<%BANNER%>

Increased Light Sensitivity in Mice Expressing a Mutant Human Rhodopsin Transgene


PAGE 1

INCREASED LIGHT SENSITIVITY IN MICE EXPRESSING A MUTANT HUMAN RHODOPSIN TRANSGENE By ALAN WHITE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Alan White

PAGE 3

For my family, and for Tom Baldwin.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank Dr. Alfred Lewin fo r his continuous friendship, support, and advice during my graduate education. I w ould also like to thank my committee members, Drs. William Hauswirth, Clay Smith, and Tere nce Flotte, for their helpful input and critical analysis as my studies progressed. Dr. Adrian Timmers was of particular help in developing methods for and performing animal injections and ERG anal ysis, and I also appreciate his cheerful encouragement. Dr. Lynn Shaw patiently overs aw much of my traini ng in the basics of molecular biology and recombinant DNA technology and for that I am grateful. I thank Dr. Quihong Li for performing the funduscopy descri bed in this dissertation. Drs. JiJing Pang, Seok Hong Min, and Marina Gorbatyuk also performed subretinal injections. Dr. Henry Baker was a great help during my devel opment as a scientific speaker, and as a student of science in general. I would like to extend special thanks to Mr. James Thomas Jr. for sharing his friendship and his knowledge of PCR and DNA analytical techniques, and for keeping the lab running smoothly, and Ms. Chrissy Stre et for managing my training grant and being my friendly liason to the Hauswirth lab. Mr. Tom Doyle was another valuable resource for animal techniques and general good input. I also extend my warmest thanks to all of my labmates and friends from othe r labs and departments, whose help, advice, and goodwill have been invaluable over the past few years.

PAGE 5

v This dissertation work could never have been completed without the heroic assistance of Ms. Joyce Conners. I would lik e to extend my warmest thanks to her for her assistance as my department graduate student secretary. Through her diligence I kept more deadlines than I missed, and was guided through the mass of bureaucratic necessities that beleaguer every graduate student. Her friendship will always be valued. Finally I would like to thank my grandpare nts, Bette Womack and Esther and Dick White, my parents, Dennis and Joan White, and my sister, Elizabeth White Spratlin, for their constant love and support throughout my life.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 Retinitis Pigmentosa.....................................................................................................1 History and Pathology...........................................................................................1 Causes of Retinitis Pigmentosa.............................................................................2 Rhodopsin.....................................................................................................................3 Rhodopsin and the Visual Cycle...........................................................................4 Rhodopsin and Retinitis Pigmentosa.....................................................................7 Animal Models of Retinitis Pigmentosa.....................................................................14 Gene Therapy for Retinitis Pigmentosa......................................................................18 Ribozymes..................................................................................................................20 Mechanistic Description......................................................................................21 AAV............................................................................................................................ 24 Production of AAV Vectors................................................................................26 AAV and Retinal Gene Therapy.........................................................................29 Project........................................................................................................................ .31 2 CHARACTERIZATION OF A NOV EL MOUSE MODEL OF RETINITIS PIGMENTOSA...........................................................................................................32 Introduction.................................................................................................................32 Materials and Methods...............................................................................................34 DNA Oligonucleotides........................................................................................34 Isolation of Genomic DNA.................................................................................34 PCR Analysis of Genomic DNA.........................................................................35 Electroretinography.............................................................................................36 Funduscopy..........................................................................................................38 Histology.............................................................................................................38

PAGE 7

vii Results........................................................................................................................ .39 Breeding Founder and Experimental Mice..........................................................39 ERG Natural History...........................................................................................40 Funduscopy..........................................................................................................40 Histology.............................................................................................................43 Discussion...................................................................................................................46 3 AAV-MEDIATED RIBOZYME TREATMENT OF mRHO+/-; hT17M MICE......50 Introduction.................................................................................................................50 Materials and Methods...............................................................................................52 RNA Oligonucleotides........................................................................................52 DNA Oligonucleotides........................................................................................52 Preparation of Synthetic R NA Ribozymes and Substrates..................................53 5' End-labeling of Deprotected Target RNAs.....................................................53 In Vitro Ribozyme Time Course Analysis..........................................................54 Ligating Ribozyme Sequences into rAAV Packaging Vectors...........................55 Subretinal Injection of r AAV Ribozyme Delivery Vectors................................58 Electroretinography.............................................................................................59 Results........................................................................................................................ .60 Ribozyme Creation..............................................................................................60 In Vitro Time Course Analysis of HRz1 and HRz3............................................61 ERG Analysis of hT17M Transgenic Mice Treated With HRz1 and HRz3.......62 Discussion...................................................................................................................65 4 INCREASED LIGHT SENSITIVIT Y IN mRHO+/-; hT17M MICE........................70 Introduction.................................................................................................................70 Materials and Methods...............................................................................................74 Retinal Illumination.............................................................................................74 Genotyping..........................................................................................................75 Electroretinography.............................................................................................75 Funduscopy..........................................................................................................75 Histology.............................................................................................................75 TUNEL Visualization of Apoptosis....................................................................76 Results........................................................................................................................ .77 High Intensity Illumination.................................................................................77 Low Intensity Illumination..................................................................................77 Apoptosis in Retinas Damaged by Low Intenisty Illumination..........................80 Funduscopic Illumination....................................................................................80 Apoptosis in Retinas Damaged by Fundus Photography....................................83 ERG Analysis of hP23H Mice Afte r High Intensity Illumination......................86 Apoptosis in hP23H Mouse Retinas After High Intensity Illumination..............89 Discussion...................................................................................................................70

PAGE 8

viii 5 RED FILTERED LIGHT FOR INJECTIONS PROTECTS AGAINST LIGHT DAMAGE IN THE mrho+/-; hT17M MOUSE..........................................................94 Introduction.................................................................................................................94 Materials and Methods...............................................................................................96 Creation of 600nm Filters....................................................................................96 Retinal Illumination.............................................................................................97 Electroretinography.............................................................................................97 Histology.............................................................................................................97 TUNEL Visualization of Apoptosis....................................................................97 Test Injections Using 600nm Filtered Light........................................................97 Results........................................................................................................................ .98 Spectrophotometric Analysis of 600nm Filters...................................................98 ERG Analysis After 600nm Retinal Illumination...............................................98 Apoptosis in Retinas Exposed to 600nm Illumination......................................100 Test Injections Using 600nm Filtered Light......................................................101 Discussion.................................................................................................................104 6 DISCUSSION...........................................................................................................107 Summary...................................................................................................................107 Mechanism of Light-Induced Photoreceptor Apoptosis...........................................110 Clinical Impact..........................................................................................................117 LIST OF REFERENCES.................................................................................................121 BIOGRAPHICAL SKETCH...........................................................................................138

PAGE 9

ix LIST OF TABLES Table page 1-1 Tissue tropism and site of isolat ion of the various AAV serotypes. ......................28 2-1 ONL averages taken from mr ho+/and mrho+/-; hT17M mice..............................45 4-1 Percent reduction in eyes illuminated 10,000 lux white light..................................88 5-1 Percent reduction in eyes illu minated with red or white light................................101

PAGE 10

x LIST OF FIGURES Figure page 1-1 Funduscopic presentation of Retinitis Pigmentosa. .................................................2 1-2 A rod photoreceptor cell.............................................................................................4 1-3 Isomerization of 11cis to alltrans retinal. ...............................................................5 1-4 The phototransduction cascade..................................................................................6 1-5 Illustration of rhodopsin as it is inserted into the outer segment ..............................8 1-6 Generic structure of a hammerhead ribozyme.........................................................22 1-7 Structure of a speci fic hairpin ribozyme................................................................. 23 1-8 Ribozyme cleavage in trans .................................................................................... 24 2-1 Agarose gel electrophoresis of PCR reactions........................................................ 35 2-2 Electroretinographic apparatus.................................................................................37 2-3 An example of an ERG tracing................................................................................37 2-4 10dB intensity ERG awave natural history.............................................................41 2-5 10dB intensity ERG bwave natural history............................................................41 2-6 10 dB intensity a-wave res ponses charted as a percentage......................................42 2-7 10 dB intensity b-wave res ponses charted as a percentage......................................42 2-8 Fundus photographs and respective 10 dB intensity ERG tracings.........................43 2-9 Representative sections of mrho+/mice (A-D) and mrho+/-; hT17M siblings......44 2-10 Tile-field mapped im age of a mouse retina..............................................................45 3-1 pXX-GS-HP MOPS 500 rAAV packaging plasmid................................................57 3-2 Dissecting scope and fiber optic light used during subretinal injections.................59

PAGE 11

xi 3-3 Cartoon depicting the subretinal injection...............................................................59 3-4 Primary structure of ribozymes HRz1 and HRz3.....................................................60 3-5 Representative PhosphorImager scan of a time course assay..................................62 3-6 Time course of HRz1 and HRz3 cleavage...............................................................62 3-7 Relative ERG responses of aand b-waves.............................................................64 3-8 10dB intensity ERG tracings from mice with one eye injected...............................66 3-9 10dB intensity ERG tracings from mice with both eyes injected........................... 67 4-1 A-wave ERG responses afte r high intensity illumination........................................78 4-2 B-wave ERG responses afte r high intensity illumination........................................78 4-3 A-wave ERG responses afte r low intensity illumination.........................................79 4-4 B-wave ERG responses afte r low intensity illumination.........................................79 4-5 TUNEL stained retinal sections fr om low intensity-illuminated mice....................81 4-6 Tile-field mapped image of the TUNEL stained retina............................................82 4-7 A and b-wave ERG responses from mice with fundus photography.......................84 4-8 Fundus pictures of mrho+/and mrho+ /-; hT17M mice at three and six weeks......85 4-9 TUNEL labeling of retinal sections from mice with right eye funduscopy.............86 4-10 10dB intensity aand b-wave ERG responses of illuminated P23H mice...............90 4-11 TUNEL labeling of retina l sections from P23H mice that were illuminated...........91 5-1 Absorbance spectrum of rhodopsin..........................................................................95 5-2 600nm red light filters..............................................................................................96 5-3 ERG amplitudes after right eye i llumination with 600nm filtered light..................99 5-4 TUNEL labeled right and left eye sections............................................................102 5-5 ERG responses measured two weeks after subretinal test injections.....................103 6-1 Eye from a human who inher ited the T17M rhodopsin mutation..........................119

PAGE 12

xii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INCREASED LIGHT SENSITIVITY IN MICE EXPRESSING A MUTANT HUMAN RHODOPSIN TRANSGENE By Alan White December 2005 Chair: Alfred Lewin Major Department: Medical Sciences--Genetics Retinitis Pigmentosa (RP) is a hete rogeneous class of retinal disorders characterized by an initial loss of peripheral and night vision, followed by loss of central and daylight vision, affecting around 1.5 milli on people worldwide. Many RP patients develop the disease because of mutations in a retinal prot ein called rhodopsin. Normal rhodopsin is a vital component of the phototransduction cascade that allows the retina to detect light, while mutant versions of this pr otein cause the cells of the retina to become sick and die, leading to blindness. Mouse models of the disease that are ge netically engineered to express mutant rhodopsin proteins are vital for studying the progression of RP and developing treatments for the disease. My work describes one of these models, which expresses a human rhodopsin transgene with a tyrosine to methionine mutation at the 17th amino acid of the protein (hT17M). We bred our line to express the T17M human transgene on a

PAGE 13

xiii background that was hemizygous null for m ouse rhodopsin (mrho+/-), which closely modeled RP mutations in human patients in expressing one copy of mutant rhodopsin and one copy of wild-type rhodopsin. We performe d electroretinographic analysis (ERG) to show that this line loses its visual responses over time. Histology confirmed that ERG attenuation was accompanied by a loss of rod photoreceptors in the retina. Unsuccessful attempts to treat the hT17M; mrho+/mice with subretinal injections of rAAV-expressed ribozymes led to the discove ry of an hT17M-specific light sensitivity that caused severe loss of aand b-wave ER G responses. Histological analysis showed a concomitant loss of photoreceptors, and TUNEL labeling of fragmented DNA in rod photoreceptor cells demonstrated that the damage was occurring via an apoptotic pathway. Attempts to reproduce this light damage phenotype in another mouse model of retinal disease that expressed a human rhodops in transgene with a proline to histidine mutation at the 23rd amino acid were unsuccessful, leading us to conclude that this light sensitivity was not common to all rhodopsin mutations. Finally, filters were developed that removed the wavelengths of light res ponsible for the retinal damage, allowing for non-damaging subretinal injections of the hT17M mice.

PAGE 14

1 CHAPTER 1 INTRODUCTION Retinitis Pigmentosa History and Pathology Retinitis pigmentosa (RP) is the comm on name for a group of retinal disorders characterized by progressive photoreceptor dege neration that culminates in loss of vision (Flannery et al., 1989; Hims et al., 2003; Farrar et al., 2002) The disease affects from 50,000 to 100,000 people in the United States and around 1.5 million people worldwide. Initial symptoms include night blindness and loss of peripheral vi sion, usually occurring during the late teens or early twenties. As the photorecepto rs continue to degrade the visual impairment progresses towards the center of the retina, eventually affecting the cone photoreceptors that are responsible for cen tral vision in bright light conditions, and resulting in the manifestation of tunnel vision, in which the visual fields of the patient constrict to less than 20o. The loss of vision is accompanied by visu al pigment depositions in the retina for which the disease is named (Figure 1-1) (Str icker et al., 2005; Farber et al., 1987). In most cases this is accompanied by a waxy pallo r to the optic nerve, attenuated retinal vasculature, and thinning of the retinal pi gmented epithelium. In many cases, abnormal ERG responses predict the development of re tinitis pigmentosa in affected individuals well before other symptoms present. For most patients the visual fields will continue to constrict until they ar e completely blind (Hum phries et al., 1992).

PAGE 15

2 Causes of Retinitis Pigmentosa Retinitis pigmentosa shows a high degree of genetic heterogeneity. Causative mutations can be subdivided into severa l genetic categories: autosomal dominant (ADRP), autosomal recessive (ARRP), Xli nked (XLRP), or syndromic (Hims et al., 2003; Farrar et al., 2002). The percent contri bution of each type of disease varies among different populations, but it is generally agreed that ADRP acc ounts for around 25% of all cases, ARRP for 20%, X-linked and syndr omic for around 8%, with the rest of the cases believed to be the result of spontaneous mutations arising in the affected individual (Diager et al., 2005). The majority of causa tive genes identified to date lead to the autosomal dominant form of RP. Figure 1-1. Funduscopic presen tation of Retinitis Pigmentosa. This series shows a normal retina (A) and a retina exhibiting evidence of Retinitis Pigmentosa (B) (Rosenfeld and Dryja, 1995). Retinitis pigmentosa can be caused by muta tions in a wide variety of genes. The first of these candidate genes to be discove red was rhodopsin, which is expressed in rod photoreceptor cells and is an integral part of the phot otransduction cascade. Subsequently a mutation in the peripherin RDS gene, which is a structural protein involved in maintaining rod photoreceptor outer segment disc morphology, was also

PAGE 16

3 shown to be linked to the disease. Mo re recently, mutations in phototransduction proteins including the and subunits of rod cGMP phosphodiesterase, the subunit of the rod cGMP-gated channel, and arrestin have been shown to lead to RP. RPE65, which is expressed in the retinal pigmented epith elium and is involved in visual pigment regeneration, as well as ROM 1, which has a st ructural role relate d to peripherin RDS, have also been identified as RP candidate ge nes. In all, 44 different loci have been identified as being associated with RP, of which 35 have been cloned (Humphries et al., 1992; Diager et al., 2005; Kennan et al., 2005). It is of interest that some of the genes associated with RP encode proteins needed in all cells for functions such as RNA splicing or purine metabolism. Why mutations in these genes lead to retinal degeneration and not other phenotypes is unknown (Bowne et al., 2002; Kennan et al., 2002; Martinez-Gimeno et al., 2003; Maita et al., 2005). Rhodopsin Opsin is a seven-transmembrane G-coupled receptor protein found in the disc membranes of the outer segments of rod and cone photoreceptor cells (Figure 1-2). The protein is oriented in the rod outer segment (ROS) memb rane such that its amino terminus is located on the inside of the disc in the intralumenal space, while the carboxyl terminus is found on the outside of the di sc in the cytoplasmic space. The amino terminus of the protein contains two glycosol ation sites, at Asn2 and Asn 15, and there is extensive association between the intralumenal portions of opsin and these carbohydrate moieties (Stenkamp et al., 2005; Hargrave a nd Mcdowell, 1992). Opsin is incredibly abundant, accounting for around 90% of the to tal protein in the rod outer segments (Daiger et al., 1995).

PAGE 17

4 Rhodopsin and the Visual Cycle In its active form, the protei n binds the visual pigment 11cis retinal to form rhodopsin. The term "rhodopsin" will be used to refer to this protein-chromophore complex for the remainder of this dissertation. The 11cis retinal is covalently bound to rhodopsin via a protonated Schiff-base at Lys296 (Bownds, 1967). The opsin protein itself does not absorb visible Figure 1-2 A rod photoreceptor cell, with e xpanded views showing outer segment disc morphology and the orientation or rhodopsin within the outer disc membrane. The chromophore 11-cis retinal is shown in a cutaway view to be bound to the interior of the rhodopsin protein. Figure adapted from Hargrave and McDowell, 1992. light, but when it is bound to 11cis retinal to form rhodopsin, the resultant molecule has a broad absorption band with a peak at around 500nm. Photons impacting upon rhodopsin provide energy that is able to temporarily convert the cis -double bond between C-11 and C-12 of 11cis retinal into a single bond, allo wing the molecule to rotate

PAGE 18

5 through 180o to an alltrans configuration (Figure 1-4). This isomerization leads to conformational changes in rhodops in that allow it to interact with downstream molecules in the phototransduction cascade. After light activation, rhodopsin shifts between two conformations, termed metarhodopsin I and metarhodopsin II. Metarhodops in II is able to transiently bind to the next protein in the cascade, transducin, whic h is a heterotrimeric G-protein consisting of Figure 1-3. Isomerization of 11cis to alltrans retinal. The C-11 to C-12 linkage about which the rotation occurs is highlighted in red. three subunits, and Each molecule of light-activated rhodopsin is able to interact with hundreds of transduc in molecules, resulting in the first of a series of signal amplifications. The subunit of transducin in its inactive state is bound to GDP. Interaction with metarhodopsin II causes the subunit to exchange its GDP for a GTP molecule, causing it to dissociate from the ot her subunits and allowing it to bind to cGMP phosphodiesterase, the next player in the cas cade. cGMP phosphodiesterase is composed of catalytic and subunits that are bound to and inhibited by two subunits.

PAGE 19

6 Interaction between activated transducin and one of the subunits releases the and subunits to hydrolyze cGMP to 5GMP. The resultant drop in the intr acellular cGMP concentration results in the closing of a few hundred to a few thousand cGMP-gated Ca2+ channels, resulting in a decrease in the intracellular concentration of Ca2+, which causes a hyperpol arization of the ROS plasma membrane. This signal is propagated along the plasma membrane to the synaptic terminus of the rod photoreceptor (Figure 1-2) resulting in a reduction in the release of glutamate. This decreases the activity of nearby bipolar cell glutam ate receptors, which in turn decreases the activation of a G-coupled receptor protein and Figure 1-4. The phototransduction cascade. Figure courtesy of Dr. Helga Kolb (Kolb et al., 2005).

PAGE 20

7 leads to an increase in the cGMP concentrati on in the bipolar cell. The increase in cGMP results in the opening of large numbers of cationic channels, resulting in bipolar cell depolarization and the generation of an act ion potential (Figure 1-4) (Hargrave and Mcdowell, 1992; Daiger et al., 1995; Maple and Wu, 1996). Inactivation of this cascade is initiated by rhodopsin ki nase, which binds to and phosphorylates metarhodopisn II. Phosphoryl ated metarhodopsin II is then able to interact with arresti n, which prevents inte ractions with transducin until metarhodopsin releases the all-trans retinal. Release of all trans retinal is thought to inhibit the reopening of the cGMP-gated ion channels. Elevencis -retinal is eventually recycled through the retinoid cycle that has steps in both photoreceptors and the retinal pigmented epithelium (McCabe et al., 2004). In the m eantime, cGMP is regenerated from GMP by the protein guanylate cyclase, and the cGMP is able to bind to and eventually reopen the cGMP-gated ion channels. This causes an influx of Ca2+ that restores the resting potential of the ROS, stimulating the release of glutamate at the synaptic terminus and terminating the light-induced signal. Rhodopsin and Retinitis Pigmentosa The first ADRP gene was identified and loca lized to the long arm of chromosome 3 in 1989 by researchers investigating the pedigree of a large Irish family with over fifty individuals reporting symptoms c onsistent with retinitis pigm entosa (Bradley et al., 1989; McWilliam et al., 1989). As RP was known to affect rod photoreceptors, and the gene encoding the ROS protein rhodopsin had recently also been localized to the long arm of chromosome 3, the race was on to identify a point mutation in rhodopsin that was associated with ADRP. In 1990, the first ADRP-causing rhodopsin mutation was

PAGE 21

8 reported, consisting of a DNA mutation (CCC to CAC) that caused histidine to be substituted for proline (P23H) at the 23rd amino acid of the prot ein (Dryja et al., 1990). Rhodopsin has subsequently become the most extensively characterized gene associated with retinitis pigmentosa. Muta tions in the rhodopsin ge ne account for around 10% of all reported cases of RP (Rivolta et al., 2002). Since discovery of the P23H mutation, around 150 different mutations of r hodopsin have been shown to cause the disease. A vast majority of these mutations lead to autosomal dominant RP, and most of these mutations are thought to cause retina l degeneration by either a toxic gain of function or a dominant negative fashion (Wilson and Wensel, 2003). Many of these mutations are illustrated in Figure 1-5. Figure 1-5. Illustration of r hodopsin as it is inserted in to the outer segment disc membrane. The image depicts some sec ondary structure and also illustrates key amino acids in which mutations lead to retinitis pigmento sa (Diager et al., 2005).

PAGE 22

9 Mutations affecting the Lys295 residu e prevent rhodopsin from binding to 11cis retinal, and cause the protein to be c onstitutively activated, leading to retinitis pigmentosa, possibly because this protein is able to continuously bind to and sequester arrestin (Berson, 1996). It is also known that mutations at the Tyr4 and Tyr17 residues abolish the glycosylation sites at the N-term inus of rhodopsin, leading to its aberrant trafficking in cell culture models, and in the case of the T17M mutation, inefficient regeneration of the protein with 11-cis reti nal (Kaushal et al., 1994). Addtionally, the cystine residues at positions 110 and 187, whic h form a conformationa lly vital disulfide bond, and the glutamate residue at codon 114, which provides the counter ion for the Schiffs base retinal linkage at codon 296, are structurally important residues that, when mutated, can lead to RP (Karnik et al., 1988; Karnik and Khorana, 1990; Daiger et al., 1995). However, the mechanisms by which th e majority of rhodopsin mutations lead to RP are not well understood. Several groups have attempted to cate gorize rhodopsin mutations by expressing rhodopsin genes containing them in cultured cells and then analyzing the resultant proteins with respect to a variety of factors. One such analysis utilized rhodopsin cDNAs engineered to contain thirty four mutati ons known to cause ADRP in patients and expressed in H293S cells. These experime nts lead to two categories of rhodopsin mutations, Class I and Class II. Class I mutations were less numerous, accounting for only six of the 34 mutations studied. Class I mutants were similar to wild-type rhodopsin expressed in the same system in terms of yield, subcellular loca lization (the plasma membrane), and regeneration with 11cis retinal, and tended to cluster in both the first transmembrane and carboxyl terminal domains of the protein. Class II mutations were

PAGE 23

10 more numerous, and were found clustered in the transmembrane and loop domains of rhodopsin. Class II mutants accumulated to si gnificantly lower than wild-type levels, regenerated poorly with 11-cis retinal, we re predominately mislocalized to the endoplasmic reticulum, and were shown to fo rm intracellular aggregates (Sung et al., 1991; Sung et al., 1993). A concurrent study involved introducing 35 mutations into a synthetic bovine rhodopsin gene and expressing them in COS cells (Kaushal and Khorana, 1994). The proteins expressed in this study were classified into three clas ses. Class I, like those of the studies of Sung and coworkers, consisted of proteins that show ed expression levels, subcellular localization (to the plasma me mbrane) and chromophore regeneration that were similar to wild-type rhodopsin. Class II consisted of proteins that showed folding defects, were mislocalized to the endoplasmic reticulum, and were not able to reconstitute with 11-cis retinal. Class III mutations also showed folding and localization defects, but were able to partiall y regenerate with 11cis retinal. Taken toge ther, these two studies provide evidence for at least two types of rhodopsin defect that can lead to ADRP. Recently, Mendes and coworkers have proposed classifying rhodopsin mutations into five groups based on additional characteristics su ch as whether they affect endocytosis or whether they remain constituively activated (Mendes et al., 2005). There are several theories concerning the mechanism of retinal degenerations caused by rhodopsin mutations. Sung and co workers note that their class II rhodopsin mutations show retention/mislocalization to the endoplasmic reticulum that is similar to that seen in other disease-cau sing mutant proteins such as Class II low density lipoprotein receptors and cystic fibrosis transmembran e conductance regulator proteins. Indeed,

PAGE 24

11 there is evidence for a variety of neurodege nerative diseases that share an accumulation of aggregated, ubiquitinated mutant proteins, suggesting that these proteins are targeted for destruction, possibly due to toxicity resulting from their expression (Wilson and Wensel, 2003; Rajan et al., 2001; Bence et al., 20 01; Illing et al., 2002). It is also thought that overloading the cellular machinery res ponsible for the removal of misfolded and toxic proteins, termed the unfolded-protei n or ER-stress response, can lead to programmed cell death (Mendes et al ., 2005; Rutkowski and Kaufman, 2004). In the case of mutations affecting the tyrosine residues at codons 4 and 17, the pathogenic mechanism may in part be explai ned by the abolishment of glycosylation at nearby residues, which has been shown to cause defective ROS membrane morphogenesis in Xenopus laevis retinas (Fliesler et al., 1985). Defects in outer segment morphogenesis are thought to lead to photor eceptor cell death, and can result from mutations in rhodopsin that prevent it from being properly transported to the OS disc membranes. It is known that the ROS shed s ~10% of its outer segment discs each day (Young and Bok, 1969). These discs are pha gocytosed by the retinal pigmented epithelium, and more must be synthesized at the base of the ROS each day to replace them (Young and Bok, 1969; Papermaster et al., 1986; Steinberg et al ., 1980). Studies of carboxy terminal rhodopsin mutations have show n that this region of the protein is important for dynein binding and cellular transp ort. Mutations in this region have been shown to lead to a mislocalization of rhodopsin in the rod photoreceptors, causing abnormal ROS disc morphogenesis, which woul d help to explain the pathogenesis of Class I rhodopsin mutations (Tai et al., 1999; Sung et al., 1994). Additionally, overwhelming synthesis of aberrantly folded rhodopsin may interfere with the subcellular

PAGE 25

12 trafficking of non-mutant rhodops in, again resulting in a hind ering or prevention of disc morphogenesis and causing a progressive shorteni ng of the ROS that eventually leads to cell death (Mendes et al., 2005; Besharse and Wetzel, 1995). While the exact mechanisms by which the various rhodopsin mutations lead to RP are still uncertain, it is clear that the ultimate fate of the affected photoreceptors is programmed cell death, or apoptosis. Apoptosis is a thoroughly documented phenomenon by which cells initiate a specific program of self-destr uction in response to intrinsic or extrinsic factors (Wenzel et al., 2005). These can include mechanical damage, toxic chemical exposure, bacterial or viral infection, and various forms of irradiation. Apoptosis is al so responsible for the progr ammed deaths of cells for developmental reasons, deaths of auto-reactive cells of the immune system, and deaths of cells that have lost growth inhibition a nd could become cancerous. Two major apoptotic pathways have been described, an extrin sic pathway involving CD95 and CD95 ligand and an intrinsic pathway involving mitochondri al damage. Activation of the aspartyl proteases, termed caspases, is common to both mechanisms, as is the ultimate release of mitochondrial cytochrome c, wh ich leads to full apoptotic ac tivation. While the extrinsic pathway plays an important role in shutti ng down the immune response, the intrinsic pathway is thought to be more important in pa thologic apoptosis such as that occurring in RP (Vermeulen et al., 2005; Mohamad et al., 2005). Additionally, caspase-independent pathways have also been described that involve effector molecules including cathepsins, calpains, granzyme A and B, and serine proteases such as AP24. Two major player s in the caspase independent pathway are apoptosis inducing factor (A IF) and PARP-1. AIF is an oxioreductase found in the

