : the official journal of the Association for Ocular Pharmacology and Therapeutics 14(1):57-66. 112 169. 170. 171. 172. Teixeira L, Scott E, Iwabe S, Dubielzig R, & Komaromy A (2013) Zonular ligament dysplasia in beagles with hereditary primary open angle glaucoma (POAG). Investigative ophthalmology & visual science 54(15). Bowd C, Weinreb RN, Williams JM, & Zangwill LM (2000) The retinal nerve fiber layer thickness in ocular hypertensive, normal, and glaucomatous eyes with optical coherence tomography. Archives of ophthalmology (Chicago, Ill. : 1960) 118(1):22-26. Falsini B, et al. (2008) Structure-function relationship in ocular hypertension and glaucoma: interindividual and interocular analysis by OCT and pattern ERG. Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie 246(8):1153-1162. Park HY & Park CK (2013) Structure-function relationship and diagnostic value of RNFL Area Index compared with circumpapillary RNFL thickness by spectral-domain OCT. Journal of glaucoma 22(2):88-97. 173. Wollstein G, et al. (2005) Optical coherence tomography longitudinal evaluation of retinal nerve fiber layer thickness in glaucoma. Archives of ophthalmology (Chicago, Ill. : 1960) 123(4):464- 470. 113