PAGE 26

13 mitochondria, and decreased levels of AIF ha ve been correlated with an increased sensitivity to oxidative stress. Activation of the caspase-independent apoptotic pathway leads to mitochondrial secretion of AIF, which is ultimately involved chromatin condensation and recruits the endonuclease E ndoG to effect chromatin degradation. PARP-1 is involved in DNA repair, and overs timulation as a result of DNA damage is thought to lead to cell deat h through metabolic depletion. Interestingly, PARP-1 is actually a target for ca spase cleavage in caspase-dependent apoptosis this is thought to be a cell strategy to re duce metabolic depletion and increase the energy available to effect orderly apoptosis through the caspase-d ependent pathway (Wenzel et al., 2005). Hallmarks of apoptosis include reduced cel l size and the appearance of bubble-like blebs on the surface of the plasma memb rane. The nuclear chromatin breaks down, leading to a diagnostic DNA laddering mor phology, and the mitochondria begin to lose integrity, releasing cytochrome c into the cy toplasm. Eventually the entire cell breaks down into small, membranous vesicles, whic h are finally engulfed by macrophages and dendritic cells that recognize the apopto tic cells (Vaux, 1993; Reme et al., 1998). The link between apoptosis and RP-related photoreceptor cell death has been well established (Reme et al., 1998). In 1994, Portera-Cailliau and coworkers investigated the mechanism of cell death in the three mouse models of retinitis pigmentosa: the rd mouse, which contains a defect in the rod cGMP phosphodiesterase gene, the rds mouse, which contains a defect in the st ructural gene peripherin, and in mice containing a Q334 termination mutation in the rhodopsin gene (P ortera-Cailliau et al ., 1994). In each mouse, DNA fragmentation, a hallmark of apoptos is was seen in the photoreceptors. In concurrent studies, researcher s showed similar evidence for apoptotic cell death in RCS

PAGE 27

14 rats, which carry defects in the ability of th e RPE to phagocytose ROS discs (Tso et al., 1994). In 1996, another study showed apoptotic cell death as the ultimate fate of rod photoreceptors in mice engineered to contain a P23H rhodopsin mutation, and a correlation was noted between increasing amounts of apoptosis and decreased ERG findings in the same animals (Naash et al., 1996). A hallmark of retinal degeneration resulting from RP is that after a certain number of rod photoreceptors have died, normal rods and cones that do not express mutant opsins begin to die as well. In 1993, chimeras be tween normal mice and mice carrying a P347S mutant rhodopsin transgene were create d in order to address this issue. In situ hybridization assays confirmed that the chimer ic retinas were made up of mixtures of adjacent mutant and non-mutant rod photorecep tor cells, while histological examination revealed that cell death was occurring simu ltaneously in both mutant and non-mutant sections of the retina (Huang et al., 1993). Th is, taken together with the observations of cone cell death resulting from a rod defect, ha s led to the belief that photoreceptor cells secrete survival factors necessary to the survival of their neighbors, and thus the deaths of mutant rod photoreceptors can have de leterious effects upon neighboring cones and non-mutant rods (Rosenfeld and Dryja, 1995; Bredesen et al., 2005). This theory is supported by experiments in which various ce ll survival factors (e.g. BDNF, CNTF, and NGF) were injected intravitreally into both na turally occurring and ar tificial models of retinal degeneration. These cellular factors sl owed the degenerations in several of the mouse lines (LaVail et al., 1998). Animal Models of Retinitis Pigmentosa The development of animal models has been vital to the study of retinitis pigmentosa. Initial lines consisted of natura lly occurring mutants that showed early onset

PAGE 28

15 photoreceptor degeneration, the first of wh ich was the rd mouse, identified in 1966, which is currently known to involve a spontan eous mutation in the beta subunit of rod cGMP phosphodiesterase (Keeler, 1966). Many other naturally occu rring mouse mutants have since been isolated including, among others, mouse models of Purkinje cell degeneration, the rds mouse, which as mentioned before contains defects in its peripherin gene, and models of Leber Congenital Amauro sis and cone photoreceptor function loss (Chang et al., 2002; Pang et al., 2005). In a ddition to the naturally occurring mutations, targeted gene disruption tec hniques have enabled research ers to create and study mouse lines which lack other integral phototransduc tion proteins, such as arrestin and the Rpe65 protein. With mutations in the rhodopsin gene account ing for over 10% of all incidences of RP, it is not surprising that a large number of mouse models involving target rhodopsin deletions and mutations have been designe d. In 1992, a mouse model was created to express a either a wild-type human rhodopsin transgene or a human rhodopsin transgene carrying the P23H mutation. Three lines expr essing the mutant transgene were created, each of which expressed the mutant rhodopsin at a different level, and although all three lines exhibited retinal degenera tion, it was observed that the rate of retinal degeneration was directly proportional to th e expression level of the mutant transgene. This model was also important in that it demonstrated that ADRP could arise from a single point mutation in a single gene (Olsson et al., 1992). Two lines expressing the normal human rhodop sin transgene were also created in the studies by Olsson and coworkers, one of which expressed rhodopsin at levels comparable to non-transgenic mice, and one that expressed five times as much rhodopsin.

PAGE 29

16 Intriguingly, the line that exhibited overexpr ession of the wild-type human transgene also showed a retinal degeneration, even t hough there was no mutation involved, which suggests that overexpression of rhodopsin can le ad to similar trafficking and aggregation problems as expression of mutant rhodopsins at normal levels. Subsequent studies on this mouse showed accumulation of mutant rhodopsin at abnormal sites in the rod photoreceptor cells (Roof et al., 1994). Another mouse model involved germline in sertion of a mouse rhodopsin transgene that was mutated to contain two silent RFLP s and three amino acid substitutions, one of which was the P23H mutation, and the othe r two being nearby non ADRP-associated amino acid substitutions that were included to provide an epitope tag. This mouse model, termed the VPP model because of the three amino acid substitutions that were introduced, showed normal expression levels of a mixture of w ild-type and mutant rhodopsin, and exhibited a slow photoreceptor de generation that mimicked that seen in human patients. Study of this mouse line reve aled that dark-reared mutant animals had significantly reduced rates of photoreceptor dege neration, with a threefold decrease in the appearance of apoptotic photorecep tor cells when compared to mutant siblings raised in twelve hour cyclic light. This suggests that light activation of mutant rhodopsin is a key causative agent in rhodopsin-mediated RP (Naas h et al., 1993; Naash et al., 1996; Goto et al., 1995; Goto et al., 1996). The exacerbation of RP symptoms by intense light exposure is a well-documented phenomenon that is crucia l to this work and will be discussed in more detail in future chapters. Furthe r work with the VPP mouse line involving immunohistochemical tracking of the mutant rhodopsin molecules revealed that the mutant opsin was correctly synthesized a nd localized, that there was normal outer

PAGE 30

17 segment disc shedding, but that there were de fective and disorganized basal discs at the connecting cilium, the site of outer segm ent disc morphogenesis, providing further evidence of a trafficking/disc morphogenesi s defect associated with some rhodopsin mutations (Wu et al., 1998; Liu et al., 1997). An interesting and very useful model of ADRP was created in 1997 via a targeted disruption of the entire rhodopsin gene. This model contained a Pol2: neomycin insertion in exon II of the endogenous mouse rhodopsin gene. Subsequent mouse lines were bred to contain either one disrupted rhodopsin gene (hemizygous null, or mRho+/mice), or two disrupted rhodopsin genes (homozygous nul l, or mRho-/mice). Hemizygous null mice showed some subcellular disorganization of the ROS, as well as shortening of the outer segments in older mice when compared to wild-type mice, but little ERG reduction or other signs of disease. Homozygous null mice, however which completely lacked expression of mouse rhodopsin, never form ed rod outer segments, showed no ERG response at eight weeks of age, and showed loss of cone photoreceptors by three months of age (Humphries et al., 1997). These mice, along with a separate rhodopsin knockout model that was subsequently developed (Lem et al., 1999), were important for several reasons. Chief among these was their useful ness at providing a ge netic background that allowed researchers to breed hemizygous null (mrho+/-) rhodopsin lin es upon which they could express rhodopsin transgenes at an allelic ratio identi cal to that seen in human patients with ADRP (i.e., one mutant copy and one wild-type copy). Further mouse lines expressing rhodopsin mutations were created in the 1990s, including one containing a Q344Ter muta tion which caused synthesis of a rhodopsin protein with an abnormally short carboxyl terminus. This mutant line exhibited

PAGE 31

18 rhodopsin accumulation in the plasma membranes of rod photoreceptor cells, demonstrating, as mentioned before, that th e carboxy terminal of rhodopsin is necessary for efficient and proper traffi cking of rhodopsin (Sung et al ., 1994). In 1998, Li et al. reported creation of mouse models of ADR P expressing human rhodopsin transgenes with either a T17M mutation (a class I mutati on that abolishes the glycosylation site at Arg15), or a P347S mutation (a class II mutati on). Both lines exhibited progressive loss of ERG response and decreasi ng thickness of the ONL. Inte restingly, treatment with Vitamin A supplementation led to partial rescue of the T17M-medaited RP, but not of the P347S animals, providing fu rther evidence that the sepa rate classes of rhodopsin mutations lead to retinitis pigmentosa via different biochemical or morphological pathways (Li et al., 1998). Gene Therapy for Retinitis Pigmentosa It has long been the goa l of medical scientists to effect therapies that act in targeted cells at the level of the gene. In cases wh ere disease symptoms are created by missing or non-functional gene products, gene replacemen t strategies could be used to introduce healthy gene products into tissues of interest In cases where there is a toxic gain of function caused by a mutant protein, therap eutic techniques designed to specifically abolish expression of the mutant gene encodi ng that protein would be warranted. Both types of gene therapy have been successfully demonstrated in animal models of retinitis pigmentosa (Hauswirth and Le win, 2000; Hauswirth et al., 2004). The first efforts at gene replacement i nvolved creating transgenic mice expressing the corrective gene and crossing these with mice showing a retinal defect. In the case of the naturally occurring rd mice, which undergo retinal degeneration due to a defect in the beta subunit of the rod cG MP phosphodiesterase gene ( PDE), the introduction of a

PAGE 32

19 functional human gene replacement through th is method restored normal photoreceptor morphology and function (Lem et al., 1992). In another study, these transgenic techniques were used to express a huma n rhodopsin transgene on the homozygous null rhodopsin (mRho-/-) mouse model of retinal degeneration. Expressi on of the functional human rhodopsin in this case resulted in re scue of photoreceptor ultrastructure and ERG response, and demonstrated th e ability of the groups target ing construct to express the transgene at therapeutic leve ls (McNally et al., 1999). Following up on the work of Lem and colle agues with the rd mouse, two groups have reported rescue of the photoreceptors th rough viral-mediated delivery of functional PDE genes. In 1996, researchers used replica tion-deficient adenovira l vectors to deliver a murine cDNA expressing the wild-type PDE gene into the subret inal space of rd mice (Bennett et al., 1996). This therapy re sulted in expression of functional PDE that resulted in a six week delay in photorecep tor degeneration. A similar strategy that utilized intravitreal injection an adeno-asso ciated viral vector delivering a wild-type PDE gene resulted in increased survival of photoreceptors and an increased ERG response in rd mice receiving the therapy (J omary et al., 1997). Gene replacement has also been effective in treati ng the retinal degenera tion seen in the naturally occurring rds mouse, which as mentioned before suffers fr om defects in the ROS structural protein, peripherin. Subretinal injecti on an adeno-associated viral vector engineered to express a wild-type peripherin gene result ed in restoration of ROS u ltrastructure and function in these mice though the result was temporary (Ali et al., 2000). Long term rescue was also seen with gene replacement designed to delive r a functional RPE65 gene to the retinas of

PAGE 33

20 the rd12 naturally occurring mouse model of the recessive retinal disease, Leber Congenital Amaurosis (Pang et al., 2005). Gene therapy designed to alleviate dominant retina l disease by suppressing expression of mutant genes has also shown efficacy in mouse models. One such study involved a rat model of retinal disease that was engineered to e xpress a P23H mutant rhodopsin transgene. Subretinal delivery of adeno-associated viral vectors engineered to deliver catalytic RNAs, or ribozymes, desi gned to selectively degrade the mutant transgene, while sparing the endogenous wild-type rat rhodopsin, was shown to result in substantial, long-term rescue of photoreceptor st ructure and fuction (Lewin et al., 1998) This rescue, which was shown to be effective to eight months age in these animals, as well as in treating animals that had already entered late stag es of the disease (LaVail et al., 2000), shows promise for the treatment of other forms of autosomal dominant disease. Ribozymes Riboymes are RNA molecules with the ability to catalyze the cleavage and joining of RNA. They were initially discovere d by Altman and Cech, who described the catalytic activity of the RNA component of RNaseP and the group I introns, respectively (Cech, 1988a; Cech, 1988b; Guerrier-Takada et al., 1983; Guerrier-Takada and Altman, 1984). Initial Group I intron cata lysis was seen to occur in cis but it was soon discovered that one could liberat e the catalytic structure of th e RNA from its substrate to generate a ribozyme with the abilit y to cleave target molecules in trans Shortly after the work of Altman and Cech, smaller catalytic structures were discovered in the sequences of certain plant pathogens which undergo site-specific, selfcatalyzed RNA cleavage as a part of their replicative process (Buzayan et al., 1988;

PAGE 34

21 Haseloff and Gerlach, 1988). These are the hammerhead and hairpin ribozymes, which are receiving much of the present attention for applications in gene therapy (Cech, 1988b; Sigurdsson and Eckstein, 1995; Phylactou et al., 1998; Citti and Rainaldi, 2005). Hammerhead ribozymes consist of three base paired stems surrounding a central catalytic core of fifteen conserved nucleotides, eleven of which are necessary for catalytic activity. The crystal structure of the hammerhead ri bozyme reveals noncanonical base pairing within the catalytic core and a magnesium binding s ite that is distal to the site of catalysis (Scott et al., 1995). The helices ar e designated I, II, and III; th e first and last helices (I and III) form via base-pairing with the targ et molecule while the middle helix (II) is responsible for stabi lizing the catalytic co re structure (Figure 1-4) (Sigurdsson and Eckstein, 1995; Pierce and Ruffner, 1998). Ha irpin ribozymes have two 5 stretches of sequence which base pair with their RNA ta rgets, forming helices I and II, followed by two downstream helices (III and IV ), which interact with one another and the substrate to comprise the catalytic core (Figure 1-5) (Jos eph et al., 1993; Berzal-H erranz et al., 1993). The hairpin ribozyme has been crystallized eith er as a protein complex or as a pure RNA, revealing internal base pairing within and e ssential interactions be tween the loop regions of the secondary structure (Rupert and Fe rre-D'Amare, 2001; Grum-Tokars et al., 2003). No magnesium is seen at the catalytic site. Mechanistic Description Ribozymes achieve their cleavage by binding to their RNA target via complementary base pairing with sequences flanking the cleavage site, folding into a specific catalytic conformation, catalyzing th e hydrolysis of the 5 phosphodiester bond at the cleavage site, and dissoci ating from the resultant clea vage products (Figure 1-6). The reaction generates two RNA fragments w ith either 5 hydroxyl or 2 cyclic

PAGE 35

22 phosphate moieties. Divalent cations, such as magnesium, are thought to aid in the folding of the ribozyme into th e proper catalytic structure. These cleavage reactions are sequence specific. The sequences flanking the catalytic core of the ribozyme must be able to form helices with sequences flanking the target cleavage site, and hydrolysis will only occur at a site containing certain combinations of nucleotides. Hammerhead ri bozymes will cleave 3 of a triplet sequence of NUX, where N is any nucle otide, U is a uridine, a nd X is any nucleotide but a Figure 1-6. Generic structur e of a hammerhead ribozyme. The requisite NUX cleavage triplet is shown in red, with the positi on of cleavage indicated by an arrow. Green nucleotides indicate conserve d ribozyme sequences necessary for catalysis.

PAGE 36

23 Figure 1-7. Structure of a specifi c hairpin ribozyme. Cleavage is indicated by an arrow. Blue nucleotides indicate conserve d ribozyme sequences necessary for catalysis guanosine. The hairpin ribozymes show greater sequence constraint, and can be engineered to cleave 5 of the G in the se quence 5-NBNGUC, where N is any nucleotide and B is G, C, or U. This theoretically allows a wide range of cleavage possibilities for therapeutic applicatio ns of ribozymes. In addition, recent experiments using larg e combinatorial libraries of ribozyme candidate molecules have shown that it is possible to achieve cleavage of almost any RNA target (Nieuwlandt, 2000; Piganeau et al ., 2001). Most importantly, targets that deviate from a ribozymes optimal cleavage sequence, especially in the area of the catalytic site, undergo reduced cleavage or none at all; even a single nucleotide difference in critical areas is enough to completely abolish cleavage activity (Werner and

PAGE 37

24 Uhlenbeck, 1995). This allows researchers to design ribozymes that are able to cleave mutant RNA transcripts while leaving wild-type messages able to integrate into the target cell genome, which makes them attractive as therapeutic effectors for the treatment of autosomal dominant disease. Figure 1-8. Ribozyme cleavage in trans A hammerhead ribozyme (black) is shown annealing to its target mRNA (red) through complementary base-pairing, cleaving it, and releasing the 5 and 3 cleavage products. Ribozymes are truly catalytic, able to catalyze the cleavage of many successive target molecules. Figure courtesy of Dr. Lynn C. Shaw. AAV Recombinant adeno-associated viral vector s offer a number of features ideal for gene therapy, including the ability to inf ect a wide variety of both non-dividing and dividing cell types (Loiler et al., 2003; Flotte, 2005; Fisher et al., 1997), the lack of pathogenicity absence of immune or inflamma tory response in transduced cells (Bennett, 2003; Hernandez et al., 1999; Conrad et al., 1996), and the ability to achieve long-term

PAGE 38

25 expression of therapeutic gene s (Guy et al., 1999). Adeno-as sociated viruses (AAV) are human parvoviruses composed of a 4.7 kb si ngle-stranded linear DNA genome packaged in a capsid composed of three structural prot eins. The genome is a model of efficiency, consisting of overlapping open reading frames (orfs) that utilize al ternative splicing and variation in translation initiation sites to e xpress several proteins from a relatively small genome. The genes encoded by the orfs (and their gene products) are termed Rep (for replication-associated proteins ) and Cap (for capsid-associated proteins). The Rep genes produce four protein products, termed Rep78, Rep68, Rep52, and Rep40, with the numbers indicating the respective sizes in ki lodaltons of each protein. The Cap genes produce three capsid-forming gene products of 87, 73, and 62 kilodaltons in size, termed VP1, VP2 and VP3, respectively, which are presen t in the viral capsid at a ratio of 1:1:10 (VP1:VP2:VP3). These genes are present in the genome between two inverted terminal repeats (TRs) consisting of 145 bases, which c ontain the viral origin of replication and are necessary for viral packaging and integratio n. The entire genome is packaged within a nonenveloped, icosahedral capsid th at is around 20nm in diameter. AAV belong to the class of viruses known as dependoviruses, because they require co-infection of a helper virus to produce produc tive infection of their own. In the latent phase of the virus it is found preferentially integr ated into a particular site on human chromosome 19 (Kotin et al., 1990). In this phase, the products of the Rep gene have actually been found to be inhibitory to AAV re plication. The latent phase of infection will persist until the infected cell is rendered permissive, which can result from certain cytotoxic exposures (including heat shock, ir radiation, cycloheximide treatment), or until the cell receives a secondary he lper virus infection, usually with adenovirus, herpesvirus,

PAGE 39

26 cytomegalovirus, or poxvirus. This co-i nfection is accompanied by the expression of helper gene products from the new viru s that help AAV achieve productive, lytic infection. In the case of secondary adenovira l infection, expression of the early set of genes (E1a, E1b, E2a, and E4) have been s hown to provide helper functions such as transcriptional transa ctivation and assistan ce in AAV mRNA accumulation and transport. The ability of AAV to achieve stable, long term integration into target genomes, coupled with a lack of association with inser tional oncogenesis, led to the initial interest in its use as a gene therapy vector. It has since become understood that integration of the virus involves interactions between the chromosomal target site, the AAV TR structures, and the Rep gene products (McLaughlin et al ., 1988; Kotin et al., 1990; Surosky et al., 1997; Weitzman et al., 1994). Since the reco mbinant AAV used for gene therapy lack the Rep gene, integration seen with these vector s is random when it is seen at all. This finding is of concern, as it raises the possibi lity of insertional activation of a protooncogene, but these concerns are somewhat alleviated by a lack of reproducible tumorigenesis in animals treated with rAAV therapy vectors (Donsante et al., 2001) and by the fact that the rAAV genomes are generally thought to persist with in the target cells in an episomal fashion (Schnepp et al., 2005). Production of AAV Vectors To make recombinant AAV vectors, therap eutic sequences are inserted into a packaging plasmid containing a cloning site and regulatory sequences flanked by the AAV TR sequences that are necessary and sufficient for packaging as recombinant AAV (McLaughlin et al., 1988). This construct is then co-transfected into HK 293 cells along with a helper plasmid that encodes both the Rep and Cap AAV genes and the adenovirus helper gene products. By this method it is possible to generate replication deficient,

PAGE 40

27 recombinant AAV containing up to 4.5 kb of therapeutic DNA sequence bounded by the AAV TRs (Hermonat et al., 1997; Zhou and Muzy czka, 1998). Providing helper gene products on a plasmid rather than by adenoviral co-infection eliminates the possibility of adenoviral contamination in the final vira l preparation. These recombinant AAV (rAAV) vectors are able to infect target ce lls and, with the in clusion of strong cis -acting promoter sequences, generate good levels of expression of the therapeutic gene products of interest for extended periods of time. Continued adva nces in the production and purification of rAAV, including the use of high-volume cell factories and effici ent, high-throughput column purification methods, have led to cons istently high yields of highly purified viral vectors (Kapturczak et al., 2001; Potter et al., 2002; Bloui n et al., 2004). Drawbacks associated with first-generation therapeu tic rAAV vectors included a delay in the expression of therapeutic genes for as mu ch as two weeks to a month, and limited transduction of certain cell types. Both issues have been greatly ove rcome to large extent through the use of pseudotyped rAAV vectors. Different serotypes of adeno-associated vi rus can show great diversity in the amino acid composition of their capsid proteins (G rimm and Kay, 2003). A lthough the original viral delivery vectors were developed from serotype 2, many more serotypes have been found to exist(Gao et al., 2002; Gao et al., 2005). The capsid dive rgence exhibited by these serotypes cause them to have profound differences in th e efficiency and speed with which they are able to transduce various cell types (Grimm and Kay, 2003; Auricchio et al., 2001; Hildinger et al., 2001; Rabinowitz et al., 2002). By using the AAV type 2 packaging vector in combination with helper plasmids providing capsid proteins from the various other serotypes it is possible to create rAAV pseudotype vectors containing the

PAGE 41

28 gene of interest and its ancillary sequen ces flanked by AAV2 TRs and packaged into capsids derived from one of the other serotypes (Grimm, 200 2; Burger et al., 2004; Choi et al., 2005; Gao et al., 2005). Vectors created in this ma nner are termed pseudotype 2/*, where the first number indicates that the packaging vector was derived from AAV serotype 2 while the asterisk represents the serotype number of the capsid proteins (thus, serotype 2 sequences packaged into a capsi d derived from AAV serotype 4 would be referred to as pseudotype 2/4). Table 1-1 summarizes the cell tropisms of the various coat protein serotypes. Table 1-1. Tissue tropism and site of isolati on of the various AAV serotypes. Bold type indicates high levels of expression, wh ile bold italic t ype denotes highest expression of all AAV serotypes. Adap ted from Hildinger and Aurricchio, 2004. Serotype Isolated In Tissue/Cell Tropsim AAV1 Cell line Muscle, eye, liver, lung AAV2 Cell line Muscle, brain, liver, eye AAV3 Cell line Not determined AAV4 Cell line Brain AAV5 Human lesion Brain, muscle, liver, lung, eye AAV6 Cell line Muscle, eye, liver, lung AAV7 Monkey Muscle, liver AAV8 Monkey Liver The usefulness of pseudotyped delivery ve ctors becomes apparent when examining the different transduction efficiencies of AAV 2/1, AAV 2/2 and AAV 2/5 when used to deliver a green fluorescent marker protein (GFP) to the retina. GFP expression is seen in both the retinal pigmented epithelium and in photoreceptor cells of retinas receiving subretinal injection of ei ther AAV 2/2 or AAV 2/5. However, the use of AAV 2/5

PAGE 42

29 resulted in a 400-fold increase in the number of transgene-expressing cells, as compared to transduction by AAV 2/2, and the number of viral genome copies per eye was thirty times higher. Additionally, AAV2/5 showed a faster onset of tran sgene expression, and was shown to achieve higher levels of tr ansgene expression. AAV2/1, in contrast, transduces cells of the retinal pigmented ep ithelium almost excl usively, being fifteen times more efficient than AAV2 and achievi ng higher levels of transgene expression (Yang et al., 2002). Clearly the selection of the proper rA AV pseudotype is of great importance in designing rAAV-mediated ge ne therapies (Dincu lescu et al., 2005; Auricchio and Rolling, 2005). Recent advances in this area have focused on directly altering individual capsid epitope s to further enhance and refi ne the selective tropism of these vector systems (Opie et al., 2003; Warrington, Jr. et al ., 2004; Gigout et al., 2005; Muzyczka and Warrington, Jr., 2005). AAV and Retinal Gene Therapy Adeno-associated vectors ar e ideally suited to deliver therapeutic DNA sequences to retinal cells. These recombinant viruses are able to achieve long term expression in retinal photoreceptor cells, retinal ganglion cells, and cells of th e retinal pigmented epithelium (Dinculescu et al., 2005; Guy et al., 1999; Flannery et al., 1997). Transduction of retinal cells by rAAV is achie ved with little or no toxic or immunogenic side-effects (Bennett, 2003). Finally, the use of the various adeno-associated viral pseudotypes with their selective tropism allows researchers to selectively target specific retinal cell types for transduction. Additionally, there are many characteristics of the retina itself that make it attractive for rAAV-mediated gene therapy. In jections into the subretinal space can be routinely performed with great speed and precision, and when properly performed lead to

PAGE 43

30 little or damage of the injected tissues (Timmers et al., 2001). Also, the eye itself is held to a certain extent to be an immune privileged site, a char acteristic which enhances the lack of an immune response to rAAV and the th erapeutic proteins or sequences that it delivers (Streilein et al., 1992; Sonoda and St reilein, 1992). Finally, the presence of two eyes in animal models ensures that there is always a built-in cont rol available to the researcher in every experimental animal, as one eye can simply remain uninjected and then be compared to the c ontralateral, treated eye. As mentioned before, rAAV gene delivery ve ctors have been used to treat wide variety of retinal disease, including th e rd, rds, rd12 mouse models of retinal degeneration, and mouse and rat models of P23H mutant rhodopsin-mediated ADRP. These vectors have also been successful com ponents of therapies designed to treat the lysosomal storage defects in the retinal pigm ented epithelium seen in the MPVII (Bosch et al., 2000; Hennig et al., 2004) and MP IIIB (Fu et al., 2002) mice, MerTK deficiency in the Royal College of Surgeons (RCS) rat (Smith et al., 2003), retinal degeneration in the naturally occurring RPE65-/Briard dog an alogue of the rd12 mouse (Acland et al., 2005; Acland et al., 2001), and retinal de generation in a mouse model of X-linked juvenile retinoschisis (Min et al., 2005). rAAV-delivered neurotroph ic factors such as GDNF (Wu et al., 2004), CNTF (Adamus et al ., 2003; Liang et al., 2001), and FGF (Lau et al., 2000; Lau and Flannery, 2003) have also been used to alleviat e retinal degeneration in various animal models of retinal disease. Finally, these vectors have been used to deliver anti-angiogenic molecules such as PE DF and angiostatin for the treatment of ocular neovascularization in animal mode ls of diabetic retinopathy, retinopathy of prmaturity, and the wet form of age-relate d macular degeneration (Raisler et al., 2002;

PAGE 44

31 Auricchio et al., 2002). Although rAAV has been shown to be effective at delivering therapy to a wide variety of organs and ti ssues, it is clearly the vector of choice for transduction of retinal cells. Project The following chapters describe experime nts designed to characterize and treat a mouse model of retinitis pigm entosa that expresses a human rhodopsin transgene with a tyrosine to methionine mutation at the 17th amino acid of the protein. I will discuss breedings designed to expr ess this transgene on a hemizygous null mouse rhodopsin background to more closely imitate the ge notype found in human RP patients. The progression of retinitis pigmentosa in this mouse line will be docmumented, as will the development of a ribozyme-mediated gene ther apy strategy to treat the mice. Finally I will discuss the results of a pilot study involving subretinal injection of rAAV expressing these therapeutic ribozymes that led to some interesting observations concerning the light sensitivity of the hT17M transgenic line, a nd strategies to circumvent this issue.

PAGE 45

32 CHAPTER 2 CHARACTERIZATION OF A NOV EL MOUSE MODEL OF RETINITIS PIGMENTOSA Introduction Human patients with the T17M rhodopsin mutation exhibit classic symptoms of autosomal dominant retinitis pigmentosa. Affected individuals report loss of peripheral and vision and night blindness, accompanied by decreased ERG response and eventually culminating in loss of centra l vision. Postmortem examination of eyes from affected patients reveals heavy deposits of bone spicul e-like pigmentation in the inferior retina that are accompanied by severe loss of phot oreceptors (Li et al., 1994). Interestingly, photoreceptors of the superior retina are re latively well-preserved, a feature that is uncommon in RP patients. These mutants are in triguing in that they reduce or abolish glycosylation at position 15 of the gene (Kaushal et al., 1994). As inhibition of glycosylation by tunicamycin cau ses defects in ROS morphology in frogs (Fliesler et al., 1985), this raises the possibility that T17M rhodopsin may not be correctly incorporated into the outer segment discs. In 1998, Li et al. described the creat ion of a mouse model of T17M rhodopsinmediated ADRP. These transgenic mice were created using a 17 kilobase human geneome fragment that included the rhodopsin gene flanked by 4.8 kilobases of upstream and 6.2 kilobases of downstream sequence. The gene contained a single nucleotide substitiution to change codon 17 from threonine to methionine. ERG analysis of these mutant animals demonstrated that they underwent complete photoreceptor degeneration

PAGE 46

33 by eight months of age. Subsequent studies with this animal model demonstrated a therapeutic effect of vitamin A supplementation, which reduced the rate of decline in aand b-wave ERG amplitudes. This partial rescue was accompanied by an increase in photoreceptor survival in the outer nuclear la yer. Parallel experiments involving a mouse line containing a P347S rhodopsin mutation di d not show vitamin A rescue, providing evidence for different paths of retinal dege neration between the two mutants (Li et al., 1998). We obtained this line from Dr. Li with the goal of developing a ribozyme therapy for T17M-mediated ADRP. It was impossible to create mice containing two copies of the human transgene (the cross led to embr yonic lethality), with the result that mice containing a 1:1 genotypic ratio of mutant to wild-type rhodo psin could not be created on a wild-type rhodopsin background. In order to establish a mouse m odel that underwent a more rapid degeneration, as well as one in which the genotype of th e model more closely mimicked the naturally occurring disease pheno type (i.e. one mutant allele and one wildtype allele), we decided to breed our hT17M animals onto a mouse rhodopsin knockout (mrho-/-) background (Lem et al., 1999). The resultant animals, which would contain one copy of mutant hT17M human rhodopsin transgene and no copies of wild-type mouse rhodopsin, could then be crossed to wild-type (mrho+/+) mi ce to produce animals that are heterozygous null at the mouse rhodops in locus (mrho+/-), of which half would also contain the hT17M muta nt rhodopsin transgene. In this chapter I describe the creation and analysis of an mrho+/-; hT17M mouse model of retinal disease. I will discuss the breeding and PCR genotyping of these

PAGE 47

34 animals, and will document the visual degeneration in this line as assayed by ERG, funduscopy, and histological examination. Materials and Methods DNA Oligonucleotides DNA oligonucleotides were ordered from Invitrogen (Palo Alto, CA), at a 50 nmole scale of synthesis. Oligonucleotides were desalte d, but otherwise unpurified by the manufacturer. The sequences were as follows: hExon 2 sense primer: 5-GAGTGCACCCTCCTTAGGCA-3 hExon 2 antisense primer: 5-TCCTGACTGGAGGACCCTAC-3 mRHO Exon 1 sense primer: 5 -CCAAGCAGCCTTGGTCTCTGTCTA-3 mRHO Exon 1 antisense primer : 5-TGTGCGCAGCTTCTTGTGGCT-3 Neo sense primer: 5-AGGATCTCCTGTCATCTCACCTTGCTCCTG-3 Neo antisense primer: 5-AAGAAC TCGTCAAGAAGGCGATAGAAGGCG-3 Isolation of Genomic DNA Genomic DNA was isolated from 0.5 cm tail snips from candidate mice using the Quiagen DNeasy Kit (Quaigen Inc, Valencia, CA), as per the manufactur ers instru ctions. In brief, the tails were digested overnight at 50oC in digestion buffer supplemented with Proteinase K. The resultant suspensi on was diluted, bound to the DNeasy column, washed twice, and eluted twi ce with 100 microliters of elut ion buffer, for a final volume of 200 microliters of genomic DNA suspension. Three microliters of this genomic DNA were used for PCR analysis, the rest was stored at either -20o or 4oC.

PAGE 48

35 PCR Analysis of Genomic DNA Endogenous mouse rhodopsin genotypes were determined using mRHO Exon 1 and Neo primers. Genomic DNA from mice that are wild-type at the mouse rhodopsin locus (mrho+/+) produce products with the mRHO Exon 1 primers, which amplify a 270 bp fragment, while the Neo primers will produce no product, as there is no Neo knockout cassette to amplify. Figure 2-1. Agarose gel electrophoresis of PCR reactions performed to determine the mouse rhodopsin genotype (left gel) or c onfirm the presence or absence of the hT17M transgene (right gel). Lane B shows amplification of the 490 bp Neo and 270 bp mRho PCR products from a mRho+ /mouse. Lanes D and J show amplification of the 290 bp human rhodopsin PCR product. Lanes E I are negative for human rhodopsin. Lanes A and C are size standards. Conversely, DNA from mice that are hom ozygous for a knockout at the mouse rhodopsin locus (mrho-/-) will produce a 490 bp fragment with the Neo primers, while producing no product when amplified with the mRHO Exon 1 primers. DNA from mice that are heterozygous null at the mouse r hodopsin locus (mrho+/-) are able to produce both fragments. Presence of the human T17M rhodopsin transgene was determined by PCR amplification of genomic DNA with the hExon 2 primers, which amplify a 290 bp fragment in the presence of the transgene (Figure 2-1).

PAGE 49

36 PCR reactions were set up in 50 l volumes as follows: 3 l (~1 g) genomic DNA, 5 l 10X PCR Buffer (Sigma, St. Louis, MO), 0.5 l 100mM dNTP mix (Sigma), 0.25 l 100mM sense primer, 0.25 l 100mM antisense primer, 0.5 l (1 Unit) Taq Polymerase (Promega, San Luis, CA), and 40.5 ml dH20. PCR was then carried out as follows: 1) 95oC for 10 minutes; 2) 95 oC for 45 seconds; 3) 54 oC for 45 seconds; 4) 72 oC for 60 seconds; 5) repeat steps 2-4 for 28 cycles; 6) 72 oC for 10 minutes. PCR amplifications were performed on a Gene Amp PCR System 2700 thermocycler (Applied Biosystems, Foster City, CA). The presence or absence of the PCR product of interest was verified by agarose gel electrophoresis. Electroretinography Mice were dark adapted overnight. A ll subsequent ERG procedures were performed under dim red light (wavelength >600nm), which does not activate rhodopsin. Mice were anesthetized with IP injecti ons of xylazine (13mg/kg) and ketamine (87mg/kg) (Phoenix Pharmaceuticals, St. Joseph, MO). The mouse corneas were anesthetized with a drop of 0.5% proparacaine HCl (Akorn, Buffalo Grove, IL), and dilated with a drop of 2.5% phenylephrine HC l (Akorn). Measurement electrodes tipped with gold wire loops were placed upon both corneas with a drop of 2.5% hypromellose (Akorn) to maintain electrode contact and co rneal hydration. A re ference electrode was placed subcutaneously in the center of th e lower scalp of the mouse, and a ground electrode was placed subcutaneously in the hind leg. The mouse rested on a homemade sliding platform that kept the anim al at a constant temperature of 37o C. The animal was positioned so that its entire head rested inside of the Ganzfeld (full-field) illumination dome of a UTAS-E 2000 Visual Electrodi agnostic System (LKC Technologies, Inc.,

PAGE 50

37 Gaithersburg, MD), as shown in Figure 2-2. Fu ll-field scotopic ERGs were measured by 10 msec flashes at an intensity of 0.9 and and 1.9 log cd m-2 at 1 minute intervals. Figure 2-2. Electroretinographic apparatus. The anesthetized animal rests on a warming tray that can slide in and out of a Ga nzfeld illumination dome. Electrodes are held in place with homemad e articulable plastic arms. Figure 2-3. An example of an ERG tracing. The X axis of this trace represents the elapsed time of the signal, while the Y axis shows the intensity of the response. The amplitudes used to calculate a and b-waves are shown

PAGE 51

38 Responses were amplified at a gain of 4,000, filtered between 0.3 to 500Hz and digitized at a rate of 2,000 Hz on two channels. Five responses were averaged at each intensity. The wave traces analyzed using UTAS-E 2000 software package (LKC Technologies, Inc.). A-waves were measured from the ba seline to the peak in the cornea-negative direction; b-waves were measured from th e cornea-negative peak to the major corneapositive peak (Figure 2-3). Funduscopy Mice were anesthetized, and their corneas anesthetized and dilated as described above. Fundus photography was performed by Dr. Quihong Li with a Kowa Genesis hand held fundus camera (Kowa Company, Lt d., Tokyo, Japan) focused through a Volk Super 66 Stereo Fundus Lens (K eeler, Berkshire, England). Two pictures of each eye were generally taken to ensure a properly focused image. Histology Mice were euthanized by an overdose of Isoflurane (Abbot, North Chicago, IL) followed by cervical dislocation. Eyes were quickly removed and fixed overnight in 4% paraformaldehyde and transferred to a so lution of phosbate buffered saline (PBS) (137mM NaCl, 10mM PO4, 2.7Mm KCl, pH 7.4). Histological sectioning and subsequent H&E staini ng was performed by UF Histol ogy core technicians. This program of sectioning resulted in twelve serial sections through the entire eye. Sections that contained the optic nerve were then photographed at 20X power using a Zeiss Axiophot Z microscope equipped with a Sony DXC-970MD 3CCD Color Vid Camera and an MCID Elite Stage, utilizing MCID (Imaging Research, Inc., Ontario, Canada) Analysis Software (Imaging Research, Inc.) th at stitched individual images together to create a tile-field composite image of the enti re retina. The images were viewed with

PAGE 52

39 Adobe Photoshop, and a radial template overlay was used to define six equivalent and equally spaced regions of the retina. From each of these areas, the mean value from three separate ONL counts was determined, and thes e regional counts were then averaged to generate a value that represented the ONL thic kness of the retina. St atistical comparisons between the transgenic and nontransgenic values were perf ormed to generate P values using the paired, one-tailed Students t-test feature of Exel spreadsheet software (Microsoft, Redmond, WA). Results Breeding Founder and Experimental Mice A breeding pair of mrho+/mice cont aining the hT17M transgene was kindly provided by Dr. Tiansen Li in 2002. However, this line ha d subsequently undergone a mating crisis due to poor fecundity and ha d become poorly characterized by the time these studies were initiated. It was uncertain which members of the colony actually contained the hT17M transgene, and of wh at respective mouse rhodopsin genotype those animals were. It was necessary to scr een the entire colony by PCR for the hT17M transgene and then work from there. Seve ral mice were found that contained the hT17M transgene; these animals were bred to mrho/mice, and their progeny were screened by PCR for the hT17M transgene, as well as fo r their respective genotype at the mouse rhodopsin locus. Eventually, mice were obtai ned that were heterozygous null for mouse rhodopsin (mrho-/-) and also contained the hT17M mutant rhodopsin transgene. Two mrho-/-; hT17M males were crossed to wild -type C57Bl6 female mice to breed the mrho+/-; hT17M animals used in these expe riments. Concurrently, mrho-/-; hT17M females were crossed to mrho-/males to maintain the hT17M transgene on a mouse rhodopsin null background.

PAGE 53

40 ERG Natural History Once these lines were breeding reliab ly, an ERG natural history study was conducted on the mrho+/-; hT17M mice and non-tr ansgenic littermates Two litters were genotyped for this study, producing eight mr ho+/animals and eight mrho+/-; hT17M siblings. These animals were subjected to ER G analysis every two weeks for the next six and a half months. Right and left eye amplitudes were averaged for each animal, and the averages of each group were plotted for each time point (Figures 2-4 and 2-5). Additionally, the ratio of the mrho+/response to the mrho+ /-; hT17M response for each time point was plotted (Figures 2-6 and 2-7). There was a definite loss of both aand bwave responses over the life of the hT17M transgenic animals compared to th eir non-hT17M littermates. The a-wave response showed particularly ear ly degradation, as the mutant response was only half that of the non-mutant siblings at one month of ag e. Mutant a-wave responses remained at around 50% of those of the non-transgenic an imals for the next 2.5 months, when they again underwent a significant dr op relative to their non-mutant siblings, to around 30% of non-mutant a-wave amplitude. The b-wave am plitudes of the transgenic animals were more robust, starting out at around 70 to 80% percent of th e non-transgenic littermates and remaining so for the next 2.5 months, wh en they began to undergo steady a steady decrease in amplitude at around the time th at the a-waves exhibi ted their drop to 30%. Funduscopy In order to see if the degeneration in dicated by the ERG natural history study would be apparent upon funduscopic examination of the retina, two mrho+/siblings, one of which contained the hT17M transgene, underw ent ERG analysis at 6 months of age. The following day, fundus photographs were take n to visualize the retinas of these

PAGE 54

41 Mutant T17M and Non-Mutant Littermate A Wave Amplitudes Over Time0 50 100 150 200 250 300 350 4001 M o 1 .5 M o 2 M o 2 .5 M o 3 M o 3.5 Mo 4 Mo 4.5 Mo 5 .5 M o 6.5 MoAge of AnimalAmplitude (uV) Rho+/Rho+/-;T17M Figure 2-4. 10 dB intensity ERG a-wave natu ral history of mrho+/and mrho+/-; hT17M mice. Mutant T17M and Non-mutant Littermate B Wave Amplitudes Over Time0 100 200 300 400 500 600 700 800 9001 M o 1.5 Mo 2 M o 2.5 Mo 3 Mo 3.5 Mo 4 M o 4.5 Mo 5.5 Mo 6.5 MoAge of AnimalAmplitude (uV) Rho +/Rho +/-;T17M Figure 2-5. 10 dB intensity ERG b-wave natu ral history of mrho+/and mrho+/-; hT17M mice. ERGs were performed at the 10dB light intensity.

PAGE 55

42 A Wave Responses: Percentage of Mutant to Non-Mutant Littermates0 10 20 30 40 50 60 70 80 90 1001 Mo 1.5 Mo 2 Mo 2. 5 Mo 3 Mo 3. 5 Mo 4 Mo 4. 5 Mo 5.5 M o 6. 5 MoAge of Animal% Mutant / Non-mutant Response Figure 2-6. 10 dB intensity a-wave responses charted as a percen tage of the average mrho+/-; hT17M response to the average response of the mrho+/siblings at each time point. B Wave Responses: Percentage of Mutant to NonMutant Littermates0 10 20 30 40 50 60 70 80 90 1001 M o 1.5 M o 2 Mo 2.5 M o 3 Mo 3 5 Mo 4 Mo 4 .5 Mo 5.5 M o 6.5 MoAge of Animal% Mutant / Non-mutant Response Figure 2-7. 10 dB intensity bwave responses charted as a percentage of the average mrho+/-; hT17M response to the average response of the mrho+/siblings at each time point.

PAGE 56

43 Figure 2-8. Fundus photographs and respective 10 dB intensity ERG tracings taken at 6 months of age. Photos and tracings from a mrho+/animal are shown on the left (A), and those of its mrho+/-; hT17M littermate are shown on the right (B). animals. The resultant images, matched with their respective ERG tracings, are shown in Figure 2-8. The fundus of the mrho+/mous e looked normal, exhibiting relatively even pigmentation and healthy-appearing retinal mo rphology, and ERG analysis of this mouse showed robust aand b-wave amplitudes. The hT17M sibling displayed a markedly depressed ERG tracing, as well as shadowy spot ting in its fundus pict ure, indicating that the retina of this mutant animal had th inned as its photoreceptors degenerated. Histology Two of the animals (hT17M mutant and non-mutant sibling) from the natural history were sacrificed at 4 months of age to obtain an early look at the histological

PAGE 57

44 results of the hT17M mediated retinal dege neration. The rest of the mice were all sacrificed at 6.5 months, as the aand b-wave loss seemed to plateau and it was feared that waiting longer would resu lt in retinas that were too degraded to provide good sections. Representative s ections from these animals showed severe, progressive degeneration of the outer nuclear layer (ONL) that is consistent with the loss of ERG response, both of which are hallmarks of retinitis pigmentosa. (Figure 2-9). Figure 2-9. Representative s ections of mrho+/mice (A-D) and mrho+/-; hT17M siblings (E-H). Sections A and E are from animal s sacrificed at 4 months of age, all other sections are from animals sacrificed at 6.5 months of age, at the completion of the natural history st udy. All hT17M animals (E-H) show severe degeneration of the outer nuclear layer (ONL). Arrows indicate ONL. In order to obtain a more concrete idea of the extent of th e ONL degeneration, tile field mapped images of the hT17m and non-hT17M retinas were examined. A semitransparent template (Figure 2-10) was used to pinpoint eight areas of each retina from which ONL thickness could be measured (in nu mbers of nuclei) and averaged. The mean

PAGE 58

45 of the counts taken from the six points was determined, and this number represents the average thickness of the ONL for that eye. For the animals sa crificed at 6.5 months of age, the ONL values from both eyes were av eraged together, while only one eye from Figure 2-10. Tile-field mapped image of a mouse retina, with a translucent overlay that defines eight evenly-spaced sections of the retina from which outer nuclei counts can be obtained. Table 2-1. ONL averages taken from mrho+ /and mrho+/-; hT17M mice sacrificed and analyzed at 4 and 6.5 months of age. ** indicates significant difference between mutant and non-mutant values with a P value of less than .001. mRho+/mice mrho+/-; hT17M mice ONL Thickness at 4 months 8.8 Rows 5.9 Rows ONL Thickness at 6.5 mont hs 7.7 Rows** 2.4 Rows** each animal at the 4 month time point was able to be analyzed. These results are summarized on Table 2-1. By this measure, hT17M mice exhibited significant thinning

PAGE 59

46 of the outer nuclear layer when compared to their non-hT17M littermates sacrificed at the 6.5 month time point. hT17M mice sacrificed and analyzed at the 4 month time point also showed a reduction in the ONL thickness. Discussion The original mrho+/+; hT17M mouse line ob tained from Dr. Tiansen Li provided a useful animal model for the study of retinal disease. These animals showed progressive photoreceptor loss that was accompanied by a loss of aand b-wave ERG amplitudes, and experiments involving treatment of th is mouse with vitamin A supplementation produced some compelling results. However, by the time we began these experiments, the line had become uncharacterized for r easons already described. A goal of the preceding experiments was to develop efficien t methods for genotyping these mice and to breed the remaining hT17M pos itive animals to a final mr ho-/background. We then wanted to use these mrho-/-; hT17M mice both for maintenance of the transgenic line and for creation of a mrho+/-; hT17M transgenic line that would be useful in developing therapy for autosomal dominant retinal disease. The PCR assays that we have developed to characterize these lines are simple and accurate. Use of the Quiagen DNeasy kit allows a researcher to easily purify substantial amounts of genomic DNA from small pieces of mouse tail in a short period of time. Two separate PCR reactions allow for determ ination of the endogenous mouse rhodopsin genotype (mrho+/+, mrho+/-, or mrho-/-), and to screen for the presence or absence of the T17M human mutant rhodopsin transgene. Th e line is now maintained on both the mrho/and mrho+/+ genetic background. The retinal degeneration displayed by mrho+/-; hT17M transgenic line is an excellent model of the vision loss observed in human patients suffering from ADRP. Bi-

PAGE 60

47 monthly ERG recordings demonstrated progr essive loss of rod photoreceptors in the hT17M transgenic animals, culminating in a se vere loss of visual response by six months of age, while the non-transgenic littermates retained normal vi sual function. The loss of ERG was accompanied by a progressive thinning of the ONL as determined from retinal sections, with mrho+/-; hT17M mice displayi ng a loss of 40% of their ONL by 4 months of age and a loss of 70% of their ONL by 6.5 mo nths of age, as compared to their nontransgenic littermates. Fundusc opic analysis of the transgen ic animals revealed marked retinal thinning by 6 months of age, with non-transgenic littermates again displaying normal retinal morphology. It is unfortunate that it has so far been impossible to breed the hT17M transgene to homozygosity. This is probably due to inse rtional inactivation of an unknown gene that is embryonic lethal if not pres ent in at least one copy. The lines have been outbred for several years now since their arrival, and future attempts to breed the transgene to homozygosity could well meet with success. At present, our breeder mice are limited to one copy of the hT17M transgene, with the result being that, at most, around 75% hT17M transgenic mice can be produced from a breedi ng of two hT17M transgenic mice. As our breeders to date have consis ted of one mrho-/-; hT17M transgenic mouse crossed with a wild-type C57BL6 mate, we have been producing litters composed of around 50% transgenic and 50% non-transgenic pups on an mrho+/background. This is a drawback in terms of requiring twice as many mous e pups to provide sufficient numbers of transgenic animals for testing therapy for ADRP. However, this drawback has proven useful in that most experimental litters c ontain several non-transge nic control mice. In

PAGE 61

48 fact, these non-transgenic li ttermates were essential in discovering an hT17M rhodopsinmediated light sensitivity that will be discussed in future chapters. Not all mouse models of retinal disease are practical for developing therapy. For example, one of the mouse lines created by Olsson et al. to express a human rhodopsin transgene containing the P 23H mutation showed nearly complete photoreceptor degeneration by 20 days of age (Olsson et al., 1992). This is too ra pid a degeneration to provide an effective test subjec t for all but the most rapidly ac ting therapies. On the other hand, an animal model that undergoes degene ration too slowly is also problematic because experiments utilizing the line can take a long time to provide useful data. The hT17M human transgene expresse d on an mrho+/+ background underwent a progressive loss of photoreceptors culmina ting in total loss of vision by around eight months of age. We decided to breed thes e animals to a mrho+/background with the goal of both creating a mouse line that cont ained one copy each of mutant human and wild-type rhodopsin genes, and of developi ng a model that would undergo a more rapid retinal degeneration while still providing a therapeutic window for treatment. The resultant hT17M; mrho+/animals did undergo a more rapid retinal degeneration, with almost complete loss of f unctional rod photoreceptors by around 6.5 months of age. Additionally, this line displayed a-wa ve and b-wave ERG responses of 100 V and 400 V, respectively, at 2.5 months of age. By 6.5 months of age these responses had degraded to an a-wave response of less than 50 V and a b-wave response of less than 200 V. Studies have shown that AAV pseudotype 2/5 can achieve significant transgene expression of mouse photorecept or cells by twelve days fo llowing subretinal injection (Yang et al., 2002; Rabinowitz et al., 2002). Assuming that our hT17M transgenic mice

PAGE 62

49 receive treatment at weaning (21 days of ag e), the line should retain ample photoreceptor function by the time one can expect expressi on of rAAV-delivered therapy designed to prolong retinal function. Th is therapeutic window should also be wide enough to be amenable to pharmacological treatment. In conclusion, the hT17M transgene ha s been maintained and expressed on mrho+/+, mrho+/-, and mrho-/backgrounds This line undergoes a progressive photoreceptor degeneration that can be m onitored by ERG measurements and that correlates with thinning of th e ONL as visualized by fundus copy and by retinal histology. The degeneration of the mrho+/-; hT17M line is practically complete by 6.5 months; yet these mice retain sufficient retinal function a nd photoreceptor survival at early ages to make them amenable to therapeutic intervention.

PAGE 63

50 CHAPTER 3 AAV-MEDIATED RIBOZYME TREATMEN T OF MRHO+/-; hT17M MICE Introduction The hT17M mutant human rhodopsin tran sgene causes the autosomal dominant form of retinitis pigmentosa, meaning that expr ession of the mutant allele is responsible for the disease. One way to treat an autosoma l dominant disorder is to selectively destroy the mRNA encoding the mutant allele in theo ry this should abo lish expression of the mutant protein and rescue the disease (Hau swirth et al., 2000). Mice bred to be hemizygous null at the rhodopsin allele (m rho+/-) show only slightly reduced ERG responses when compared to mice that contain two wild type copies of the gene (Humphries et al., 1997). T hus in our mrho+/-; hT17M line, removal of the mutant human mRNA while leaving the wild-type mous e message intact should protect against the vision loss associated with th e mutant rhodopsin gene product. Ribozymes are ideally suited for the treatmen t of autosomal dominant disease. As discussed before, the sequence specificity of ribozyme cleavage is stringent enough that a ribozyme can often be designed to discrimina te between mutant and wild-type target RNAs that differ by a single nucleotide. This makes it feasible to treat certain dominant diseases arising from point mutations with ri bozymes that selectively degrade the mutant RNA, thus selectively reducing or abolishing expression of the mutant protein. It was recently shown that the use of such a ribozym e that targeted the reduction of a P23H mutant rhodopsin message resulted in rescue of vision in a rat model of ADRP (Lewin et

PAGE 64

51 al., 1998; LaVail et al., 2000; Dren ser et al., 1998). We wanted to explore the efficacy of a similar ribozyme-mediated therapy for the threonine to methionine at position 17 (T17M) mutation carried in our rho+/-; hT17M mouse line. To this end we designed ribozymes to se lectively cleave human rhodopsin mRNAs. In most forms of ribozyme gene therapy for autosomal dominant disease, the ribozymes are designed to select between mutant a nd non-mutant messages because of sequence differences at the site of the causative mu tation. This means that the therapeutic ribozyme must be specifically tailored to cleave the messa ge at the mutation site, and often this target site is less than ideal for ribozyme-mediated cleavage, if it is susceptible to cleavage at all. In the hT17M transgen ic mouse model, expression of a mutant human gene on a mouse genetic background is responsi ble for the disease, so we were able to design ribozymes that were targeted to idea l cleavage sites in th e human rhodopsin gene that differed from the endogenous, wild-type mouse sequence, rather than having to design a ribozyme to a target site that was re stricted to the site of the mutation. Using this strategy we were able to create two hi ghly active ribozymes that cleaved the human rhodopsin message but should theoretica lly leave the mouse message intact. Once an effective ribozyme has been create d, the next step in developing a therapy for retinal disease is to deliver the ribozyme s to the pertinent retinal cells. For this purpose we used recombinant adeno-associat ed virus (AAV) delivery vectors. As previously discussed, AAV is extremely well su ited for retinal gene transfer (Flannery et al., 1997). For our experiments we decided to use AAV pseudotype 5, as this pseudotype has demonstrated preferentially high levels of transduction in photor eceptor cells, as well as a rapid onset of transgen e expression (Yang et al., 2002). To drive the expression of

PAGE 65

52 our human rhodopsin-specific ribozymes we chose the MOPS500 promoter, which has been shown to achieve impressive levels of transgene expression in the rod photoreceptors of rats and mice (Flannery et al., 1997). The combination of AAV pseudotype 5 vectors delivering ribozymes under the control of the MOPS500 promoter helps ensure high levels of ribozyme expression that are specific to the rod photoreceptors of treated animals. In this chapter, I describe the creation and in vitro testing of two ribozymes designed to specifically cleave the human rhodopsin transgene. I will discuss how these ribozymes were cloned in specialized plas mids for packaging as recombinant AAV. Finally, I will detail the subretinal injection technique and the result s of delivering this virus to the subretinal space of mr ho+/-; hT17M transgenic animals. Materials and Methods RNA Oligonucleotides RNA nucleotides were ordered from Dharm acon Research Inc. (Boulder, CO), at the 50 molar scale. The sequences were as follows: Rz1: 5-CCGAACUGAUGA GCCGUUCGCGGCGAAACGAAG-3 Rz3: 5-GUGAACUGAUGAGC CGUUCGCGGCGAAACGAGC-3 Target1: 5-CUUCGUCUUCGG-3 Target 3: 5-GCUCGUCUUCAC-3 DNA Oligonucleotides DNA oligonucleotides were obtained from Invitrogen at the 40 nmolar scale of preparation. Oligonucleotides used for cloning had 5 phosphate groups chemically added by the manufacturer, and were purified by desalting. Their sequences were as follows:

PAGE 66

53 Rz1 Cloning Sense: 5AGCTT CCGAACTGATGAGCCGTTCGCGGCGAAACGAAG ATGCA -3 Rz1 Cloning Antisense 5T CTTCGTTTCGCCGCGAACGGCTCATCAGTTCGG A -3 Rz3 Cloning Sense 5AGCTT GTGAACTGATGAGCCGTTCGCGGCGAAACGAGC ATGCA -3 Rz3 Cloning Antisense 5T GCTCGTTTCGCCGCGAA CGGCTCATCAGTTCAC A -3 Red nucleotides indicate restriction sites for HindIII (AAGCTT) and NsiI (ATGCAT). Preparation of Synthetic RNA Ribozymes and Substrates All hammerhead ribozymes as well as the substrate RNAs were purchased from Dharmacon Research, Inc. (Boulder, CO). The RNA oligonucleotides were chemically synthesized with an acid-l abile orthoester protecting group (to reduce ribonuclease degradation) on the 2-hydroxyl (2-ACE) that must be deprotected by incubation at pH 3.8 at 60oC according to the manufacturers protoc ol prior to use. Deprotected RNA oligos were suspended at a final concen tration of 300 picomoles per microliter. 5' End-labeling of Deprotected Target RNAs Prior to in vitro kinetic analysis, the ta rget RNAs were 5 end-labeled with [ 32P]ATP (ICN, Irvine, CA) using T4 Polynucleot ide Kinase (T4 PNK) (Promega, Madison, WI). A typical reaction contained: 2 l target RNA oligo (20 picomoles total), 1 l of 10X PNK Buffer [700 mM Tris-HCl (pH 7.6 at 25oC), 100 mM MgCl2], 1 l RNasin

PAGE 67

54 (Promega), 1 l 0.1M DTT, 3 l H2O, 1 l [ 32P]-ATP, and 1 l T4 PNK. Reactions were incubated at 37oC for 30 minutes, and then 90 l of H2O was added. The mixture was then incubated at 65 for 5 minut es to inactivate the T4 PNK. Two Phenol:Chloroform:Isoamyl Alcohol extr actions were then performed, and 90 l of the aqueous phase was purified over a Sephadex G25 Spin Column (Pharmacia, Piscataway, NJ) to separate the labeled target molecule from unincorporated radionucleotides. The resulting labeled target solution was at a final concentration of 0.2 picomoles per microliter. In Vitro Ribozyme Ti me Course Analysis Separate ribozyme and target mixes were made. The ribozyme mix consisted of: 13 l 400 mM Tris-HCl (pH 7.45), 1 l ribozyme (diluted to 2 pmol/ l), and 80 l H2O. The target mix consisted of 1 l radiolabeled target RNA, (diluted to 0.2 pmol/ l), 1 l cold target RNA (diluted to 20 pmol/ l), and 8 l of H2O. After these two mixes were created, the ribozyme mix was heated to 65oC for two minutes, and then cooled to room temperature for at least ten minutes. 13 l of a 1:10 RNasin:0.1M DTT mix and 13 l of 50 mM MgCl2 (for a final reaction con centration of 5 mM Mg Cl2) were then added, and the solution was equilibrated at 37oC for ten minutes. For the reaction, the 10 l target mix was added to the ribozyme mix, the solution was mixed thoroughly by vi gorous pipetting, and 10 l of the reaction mix was immediately added to 10 ul formamide stop buffer (90% formamide, 50 mM EDTA, 0.05% bromophenol blue, and 0.05% xylene cya nol), and placed on ice (this was the 0 minute time point), and the remaini ng reaction mix was incubated at 37oC. At subsequent intervals of 1, 2, 4, 8, 16, 32, 64, and 128 minutes after the start of the

PAGE 68

55 reaction 10 l of the reaction mix was removed and likewise added to 10 l formamide stop buffer and placed on ice. The reaction/formamide stop buffer mix from each time point was then heat denatured at 95oC for five to ten minutes, then placed on ice for five minutes, and then separated on a 10% polyacrylamide / 8M urea ge l. The gel was fixed in 2 L of fixation solution (40% methanol, 10% ace tic acid, 3% glycerol) for th irty minutes, dried, exposed to radioanalytic phosphorescent screens, a nd analyzed using a Molecular Dynamics PhosphoImager system and ImageQuant so ftware (Molecular Dynamics, Sunnyvale, CA). The percentage of substrate cleaved in each sample was determined from the ratio of radioactivity in the 5-end labeled cleavage produc t (P) to the sum of the radioactivity in the 5-end labeled cleavage product and the substrate band (S): % Cleavage = P/P+S. Using Excel (Microsoft, Redmond, WA) the pe rcentage substrate cleaved was then plotted as a function of time to generate a graphical representation of the cleavage time course. Ligating Ribozyme Sequences into rAAV Packaging Vectors Complementary DNA oligonucleotides encodi ng the sense and antis ense strands of HRz1 and HRz3 were ordered. In ad dition to the ribozyme sequence these oligonucleotides contained sequences (shown ab ove in red) appended to their 5 and 3 ends so that when they annealed they formed the sticky overhangs corresponding to HindIII at the 5 end and NsiI at the 3 ends This allowed the oligonucleotides to be ligated into the rAAV packaging vector pXX-GS-HP-MOPS500 (F igure 3-1) at a multiple cloning site containing a HindIII restri ction site upstream of an NsiI restriction

PAGE 69

56 site. The oligonucleotides came from the supplier with 5-PO4 groups already attached to allow immediate ligation into the packaging vector. To linearize pXX-GS-HP-MOPS 500, 5 g of plasmid DNA was digested with HindIII for three hours at 37oC. The DNA was then ethanol precipitated, resuspended, and digested with NsiI for three hours at 37oC. After the two digestions were complete, the resultant fragment was run on a 1% agar ose gel, and the digested plasmid band was visualized by ethidium bromide staining under UV illumination. The band was excised from the gel and the DNA was purified using a freeze squeeze technique (Sugden et al., 1975). In brief, the gel fragment was crushe d and mixed in an equal volume of phenol in a 1.5 mL Eppendorf tube. The tube was then incubated at -80oC for 2 hours. Next the tube was spun at a speed of 13,000 rpm for 10 minutes, and the aqueous solution was removed. The aqueous solution was then extracted with an equal volume of phenol:chloroform:isoamyl alcohol, ethanol precipitated, and the resultant purified plasmid pellet resuspended at a concentration of 0.5 g/ l. To ligate the ribozyme-encoding oligonucle otides with the lin earized packaging vector, the complementary oligonucleotides were mixed together for a final concentration of 20 picomoles each in a vol ume of 4 microliters. To facilitate proper annealing, the oligos were heated to 95oC for five minutes, and then allo wed to slowly cool to room temperature. 0.5 g of linearized plasmid (1 l) was then added to the mixture along with 5 l of 5X ligation buffer [250 mM Tris (pH 7.5), 50 mM MgCl2, 5mM ATP], 1 l 25 mM DTT, 3 l PEG 4000, 10 l H2O, and 1 l T4 DNA ligase (Promega). Reactions were incubated at 25oC overnight, and then 2 l was transformed into 50 l of electrocompetent E. coli by electroporation using a Bio-Rad Gene Pulser II

PAGE 70

57 electroporation apparatus (Bio -Rad, Hercules, CA), utilizing 0.1 mm electroporation cuvettes (USA Scientific, Ocala, FL). The E. coli was plated on LB plates containing ampicillin, and the resultan t transformants were picked, their DNA isolated, and sequenced for the proper ribozyme insert. Figure 3-1. pXX-GS-HP MOPS 500 rAAV packaging plasmid. Human rhodopsin specific ribozymes were cloned into th e HindIII-NsiI restriction sites as indicated. Once each ribozyme was successfully cloned into pXX-GS-HP MOPS 500, 700ug of DNA was produced from a 1 L E. coli culture, purified vi a cesium banding on an ultracentrifuge, and finally packaged as recombinant AAV type 5 at the UF Ophthalmology Packaging Core.

PAGE 71

58 Subretinal Injection of rAAV Ribozyme Delivery Vectors At weaning age (21-24 days) litters were re moved from their parents and their right eyes were dilated with 1% atropine sulfate solution (Bausch and Lomb, Tampa, FL). The next morning, the right eyes were again dilate d with 1% atropine sulfate, and again an hour before the injection proce dure, at which time the eyes also received a drop of 2.5% phenylephrine HCl and 0.5% proparacaine HCl. An hour after this final dilation, the animals were anesthetized by ketamine/xylazi ne injection and agai n treated with a drop each of 1% atropine sulfate, 2.5% phenyle phrine HCl, and 0.5% proparacaine HCl. The right eyes of these animals th en received a drop of 2.5% hypr omellose to aid in retinal visualization and to help keep the retina hydr ated. Injections were visualized with a Nikon SM2800 (Nikon, Melville, NY) dissecting mi croscope, with illumination provided by a Southern Micro Instruments 150 Watt fibe r optic light source with Schott Fostec fiber optic arms (Southern Micron Instruments, Marietta, GA) (Figure 3-2), which at full power provided an intensity of illuminati on of around 10,000 lux. A hole was placed in the inferior cornea of the eye with a 28 gauge needle. A blunt 32 gauge needle was then inserted into the hole, the tip of the needle was rotate d around the lens, and pushed through the retina until it came to rest at the sclera, which could be visualized by the eye sinking back into the socket. 0.5 l of rAAV suspension was then slowly delivered into the subretinal space over a period of 20-30 second s. This injection strategy is depicted in Figure 3-3. VPP antibiotic ointment (Akorn) was placed upon both eyes to maintain hydration and prevent infection in the injected eyes, and the animals were ear marked and 0.5 cm sections of tail tip were removed fo r genotyping as described previously. The animals were then allowed to recover on a warming plate at 37oC.

PAGE 72

59 Figure 3-2. Dissecting scope a nd fiber optic light used durin g subretinal injections and experimental retinal illumination. Figure 3-3. Cartoon depicting th e subretinal injection. The blunt injection needle is shown passing through the cornea, around th e lens, and into the subretinal space (right). Once positioned thus, the rAAV solution can be delivered to the subretinal space (left), resulting in a local ized retinal detachment that resolves itself over time as the virus spreads laterally from the site of injection (red arrow). Figure courtesy of Dr. Lynn C. Shaw. Electroretinography Electroretinographic analysis of ribozyme-treated animals was performed as described above.

PAGE 73

60 Results Ribozyme Creation Our therapeutic plan called for the creati on of a ribozyme that would specifically cleave the mutant human rhodopsin mRNA while leaving the endogenous, wild-type mouse mRNA intact. A benefit of this strategy is that it allowed us to consider all possible ribozyme cleavage sites in areas of the human gene that showed polymorphisms with the mouse gene. We were also able to look for the specific cleavage site GUCUU. It has been consistently demonstrated that hammerhead ribozymes cleave more efficiently at the GUC target site than at a ny other (Shimayama et al., 1995). It has also been reported that hammerhead ribozyme cl eavage can be enhanced when the triplet target sequence is followed by a UU or UA di nucleotide (Clouet-d'Orval and Uhlenbeck, 1997). Figure 3-4. Primary structure of ribozymes HR z1 and HRz3 (green), shown paired with their human (black) and mous e (red) target sequences. Polymorphisms between the mouse and human rhodopsin genes are boxed. The human rhodopsin gene contains 19 GUC s ites in its reading frame. Of these, three contained a UU dinucleotide directly fo llowing the cleavage si te, and two of these GUCUU sites contained single nucleotide polymorphisms between the endogenous mouse gene and the mutant human transgene. The first of these sites is a GUCUU site beginning at nucleotide 310 as measured from the start of the coding sequence for the

PAGE 74

61 gene. The second of these begins at nucleotid e 679. The predicted st ructure of these two areas of the mRNA showed no serious energe tically favorable sec ondary structure that would inhibit proper ribozyme binding to th eir respective target sites. The two ribozymes, named HRz1 and HRz3, are illustrated in Figure 3-4. In Vitro Time Course Anal ysis of HRz1 and HRz3 In vitro cleavage analysis showed HRz1 and HRz3 to be efficient at cleaving the human rhodopsin mRNA. Reactions were pe rformed under a condition of 10-fold excess of substrate relative to ribozyme (10nM to 1nM). Both ribozymes achieved 20% substrate cleavage in one minute in a reaction mixture containing 20mM MgCl2. Magnesium is necessary in cell free reactions to promote the folding of the ribozyme, but is not required in cells. Figure 3-5 shows a representative time c ourse cleavage reaction performed by incubating the HRz1 RNA oli gonucleotide with its 12 nucleotide, 5 endlabeled target as described. These reactio ns generate two bands when separated on polyacrylamide gels. The top band is the ra dioactively-labeled, uncut 12 nucleotide RNA target molecule, while the bottom band is the 7 nucleotide 5 cleavage product (as only the 5 end of the RNA target oligonucleotide was labeled, the 5 nucleotide 3 cleavage product was not detectable by autoradiography). Phosphorimag er analysis was used to determine the relative intensity of these two bands, which in turn were used to calculate the time course cleavage rates as describe d. Identical time course reactions were performed using the HRz3 ribozyme/target combination (data not shown). Graphical representation of the time course reactions of HRz1 and HRz3, illu strated in Figure 3-6, confirmed the activity of our ribozyme selectio ns and prompted us to initiate efforts to treat the hT17M mice with rAAV expressing these ribozymes.

PAGE 75

62 Figure 3-5. Representative PhosphorImager s can of a time course assay showing HRz1 cleavage of a 12 nucleotide synthe tic human rhodopsin target RNA. Figure 3-6. Time course of HRz1 and HRz3 cleavage. The percent cleavage of each 12 nucleotide synthetic target by its respective ribozyme is plotted as a function of time. Both ribozymes are able to achieve 20% cleavage in a minute or less in the presence of 20mM MgCl2. ERG Analysis of hT17M Transgenic Mice Treated With HRz1 and HRz3 Several features of the pXX-MOPS-GS-HP recombinant AAV packaging vector used in these studies merit attention (Fi gure 3-1). The plasmid contains a multiple cloning site under the contro l of the MOPS 500 promoter, which has been shown HRz1 Timecourse at 20mM MgCl20 10 20 30 40 50 60 70 80 90 100 020406080100120 Time in MinutesPercent Cleavage HRz3 Timecourse at 20mM MgCl20 10 20 30 40 50 60 70 80 90 100 020406080100120 Time in MinutesPercent Cleavage

PAGE 76

63 previously to induce high levels of transg ene expression in mous e and rat photoreceptor cells (Flannery et al., 1997). MOPS500 cons ists of 483 bp of the mouse opsin proximal promoter, including 70 bp of the 5 untrans lated region of the mRNA coding sequence. Immediately following the promoter is an SV40 intron (SD/SA), which has also been shown to increase expression of RNAs by pr omoting nuclear export via the spliceosomal pathway (Bertrand et al., 1997). Next is the multiple cloning site, in which either HRz1 or HRz3 was inserted into the HindIII/NsiI junction as described. Following this is a downstream hairpin ribozyme that generate s well-defined 3 ends for the ribozyme transcript, reducing the possibility of the ribozyme interacting w ith excess downstream sequence in such a way as to cause it to fold into an inactive or inac cessible (to the target mRNA) conformation (Altschuler et al., 1992). The vector also contains an ampicillin antibiotic resistance gene to aid with bacterial cloning, a neomycin resistance ge ne enabling selection in mammalian cells using the antibiotic G418, and a GFP marker gene under the expre ssion control of an internal ribosomal entry site (IRES). The entire ribozym e expression, GFP marker, and neomycin resistance cassettes are contained wi thin inverted terminal repeat sequences (ITRs) that are necessary for these various elements to be packaged as recombinant AAV The resultant viruses were purified to a titer of 2x1013 genome copies/ml (HRz1) and 1x1013 genome copies/ml (HRz3). Our initial treatment attempt involved subretinally injecti ng the right eyes of a litter consisting of five mrho+/-; hT17M transgenic mice and two mrho+/siblings with 0.5 l of rAAV expressing HRz1. The injections we re performed at 21 days of age, and the animals underwent ERG analysis one and a half months later to assay rescue of the

PAGE 77

64 injected eye. In all five hT17M mice, the injected eyes showed substantial ERG reduction relative to the uninjected eyes. Th e non-transgenic, mrho+/siblings were not affected. These results are summarized in Fi gures 3-7 and 3-8. Injection of recombinant AAV expressing the HRz3 ribozyme produced sim ilar results: significant depression of both aand b-wave ERG responses in the in jected eyes of the animals containing the hT17M transgene, but not in th eir non-transgenic littermates that were also injected. Relative ERG Responses of Mice Receiving Subretinal HRZ1 Injections0 20 40 60 80 100 120 20dB A20dB B10dB A10dB B% R/L Eye ERG Response T17M Mice non Rho+/Mice Figure 3-7. Relative ERG respons es of aand b-waves at 20 and 10dB flash intensities. Each bar represents the average of the ra tio of right to left eye ERG responses for all animals in that group. These mice were analyzed at one month after subretinal injection of the right eye with 0.5 micr oliters of rAAV expressing HRz1. Subretinal injection led to sign ificantly greater damage in the hT17M transgenic animals (dark green bars) th an in their non-tran sgenic littermates (light blue bars). Data generated during the creation of anot her mouse model of retinitis pigmentosa expressing a human rhodopsin transgene contai ning a P23H mutation demonstrated that such mice expressing two copies of wild type rhodopsin (mrho+/+) degenerated more slowly than those expressing onl y one copy (mrho+/-). This le d us to attempt to treat the

PAGE 78

65 hT17M model, which was expressed on a mrho+/background, with recombinant AAV expressing wild-type mouse rhodopsin on the pr emise that increasing the level of normal rhodopsin might dilute the impact of the mutant transgene. These particular animals also received contralateral inj ections of recombinant AAV e xpressing only the GFP marker protein to control for any resc ue that might result from an ocular response to injection damage. These injections led to ERG reduc tions in the mutant mice (Figure 3-9) that were similar to those seen with the ribozyme in jections, only they were seen in both eyes of the hT17M transgenic animals, as both eyes were injected. Discussion We developed ribozymes with the ab ility to efficiently cleave the mRNA associated with a human rhodopsin transgene, which could be useful as therapeutic reagents for treatment of th e hT17M mouse model of RP. The ribozymes were designed to allow them to discriminate between endogenous, wild-type mouse rhodopsin and the mutant human transgene in such a way as to abolish expression of the mutant RNA while leaving the wild-type RNA intact (Figure 34). HRz3 contained a mismatch with the mouse target at the first nucleotide upstr eam of the cleavage tr iplet, a site where mismatches between the ribozyme and its targ et sequence has been shown to abolish the catalytic step of hammerhead ribozyme cleavage in vitro (Werner and Uhlenbeck, 1995). HRz1 contained a mismatch w ith the mouse target that is located three nucleotides downstream of the cleavage trip let, and although sequence differe nces in this area are not though to severely reduce the ca talytic step of ribozyme cleavage, they are thought to cause sufficient disruption of ribozyme bindi ng to allow preferential cleavage of a perfectly matched target sequence. In vitro analysis showed these ribozymes to be catalytically efficient, with around 20% cl eavage of a 12nt target RNA sequence

PAGE 79

66 Figure 3-8. 10dB intensity ERG tracings from mice receiving subretin al injections of their right eyes with rAAV expressing HRz1. mrho+/mice (A and B) show normal ERG responses in both left and right eyes, while the mrho+/-; hT17M transgenic mice (C-G) show substantial reduction in the ERG response of the injected eye. These tracings were used to generate the data in Figure 3-7. E F G C A B D

PAGE 80

67 Figure 3-9. 10dB intensity ERG tracings from mice receiving subretin al injections of both eyes. The right eyes were injected with r AAV delivering a wild type mouse rhodopsin transgene, while the left eyes were injected with control rAAV delivering the GFP marker gene. mro+/mice (A and B) show normal ERG responses in both left and ri ght eyes, while the mrho+/-; hT17M transgenic mice (C,D,E, and F) show substantial reduction in the ERG response of the injected eyes. C D E F A B

PAGE 81

68 observed at the one minute time point at a MgCl2 concentration of 20mM. This time course cleavage rate was comparable to that demonstrated by other hammerhead ribozymes in our laboratory that have been shown to be effective in animal models (Lewin et al., 1998; Fritz et al., 2002; Liu et al., 2005; Gorbatyuk et al., 2005). The ribozymes were cloned and packag ed as recombinant AAV (rAAV), with expression controlled by a rhodopsin-specific promoter sequence. Ancillary sequences were included in these viral vectors that s hould allow for efficient expression of ribozyme molecules in the target photoreceptor cel ls. Downstream, self-cleaving ribozyme sequences were also included to generate prec ise 3 ends for the therapeutic ribozymes. rAAV viral vectors expressing bot h ribozymes were purified to a high titer, and injections were performed to deliver 0.5 l of the HRz1-expressing virus (1.0 x 1010 genome copies per injection) to the subretin al space of mrho+/-; hT17M tr ansgenic animals, along with their non-transgenic littermates. Subse quent PCR genotyping was performed to determine the mrho and hT17M genotype of each animal. Unfortunately, these injections actually resulted in significant retinal damage instead of the intended rescue. ERG anal ysis of mice containing the mutant human transgene performed one month post-injection revealed severe attenuation of both aand bwave ERG responses (Figur e 3-8). The non-transgenic li ttermates were unaffected. Repetition of these experiments with several subsequent litters using either HRz1 or HRz3 as the delivered ribozyme produced the same results severe damage to the retina as determined by ERG analysis one month after inje ction. An attempt was also made to achieve treatment by injecting vectors de signed to produce a surplus of wild-type rhodopsin in the retinas of the hT17M transgen ic animals. This experiment utilized a

PAGE 82

69 control injection of rAAV designed to deliver a GFP marker protei n to the photoreceptors of the contralateral eye. Th is paired injection experiment led to severe ERG attenuation of BOTH eyes of the transgenic animals, wh ile again the non-transgenic littermates were unaffected (Figure 3-9). These results were surprising. Although a slight amount of ERG attenuation is often seen in animals following subretinal inj ection, it is usually neither as severe nor as prolonged as the reduction observed in th ese experiments (Timmers et al., 2001). Occasionally bad injections ar e accidentally performed that can cause severe retinal damage, but these are rare when the techniqu e is performed, as ours was, by experienced personnel. Attempts to treat our hT17M tran sgenic mouse line invariably resulted in grossly high failure rates, a nd it eventually became evident that the severe injection damage was always seen in mice containing the hT17M transgene and not in their nontransgenic littermates. This led us to suspect a defect in injection tolerance in the hT17M mice and to design experiments to determine which aspects of the subretinal injection technique were responsible for such a severe loss of ERG response.

PAGE 83

70 CHAPTER 4 INCREASED LIGHT SENSITIVITY IN mRHO+/-; hT17M MICE Introduction Repeated attempts were made to trea t hT17M-mediated retinal degeneration in mice with subretinal injection of rAAV delivering either human rhodopsin-specific ribozymes or wild-type mouse rhodopsin transgen es. Analysis of treated animals showed severely reduced ERG responses in the injected eyes. At first this was thought to be the result of injection damage, but it was s oon observed that the ERG reduction was seen only in mice carrying the hT17M mutant rhodops in transgene. After this trend was shown to repeat itself in several expe rimental groups, we hypothesized that the mechanism of injection itself was detrimental to the visual response of the hT17M mutant animals and not to their non-tran sgenic siblings. The subre tinal injection technique had two components that seemed likely candida tes for the damage: the introduction of a virus-containing solution into the subretinal sp ace and the use of bright fiber optic light to illuminate the extremely dilated eyes of these animals during the actual procedure. Subretinal injection is known to create a retinal detachment that resolves itself over time as the injected solution (in this case containing rAAV expressing the human rhodopsin-specific ribozyme) is removed from the eye (Timmers et al., 2001). If the technique is not performed prope rly, it is possible to damage th e retina so severely that the visual response is affected. It is al so possible that mice expressing hT17M mutant rhodopsin are for some reason unable to resolv e their retinal detachments, leading to

PAGE 84

71 decreased ERG response following what would normally have been a successful subretinal injection. Injecting the transgenic mice w ith saline solution resulted in a reduction in the ERG response similar to that observed following subre tinal injections of rAAV solutions, suggesting retinal damage (data not shown). Light-mediated retinal damage (LMD) has b een studied extensively since the first reports of its occurrence in laboratory rats four decades ago (Noell et al., 1966). It has since been documented in various other la boratory animals, including pigmented and non-pigmented fish (Penn, 1985), mice (LaVai l et al., 1987), rabbits (McKechnie and Johnson, 1977), dogs (Cideciyan et al., 2005), an d monkeys (Lawwill et al., 1980). In many instances, simply maintaining thes e animals under continuous room-level illumination can lead to severe ly attenuated ERG responses th at correlate with a loss of rod photoreceptor cells. Although the phenomenon has been extensively studied, the exact mechanism by which light damage leads to photoreceptor cell death is still not fully understood (Wenzel et al., 2005). All forms of light-induced retinal damage have two things in common: rhodopsin is the initial effector molecule, and the ultimate fate of the damaged photoreceptor cells is death by apoptosis. There is significant evidence to dem onstrate the involvement of rhodopsin in LMD. Studies of the RPE65 knockout mouse m odel show that this mouse is completely resistant to light damage. Since RPE 65 is involved in regeneration of 11cis retinal from alltrans retinal, this observation le d researchers to conclude that reconstituted rhodopsin (i.e. opsin bound to the 11cis retinal chromophore) is th e initial mediator of LMD (Grimm et al., 2000b). Protection against light damage can also be achieved by inhibiting rhodopsin reconstitution through pha rmacological means. Intravitreal

PAGE 85

72 treatment with 11cis retinoic acid, which presumably competes with 11cis retinal binding to RPE-65 and thus inhibits reconst itution of active rhodopsi n, has been shown to protect against light damage in rats (Sievi ng et al., 2001). Admini stration of halothane anesthesia has also been show n to block regeneration of 11cis retinal, and such treatment prior to light damaging light exposure can prev ent LMD in albino rats and mice (Keller et al., 2001). Apoptotic involvement in light mediat ed retinal damage was convincingly demonstrated in albino rats through TUNEL la beling of fragmented DNA in the affected retinas (Aonuma et al., 1999). In 1998, studies involving p53 knockout mice demonstrated that they were not resistant to LMD, indi cating that the gene was not involved in the light-induced apoptotic pathwa y (Lansel et al., 1998). Subsequent studies reported involvement of the apoptotic effector molecule c-fos, which is a member of the AP1 transcription factor complex, in light-med iated apoptosis, and elevated levels of AP1 have been demonstrated in several models of acute LMD (Reme et al., 1998; Wenzel et al., 2002; Naash et al., 1996; Cideciyan et al., 2005). Genetic mutations have been shown to incr ease light sensitivity in the retina. In 1987, LaVail and coworkers reported increased light sensitivity in Balb/c mice as compared to C57BL6 mice (LaVail et al., 1987). Matings between the two strains produced F1 progeny that showed an in termediate phenotype, demonstrating a segregating genetic trait as the mediator of light sensitivit y. It is now known that this is the RPE65 gene, and that a polymorphism at codon 450 (leucine in Balb/c mice and methionine in C57BL6 mice) leads to increa sed RPE65 activity in the leucine variant, leading to accelerated rhodopsin regenera tion which, as discussed above, leads to

PAGE 86

73 increased susceptibility to light damage (W enzel et al., 2001b; Danciger et al., 2004). However, these results could not be duplicated in rat models of LMD (Beatrice et al., 2003). Mice with mutations in the SOD 1 gene also show increased susceptibility to light damage (Mittag et al., 1999), suggesting a role for reactive oxygen species as mediators of damage. As it has been demonstrated that rhodopsin is the initial mediat or of retinal light damage, it is perhaps unsurprising that mutations affecting this gene have been associated with increased susceptibility to such damage. Both th e P23H and S334ter rhodopsin mutations have been associated with increase d light sensitivity in rat and mouse models of ADRP (Ranchon et al., 2003; Wang et al ., 1997). Indeed, dark-reared mice containing the P23H mutation show substantial reduction in the rate of retinal degeneration when compared to P23H litters raised in normal cyclic light, suggesting that cell death as a result of ADRP could share at least some apoptotic mechanisms with certain forms of light-induced damage (Naash et al., 1996). Recently a dog model of human retinitis pigmentosa containing a tyrosine to arginine mutation at rhodopsin codon 4 has been associated with an ex treme sensitivity to light damage (Cideciyan et al., 2005). Thes e animals were subjected to focused retinal light exposures with light inte nsities that were 1500 to 6000 times less intense than those typically used in animal models of LMD. The researchers reported significant loss of retinal thickness at the site of light ex posure, as measured by optical coherence tomography. This result is intriguing, as both T4R and T17M mutations have been shown to affect glycosylation at the amino terminus of rhodopsin (K aushal et al., 1994; Zhu et al., 2004).

PAGE 87

74 Given the association of rhodopsin mutations with LMD, and considering that the subretinal injection te chnique utilized in our studies invo lved shining bright fiber optic light into the severely dilated retinas of anes thetized animals, experiments were designed to test whether this intense exposure to light could have a damaging effect on the retinas of mice containing the hT17M mutant rhodopsin transgene. This chapter describes those experiments. Materials and Methods Retinal Illumination Breedings were arranged as described above to create mrho+/litters, of which a portion would also contain the hT17M mutant rhodopsin transgene. At weaning age (2124 days) these litters were removed from thei r parents and their righ t eyes were dilated with 1% atropine. The next morning, the right eyes were again dilated with 1% atropine, and again an hour before the illumination proced ure, at which time the eyes also received a drop of 2.5% phenylephrine and 0.5% propara caine HCl. An hour after this final dilation, the animals were anesthetized and ag ain treated with a drop each of 1% atropine, 2.5% phenylephrine, and 0.5% proparacaine HCl. The right eyes of these animals then received a drop of 2.5% hypomellose to aid in retinal visualization a nd to help keep the retina hydrated, and were illuminated with a Southern Micro Instruments 150 Watt fiber optic light source with Schott Fostec fiber optic arms at an intensity of 10,000 or 5,000 lux for a period of 2.5 minutes. Light intensi ties were measured with an Extech Data Logging Light Meter (Extech, Waltham, MA). Retinas were visualized under a dissecting microscope, as describe d previously for the subretinal injections, to ensure that the pupils remained dilated and that the light remained focused on the retina throughout the duration of the experiment. Both eyes of the animals then received smears of VPP

PAGE 88

75 ointment, and the mice were allowed to recove r on a warming tray. This experiment was designed to closely mimic the subretinal injection protocol in all ways except for the actual injection. Genotyping Tail snips were taken from the mice while they were anesthetized for retinal illumination, funduscopy, or in the case of the animals used for histological examination, after they were sacrificed. Genomic DNA is olation and PCR analysis was performed as described above to identify animals c ontaining the hT17M mutant human rhodopsin transgene. Electroretinography Electroretinography was performed as desc ribed above. Statistical comparisons between the illuminated and non-illumated eyes were performed to generate P values using the paired, one-tailed Students t-test feature of Exel spreadsheet software (Microsoft, Redmond, WA). Funduscopy Funduscopy was performed as described above. Histology Animals were sacrificed by overdose of Isoflurane, followed by cervical dislocation. Eyes were enucleated, and a sm all hole was placed in the cornea with an insulin needle. They were then fixed overnight at 4oC in freshly-made 4% paraformaldehyde. The next day they were in cubated in solutions of sucrose diluted in phosphate buffer (pH 7.4) at concen trations of 7% (2 hours at 4oC), 15% (2 hours at 4oC), and 30% (overnight at 4oC), for cryoprotection. After the final incubation, the eyes were

PAGE 89

76 suspended in 15x15x5mm disposable base mo lds (Electron Microscopy Sciences, Ft. Washington, PA) in Tissue Tek OCT Co mpound Embedding Medium (Sakura Finetek, Torrance, CA) such that the cornea and optic nerve formed an axis parallel to the bottom of the mold, with the cornea to the front. The blocks were then frozen in isopentane at a temperature of -40oC. Frozen eyes were stored at -80oC. 12-14 micron retinal sections were then obtained from these frozen ey es using a Microm H550 (Microm, Walldorf, Germany) cryostat, with partic ular care taken to obtain sections around the optic nerve. Fisherbrand (Fisher Scientific, Pittsburgh, PA) Superfrost Plus microscope slides of size 75x25x1.0mm were used to collect the sect ions, which were stored at -80oC. TUNEL Visualization of Apoptosis DNA fragmenting, a characteristic of apopt osis, was detected using a terminal deoxynucleotide-mediated nick end-labeling (T UNEL) assay. For these experiments, the In Situ Cell Death Detection Kit, TMR red (Roche Applied Science, Mannheim Germany) was used, as per the manufacturers instructions. In br ief, sections were thawed for twenty minutes, and then washed twice for five minutes in room temperature 1X PBS. Next, the sections were permeabli zed in a solution of 0.1% sodium citrate and 0.1% Triton X-100 detergent for two minutes on ice. The sections were then washed twice for ten minutes with 1X PBS at r oom temperature. A hydrophobic slide marker pen (Daido Sangyo Co., Ltd., Tokyo, Japan) was us ed to surround the retinal sections so that the TUNEL reagents would not leak off. TUNEL label mix was subsequently added to the slides as per the manufacturers inst ructions. Cover slips (size 24x60mm, Fisher Scientific) were added, and the sect ions incubated in the dark at 37 oC in a humid chamber. After an hour, the cover slips were gently removed, the se ctions were rinsed three more times in room temperature 1X PBS for 10 minutes each wash, dried briefly,

PAGE 90

77 mounted with Vectashield Mounting Medi um with DAPI (Vector Laboratories, Burlingame, CA), and re-covered. Cover slip s were sealed to the slides with Sally Hansen Double Duty Nail Polish (Del Laborat ories, Inc., Farmingdale, NY). Apoptotic cells fluoresced red when visualized and photographed on a Zeiss Axiokop 2 mot plus microscope utilizing Axiovision 4 software (Zeiss International). Results High Intensity Illumination The right eyes of an experimental gr oup consisting of six mrho+/mice and six mrho+/-; hT17M mice, aged 21-23 days, were illuminated as descri bed with light of 10,000 lux intensity for 2.5 minutes. A peri od of 2.5 minutes was chosen for the illumination because it is the approximate tim e an animal is exposed during an actual injection procedure. Electr oretinographic analysis was th en performed on the mice at intervals of 1, 3, and 5 weeks, and the results plotted (Figures 4-1 and 4-2). The results demonstrate a significant decrease of around 30% in both aand b-wave responses at each time point in the illuminated eyes of animals containing the hT17M transgene, but not their non-transgenic siblings. Low Intensity Illumination One possible way to reduce the damage caused by retinal illumination during subretinal injections would be to reduce the intensity of th e fiber optic light used to visualize the retina. To test whether a reduction in intensity could decrease or eliminate light-induced retinal degeneration, nine mr ho+/and eight mrho+/-; hT17M mice were subjected to illumination as described, with th e intensity of illumination reduced to 5,000 lux. Electroretinographic analysis was perf ormed on the mice at intervals of 1, 3, and five weeks after illumination, and the results pl otted (Figures 4-3 and 4-4). Low intensity

PAGE 91

78 Figure 4-1. A-wave ERG res ponses at one, three, and five weeks after 10,000 lux intensity, 2.5 minute illu mination of mrho+/and mrho+/-;T17M mice. Asterisks indicate a difference between th e right eye and the left eye with a P value of less than 0.05. Figure 4-2. B-wave ERG res ponses at one, three, and five weeks after 10,000 lux intensity, 2.5 minute illu mination of mrho+/and mrho+/-;T17M mice. Asterisks indicate a difference between th e right eye and the left eye with a P value of less than 0.05. 2.5 Min Dissecting Scope Light Exposure 10dB A Waves 0 50 100 150 200 250Response in uV T17M Left T17M Right Rho +/Left Rho+/Right **1 Week 3 Weeks 5 Weeks 2.5 Min Dissecting Scope Light Exposure 10dB B Waves0 100 200 300 400 500 600 700 800Response in uV T17M Left T17M Right Rho +/Left Rho+/Right * *1 Week 3 Weeks 5 Weeks

PAGE 92

79 Figure 4-3. A-wave ERG res ponses at one, three, and five weeks after 5,000 lux intensity, 2.5 minute illu mination of mrho+/and mrho+/-;T17M mice. Asterisks indicate a difference between th e right eye and the left eye with a P value of less than 0.05. Figure 4-4. B-wave ERG res ponses at one, three, and five weeks after 5,000 lux intensity, 2.5 minute illu mination of mrho+/and mrho+/-;T17M mice. Asterisks indicate a difference between th e right eye and the left eye with a P value of less than .05. 2.5 Min Dissecting Light Exposure 10dB A waves0 50 100 150 200 250 300ERG Response (uV) T17M Left T17M Right Rho+/Left Rho +/Right 2.5 Min Dissecting Light Exposure 10dB B waves0 100 200 300 400 500 600 700 800ERG Response (uV) T17M Left T17M Right Rho+/Left Rho +/Right **1 Week 3 Weeks 5 Weeks 1 Week 3 Weeks 5 Weeks* *

PAGE 93

80 illumination led to a 20-30% decrease in both aand b-wave amplitudes at all three time points. These results closely resembled th e results observed afte r the similar, highintensity illumination experiments. Apoptosis in Retinas Damaged by Low Intenisty Illumination Light damage to the retina has been extens ively studied. It has been demonstrated that light-induced retinal damage is caused by apoptosis of photoreceptor cells (Hao et al., 2002; Farrar et al., 2002). It is this cell death that is responsible for the depressed ERG responses that are seen in light-damaged animals. To test whether apoptotic cell death was the cause of the ERG reduction that was seen in the retinal illumination experiments, two mrho+/and three mrho+ /-; hT17M mice were subjected to 5000 lux illumination as described previously. The anim als were sacrificed after 24 hours, their eyes were fixed and sectioned, and TUNEL labe ling was used to visualize apoptosis. The results, shown in Figures 4-5 and 4-6, show evidence of intense photoreceptor cell death in the illuminated retinas of the animals contai ning the hT17M transgene, but not in their non-transgenic littermates. In order to document the extent of photor eceptor apoptosis, pan-retinal images were obtained from the TUNEL stained sections using tile-field mapping of 20X images on a Zeiss Axiophot Z microscope equipped with a Sony DXC-970MD 3CCD Color Vid Camera and an MCID Elite Stage, utiliz ing MCID (Imaging Research, Inc., Ontario, Canada) Analysis Software (Imaging Researc h, Inc.). The results, shown in Figure 4-6, show that the photoreceptor damage is panretinal, rather than tightly localized. Funduscopic Illumination Funduscopic examination is one of the most common ophthalmologic procedures, often performed either as part of a routine phys ical or complete eye examination to detect

PAGE 94

81 Figure 4-5. TUNEL stained re tinal sections from mrho+ /and mrho+/-; hT17M mice whose right eyes were illuminated wi th 5000 lux white light for 2.5 minutes. Sections of mrho+/mice (rows A a nd B) show no evidence of apoptosis, while sections from mrho+/ -; hT17M mice (rows C, D, and E) show extensive apoptosis, as evidenced by large numbers of red-labeled photoreceptor nuclei. In each row, left and right eye sections are from the same experimental animal. C E D A B Right Left

PAGE 95

82 Figure 4-6. Tile-field mapped image of th e TUNEL stained retina of an eye from a mrho+/-; hT17M animal that was illuminated for 2.5 minutes with 5,000 lux light. Arrows indicate pan-retinal re d fluorescence indicative of apoptotic photoreceptors. and evaluate symptoms of eye disease, such as glaucoma or retinal detachment, or if diabetes, hypertension, or othe r vascular disease is suspec ted. The characteristic bone spicule deposits associated with retinitis pi gmentosa are among the indicators that a patient presenting with reduced visual fiel ds and impaired night vision is actually suffering from RP. During this procedure, the back of the retina is visualized through the dilated iris of the patient usi ng a bright white light. If phot ographs of the retina are taken, they too must utilize intense flashes of li ght to record their images. It has been demonstrated that funduscopic examination is damaging to th e retinas of dogs containing a T4R rhodopsin mutation (Cideciyan et al., 2005). In order to determine if funduscopic exam ination and photography of the retinas of mice carrying the hT17M human mutant rhodopsin transgene wa s harmful, eight mrho+/mice, four with the hT17M transgene and f our that were non-transgenic, had two fundus pictures taken of their right eyes at three a nd six weeks of age. On e week after each set

PAGE 96

83 of photographs, electroretinogra phy was performed as describe d, and the results averaged and plotted (Figure 4-7). These ERG recordi ngs show clearly depr essed aand b-wave amplitudes in the hT17M transgenic mice following both sets of funduscopic photography, while their non-transgen ic littermates were unaffected. The first set of fundus pictur es show no evidence of retinitis pigmentosa in either the mrho+/mice or their mrho+/-; hT17M litte rmates. However the second set of fundus images, taken at six weeks of age, reveal punctate regions of th e retina in the hT17M transgenic animals, suggesting loss of pigm ent in the retina. The images of the nontransgenic mice looked normal. These results are summarized in Figure 4-8. Apoptosis in Retinas Damaged by Fundus Photography It seemed reasonable to assume that the depression of ERG response seen in hT17M transgenic mice following fundus photography would be accompanied by photoreceptor apoptosis, as was noted with the animal s damaged by low-intensity fiber optic illumination. In order to confirm this, two mrho+/littermate mice, one containing the hT17M mutant rhodopsin transgene and one th at was non transgenic, were subjected to fundus photography of the right eye at 21 days of age. One day later, the animals were sacrificed, and their eyes were enucleated, fi xed and sectioned. S ections containing the optic nerve were then TUNEL stained to visualize apoptotic cells, with DAPI counterstain to reveal retina l morphology. The results, illust rated in Figure 4-9, show that fundus photography clearl y induced apoptosis in the rod photoreceptor cells of mrho+/-; hT17M mice, but not in their non-hT17M littermates.

PAGE 97

84 Figure 4-7. A and b-wave ERG responses of mrho+/and mrho+/-; hT17M mice subjected to fundus photography at 3 week s (first series) and 6 weeks (second series) of age. indicates significant difference between right and left eyes with a P value of less than 0.05. A Wave ERG Following R Eye Funduscopy0 50 100 150 200 250ERG Response in uV Rho +/-;T17M Left Rho +/-;T17M Right Rho+/Left Rho+/Right First Series Second Series* B Wave ERG Following R Eye Funduscopy0 100 200 300 400 500 600 700ERG Response in uV Rho +/-;T17M Left Rho +/-;T17M Right Rho+/Left Rho+/Right First Series Second Series

PAGE 98

85 Figure 4-8. Fundus pictures of mrho+/(left sets) and mrho+ /-; hT17M (right sets) mice at three and six weeks of age. Evidence of retinal degenerati on is seen in the six week set of hT17M transgenic mice. Week 3 Week 6 Week 3 Week 6 mrho+/mrho+/-;hT17M

PAGE 99

86 Figure 4-9. TUNEL labeling of retinal sections from mice whose right eyes underwent fundus photography at 21 days of age. Sections of mrho+/mice (bottom row) show no evidence of apoptosis, wh ile sections from mrho+/-; hT17M mice (top rows) show extensive apoptosi s, as evidenced by large numbers of red-labeled photoreceptor nuclei. Images in the top row were taken at 20X magnification; those in the bottom two rows were taken at 40X magnification. ERG Analysis of hP23H Mice Aft er High Intensity Illumination As has been discussed, rats and mice c ontaining a human rhodopsin transgene that contains a proline to histidine mutation at c odon 23 have been extensively used to study and model retinal disease. Mice that are bred to be hemizygous null at the mouse rhodopsin locus and that also contain th e hP23H transgene undergo rapid retinal LeftRight mrho+/-; hT17M+ mrho+/-; hT17M+ mrho+/-;

PAGE 100

87 degeneration with concomitant photoreceptor lo ss, culminating with a 60% loss of ERG response at three months of age. This model has been successfully treated with ribozymes designed to specifically cleave th e mutant human transgene (F. Fritz UF doctoral dissertation and unpublished results). Since procedures used for subretinal injec tion of both lines of mice are identical, it was important to explore the effect of focu sed fiber optic light on the retinas of mice containing the hP23H transgene. If the hP23H mice showed light sensitivity similar to the hT17M transgenics, it could have a ne gative impact on the ultimate success of treatment with subretinal ribozyme injections. It was also of interest to determine whether these two closely located mutations, which lead to different rates of retinal degeneration, would respond differently to intense retina l illumination. Our initial experiments examined the effect of high intensity fiber optic light on the ERG responses of mrho+/-;hP23H mice. Eight rhodopsin hemizygous mice containing the hP23H transgene and their five non-transg enic siblings were exposed in their right eyes to 2.5 minutes of 10,000 lux intensity lig ht, as described above. The animals then underwent ERG analysis at one, three, and fi ve weeks post-illumination to determine the effect of the light exposure upon their visual response. Th ese data are summarized in Figure 4-10. Although there were significant depressions of aand bwave ERG amplitudes in the hP23H transgenic mice after illuminati on, the differences averaged around a 10% reduction of aand bwave responses at 1 week after illumination, as compared to around 26% depression in the hT17M transgen ic mice. Furthermore, the hT17M mice continued to exhibit reducti on in aand b-wave ERG responses at five weeks post

PAGE 101

88 illumination, while the hP23H transgenic mice showed only a statistically insignificant 5% depression of both aand bwave amplit udes at this time poi nt. Additionally, the non-transgenic littermates used in these experiments also exhibited an initial ERG reduction of around 5 to 10%, one of which was actually statistically significant, at the initial ERG time point (Table 4-1). The fact that the hP23H transgenic animals were not more sensitive to high intensit y retinal illumination than th eir non-transgenic littermates taken together with the transient, moderate severity of the phenotype when compared to the hT17M transgenic mice seems to indicate that the hP23H mutation does not confer profound sensitivity to high intensity retinal irra diation of the type used to facilitate subretinal injections or fundus photography. Table 4-1. Percent reduction in eyes illu minated 10,000 lux white light. Illumination effects are shown for mrho+ /-; hP23H mice (left sets ), mrho+/-; hT17M mice (center sets), and mrho+/littermates (right sets). Numbers represent the percent difference between the right (illuminated) and left (unilluminated) eyes. A + sign indicates that the right eye response was greater than the left eye response. indicates significant differences with a P-value of less than 0.05. mho+/-; hP23H mrho+/-; hT17M mrho+/a-wave b-wave a-wave b-wave a-wave b-wave 1 Week 14%* 7%* 22% 29%* +4% +2% 3 Weeks 14%* 7%* 20%* 32%* 1% 4% 5 Weeks 4% +3% 43%* 28%* +5% 0% n=8, n=6, n=11

PAGE 102

89 Apoptosis in hP23H Mouse Retinas A fter High Intensity Illumination The results discussed above seemed to demonstrate that hP23H transgenic mice were not susceptible to light damage in a manner similar to mice containing the hT17M transgene. To confirm this result, hist ology and TUNEL staining were performed as described previously on four mrho+/-;hP23H and one mrho+/mice were illuminated with 10,000 lux fiber optic light in their right eyes for 2.5 minutes. hP23H transgenic mice exhibited no increase in a poptosis in their illuminated eyes either compared to their own unilluminated left eyes, or the illuminate d right eye of a non transgenic littermate (Figure 4-11). A positive control involving DNase treatment prior to TUNEL labeling was performed on a section from a non transgen ic, non-illuminated eye in order to ensure that the lack of TUNEL-positive photoreceptor s was not due to any problems with the TUNEL kit itself. These results support the conclusion that hP23H transgenic mice are not as sensitive to retinal light damage as their hT17M transgenic cousins. Discussion Investigation of the injection dama ge phenomenon observed in the hT17M transgenic mouse line led to the discovery of an acute light sensitivity in these transgenic animals. Illumination for 2.5 minutes w ith both 10,000 and 5,000 lux white fiber optic light caused severe attenua tion of the ERG response in these mice, with around a 30% reduction seen in both aand b-waves out to five weeks post-illu mination. Histological analysis followed by TUNEL labeling revealed significant amounts of apoptosis localized to the photoreceptor cells of th e ONL in the illuminated eyes of hT17M mice. This result supports the conclusion that light exposure of the duration and intensity used for subretinal injections caused a poptotic photoreceptor cell death.

PAGE 103

90 Figure 4-10. 10dB intensity aand b-wave ER G responses at one, three, and five weeks after 2.5 minute, 10,000 lux illuminati on of mrho+/and mrho+/-;hP23H mice. indicates a right to left eye difference with a P value of less than 0.05. Rho+/and P23H Mice with High White Light Illumination B Waves0 200 400 600 800ERG Response (uV) P23H Left P23H Right Rho+/Left Rho+/Right 1 Week 3 Weeks 5 Weeks Rho+/and P23H Mice with High White Light Illumination A Waves0 50 100 150 200 250 300ERG Response (uV) P23H Left P23H Right Rho+/Left Rho+/Right 1 Week 3 Weeks 5 Weeks* * *

PAGE 104

91 Figure 4-11. TUNEL labeling of mouse retinal s ections from mice that were irradiated in the right eye with 10,000 lux white fiber opt ic light at 21 days of age. Rows A-D are from mrho+/-;hP23H mice. Ro w E contains (left) a DNase treated positive control eye from an mrho+/;hP23H mouse and (right) a TUNEL labeled, illuminated right ey e from a non-hP23H mouse. LeftRight A B C D E

PAGE 105

92 In 1978, researchers reported retinal da mage in owl monkeys resulting from intraocular illumination with fiber optic whit e light similar to that used to perform vitrectomies in human patients (Fuller et al ., 1978). These findings coupled with other examples of light-mediated damage in normal, healthy animals, led to concerns that other common ophthalmologic techniques that involved retinal illumination, such as funduscopic examination and photography, could also be damaging (Lanum, 1978). Fundus photography is a routine procedure that involves photogr aphing the retina through the magnifying lens of an apparatu s called a funduscope. As the procedure involves bright flashes of white light focu sed on a dilated retina, we explored the possibility that this procedur e could cause damage in the light sensitive hT17M mice. ERG analysis confirmed that fundus photogra phy resulted in signi ficant photoreceptor damage in the hT17M transgenic mice, but not in their non-transgenic littermates. Analysis of TUNEL stained retinal sections following the procedure revealed that this damage was the result of wide spread photoreceptor apoptosis. We also wanted to explore whether this light sensitivity would be observed in a mouse model of ADRP that contained a P23H mutant human rhodopsin transgene. ERG analysis demonstrated that th is was not the case, as the hP23H transgenic animals showed only a 5% decrease in the a-wa ve and b-wave responses of the illuminated eyes after 2.5 minute illumination of 10,000 lux intensity white fiber optic light at one week after the procedure. There was no si gnificant aor b-wave differe nce between illuminated and non-illuminated eyes observed at five weeks after the procedure. Analysis of TUNEL stained retinal sections from hP23H mice 24 hours postillumination revealed no

PAGE 106

93 significant photoreceptor a poptosis, supporting the ERG re sults. These experiments suggest that the extreme light sensitivity obs erved in the hT17M tr ansgenic mice is not common to all rhodopsin mutants. Analysis of other rhodopsin mutants using these illumination parameters would be helpful in determining whether this is a rare or common feature of mutations that cause ADRP. This light sensitivity, while intriguing from a scientific point of view, presents a problem for our therapeutic strategy. Prev ious results of rAAV-delivered, ribozymemediated therapy for ADRP in a rat model containing a P23H mutant human rhodopsin transgene resulted in a 30% rescue of ERG re sponse. Such a ther apeutic outcome, if it could be achieved in our hT17M mouse line, would be comple tely masked by the retinal damage produced by the light exposure asso ciated with our s ubretinal injection technique. Reducing the intensity of illumination by 50% (from 10,000 to 5,000 lux) did not alleviate the damage, and after damage was seen in the hT17M mice following the incredibly brief light flashes associated with fundus photography, it became clear that special measures would be required to overcome this therapeutic obstacle.

PAGE 107

94 CHAPTER 5 RED FILTERED LIGHT FOR INJECTIONS PROTECTS AGAINST LIGHT DAMAGE IN THE mrho+/-; hT17M MOUSE Introduction Results from the previous chapter clearly demonstrate a mutation-specific sensitivity to light exposure in hT17M transgenic animals. This sensitivity lead to a 2030% decrease in ERG amplitudes after exposur e to white light of the type used to visualize the retina during s ubretinal injection of ther apeutic rAAV. Lowering the intensity of this light did not protect agains t retinal damage. Since our ability to treat these animals with ribozyme-expressing, r ecombinant AAV depends upon a subretinal injection protocol that is not harmful to the visual response, we decided to explore ways to modify our current injecti on technique to make it less harm ful to the hT17M transgenic mice. Rhodopsin absorbs light at a peak wavele ngth of 500nm. As one moves towards longer or shorter wavelengths of light, the ab ility of rhodopsin to ab sorb light decreases markedly (Figure 5-1). Red filtered light contains primarily wavelengths above 600nm, which do not activate rhodopsin, or activate it only poorly (F igure 5-1). This is the rationale behind using red-filtered headlamp s while handling dark-adapted animals that are to undergo ERG analysis. As mentioned before, several studies have implicated rhodopsin as the effector molecule in animal models of light-mediated retinal damage. The damage we have described in the preced ing chapters affected only mice expressing

PAGE 108

95 mutant rhodopsin; it therefore seemed reasona ble to assume that activation of mutant rhodopsin by light was the key causative agent of LMD in the rho+/-; hT17M mice. In order to create an injection pr otocol that would allow us to visualize the injected eye while not activating the mutant rhodopsin, we decided to fi lter the light we used for subretinal injections like th at used in our darkroom head lamps, so that it contained wavelengths longer than 600nm. Figure 5-1. Absorbance spectrum of rhodopsin (black), shown together with the measured transmittance spectrum of the re d plastic used to create the 600nm red light filters. Note the small overlap of spectra at around 600nm. This chapter describes the creation a nd characterization of 600n m red light filters for use in subretinal injections of animals sensitive to light damage. The architecture and Rhodopsin Red Filter 100% 75% 50% 25% 0% 400500 600700 Wavelength (nm)Absorbance / Transmittance

PAGE 109

96 absorbance characteristics of the filters will be discussed, as will experiments designed to test the new filters for their ability to illumi nate the retina sufficiently for successful subretinal injections while re ducing or eliminating the retin al damage associated with exposure to unfiltered fiber optic light. Materials and Methods Creation of 600nm Filters Plastic photographic filters of the Cokin 003 variety were obtained from a local photographic supply store. The filters were analyzed in a Hewlett-Packard 8452A Diode Array Spectrophotometer (Hewlett-Packard Comp any, Palo Alto, CA) to determine their absorbance/transmittance spectrum, which is shown in Figure 5-1. A Dremel Minimite model 750 modeling tool (Dremel, Racine, WI) was used to excise sections of these filters that were around 2cm in diameter. These sections were then glued to cylindrical sections cut from 15 ml plastic test tubes (Sarstedt Inc., Newton, N.C.). The resulting filters, shown in Figure 5-2 (left image), could then be attached to the ends of the arms of Figure 5-2. 600nm red light filters (left) are attached to the fiber optic light sources used for subretinal injections to create an apparatus (right) that filters out harmful wavelengths of light duri ng subretinal injections.

PAGE 110

97 the fiber optic light source used for subretinal injections, creating an apparatus that gave proper injection illumination while filteri ng out light with wavelengths below 600nm (Figure 5-2, right image). Retinal Illumination Although the red filters reduced the intensity of transmitted light, at full power they were still able to produce il lumination of an intensity of 5,000 lux. This intensity was used for retinal illumination as described previously, with the right eyes of the experimental mice exposed to 5,000 lux inte nsity red-filtered (greater than 600nm wavelength) light for a duration of 2.5 minutes. Electroretinography Electroretinography was performed as desc ribed above. Statistical comparisons between the illuminated and non-illumated eyes were performed to generate P values using the paired, one-tailed Students t-test feature of Exel spreadsheet software (Microsoft, Redmond, WA). Histology Histology was performed as described above. TUNEL Visualization of Apoptosis TUNEL labeling of retinal sections and visualizati on of apoptotic photoreceptor cell death was performed as described above. Test Injections Using 600nm Filtered Light Injection procedures were as describe d, with the excepti on that the 600nm red filters were attached to the fiber optic light source as depicted in Figure 5-2. Mice of

PAGE 111

98 weaning age (21-24 days of age) received 0.5 l subretinal injections of Lactated Ringers solution (Abbott Laboratorie s, Abbott Park, IL). Results Spectrophotometric Analysis of 600nm Filters In order to ensure that the red photogra phic plastic we obtained for the purpose of creating our injection filters actually filtered out light below 600nm, the transmittance spectrum of the plastic was determined. The results, which are shown merged with the known absorbance spectrum of rhodopsin in Figu re 5-1, indicate that they efficiently excluded light with wavelengths below 600nm. Most importantly, the spectrum contained no holes below 600n m, which would appear as sp ikes of light at selected wavelengths below 600nm, and would indicate imperfections of the filters that would pass light of wavelengths which could activat e rhodopsin (and thus damage the retina). Additionally, the 600nm transmittance cutoff wa s a sharp, almost vertical line, as opposed to a gradual curve up to a 600nm cutoff, which shows that these filters only pass a small amount of light with wavelengths below immediately below 600nm that can overlap with the rhodopsin absorbance spectrum (and thus activate rhodopsin). ERG Analysis After 600n m Retinal Illumination After determining the transmittance characteristics of and building the 600nm red light filters, the effects of red-shifted illumination on the retinas of mice containing the hT17M transgene were tested. The intensity of light transmitted th rough the filters when the source was turned to full power was 5,000 lux. Therefore, using th e red filters at the full power setting created illumination conditions that were identical to those in the low white light study that led to re tinal damage, with the exception that all the light in this study was filtered to be of a wavelength greater than 600nm.

PAGE 112

99 A large number of experimental animals were used in these experiments in order to ensure that any protective effect seen from light filtration was reproducible and statistically significant. Thirty four mice were analyzed in all, of which fourteen were mrho+/-; hT17M mutant animals and twenty were mrho+/control siblings. The animals were subjected as described to 2.5 minute illumination in the right eye with 5,000 lux Figure 5-3. A wave (top) and B wave (bottom) amplitudes at one, three, and five weeks after right eye illumination with 600nm filtered light. 14 hT17M transgenic mice and 20 non-transgenic littermates were analyzed. Asterisks indicate difference between right and left ey es with a P value of less than 0.05. Rho +/and T17M Red Light Illuminated A Waves 0 50 100 150 200 250 300ERG Response (uV) T17M Left T17M Right Rho+/Left Rho +/Right Rho +/and T17M Red Light Illuminated B Waves 0 100 200 300 400 500 600 700ERG Response (uV) T17M Left T17M Right Rho+/Left Rho +/Right 1 Week 3 Weeks 5 Weeks * * * 1 Week 3 Weeks 5 Weeks

PAGE 113

100 intensity light passed through the 600nm f ilters. The animals then underwent ERG analysis at one, three, and five weeks afte r illumination, and the average aand b-wave maximum amplitudes were graphed and comp ared between hT17M and non-hT17M sets. The results, shown in figure 5-3, dem onstrate that the 600nm filters provide substantial protection against light induced damage in the hT17M transgenic mice. Although there was a statistically significant reduction in aand b-wave amplitudes at one and three weeks post illumination, these effects were not as substantial ( an average of 15% difference in a-wave responses and 8% difference in b-wave responses) when compared to the previously described eff ect of 5,000 lux illumination with unfiltered light in the same line, which resulted in an average a-wave re duction of 35% and an averate b-wave reduction of 24% at these time points. The damage was also transient, and at the five week ERG measurement th e right and the left eye of the hT17M transgenic animals showed identical aand b-wave ERG responses (Table 5-1). Apoptosis in Retinas Exposed to 600nm Illumination In order to ensure that the protection seen at the ERG level was mirrored by a corresponding lack of apoptos is in rod photoreceptors, histological sectioning and TUNEL labeling was performed. Five mrho+ /-; hT17M mice and one mrho+/littermate underwent 2.5 minute right eye illumination with 5,000 lux of 600nm filtered light, as described above. A day later, the animals were sacrificed, their eyes fixed and sectioned, and sections containing the optic nerve were TUNEL labeled as described in order to identify apoptotic photoreceptors. The results of this experiment, illustrated in Figure 54, demonstrate that 600nm illumination does not lead to a substantial increase of apoptotic photoreceptor cells in the illuminated right eyes of hT17M transgenic mice.

PAGE 114

101 Test Injections Using 600nm Filtered Light Having demonstrated through ERG analys is and TUNEL labeling of retinal sections the protective effect of filtering wa velengths below 600nm from the light used for retinal illumination, we decided to attempt test injections under th e red-filtered retinal Table 5-1. Percent reduction in eyes illumina ted with red or white light. Illumination effects are shown for both mrho+/mice (left sets) and mrho+/-; hT17M littermates (right sets). Numbers repr esent the percent difference between the right (illuminated) and left (unilluminated) eyes. A + sign indicates that the right eye response was greater than th e left eye response. indicates significant differences with a P-value of less than 0.05. mrho+/mrho+/-; hT17M white a-wave red a-wave white b-wave red b-wave white a-wave red a-wave white a-wave red a-wave 1 Week 0% 4% +4% 4% 39%* 16%* 30%* 9%* 3 Weeks +8% 0% +3% +1% 31%* 13%* 18%* 6%* 5 Weeks +5% 5%* +3% 7%* 28%* 1% 15%* +2% n=9, n=20, n=14, n=8 illumination. Five mice received subretinal injections of 0.5 l Lactated Ringers solution under these conditions. Tail snip s were obtained while the animals were under anesthesia for the procedure. Subsequent DNA isolation and genotyping showed th at three of the

PAGE 115

102 Figure 5-4. TUNEL labeled right and left eye se ctions from eyes that were subjected to 2.5 minute illumination with 5000 lux, 600nm filtered light. Images in rows A-E are from mrho+/-; hT17M mice; imag es in row F are from a mrho+/non transgenic littermate control mouse. Refer to chapter 4 for comparison with images from hT17M mouse eyes illuminated with unfiltered light. Left Eye Right Eye A B C D E F

PAGE 116

103 Figure 5-5. ERG responses measured two w eeks after subretinal test injections performed under 600nm filtered fiber op tic illumination. A-C are tracings from mrho+/-; hT17M transgenic mice. D and E are from non-transgenic littermates. Although the inje ctions shown in panels C and D appeared to be failures, meaning that the procedur e itself caused physical damage that resulted in a total loss of aand bwave responses by two weeks, the rest, including two performed upon hT17M tran sgenic mice, were well-tolerated. C E D A B

PAGE 117

104 five animals were mrho+/-; hT17M transgen ic animals, while the other two did not contain the mutant human transgene. ER G analysis was performed two weeks after injection. The results, illustrated in Figure 5-5, demonstrate the feas ibility of performing subretinal injections under this type of illumination. Although there were two failed injections that led to total loss of ERG response, thes e were seen in only two of the fiveanimals, and involved both an hT17M transgenic mouse and a non-transgenic littermate. Most importantly, two of the hT17M transgenic mice tolerated subretinal injection with preservation of ERG response th at was far better than that seen using unfiltered white light. Discussion In 1977, Adrian et al. described the creation of a brownish ophthalmic filter which absorbs the short wavelengths preferentially, thus protecting the rods primarily in an attempt to protect RP patients from light-m ediated damage (Adrian et al., 1977). The authors go on to state Whether or not use of th ese filters will be effi cacious has yet to be determined and will require careful experime ntation and the accumulation of clinical experience. Almost thirty years later, we can say that in the case of the damage resulting from retinal illumination in the hT 17M transgenic mouse model of ADRP, such filtration can be very effective. We created filters using phot ographic plastic that would only pass red-shifted light, containing wavelengths longer th an 600nm. The 600nm filters passed light of a sufficient level to allow an experienced researcher to perform a subretinal in jection. However, red light illumination at the 5,000 lux inte nsity did not result in persistent electroretinographic aor b-wave attenuation. Additionally, the retinal histology of animals illuminated with this type of light appeared normal at 24 hours following the

PAGE 118

105 procedure, and there was lit tle increase in apoptotic photoreceptors relative to the unilluminated eye, as assayed by TUNEL labeling of retinal sections. We utilized these filters to provide illu mination during a preliminary experiment involving subretinal Lactated Ri ngers injections in five mi ce, three of which contained the hT17M transgene. ERG analysis perfor med two weeks after the injection revealed that although there were two fa iled injections, one in a mutant animal and one in a nontransgenic littermate, the other three mice, of which two contained the hT17M transgene, maintained good aand b-wave responses relati ve to the uninjected eye. This is an important result, as these are the first subretinal injections in this line that did not result in substantial retinal damage. The results dem onstrate the usefulness of the 600nm filters in preventing the damage caused by intense retinal illumination during s ubretinal injection. Future experiments will use these filters to facilitate subretinal injections of rAAV delivering ribozyme therapy in an attempt to treat the hT17M-mediated retinal degeneration. At least one other rhodopsin mutation, tyrosine to arginine at position four, has been shown to confer light sensitivity in dogs that is similar to that seen in our hT17M mouse line (Cideciyan et al., 2005). This is intriguing because of the fact that both mutations result in abolished N-termial r hodopsin glycosylation (K aushal et al., 1994; Zhu et al., 2004). It would be interesting to see if the filters deve loped in our laboratory would protect against light damage observed in the T4R dog model. It remains to be seen if other rhodopsin mutations will behave si milarly, although it does not appear that all will do so, as demonstrated by the experiments involving hP23H transgenic mice detailed in the previous chapter. However, gi ven the large number of rhodopsin mutations

PAGE 119

106 associated with ADRP, coupled with the ra nge of degenerative phe notypes conferred by them, it seems reasonable to predict that some of these mutations will also confer some degree of light sensitivity. Here we have described the creation of filters designed to prevent the lightmediated retinal damage associated with subretinal injections. The filters are inexpensive, simple to create, and provide su fficient illumination for successful subretinal injection. Given that the goal of gene ther apy for retinal disease is to effect the preservation of rod and cone photoreceptors the use of such f ilters when attempting subretinal injections intended to treat animal models of ADRP that involve rhodopsin mutations would seem a useful and desirable precautionary measure.

PAGE 120

107 CHAPTER 6 DISCUSSION Summary The preceding chapters detailed the crea tion of a novel hT17M transgenic mouse model of retinitis pigmentosa. Although th e hT17M transgene had been previously studied in a mouse background containi ng two copies of nonmutant rhodopsin (mrho+/+), to our knowledge this was the firs t examination of a mouse line in which the transgene was expressed on a hemizygous null rhodopsin background (mrho+/-). Expressing the mutant human rhodopsin transgen e in the presence of only one wild type mouse rhodopsin allele provides a closer a pproximation of autosomal dominant retinitis pigmentosa as it occurs in human populations, in which patients will most often present with one normal and one mutant copy of the rhodopsin gene. The mrho+/-; hT17M mice in this study s howed significant reduc tion in the outer nuclear layer (ONL), with concomitant reduc tion of aand b-wave ERG amplitudes to almost undetectable levels by around six a nd a half months. Si gnificant drop in ERG amplitudes was seen as early as 1 month of age. Furthermore, evidence of retinal damage was noted as early as three months of age when fundus photography was used to visualize the retinal morphology of the mutant mice. However, substantial ERG response and ONL thickness was still seen at 4.5 months, providing a window for therapeutic intervention that should have made the mrho+/-; hT17M line a good candidate model for testing recombinant AAV-delivered ribozyme th erapy. It was unfortunate that after

PAGE 121

108 creating a pair of highly activ e hammerhead ribozymes targeted to the mutant human transgene, packaging them and purifying and concentr ating them as high-titer recombinant AAV, we were unable to test the e ffectiveness of this therapy because of the significant damage caused during vector deliver y. However, exploring possible reasons for mutation-specific sensitivity to subretinal injection damage led us to discover the extreme light sensitivity of the mice in ques tion. This was an intriguing phenomenon in its own right, and one that may have signi ficant therapeutic implications, as will be discussed. The hT17M transgenic mice showed signi ficant reduction in bot h aand b-wave amplitudes of eyes that were illuminated with either 5,000 or 10,000 lux intensity white light for a period of 2.5 minutes. This reduc tion was seen as early as one week and persisted for at least five weeks follo wing illumination. Non-transgenic, mrho+/littermate mice were unaffected by this light exposure. Light e xposure due to fundus photography of the mutant mice likewise led to a reduction of ERG amplitude in the photographed eyes, while again th e non-transgenic mice showed no measurable ill effects from this procedure. TUNEL analysis of re tinal sections from mu tant mice exposed to both fiber optic illumination and funduscopic examination and photography revealed that there was significant photoreceptor apoptosis that was accompanied by a decrease in ONL thickness by as early as 24 hours following e ither procedure. These assays revealed no apoptosis in non-transgenic littermates su bjected to identical retinal illumination. These observations strongly support the conc lusion that the hT17M mutant rhodopsin gene confers significant light sensitivity to transgenic mice that express it.

PAGE 122

109 These observations caused us to explor e ways to more successfully perform subretinal injections of hT17M transgenic mice without causing light-mediated retinal damage. As the mutant rhodopsin transgene was required for the damage to occur, it seemed reasonable to assume that it was th e absorption of light by the mutant rhodpsin molecule that was the cause of the damage Since rhodopsin absorbs in the visible spectrum with a peak exhaustion coefficient of 500nm, we decided to illuminate our hT17M mice with 2.5 minutes of 5,000 lux inte nsity light that was filtered to pass no wavelengths below 600nm. Transgenic mice that were illuminated in this manner showed little or no ERG damage, and this was reflected by a lack of significant apoptosis in the illuminated retinas, as assayed by TUNEL labeling of retinal sections. The redfiltered light affords sufficient retinal visualization for experienced personnel to perform subretinal injections with no loss of speed, although the efficacy of injections performed under this type of illumination remains to be demonstrated. Recently, a mixed litter of mrho+/and mrho+/-; hT17M mice was inject ed with lactated ringers solution, and although two animals (one transgenic, one nont ransgenic) received significant injection damage, the other five animals, including two hT17M transgenic mice, showed good ERG responses relative to their uninjected ey es when analyzed at 2 weeks post injection. Light sensitivity also appear ed to be mutation specific. A second mouse model of ADRP was available for study, this one engine ered to express a P23H mutant human rhodopsin transgene on an mrho+/genetic b ackground. Despite exhibiting what seemed to be a more severe form of ADRP, the hP 23H transgenic mice were not susceptible to light-induced retinal damage when exposed to 10,000 lux fiber optic illumination for 2.5 minutes, as measured by both ERG analysis and TUNEL labeling of retinal sections.

PAGE 123

110 To summarize, we have bred the hT17M human rhodopsin transgene onto a hemizygous null (mrho+/-) mouse rhodopsin b ackground. ERG and histological analysis of this line show it to be a good model of autosomal dominant retinitis pigmentosa, with significant photoreceptor degeneration by 6.5 mo nths that is preceded by preservation of retinal function for a duration sufficient to provide a good therapeutic window for treatment. While the difficulties caused by il lumination with intense white light during subretinal injection we re unfortunate, these setbacks a llowed us to identify an extreme light sensitivity associated with the hT17M transgene. The use of 600nm red filters during subretinal injection of hT17M transgenic mice prevents the light-mediated damage, and will facilitate future experi ments designed to introduce rAAV-delivered, ribozyme-mediated therapy to th e retinas of these animals. Mechanism of Light-Induced Photoreceptor Apoptosis Light-mediated retinal damage is a phe nomenon that has received considerable study. Although there are many different theories as to the exact mechanism of LMD, all share two central points: first, rhodopsin is the initial mediat or of the damage (Grimm et al., 2000b; Sieving et al., 2001; Keller et al., 2001) and, second, apoptot ic cell death is the ultimate fate of the affected photoreceptors (Aonuma et al., 1999). Apoptotic cell death is also the causative event in the retinal de generation seen in patients suffering from ADRP arising from mutations in the rhodopsin gene. This has led researchers to study the pathways involved in light-mediated photoreceptor apoptosis with the goal of achieving a better understanding of apoptotic ph otoreceptor death in patients suffering from rhodopsin-mediated ADRP. In 2002, Hao and coworkers reported eviden ce for at least two apoptotic pathways involved in light-mediated retinal degene ration. One pathway, termed the acute

PAGE 124

111 pathway, was induced by 10 minute illumination of Balb/c mice with white light at an intensity of 5,000 lux, and was shown to be in dependent of transducin activity, meaning that the phototransduction cascade was not re quired for the induc tion of apoptosis observed in photoreceptor layers of these mi ce. They reasoned the acute pathway is caused by activated rhodopsin or its photo-bl eached products, based on the fact that mice deficient in both rhodopsin kina se and arrestin, which are in volved in the in activation of rhodopsin, are extremely sensitiv e to acute light exposure. This type of apoptosis was also found to be dependent upon expression of the transcription factor AP-1. A second light damage pathway, termed the low-inten sity pathway, was noted in animals with defects in either arrestin or rhodopsin kinase, which are involved in inactivation of photoactivated rhodopsin; these mice were s hown to undergo retinal degeneration upon prolonged exposure to normal, cyclic room li ght. In contrast to the acute pathway, transducin activity was cent ral to the low-intensity apoptotic pathway, and mice lacking a functional transducin gene were protected from this form of LMD. AP-1 expression was shown to be uninvolved with the low-intensity pathway (Hao et al., 2002). A possible mechanism for photoreceptor apop tosis resulting from the low-intensity pathway can be found in the equivalent-li ght hypothesis of Fa in and Lisman, who postulated that photoreceptor a poptosis resulting fr om ADRP could be the result of constant activation of the visu al cycle in both the presence and absence of light (Fain and Lisman, 1993). Such a defect would result in depressed intracellular levels of Ca2+, which has been shown to cause death in cu ltured neuronal cells (Woodruff et al., 2004; Fain and Lisman, 1999). Although the equiva lent-light hypothesis cannot explain all

PAGE 125

112 forms of light-mediated apoptosis, as eviden ced by the transducin-independent apoptosis seen in the acute model of Hao and co lleagues, it is possibl e that depressed Ca2+ levels are involved in the transducin-dependent photorecepto r apoptosis observed in the low intensity pathway. This is supported by expe riments showing that low-intensity lightinduced apoptosis in Balb/c mice is bloc ked by treatment with the calcium channelblocker Dcis -diltiazem, which presumably prevents efflux of Ca2+ from the cell via the Na2+ / C2+ exchanger following constitu tive loss of cGMP-mediated Ca2+ influx (Woodruff et al., 2004; Donovan and Cotter, 2002). AP-1 is a transcription fa ctor involved in light-media ted retinal apoptosis, and exists as a heterodimer with proteins from the Jun/ Fos families. Retinal exposure to damaging white light has been shown to lead to increased levels of mRNA for two of these proteins, c-Fos and c-Jun in mice (Gri mm et al., 2000a). E xperiments utilizing DNA microarrays to analyze retinal gene expression following light exposure have similarly confirmed upregulation of AP-1 (Che n et al., 2004). In ot her work, transgenic c-fos knockout mice ( c-fos -/-) were shown to be highly re sistant to light damage (Hafezi et al., 1997). AP-1 upregulation is also not ed in the acute light damage model of Hao et al. (Hao et al., 2002), and following damaging retinal illumination in the T4R dog (Cideciyan et al., 2005). It remains to be de termined if a similar AP-1 induction can be observed following light damage in the mrho+/-; hT17M mouse line. As discussed earlier, apoptot ic pathways can also be classified as caspasedependent or caspase-independe nt. Caspase-dependent path ways involve the activation of caspase proteins, which normally exist in an inactive form termed procaspases. These are cleaved by activated cas pases into smaller, active s ubunits which in turn cleave

PAGE 126

113 and activate other caspases in a highly orga nized cascade that lead s to photoreceptor cell death. The caspase-dependent pathway has been shown to result from both an extrinsic pathway, which is initiated by the binding of extracellular ligands to membrane receptors such as Fas/CD95 or TNF and an intrinsic pathway invol ving mitochondria l release of cytochrome c (Wenzel et al., 2005). Caspase-indepe ndent apoptosis de scribes apoptotic pathways in which involvement of caspases has not been demonstr ated. Components of the caspase-independent pathway include cath epsins, calpains, granzymes A and B, and serine proteases like AP24 (Wenzel et al., 2005 ). Both pathways have been observed in animal models of retinal de generation (Doonan et al., 2003), and it would be interesting to determine which is operating in photor eceptors of the mrho +/-; hT17M mouse in response to light damage. Numerous attempts have been made to overcome light-induced apoptosis with pharmacological agents (for an excellen t review, see Wenzel et al., 2005). Dexamethasone treatment, which results in induction of glucocorticoid activity, has shown to lower AP-1 levels and rescue light-induced damage in the acute model of Hao et al.(Wenzel et al., 2001a; Hao et al., 2002). Administration of halothane anesthesia, which has been shown to limit rhodopsin regeneration following photoactivation, is also protec tive against light-mediated dama ge in albino mice and rats (Keller et al., 2001). In 2001, Cao et al. repo rted the protective e ffect of intravitreal injection of pigment epitheliumderived growth factor and basi c fibroblast growth factor in albino Sprague-Dawley rats (Cao et al., 2001 ). Intraperitoneal injection of phenyl-Ntert-butylnitrone, which is known to possess an tioxidant properties, also prevented light damage in Sprague-Dawley rats (Ranchon et al., 2001), but such treatment was unable to

PAGE 127

114 rescue photoreceptor degeneration in rat models of ADRP that were transgenic for P23H or S334ter mutant rhodopsin(Ranchon et al ., 2003). In fact, although many agents are known to prevent the retinal degeneration induced by light damage, none of these treatments to date has been successful at ad ditionally preventing re tinal degeneration in animal models of retinitis pigmentosa (Wenzel et al., 2005). It is possible that further understanding of the apoptotic pathways involved in photoreceptor death resulting from both light-induced damage and retinitis pigmento sa will lead to the discovery of agents or combinations of agents that can perform both therapeutic functions. The experiments described above provide in sight into possible means of unraveling the mechanism of apoptotic cell death induced by light exposure in the mrho+/-; hT17M transgenic mice. Crossing th is line with transducin knockou t mice, for example, would enable us to determine whether the damage was transducin-dependent, and possibly Ca2+ related, as seen in the low -intensity pathway of Hao et al. On the other hand, the observation of elevated leve ls of the AP-1 transcrip tion factor or rescue by dexamethasone would lead to the classifica tion of the hT17M-mediated light sensitivity as more of an acute model of light-indu ced photoreceptor apoptosis. If the hT17M degeneration involves the acute pathway, then crossing this line to the RPE65 knockout mouse line would be expected to reduce or pr event retinal light damage. Similarly, the success or failure of treatment with pharmacological agents such as Dcis -diltiazem, phenyl-N-tert-butylnitron e, halothane, or 13cis retinoic acid at preventing light damage would assist in understanding this apoptotic pathway, c ontributing to the overall understanding of the mechanis ms of photoreceptor apoptosis and possibly leading, as discussed, to more effective treatments for retinitis pigmentosa.

PAGE 128

115 On the other hand, it is possible that th e aforementioned experiments would have little success in elucidating the mechanism of the light sensitivity conferred by hT17M transgene expression. The evidence descri bed here indicates that hT17M transgene expression results in one of the most li ght-sensitive mouse retinal phenotypes yet described. To illustrate this point, consider the Balb/c model of light sensitivity, which owes its phenotype to the le ucine at the polymorphic codon 450 of the RPE65 gene (Wenzel et al., 2001b). The threshold of li ght-mediated damage in this animal, as measured by nucleosome release, is between 10 and 15 minutes of exposure to white light of 13,000 lux intensity (Wenzel et al ., 2005). In our model of light-mediated damage, significant apoptosis was observed af ter a much less intense light exposure of 5,000 lux for 2.5 minutes. Additionally, we obs erved significant photoreceptor apoptosis in the hT17M mice that resulted from the ex tremely brief light exposure associated with fundus photography. The severity of the light sensitivity seen in the hT17M mouse line raises the possibility that it is the result of apoptotic path ways that differ from those previously described. It is intriguing that the T17M rhodopsin mu tation abolishes the glycosylation site at position 15 of rhodopsin (Kaushal et al., 1994). It shares this feature with the T4R rhodopsin mutation, which has been also been associated with extreme sensitivity to light-mediated retinal damage in a dog mode l of ADRP (Cideciyan et al., 2005; Zhu et al., 2004). Experiments involving the inhibiti on of rhodopsin glycosylation have been shown to lead to defects in the morphogenesi s of photoreceptor outer segments (Fliesler et al., 1985). It seems reasona ble to postulate that inhibi tion of rhodopsin glycosylation by the T4R or T17M mutations results in a protein that is uniquely affected by light

PAGE 129

116 exposure, leading to photoreceptor cell death. It is possible that li ght exposure increases the rate of aberrant disc morphogenesis in these animals, which would be expected to lead to result in photoreceptor cell death (Mendes et al., 2005; Be sharse and Wetzel, 1995). Light-induced, toxic accumulation of non-glyosylated rhodopsin could also trigger the unfolded-protein or ER-stress re sponse, which has been shown to lead to apoptosis (Mendes et al., 2005; Rutkowski and Kaufman, 2004). When considering other possible mechan isms for light-mediated photoreceptor damage in the hT17M transgenic mice, it b ears repeating that Sung and coworkers note that their class II r hodopsin mutations show retention/mi slocalization to the endoplasmic reticulum similar to that seen with other di sease-causing mutant proteins such as Class II low density lipoprotein receptors and cyst ic fibrosis transmembrane conductance regulator proteins (Sung et al ., 1991; Sung et al., 1994). As T 17M was characterized as a class II mutation in those studies, it is quite possible that the mutant rhodopsins expressed by our mrho+/-; hT17M mice are improperly re tained in the inner segments of rod photoreceptor cells. Rhodopsin, as was mentio ned, is a G-coupled receptor protein, and it is possible that photoactivated rhodopsin is ab le to interact with other heterotrimeric G proteins besides transducin. In response to certain intensitie s of illumination, these mislocalized T17M rhodopsins could be re sponsible for widespread activation of heterotrimeric G proteins in improper cel lular locations, with possible deleterious consequences for the photorecept or cell. At least one such G protein has been shown to exist in the outer segment (Peng et al., 1997). Ironically, the very light sensitivity that has slowed our efforts to treat the hT17M mouse line with rAAV-delivered, ribozyme medi ated gene therapy could make this an

PAGE 130

117 excellent model for the rapid identification of compounds that are able to inhibit or prevent the retinal degeneration associated with ADRP. Accumu lation of misfolded protein and the ultimate apoptos is of affected photoreceptors are two target s for potential pharmacological intervention with either anti -apoptotic treatments or agents that can accelerate the clearance of misf olded proteins. If such drugs could be shown to be effective for long term prevention of retinal degeneration in the mrho+/-; hT17M line, as assayed by periodic electroretinograpic a nd histologic examination, there is the possibility that their introduction could also rescue th e acute light sensitivity seen in these animals. If such is the case, then the eff ectiveness of similar types of drugs could be rapidly screened in the mrho+/-; hT17M line in a matter of days rather than the months it would take to observe rescue of the ADRPmediated degeneration. The presence or absence of light-induced photor eceptor apoptosis following re tinal light exposure can be determined by TUNEL labeling of histologi cal sections in around a week. More quantitative results could be obtained in around 48 hours by using the nucleosome release assay described by Hao et al.(Hao et al., 2002). Indeed, we plan to employ this assay for future experiments involving light-mediated retinal damage in th e mrho+/-; hT17M line because of its speed and the qua ntitative nature of the data that it generates. These experiments will hopefully increase our understanding of the apoptotic mechanisms involved in retinal degeneration. Clinical Impact Mutation-specific light damage is an impor tant concern for clinical practitioners advising and treating patients w ith retinitis pigmentosa. F unduscopic examination is one of the simplest and most co mmon of all ophthalmologic tech niques, and the data reported here indicate that it is likely that this type of examination could result in retinal

PAGE 131

118 illumination with intensities of white light th at are extremely deleterious to the visual acuity of RP patients containing certain rhodops in mutations, like T17M. Retinal surgery to resolve cataracts, detached retinas, or perform vitrectom ies can also expose the retina to intense white light of the type shown here to cause retinal damage and exacerbation of RP photoreceptor degeneration (Lawwill et al ., 1980; Michels et al ., 1987; Meyers and Bonner, 1982; McKechnie and Ghafour, 1982). The fact that some forms of rhodopsin RP mutations, like P23H, exhibit no increased sensitivity to acute light dama ge while others, like the T17M and T4R mutation show exquisite sensitivity, make it a ll the more necessary for the vision research community to urge extreme caution when dea ling with light exposure for RP patients. The fact that an ophthalmologist or opt ometrist has seen no ill effects following funduscopy or other procedures involving patients with re tinitis pigmentosa does not mean that the light exposure was harmless, or if it was harmless, that it will not be damaging to future RP patients with different rhodopsin mutations. In fact, there could be many who would experience se vere visual impairment following these examinations. Because of the heterogeneous nature of RP and the varying rates with which it affects different patients with the same rhodopsin mu tations it would be easy for a clinician to conclude that a large drop in visual acuity reported by an RP patient following ophthalmologic procedures involving intense re tinal illumination was the result of the patient suffering a rapid onset form of the di sease rather than damage incurred from the examination procedure(s). Without prospect ive studies aimed at determining whether human patients suffering from RP caused by certain extreme light sensitive rhodopsin

PAGE 132

119 Figure 6-1. Eye from a human who suffered from ADRP caused by the T17M rhodopsin mutation. The eye is oriented with th e superior portion at the top of the photograph. Evidence of retinitis pigmen tosa, including bone spicule deposits and drusen bodies are located almost excl usively in the inferior retina(Li et al., 1994). mutations suffer greater retinal damage from such procedures than patients with other rhodopsin mutations, as we have shown to be the case in the mouse lines used in these experiments, this question cannot be clearly answered. In any event, patients with RP and other types of retinal degeneration should be encouraged to avoid light exposure whenev er possible by staying indoors on sunny days and wearing personal eye protection in the form of sunglasses or ha ts whenever outdoor activity cannot be avoided. The importan ce of incidental light exposure in ADRP associated with the T17M mutation is driven home by its clinical presentation: Retinal

PAGE 133

120 degeneration is frequently observed primarily in the inferior retina (Figure 6-1). The image shown in this figure is of an eye excised from a deceased human patient who suffered from ADRP caused by the T17M rhodops in mutation. This eye exhibits severe photoreceptor damage resulting from ADRP, as evidenced by the multitude of bone spicule deposits and drusen bodies. Interes tingly, these telltale signs of retinitis pigmentosa are located almost exclusively in the inferior portion of the retina, while the superior retina appears relatively normal. Lifelong exposure to sunlight and overhead illumination specifically irradiated the inferior portion of the patients retina and may thus have accelerated the rate of degeneration in that portion of the eye. It is difficult to overstate the importance of limiting retinal i llumination by intense white light in patients suffering from autosomal dominant retinitis pigmentosa and other forms of retinal disease.

PAGE 134

121 LIST OF REFERENCES Acland,G.M., Aguirre,G.D., Bennett,J., Alem an,T.S., Cideciyan,A.V., Bennicelli,J., Dejneka,N.S., Pearce-Kelling,S.E ., Maguire,A.M., Palczewski,K., Hauswirth,W.W., and Jacobson,S.G. (2005) Long-Term Restoration of Rod and Cone Vision by Single Dose rAAV-Mediated Gene Transfer to the Retina in a Canine Model of Childhood Blindness. Mol. Ther. Acland,G.M., Aguirre,G.D., Ray,J., Zhang,Q., Aleman,T.S., Cideciyan,A.V., PearceKelling,S.E., Anand,V., Zeng,Y., Maguire ,A.M., Jacobson,S.G., Hauswirth,W.W., and Bennett,J. (2001). Gene therapy restor es vision in a canine model of childhood blindness. Nat. Genet. 28 92-95. Adamus,G., Sugden,B., Shiraga,S., Timmers ,A.M., and Hauswirth,W.W. (2003). Antiapoptotic effects of CNTF gene tr ansfer on photoreceptor degeneration in experimental antibody-induced retinopathy. J. Autoimmun. 21 121-129. Adrian,W., Everson,R.W., and Schmidt,I. (1977) Protection against photic damage in retinitis pigmentosa. Adv. Exp. Med. Biol. 77 233-247. Ali,R.R., Sarra,G.M., Stephens,C., Alwis,M. D., Bainbridge,J.W., Munro,P.M., Fauser,S., Reichel,M.B., Kinnon,C., Hunt,D.M., Bhatt acharya,S.S., and Thrasher,A.J. (2000). Restoration of photoreceptor ultrastructure and functi on in retinal degeneration slow mice by gene therapy. Nat. Genet. 25 306-310. Altschuler,M., Tritz,R., and Hampel,A. (1992) A method for generati ng transcripts with defined 5' and 3' termini by autolytic processing. Gene 122 85-90. Aonuma,H., Yamazaki,R., and Watanabe,I. (1999). Retinal cell death by light damage. Jpn. J. Ophthalmol. 43 171-179. Auricchio,A., Behling,K.C., Maguire,A.M., O'Connor,E.M., Bennett,J., Wilson,J.M., and Tolentino,M.J. (2002). Inhibition of retina l neovascularization by intraocular viralmediated delivery of anti-angiogenic agents. Mol. Ther. 6 490-494. Auricchio,A., Kobinger,G., Anand,V., Hild inger,M., O'Connor,E., Maguire,A.M., Wilson,J.M., and Bennett,J. (2001). Exchange of surface proteins impacts on viral vector cellular specificity a nd transduction characteristics: the retina as a model. Hum. Mol. Genet. 10 3075-3081. Auricchio,A., and Rolling,F. (2005). Adeno-asso ciated viral vectors for retinal gene transfer and treatment of reti nal diseases. Curr. Gene Ther. 5 339-348.

PAGE 135

122 Beatrice,J., Wenzel,A., Reme,C.E., and Gr imm,C. (2003). Increased light damage susceptibility at night does not corr elate with RPE65 levels and rhodopsin regeneration in rats. Exp. Eye Res. 76 695-700. Bence,N.F., Sampat,R.M., and Kopito,R.R (2001). Impairment of the ubiquitinproteasome system by protein aggregation. Science 292 1552-1555. Bennett,J. (2003). Immune response following in traocular delivery of recombinant viral vectors. Gene Ther. 10 977-982. Bennett,J., Tanabe,T., Sun,D., Zeng,Y., Kjeldbye,H., Gouras,P., and Maguire,A.M. (1996). Photoreceptor cell rescue in retinal degeneration (rd) mice by in vivo gene therapy. Nat. Med. 2 649-654. Berson,E.L. (1996). Retinitis pigmentosa: unfoldi ng its mystery. Proc. Natl. Acad. Sci. U. S. A 93 4526-4528. Bertrand,E., Castanotto,D., Zhou,C., Carbonne lle,C., Lee,N.S., Good,P., Chatterjee,S., Grange,T., Pictet,R., Kohn,D., Engelke,D., and Rossi,J.J. (1997). The expression cassette determines the functional activ ity of ribozymes in mammalian cells by controlling their intracellu lar localization. RNA. 3 75-88. Berzal-Herranz,A., Joseph,S., Chowrira,B.M ., Butcher,S.E., and Burke,J.M. (1993). Essential nucleotide sequences and seconda ry structure elements of the hairpin ribozyme. EMBO J. 12 2567-2573. Besharse,J.C., and Wetzel,M.G. (1995). Immuno cytochemical localization of opsin in rod photoreceptors during periods of rapi d disc assembly. J. Neurocytol. 24 371-388. Blouin,V., Brument,N., Toublanc,E., Raimbaud, I., Moullier,P., and Salvetti,A. (2004). Improving rAAV production and purification: towards the definition of a scaleable process. J. Gene Med. 6 Suppl 1 S223-S228. Bosch,A., Perret,E., Desmaris,N., and Hear d,J.M. (2000). Long-term and significant correction of brain lesions in adult mu copolysaccharidosis type VII mice using recombinant AAV vectors. Mol. Ther. 1 63-70. Bownds,D. (1967). Site of attachment of retinal in rhodopsin. Nature 216 1178-1181. Bowne,S.J., Sullivan,L.S., Blanton,S.H., Cepko,C.L., Blackshaw,S., Birch,D.G., Hughbanks-Wheaton,D., Heckenlively,J.R., a nd Daiger,S.P. (2002). Mutations in the inosine monophosphate dehydrogenase 1 gene (IMPDH1) cause the RP10 form of autosomal dominant retinitis pigmentosa. Hum. Mol. Genet. 11 559-568. Bradley,D.G., Farrar,G.J., Sharp,E.M., Kenna,P., Humphries,M.M., McConnell,D.J., Daiger,S.P., McWilliam,P., and Humphries,P. (1989). Autosomal dominant retinitis pigmentosa: exclusion of the gene from the short arm of chromosome 1 including the region surrounding the rhesus locus. Am. J. Hum. Genet. 44 570-576.

PAGE 136

123 Bredesen,D.E., Mehlen,P., and Rabizadeh,S. (2005). Receptors that mediate cellular dependence. Cell Death. Differ. 12 1031-1043. Burger,C., Gorbatyuk,O.S., Velardo,M.J ., Peden,C.S., Williams,P., Zolotukhin,S., Reier,P.J., Mandel,R.J., and Muzyczka,N (2004). Recombinant AAV viral vectors pseudotyped with viral capsids from se rotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system. Mol. Ther. 10 302-317. Buzayan,J.M., Feldstein,P.A., Segrelles,C., and Bruening,G. (1988). Autolytic processing of a phosphorothioate dieste r bond. Nucleic Acids Res. 16 4009-4023. Cao,W., Tombran-Tink,J., Elias,R., Sezate, S., Mrazek,D., and McGinnis,J.F. (2001). In vivo protection of photoreceptors from light damage by pigment epitheliumderived factor. Invest Ophthalmol. Vis. Sci. 42 1646-1652. Cech,T.R. (1988a). Conserved sequences and st ructures of group I in trons: building an active site for RNA catalysis--a review. Gene 73 259-271. Cech,T.R. (1988b). Ribozymes and th eir medical implications. JAMA 260 3030-3034. Chang,B., Hawes,N.L., Hurd,R.E., Davisson,M .T., Nusinowitz,S., and Heckenlively,J.R. (2002). Retinal degeneration mutants in the mouse. Vision Res. 42 517-525. Chen,L., Wu,W., Dentchev,T., Zeng,Y., Wang,J ., Tsui,I., Tobias,J.W., Bennett,J., Baldwin,D., and Dunaief,J.L. (2004). Light damage induced changes in mouse retinal gene expression. Exp. Eye Res. 79 239-247. Choi,V.W., McCarty,D.M., and Samulski,R.J (2005). AAV hybrid serotypes: improved vectors for gene delivery. Curr. Gene Ther. 5 299-310. Cideciyan,A.V., Jacobson,S.G., Aleman,T.S., Gu,D., Pearce-Kelling,S.E., Sumaroka,A., Acland,G.M., and Aguirre,G.D. (2005). In vivo dynamics of retinal injury and repair in the rhodopsin mutant dog model of human retinitis pigmentosa. Proc. Natl. Acad. Sci. U. S. A 102 5233-5238. Citti,L., and Rainaldi,G. (2005). Synthetic hamm erhead ribozymes as therapeutic tools to control disease genes. Curr. Gene Ther. 5 11-24. Clouet-d'Orval,B., and Uhlenbeck,O.C. (1997) Hammerhead ribozymes with a faster cleavage rate. Biochemistry 36 9087-9092. Conrad,C.K., Allen,S.S., Afione,S.A., Re ynolds,T.C., Beck,S.E., Fee-Maki,M., BarrazzaOrtiz,X., Adams,R., Askin,F.B., Carter,B .J., Guggino,W.B., and Flotte,T.R. (1996). Safety of single-dose administration of an adeno-associated virus (AAV)-CFTR vector in the primate lung. Gene Ther. 3 658-668.

PAGE 137

124 Daiger,S.P., Sullivan,L.S., and Rodriguez,J. A. (1995). Correlation of Phenotype with Genotype in Inherited Retinal Degenera tion. Behavioral a nd Brain Sciences 18 452-467. Danciger,M., Lyon,J., Worrill,D., Hoffma n,S., Lem,J., Reme,C.E., Wenzel,A., and Grimm,C. (2004). New retinal light damage QTL in mice with the light-sensitive RPE65 LEU variant. Mamm. Genome 15 277-283. Diager, S. P., Sullivan, L. S., and Browne, S. J. RetNet. internet 1-11-2005. Ref Type: Electronic Citation Dinculescu,A., Glushakova,L., Min,S.H., and Hauswirth,W.W. (2005). Adeno-associated virus-vectored gene therapy for retinal disease. Hum. Gene Ther. 16 649-663. Donovan,M., and Cotter,T.G. (2002). Caspase-in dependent photoreceptor apoptosis in vivo and differential expression of apoptot ic protease activa ting factor-1 and caspase-3 during retinal development. Cell Death. Differ. 9 1220-1231. Donsante,A., Vogler,C., Muzyczka,N., Craw ford,J.M., Barker,J., Flotte,T., CampbellThompson,M., Daly,T., and Sands,M.S. (2001). Observed incidence of tumorigenesis in long-term rodent st udies of rAAV vectors. Gene Ther. 8 13431346. Doonan,F., Donovan,M., and Cotter,T.G. (2003) Caspase-independent photoreceptor apoptosis in mouse models of re tinal degeneration. J. Neurosci. 23 5723-5731. Drenser,K.A., Timmers,A.M., Hauswirth,W .W., and Lewin,A.S. (1998). Ribozymetargeted destruction of RNA associat ed with autosomal-dominant retinitis pigmentosa. Invest Ophthalmol. Vis. Sci. 39 681-689. Dryja,T.P., McGee,T.L., Reichel,E. Hahn,L.B., Cowley,G .S., Yandell,D.W., Sandberg,M.A., and Berson,E.L. (1990). A point mutation of the rhodopsin gene in one form of retinitis pigmentosa. Nature 343 364-366. Fain,G.L., and Lisman,J.E. (1993). Photorecep tor degeneration in vi tamin A deprivation and retinitis pigmentosa: the equiva lent light hypothesis. Exp. Eye Res. 57 335340. Fain,G.L., and Lisman,J.E. (1999). Light, Ca2+ and photoreceptor death: new evidence for the equivalent-light hypothesis from a rrestin knockout mice. Invest Ophthalmol. Vis. Sci. 40 2770-2772. Farber,D.B., Flannery,J.G., Bird,A.C., Shuste r,T., and Bok,D. (1987). Histopathological and biochemical studies on donor eyes aff ected with retinitis pigmentosa. Prog. Clin. Biol. Res. 247 53-67.

PAGE 138

125 Farrar,G.J., Kenna,P.F., and Humphries,P. (2002) On the genetics of retinitis pigmentosa and on mutation-independent approaches to therapeutic intervention. EMBO J. 21 857-864. Fisher,K.J., Jooss,K., Alston,J., Yang,Y., H aecker,S.E., High,K., Pathak,R., Raper,S.E., and Wilson,J.M. (1997). Recombinant adenoassociated virus for muscle directed gene therapy. Nat. Med. 3 306-312. Flannery,J.G., Farber,D.B., Bird,A.C., and Bok,D. (1989). Degene rative changes in a retina affected with autosomal dominant retinitis pigmentosa. Invest Ophthalmol. Vis. Sci. 30 191-211. Flannery,J.G., Zolotukhin,S., Vaquero,M .I., LaVail,M.M., Muzyczka,N., and Hauswirth,W.W. (1997). Effici ent photoreceptor-targeted gene expression in vivo by recombinant adeno-associated virus. Proc. Natl. Acad. Sci. U. S. A 94 69166921. Fliesler,S.J., Rayborn,M.E., and Hollyfiel d,J.G. (1985). Membrane morphogenesis in retinal rod outer segments: inhibi tion by tunicamycin. J. Cell Biol. 100 574-587. Flotte,T.R. (2005). Recent developments in recombinant AAV-mediated gene therapy for lung diseases. Curr. Gene Ther. 5 361-366. Fritz,J.J., White,D.A., Lewin,A.S., and Hauswirth,W.W. (2002). Designing and characterizing hammerhead ribozymes fo r use in AAV vector-mediated retinal gene therapies. Methods Enzymol. 346 358-377. Fu,H., Samulski,R.J., McCown,T.J., Picornell, Y.J., Fletcher,D., and Muenzer,J. (2002). Neurological correction of lysosomal st orage in a mucopolysaccharidosis IIIB mouse model by adeno-associated virusmediated gene delivery. Mol. Ther. 5 4249. Fuller,D., Machemer,R., and Knighton,R.W. (1978). Retinal damage produced by intraocular fiber optic lig ht. Am. J. Ophthalmol. 85 519-537. Gao,G., Vandenberghe,L.H., and Wilson,J.M. (2005). New recombinant serotypes of AAV vectors. Curr. Gene Ther. 5 285-297. Gao,G.P., Alvira,M.R., Wang,L., Calcedo,R., Johnston,J., and Wilson,J.M. (2002). Novel adeno-associated viruses from rhesus m onkeys as vectors for human gene therapy. Proc. Natl. Acad. Sci. U. S. A 99 11854-11859. Gigout,L., Rebollo,P., Clement,N., Warringt on,K.H., Jr., Muzyczka,N., Linden,R.M., and Weber,T. (2005). Altering AAV tropism with mosaic viral capsids. Mol. Ther. 11 856-865.

PAGE 139

126 Gorbatyuk,M.S., Pang,J.J., Thomas,J., Jr., Hauswirth,W.W., and Lewin,A.S. (2005). Knockdown of wild-type mouse rhodopsin using an AAV vectored ribozyme as part of an RNA replacement approach. Mol. Vis. 11 648-656. Goto,Y., Peachey,N.S., Ripps,H., and Naash,M .I. (1995). Functional abnormalities in transgenic mice expressing a mutant rhodopsin gene. Invest Opht halmol. Vis. Sci. 36 62-71. Goto,Y., Peachey,N.S., Ziroli,N.E., Seiple,W.H., Gryczan,C., Pepperberg,D.R., and Naash,M.I. (1996). Rod phototransduction in transgenic mice expressing a mutant opsin gene. J. Opt. Soc. Am. A Opt. Image Sci. Vis. 13 577-585. Grimm,C., Wenzel,A., Hafezi,F., and Reme,C.E (2000a). Gene expression in the mouse retina: the effect of damaging light. Mol. Vis. 6 252-260. Grimm,C., Wenzel,A., Hafezi,F., Yu,S ., Redmond,T.M., and Reme,C.E. (2000b). Protection of Rpe65-deficient mice identif ies rhodopsin as a mediator of lightinduced retinal degeneration. Nat. Genet. 25 63-66. Grimm,D. (2002). Production methods for ge ne transfer vectors based on adenoassociated virus serotypes. Methods 28 146-157. Grimm,D., and Kay,M.A. (2003). From virus e volution to vector revolution: use of naturally occurring serotypes of adeno-associated virus (AAV) as novel vectors for human gene therapy. Curr. Gene Ther. 3 281-304. Grum-Tokars,V., Milovanovic,M., and Wedeki nd,J.E. (2003). Crystal lization and X-ray diffraction analysis of an all-RNA U39C mutant of the minimal hairpin ribozyme. Acta Crystallogr. D. Biol. Crystallogr. 59 142-145. Guerrier-Takada,C., and Altman,S. (1984). Catalytic activity of an RNA molecule prepared by transcription in vitro. Science 223 285-286. Guerrier-Takada,C., Gardiner,K., Marsh,T ., Pace,N., and Altman,S. (1983). The RNA moiety of ribonuclease P is the ca talytic subunit of the enzyme. Cell 35 849-857. Guy,J., Qi,X., Muzyczka,N., and Hauswirth,W.W (1999). Reporter expression persists 1 year after adeno-associated virus-mediated gene transfer to the optic nerve. Arch. Ophthalmol. 117 929-937. Hafezi,F., Steinbach,J.P., Marti,A., Munz ,K., Wang,Z.Q., Wagner,E.F., Aguzzi,A., and Reme,C.E. (1997). The absence of c-fos pr events light-induced apoptotic cell death of photoreceptors in retinal de generation in vivo. Nat. Med. 3 346-349. Hao,W., Wenzel,A., Obin,M.S., Chen,C.K ., Brill,E., Krasnoperova,N.V., EversoleCire,P., Kleyner,Y., Taylor,A., Simon,M .I., Grimm,C., Reme,C.E., and Lem,J. (2002). Evidence for two apoptotic pathways in light-induced retinal degeneration. Nat. Genet. 32 254-260.

PAGE 140

127 Hargrave,P.A., and Mcdowell,J.H. (1992). R hodopsin and Phototransduction A Model System for G-Protein-Linked Receptors. Faseb Journal 6 2323-2331. Haseloff,J., and Gerlach,W.L. (1988). Simp le RNA enzymes with new and highly specific endoribonuclease activities. Nature 334 585-591. Hauswirth,W.W., LaVail,M.M., Flannery,J .G., and Lewin,A.S. (2000). Ribozyme gene therapy for autosomal dominant retinal disease. Clin. Chem. Lab Med. 38 147-153. Hauswirth,W.W., and Lewin,A.S (2000). Ribozyme uses in retinal gene therapy. Prog. Retin. Eye Res. 19 689-710. Hauswirth,W.W., Li,Q., Raisler,B., Ti mmers,A.M., Berns,K.I., Flannery,J.G., LaVail,M.M., and Lewin,A.S. (2004). Range of retinal dis eases potentially treatable by AAV-vectored gene therapy. Novartis. Found. Symp. 255 179-188. Hennig,A.K., Ogilvie,J.M., Ohlemiller,K.K., Timmers,A.M., Hauswirth,W.W., and Sands,M.S. (2004). AAV-mediated intravitre al gene therapy reduces lysosomal storage in the retinal pigmented epitheliu m and improves retinal function in adult MPS VII mice. Mol. Ther. 10 106-116. Hermonat,P.L., Quirk,J.G., Bishop,B.M., and Han,L. (1997). The packaging capacity of adeno-associated virus (AAV) and the potential for wild-type-plus AAV gene therapy vectors. FEBS Lett. 407 78-84. Hernandez,Y.J., Wang,J., Kearns,W.G., Loile r,S., Poirier,A., and Flotte,T.R. (1999). Latent adeno-associated virus infecti on elicits humoral but not cell-mediated immune responses in a nonhuman primate model. J. Virol. 73 8549-8558. Hildinger,M., Auricchio,A., Gao,G., Wang,L ., Chirmule,N., and Wilson,J.M. (2001). Hybrid vectors based on adeno-associat ed virus serotypes 2 and 5 for muscledirected gene transfer. J. Virol. 75 6199-6203. Hims,M.M., Diager,S.P., and Inglehearn,C .F. (2003). Retinitis pigmentosa: genes, proteins and prospect s. Dev. Ophthalmol. 37 109-125. Huang,P.C., Gaitan,A.E., Hao,Y., Pette rs,R.M., and Wong,F. (1993). Cellular interactions implicated in the mech anism of photoreceptor degeneration in transgenic mice expressing a mutant rhodopsin gene. Proc. Natl. Acad. Sci. U. S. A 90 8484-8488. Humphries,M.M., Rancourt,D., Farrar,G.J., Kenna,P., Hazel,M., Bush,R.A., Sieving,P.A., Sheils,D.M., McNally,N ., Creighton,P., Erven,A., Boros,A., Gulya,K., Capecchi,M.R., and Humphries, P. (1997). Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat. Genet. 15 216-219. Humphries,P., Kenna,P., and Farrar,G.J. (1992) On the Molecular-Genetics of RetinitisPigmentosa. Science 256 804-808.

PAGE 141

128 Illing,M.E., Rajan,R.S., Bence,N.F., and Kop ito,R.R. (2002). A rhodopsin mutant linked to autosomal dominant retinitis pigmentosa is prone to aggregate and interacts with the ubiquitin proteasome system. J. Biol. Chem. 277 34150-34160. Jomary,C., Vincent,K.A., Grist,J., Neal,M .J., and Jones,S.E. (1997). Rescue of photoreceptor function by AAV-mediated ge ne transfer in a mouse model of inherited retinal dege neration. Gene Ther. 4 683-690. Joseph,S., Berzal-Herranz,A., Chowrira,B.M., Butcher,S.E., and Burke,J.M. (1993). Substrate selection rules for the hairpin ribozyme determined by in vitro selection, mutation, and analysis of mism atched substrates. Genes Dev. 7 130-138. Kapturczak,M.H., Flotte,T., and Atkinson,M.A (2001). Adeno-associated virus (AAV) as a vehicle for therapeutic gene delivery: improvements in vector design and viral production enhance potential to prolong graf t survival in pancreatic islet cell transplantation for the reversal of type 1 diabetes. Curr. Mol. Med. 1 245-258. Karnik,S.S., and Khorana,H.G. (1990). Asse mbly of functional rhodopsin requires a disulfide bond between cysteine residues 110 and 187. J. Biol. Chem. 265 1752017524. Karnik,S.S., Sakmar,T.P., Chen,H.B., and K horana,H.G. (1988). Cysteine residues 110 and 187 are essential for the formation of correct structure in bovine rhodopsin. Proc. Natl. Acad. Sci. U. S. A 85 8459-8463. Kaushal,S. and Khorana,H.G. (1994). Stru cture and function in rhodopsin. 7. Point mutations associated with autosomal domi nant retinitis pigmentosa. Biochemistry 33 6121-6128. Kaushal,S., Ridge,K.D., and Khorana,H.G. (1994). Structure and function in rhodopsin: the role of asparagine-linked glycosyl ation. Proc. Natl. Acad. Sci. U. S. A 91 4024-4028. Keeler,C. (1966). Retinal degeneration in the mouse is rodless retina. J. Hered. 57 47-50. Keller,C., Grimm,C., Wenzel,A., Hafezi,F., a nd Reme,C. (2001). Protective effect of halothane anesthesia on retinal light da mage: inhibition of metabolic rhodopsin regeneration. Invest Ophthalmol. Vis. Sci. 42 476-480. Kennan,A., Aherne,A., and Humphries,P. (2005) Light in retinitis pigmentosa. Trends Genet. 21 103-110. Kennan,A., Aherne,A., Palfi,A., Humphries, M., McKee,A., Stitt,A., Simpson,D.A., Demtroder,K., Orntoft,T., Ayuso,C., Kenna ,P.F., Farrar,G.J., and Humphries,P. (2002). Identification of an IMPDH1 muta tion in autosomal dominant retinitis pigmentosa (RP10) revealed following comparative microarray analysis of transcripts derived from retinas of wild-t ype and Rho(-/-) mice. Hum. Mol. Genet. 11 547-557.

PAGE 142

129 Kolb, H., Fernandez, E., and Nelson, R. Webvi sion: The Organization of the Retina and Visual System. Internet 2005. Ref Type: Electronic Citation Kotin,R.M., Siniscalco,M., Samulski,R .J., Zhu,X.D., Hunter,L., Laughlin,C.A., McLaughlin,S., Muzyczka,N., Rocchi,M., a nd Berns,K.I. (1990). Site-specific integration by adeno-associated viru s. Proc. Natl. Acad. Sci. U. S. A 87 22112215. Lansel,N., Hafezi,F., Marti,A., Hegi,M., Re me,C., and Niemeyer,G. (1998). The mouse ERG before and after light damage is independent of p53. Doc. Ophthalmol. 96 311-320. Lanum,J. (1978). The damaging effects of lig ht on the retina. Empirical findings, theoretical and practical imp lications. Surv. Ophthalmol. 22 221-249. Lau,D., and Flannery,J. (2003). Viral-mediated FGF-2 treatme nt of the constant light damage model of photoreceptor degeneration. Doc. Ophthalmol. 106 89-98. Lau,D., McGee,L.H., Zhou,S., Rendahl,K.G., Manning,W.C., Escobedo,J.A., and Flannery,J.G. (2000). Retinal degeneration is slowed in transgenic rats by AAVmediated delivery of FGF-2. I nvest Ophthalmol. Vis. Sci. 41 3622-3633. LaVail,M.M., Gorrin,G.M., Repaci,M.A., Th omas,L.A., and Ginsberg,H.M. (1987). Genetic regulation of light damage to photor eceptors. Invest Ophthalmol. Vis. Sci. 28 1043-1048. LaVail,M.M., Yasumura,D., Matthes,M.T., Dren ser,K.A., Flannery,J.G., Lewin,A.S., and Hauswirth,W.W. (2000). Ribozyme rescue of photoreceptor cells in P23H transgenic rats: long-term su rvival and late-stage therapy. Proc. Natl. Acad. Sci. U. S. A 97 11488-11493. LaVail,M.M., Yasumura,D., Matthes,M.T., La u-Villacorta,C., Unoki,K., Sung,C.H., and Steinberg,R.H. (1998). Protec tion of mouse photoreceptors by survival factors in retinal degenerations. Invest Ophthalmol. Vis. Sci. 39 592-602. Lawwill,T., Crockett,R.S., Currier,G., and Rosenberg,R.B. (1980). Review of the macaque model of light damage with im plications for the use of ophthalmic instrumentation. Vision Res. 20 1113-1115. Lem,J., Flannery,J.G., Li,T., Applebury,M .L., Farber,D.B., and Simon,M.I. (1992). Retinal degeneration is resc ued in transgenic rd mice by expression of the cGMP phosphodiesterase beta subunit. Proc. Natl. Acad. Sci. U. S. A 89 4422-4426.

PAGE 143

130 Lem,J., Krasnoperova,N.V., Calvert,P.D ., Kosaras,B., Cameron,D.A., Nicolo,M., Makino,C.L., and Sidman,R.L. (1999). Morphological, physiological, and biochemical changes in rhodopsin knockout mice. Proc. Natl. Acad. Sci. U. S. A 96 736-741. Lewin,A.S., Drenser,K.A., Haus wirth,W.W., Nishikawa,S., Yasumura,D., Flannery,J.G., and LaVail,M.M. (1998). Ribozyme rescue of photoreceptor cells in a transgenic rat model of autosomal dominant retinitis pigmentosa. Nat. Med. 4 967-971. Li,T., Sandberg,M.A., Pawlyk,B.S., Rosner,B., Ha yes,K.C., Dryja,T.P., and Berson,E.L. (1998). Effect of vitamin A supplementa tion on rhodopsin mutants threonine-17 --> methionine and proline-347 --> serine in tr ansgenic mice and in cell cultures. Proc. Natl. Acad. Sci. U. S. A 95 11933-11938. Li,Z.Y., Jacobson,S.G., and Milam,A.H. (1994). Autosomal dominant retinitis pigmentosa caused by the threonine-17-me thionine rhodopsin mutation: retinal histopathology and immunocyt ochemistry. Exp. Eye Res. 58 397-408. Liang,F.Q., Dejneka,N.S., Cohen,D.R., Krasnoperova,N.V., Lem,J., Maguire,A.M., Dudus,L., Fisher,K.J., and Bennett,J. ( 2001). AAV-mediated delivery of ciliary neurotrophic factor prol ongs photoreceptor survival in the rhodopsin knockout mouse. Mol. Ther. 3 241-248. Liu,J., Timmers,A.M., Lewin,A.S., and Haus wirth,W.W. (2005). Ribozyme Knockdown of the {gamma}-Subunit of Rod cGMP P hosphodiesterase Alters the ERG and Retinal Morphology in Wild-Type Mice. Invest Ophthalmol. Vis. Sci. 46 38363844. Liu,X., Wu,T.H., Stowe,S., Matsushita,A., Arikawa,K., Naash,M.I., and Williams,D.S. (1997). Defective phototransductive disk membrane morphogenesis in transgenic mice expressing opsin with a mutated N-terminal domain. J. Cell Sci. 110 ( Pt 20) 2589-2597. Loiler,S.A., Conlon,T.J., Song,S., Ta ng,Q., Warrington,K.H., Agarwal,A., Kapturczak,M., Li,C., Ricordi,C., Atkinson,M.A., Muzyczka,N., and Flotte,T.R. (2003). Targeting recombinant adeno-associ ated virus vectors to enhance gene transfer to pancreatic islets and liver. Gene Ther. 10 1551-1558. Maita,H., Kitaura,H., Ariga,H., and Iguchi-A riga,S.M. (2005). Association of PAP-1 and Prp3p, the products of causative genes of do minant retinitis pigm entosa, in the trisnRNP complex. Exp. Cell Res. 302 61-68. Maple,B.R., and Wu,S.M. (1996). Synaptic inpu ts mediating bipolar cell responses in the tiger salamander retina. Vision Res. 36 4015-4023.

PAGE 144

131 Martinez-Gimeno,M., Gamundi,M.J., Hernan,I ., Maseras,M., Milla,E., Ayuso,C., GarciaSandoval,B., Beneyto,M., Vilela,C., Baig et,M., Antinolo,G., and Carballo,M. (2003). Mutations in the pre-mRNA splic ing-factor genes PRPF3, PRPF8, and PRPF31 in Spanish families with autoso mal dominant retinitis pigmentosa. Invest Ophthalmol. Vis. Sci. 44 2171-2177. McCabe,S.L., Pelosi,D.M., Tetreault,M., Miri,A., Nguitragool,W., Kovithvathanaphong,P., Mahajan,R., a nd Zimmerman,A.L. (2004). All-transretinal is a closed-state i nhibitor of rod cyclic nucleo tide-gated ion channels. J. Gen. Physiol 123 521-531. McKechnie,N.M. and Ghafour,I.M. (1982). Poten tial retinal light damage from the use of therapeutic instruments. Trans. Ophthalmol. Soc. U. K. 102 (Pt 1) 140-146. McKechnie,N.M., and Johnson,N.F. (1977). Ligh t damage to the retina. Albrecht. Von. Graefes Arch. Klin. Exp. Ophthalmol. 203 283-292. McLaughlin,S.K., Collis,P., Hermonat,P.L., a nd Muzyczka,N. (1988). Adeno-associated virus general transduction vectors: analys is of proviral structures. J. Virol. 62 1963-1973. McNally,N., Kenna,P., Humphries,M.M ., Hobson,A.H., Khan,N.W., Bush,R.A., Sieving,P.A., Humphries,P., and Farrar,G .J. (1999). Structural and functional rescue of murine rod photoreceptors by human rhodopsin transgene. Hum. Mol. Genet. 8 1309-1312. McWilliam,P., Farrar,G.J., Kenna,P., Bradley,D.G., Humphries,M.M., Sharp,E.M., McConnell,D.J., Lawler,M., Sheils,D., Rya n,C., and (1989). Autosomal dominant retinitis pigmentosa (ADRP): localizati on of an ADRP gene to the long arm of chromosome 3. Genomics 5 619-622. Mendes,H.F., van der,S.J., Chapple,J.P., a nd Cheetham,M.E. (2005). Mechanisms of cell death in rhodopsin re tinitis pigmentosa: implications for therapy. Trends Mol. Med. 11 177-185. Meyers,S.M., and Bonner,R.F. (1982). Yellow filter to decrease the risk of light damage to the retina during vitrec tomy. Am. J. Ophthalmol. 94 677. Michels,M., Dawson,W.W., Fe ldman,R.B., and Jarolem,K. (1987). Infrared. An unseen and unnecessary hazard in ophthalmic devices. Ophthalmology 94 143-148. Min,S.H., Molday,L.L., Seeliger,M.W., Di nculescu,A., Timmers,A.M., Janssen,A., Tonagel,F., Tanimoto,N., Weber,B.H., Molday,R.S., and Hauswirth,W.W. (2005). Prolonged recovery of retinal structure/f unction after gene therapy in an Rs1hdeficient mouse model of x-linked ju venile retinoschisis. Mol. Ther. 12 644-651. Mittag,T.W., Bayer,A.U., and La VAIL,M.M. (1999). Light-induced retinal damage in mice carrying a mutated SOD I gene. Exp. Eye Res. 69 677-683.

PAGE 145

132 Mohamad,N., Gutierrez,A., Nunez,M., Cocca, C., Martin,G., Cricco,G., Medina,V., Rivera,E., and Bergoc,R. (2005). Mitoc hondrial apoptotic pathways. Biocell 29 149-161. Muzyczka,N., and Warrington,K.H., Jr. (2005). Cu stom adeno-associated virus capsids: the next generation of recombinant vect ors with novel tropism Hum. Gene Ther. 16 408-416. Naash,M.I., Hollyfield,J.G., Al Ubaidi,M.R., and Baehr,W. (1993). Simulation of human autosomal dominant retinitis pigmentosa in transgenic mice expressing a mutated murine opsin gene. Proc. Natl. Acad. Sci. U. S. A 90 5499-5503. Naash,M.L., Peachey,N.S., Li,Z.Y., Gryczan,C.C ., Goto,Y., Blanks,J., Milam,A.H., and Ripps,H. (1996). Light-induced acceleration of photoreceptor degeneration in transgenic mice expressing mutant rhodops in. Invest Ophthalmol. Vis. Sci. 37 775782. Nieuwlandt,D. (2000). In vitro selection of functional nucleic acid sequences. Curr. Issues Mol. Biol. 2 9-16. Noell,W.K., Walker,V.S., Kang,B.S., and Berm an,S. (1966). Retinal damage by light in rats. Invest Ophthalmol. 5 450-473. Olsson,J.E., Gordon,J.W., Pawlyk,B.S., Roof,D ., Hayes,A., Molday,R.S., Mukai,S., Cowley,G.S., Berson,E.L., and Dryja,T.P. (1992). Transgenic mice with a rhodopsin mutation (Pro23His): a mouse model of autosomal dominant retinitis pigmentosa. Neuron 9 815-830. Opie,S.R., Warrington,K.H., Jr., Agba ndje-McKenna,M., Zolotukhin,S., and Muzyczka,N. (2003). Identification of amino acid residues in the capsid proteins of adeno-associated virus type 2 that contribute to he paran sulfate proteoglycan binding. J. Virol. 77 6995-7006. Pang,J.J., Chang,B., Kumar,A., Nusinowitz,S., N oorwez,S.M., Li,J., Rani,A., Foster,T.C., Chiodo,V.A., Doyle,T., Li,H., Malhotra,R ., Tusner,J., Mcdowell,J.H., Min,S.H., Li,Q., Kaushal,S., and Haus wirth,W.W. (2005). Gene Therapy Restores VisionDependent Behavior as Well as Retinal Structure and Function in a Mouse Model of RPE65 Leber Congenital Amaurosis. Mol. Ther. Papermaster,D.S., Schneider,B.G., DeFoe,D., and Besharse,J.C. ( 1986). Biosynthesis and vectorial transport of opsin on vesicles in retinal rod photorecept ors. J. Histochem. Cytochem. 34 5-16. Peng,Y.W., Rhee,S.G., Yu,W.P., Ho,Y.K., Schoen,T., Chader,G.J., and Yau,K.W. (1997). Identification of components of a phosphoinositide signaling pathway in retinal rod outer segments. Proc Natl. Acad. Sci. U. S. A 94 1995-2000.

PAGE 146

133 Penn,J.S. (1985). Effects of continuous li ght on the retina of a fish, Notemigonus crysoleucas. J. Comp Neurol. 238 121-127. Phylactou,L.A., Kilpatrick,M.W., and Wood,M.J. (1998). Ribozymes as therapeutic tools for genetic disease. Hum. Mol. Genet. 7 1649-1653. Pierce,M.L., and Ruffner,D.E. (1998). Construc tion of a directed hammerhead ribozyme library: towards the identification of optim al target sites for antisense-mediated gene inhibition. Nucleic Acids Res. 26 5093-5101. Piganeau,N., Thuillier,V., and Famulok,M. ( 2001). In vitro select ion of allosteric ribozymes: theory and experiment al validation. J. Mol. Biol. 312 1177-1190. Portera-Cailliau,C., Sung,C.H., Nathans,J., a nd Adler,R. (1994). Apoptotic photoreceptor cell death in mouse models of retinitis pigmentosa. Proc. Natl. Acad. Sci. U. S. A 91 974-978. Potter,M., Chesnut,K., Muzyczka,N., Flotte,T ., and Zolotukhin,S. (2002). Streamlined large-scale production of recombinant ad eno-associated viru s (rAAV) vectors. Methods Enzymol. 346 413-430. Rabinowitz,J.E., Rolling,F., Li,C., Conrath,H ., Xiao,W., Xiao,X., and Samulski,R.J. (2002). Cross-packaging of a single adenoassociated virus (AAV) type 2 vector genome into multiple AAV serotypes enables transduction with broad specificity. J. Virol. 76 791-801. Raisler,B.J., Berns,K.I., Grant,M.B., Be liaev,D., and Hauswirth,W.W. (2002). Adenoassociated virus type-2 expression of pigmented epithelium-derived factor or Kringles 1-3 of angiostatin reduce retinal neovasculariza tion. Proc. Natl. Acad. Sci. U. S. A 99 8909-8914. Rajan,R.S., Illing,M.E., Bence,N.F., and Kopito ,R.R. (2001). Specificity in intracellular protein aggregation and incl usion body formation. Proc. Natl. Acad. Sci. U. S. A 98 13060-13065. Ranchon,I., Chen,S., Alvarez,K., and Anderson,R .E. (2001). Systemic administration of phenyl-N-tert-butylnitrone protects the retina from light damage. Invest Ophthalmol. Vis. Sci. 42 1375-1379. Ranchon,I., LaVail,M.M., Kotake,Y., and A nderson,R.E. (2003). Free radical trap phenyl-N-tert-butylnitrone protects against light damage but does not rescue P23H and S334ter rhodopsin transgenic rats fr om inherited retinal degeneration. J. Neurosci. 23 6050-6057. Reme,C.E., Grimm,C., Hafezi,F., Marti,A., and Wenzel,A. (1998). Apoptotic cell death in retinal degenerations. Prog. Retin. Eye Res. 17 443-464.

PAGE 147

134 Rivolta,C., Sharon,D., DeAngelis,M.M., and Dryj a,T.P. (2002). Retinitis pigmentosa and allied diseases: numerous diseases, genes, and inheritance patterns. Hum. Mol. Genet. 11 1219-1227. Roof,D.J., Adamian,M., and Hayes,A. (1994). Rhodopsin accumulation at abnormal sites in retinas of mice with a human P23H rhodopsin transgene. Invest Ophthalmol. Vis. Sci. 35 4049-4062. Rosenfeld,P.J., and Dryja,T.P. (1995). Molecu lar Genetics of Retinitis Pigmentosa and Related Retinal Degenerations. In Molecu lar Genetics of Ocular Disease, WileyLiss Inc.). Rupert,P.B., and Ferre-D'Amare,A.R. (2001). Crystal structure of a hairpin ribozymeinhibitor complex with implications for catalysis. Nature 410 780-786. Rutkowski,D.T., and Kaufman,R.J. (2004). A trip to the ER: coping with stress. Trends Cell Biol. 14 20-28. Schnepp,B.C., Jensen,R.L., Chen,C.-L., Johnson,P.R., and Clark,K.R. (2005). Characterization of Adeno-Associated Virus Genomes Isolated from Human Tissues. J. Virol. 79 14793-14803. Scott,W.G., Finch,J.T., and Klug,A. (1995). The crystal structure of an all-RNA hammerhead ribozyme: a proposed mechanism for RNA catalytic cleavage. Cell 81 991-1002. Shimayama,T., Nishikawa,S., and Taira,K. ( 1995). Generality of the NUX rule: kinetic analysis of the results of systematic muta tions in the trinucleotide at the cleavage site of hammerhead ribozymes. Biochemistry 34 3649-3654. Sieving,P.A., Chaudhry,P., Kondo,M., Provenz ano,M., Wu,D., Carlson,T.J., Bush,R.A., and Thompson,D.A. (2001). Inhibition of the visual cycle in vivo by 13-cis retinoic acid protects from light damage and provi des a mechanism for night blindness in isotretinoin therapy. Proc. Natl. Acad. Sci. U. S. A 98 1835-1840. Sigurdsson,S.T., and Eckstein,F. (1995). Stru cture-function relations hips of hammerhead ribozymes: from understanding to applications. Trends Biotechnol. 13 286-289. Smith,A.J., Schlichtenbrede,F.C., Tschernutte r,M., Bainbridge,J.W., Thrasher,A.J., and Ali,R.R. (2003). AAV-Mediated gene tran sfer slows photoreceptor loss in the RCS rat model of retinitis pigmentosa. Mol. Ther. 8 188-195. Sonoda,Y., and Streilein,J.W. (1992). Orthot opic corneal transpla ntation in mice-evidence that the immunogenetic rules of rejection do not apply. Transplantation 54 694-704. Steinberg,R.H., Fisher,S.K., and Anders on,D.H. (1980). Disc morphogenesis in vertebrate photorecepto rs. J. Comp Neurol. 190 501-508.

PAGE 148

135 Stenkamp,R.E., Teller,D.C., and Palczewski, K. (2005). Rhodopsin: a structural primer for G-protein coupled receptors. Arch. Pharm. (Weinheim) 338 209-216. Streilein,J.W., Wilbanks,G.A ., and Cousins,S.W. (1992). Immunoregulatory mechanisms of the eye. J. Neuroimmunol. 39 185-200. Stricker,H.M., Ding,X.Q., Quiambao,A., F liesler,S.J., and Naash,M.I. (2005). The Cys214-->Ser mutation in peripherin/rds causes a loss-of-func tion phenotype in transgenic mice. Biochem. J. 388 605-613. Sugden,B., De Troy,B., Roberts,R.J., and Sambrook,J. (1975). Agarose slab-gel electrophoresis equipment. Anal. Biochem. 68 36-46. Sung,C.H., Davenport,C.M., and Nathans,J. (1 993). Rhodopsin mutations responsible for autosomal dominant retinitis pigmentosa. Clustering of functional classes along the polypeptide chain. J. Biol. Chem. 268 26645-26649. Sung,C.H., Makino,C., Baylor,D., and Nathan s,J. (1994). A rhodopsin gene mutation responsible for autosomal dominant retinitis pigmentosa results in a protein that is defective in localization to the photoreceptor outer segment. J. Neurosci. 14 58185833. Sung,C.H., Schneider,B.G., Agarwal,N., Pape rmaster,D.S., and Nathans,J. (1991). Functional heterogeneity of mutant rhodops ins responsible for autosomal dominant retinitis pigmentosa. Proc. Natl. Acad. Sci. U. S. A 88 8840-8844. Surosky,R.T., Urabe,M., Godwin,S.G., McQu iston,S.A., Kurtzman,G.J., Ozawa,K., and Natsoulis,G. (1997). Adeno-associated viru s Rep proteins target DNA sequences to a unique locus in the human genome. J. Virol. 71 7951-7959. Tai,A.W., Chuang,J.Z., Bode,C., Wolfru m,U., and Sung,C.H. (1999). Rhodopsin's carboxy-terminal cytoplasmic tail acts as a membrane receptor for cytoplasmic dynein by binding to the dynein light chain Tctex-1. Cell 97 877-887. Timmers,A.M., Zhang,H., Squitieri,A., and Gonzalez-Pola,C. (2001). Subretinal injections in rodent eyes : effects on electrophysiology a nd histology of rat retina. Mol. Vis. 7 131-137. Tso,M.O., Zhang,C., Abler,A.S., Chang,C .J., Wong,F., Chang,G.Q., and Lam,T.T. (1994). Apoptosis leads to photoreceptor degeneration in inherited retinal dystrophy of RCS rats. Invest Ophthalmol. Vis. Sci. 35 2693-2699. Vaux,D.L. (1993). Toward an understanding of the molecular mechanisms of physiological cell death. Proc. Natl. Acad. Sci. U. S. A 90 786-789. Vermeulen,K., Van Bockstaele,D.R., and Berneman,Z.N. (2005). Apoptosis: mechanisms and relevance in cancer. Ann. Hematol. 84 627-639.

PAGE 149

136 Wang,M., Lam,T.T., Tso,M.O., and Naash,M.I. (1997). Expression of a mutant opsin gene increases the susceptibility of the retina to light dama ge. Vis. Neurosci. 14 55-62. Warrington,K.H., Jr., Gorbatyuk,O.S., Harris on,J.K., Opie,S.R., Zolotukhin,S., and Muzyczka,N. (2004). Adeno-associated virus type 2 VP2 capsid protein is nonessential and can tolerate la rge peptide insertions at its N terminus. J. Virol. 78 6595-6609. Weitzman,M.D., Kyostio,S.R., Kotin,R.M., a nd Owens,R.A. (1994). Adeno-associated virus (AAV) Rep proteins mediate comp lex formation between AAV DNA and its integration site in human DNA. Pr oc. Natl. Acad. Sci. U. S. A 91 5808-5812. Wenzel,A., Grimm,C., Samardzija,M., and Re me,C.E. (2005). Molecular mechanisms of light-induced photoreceptor apoptosis and ne uroprotection for retinal degeneration. Prog. Retin. Eye Res. 24 275-306. Wenzel,A., Grimm,C., Seeliger,M.W., Jaissle, G., Hafezi,F., Kretschmer,R., Zrenner,E., and Reme,C.E. (2001a). Prevention of phot oreceptor apoptosis by activation of the glucocorticoid receptor. Inve st Ophthalmol. Vis. Sci. 42 1653-1659. Wenzel,A., Iseli,H.P., Fleischmann,A., Hafezi,F., Grimm,C., Wagner,E.F., and Reme,C.E. (2002). Fra-1 substitutes for c-Fo s in AP-1-mediated signal transduction in retinal apoptosis. J. Neurochem. 80 1089-1094. Wenzel,A., Reme,C.E., Williams,T.P., Haf ezi,F., and Grimm,C. (2001b). The Rpe65 Leu450Met variation increases retinal resistance agains t light-induced degeneration by slowing rhodopsin regeneration. J. Neurosci. 21 53-58. Werner,M., and Uhlenbeck,O.C. (1995). The effect of base mismatches in the substrate recognition helices of hammerhead ribozym es on binding and catalysis. Nucleic Acids Res. 23 2092-2096. Wilson,J.H., and Wensel,T.G. (2003). The natu re of dominant mutations of rhodopsin and implications for gene therapy. Mol. Neurobiol. 28 149-158. Woodruff,M.L., Lem,J., and Fain,G.L. (2004). Early receptor current of wild-type and transducin knockout mice: photosensitivity and light-induced Ca2+ release. J. Physiol 557 821-828. Wu,T.H., Ting,T.D., Okajima,T.I., Pepperber g,D.R., Ho,Y.K., Ripps,H., and Naash,M.I. (1998). Opsin localization and rhodopsin photochemistry in a transgenic mouse model of retinitis pigmentosa. Neuroscience 87 709-717. Wu,W.C., Lai,C.C., Chen,S.L., Sun,M.H., Xi ao,X., Chen,T.L., Tsai,R.J., Kuo,S.W., and Tsao,Y.P. (2004). GDNF gene therapy attenuat es retinal ischemic injuries in rats. Mol. Vis. 10 93-102.

PAGE 150

137 Yang,G.S., Schmidt,M., Yan,Z., Lindbloom,J.D., Harding,T.C., Donahue,B.A., Engelhardt,J.F., Kotin,R., and Davidson,B.L (2002). Virus-mediated transduction of murine retina with ade no-associated virus: effect s of viral capsid and genome size. J. Virol. 76 7651-7660. Young,R.W., and Bok,D. (1969). Pa rticipation of the retinal pigment epithelium in the rod outer segment renewal process. J. Cell Biol. 42 392-403. Zhou,X., and Muzyczka,N. (1998). In vitro pack aging of adeno-associated virus DNA. J. Virol. 72 3241-3247. Zhu,L., Jang,G.F., Jastrzebska,B., Filipe k,S., Pearce-Kelling,S.E., Aguirre,G.D., Stenkamp,R.E., Acland,G.M., and Palczewski,K. (2004). A naturally occurring mutation of the opsin gene (T4R) in dogs a ffects glycosylation a nd stability of the G protein-coupled receptor. J. Biol. Chem. 279 53828-53839.

PAGE 151

138 BIOGRAPHICAL SKETCH Alan White was born in Atlanta, Georgi a, in 1974, the son of Dennis and Joan White. Soon after, the family moved to Rale igh, North Carolina, where Alans sister Elizabeth was born in 1978. The family retu rned to the Atlanta area in 1981, where his parents live to this day. Alan graduated with honors from Parkview High School in 1992, where he lettered in football, wrestling, and tr ack, and was the male recipient of the schools Scholar Athlete Award. He was also a member of th e Beta Club, and was el ected to the student council during his junior and senior years. Following high school, Alan attended the Un iversity of Georgia from 1992 to 1996, participating in the univers itys honors program and majoring in genetics. His studies were supported by a scholarship from th e Gwinnett County Bulldog Club, a Georgia State Hope Scholarship, and a National Merit Scholarship. Hi s summers were spent as a lifeguard for Gwinnett County Parks and Recrea tion. He graduated in June of 1996 with a Bachelor of Science in genetics. In the Fall of 1997, he entered into his gradua te studies at the University of Florida as a student in the Interdisciplinary Program in Biomedical Sciences, and has pursued his dissertation research in the laboratory of Dr. Alfred Lewin. Following completion of his Ph.D., he plans to continue hi s studies in the field of visi on research with a postdoctoral fellowship in the lab of Dr. Shalesh Kaushal.


Permanent Link: http://ufdc.ufl.edu/UFE0013097/00001

Material Information

Title: Increased Light Sensitivity in Mice Expressing a Mutant Human Rhodopsin Transgene
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013097:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013097/00001

Material Information

Title: Increased Light Sensitivity in Mice Expressing a Mutant Human Rhodopsin Transgene
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013097:00001


This item has the following downloads:


Full Text












INCREASED LIGHT SENSITIVITY IN MICE EXPRESSING A MUTANT HUMAN
RHODOPSIN TRANSGENE















By

ALAN WHITE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Alan White

































For my family, and for Tom Baldwin.















ACKNOWLEDGMENTS

I would like to thank Dr. Alfred Lewin for his continuous friendship, support, and

advice during my graduate education. I would also like to thank my committee members,

Drs. William Hauswirth, Clay Smith, and Terence Flotte, for their helpful input and

critical analysis as my studies progressed.

Dr. Adrian Timmers was of particular help in developing methods for and

performing animal injections and ERG analysis, and I also appreciate his cheerful

encouragement. Dr. Lynn Shaw patiently oversaw much of my training in the basics of

molecular biology and recombinant DNA technology and for that I am grateful. I thank

Dr. Quihong Li for performing the funduscopy described in this dissertation. Drs. JiJing

Pang, Seok Hong Min, and Marina Gorbatyuk also performed subretinal injections. Dr.

Henry Baker was a great help during my development as a scientific speaker, and as a

student of science in general.

I would like to extend special thanks to Mr. James Thomas Jr. for sharing his

friendship and his knowledge of PCR and DNA analytical techniques, and for keeping

the lab running smoothly, and Ms. Chrissy Street for managing my training grant and

being my friendly liason to the Hauswirth lab. Mr. Tom Doyle was another valuable

resource for animal techniques and general good input. I also extend my warmest thanks

to all of my labmates and friends from other labs and departments, whose help, advice,

and goodwill have been invaluable over the past few years.









This dissertation work could never have been completed without the heroic

assistance of Ms. Joyce Conners. I would like to extend my warmest thanks to her for

her assistance as my department graduate student secretary. Through her diligence I kept

more deadlines than I missed, and was guided through the mass of bureaucratic

necessities that beleaguer every graduate student. Her friendship will always be valued.

Finally I would like to thank my grandparents, Bette Womack and Esther and Dick

White, my parents, Dennis and Joan White, and my sister, Elizabeth White Spratlin, for

their constant love and support throughout my life.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ....................................................... ............ ....... ....... ix

LIST OF FIGURES ............................... ... ...... ... ................. .x

ABSTRACT .............. ..................... .......... .............. xii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

R etinitis Pigm entosa ............................................... ..... ....... ............ .. 1
H history and Pathology ................................................... .... .............. .
Causes of R etinitis Pigm entosa ................................... ............................. ....... 2
R h odop sin ........................... .............................................. 3
Rhodopsin and the V isual Cycle ........................................ ....................... 4
Rhodopsin and Retinitis Pigm entosa................................... ...................... 7
Animal Models of Retinitis Pigmentosa............... ..................... ...............14
Gene Therapy for Retinitis Pigm entosa................................... ....................... 18
R ibozym es .........................................20
M echanistic Description ........................................ ....... ...................... 21
A A V ...................................................................................................................... 2 4
Production of A A V V ectors ................................... ........................................... 26
AAV and Retinal Gene Therapy ........................................ ...... ............... 29
P roj ect ..................................................................... ............ ...... .. 3 1

2 CHARACTERIZATION OF A NOVEL MOUSE MODEL OF RETINITIS
PIGMENTOSA............ .................. .. ... .... ..................32

In tro du ctio n ...................................... ................................................ 3 2
M materials and M methods ....................................................................... ..................34
D N A O ligonucleotides ............................................... ............................. 34
Isolation of G enom ic D N A ........................................ ........................... 34
PCR Analysis of Genomic DNA ..................................................................... 35
E lectroretinography .............................. ........................ .. ........ .... ............36
F u n du scopy ...................................... ............................................ ... 3 8
H isto lo g y .............................................................................................................3 8









R results ............... .. .... ..... ......... .. ........ .................... 39
Breeding Founder and Experimental M ice............................... ............... 39
E R G N atu ral H history ........................................ ............................................4 0
Funduscopy........... .... ............................. ......... ..... ......... 40
H isto lo g y ....................................................... 4 3
Discussion ....................................................46

3 AAV-MEDIATED RIBOZYME TREATMENT OF mRHO+/-; hT17M MICE...... 50

In tro d u ctio n ........................................................................................................... 5 0
M materials and M methods ....................................................................... ..................52
R N A O ligonucleotides ............................................... ............................. 52
D N A O ligonucleotides .............................................................. .............. 52
Preparation of Synthetic RNA Ribozymes and Substrates..............................53
5' End-labeling of Deprotected Target RNAs .............................................53
In Vitro Ribozyme Time Course Analysis............................................. 54
Ligating Ribozyme Sequences into rAAV Packaging Vectors...........................55
Subretinal Injection of rAAV Ribozyme Delivery Vectors ..............................58
E lectroretinography .............................. ........................ .. ........ .... ............59
R e su lts ......... ... ....... .. ............. .. ....................................................6 0
Ribozym e Creation ................ ...................... 60
In Vitro Time Course Analysis of HRzl and HRz3 ..................... ........................61
ERG Analysis of hT17M Transgenic Mice Treated With HRzl and HRz3 .......62
Discussion .............. ............ .... ...... ........ ................ ........... 65

4 INCREASED LIGHT SENSITIVITY IN mRHO+/-; hT17M MICE ..... ........ 70

In tro d u ctio n .......................... .. .............. ....................... ................ 7 0
M materials and M methods ....................................................................... ..................74
R etinal Illum nation ......... ............................................................ ... .... ....... 74
G enotyping ........................................................................75
E lectroretinography .............................. ........................ .. ........ .... ............75
Funduscopy ............. ..... ........................................ .... ..... 75
H isto lo g y ............... .................................................................7 5
TUNEL Visualization of Apoptosis ............. ......... ......... ............... 76
Results ....................... ...................................77
H igh Intensity Illum nation ........................................ ........................... 77
Low Intensity Illum nation .................................... ......... .. ..................... 77
Apoptosis in Retinas Damaged by Low Intenisty Illumination ........................80
Funduscopic Illum nation .................. ........... .................................... ... ..... 80
Apoptosis in Retinas Damaged by Fundus Photography ...................................83
ERG Analysis of hP23H Mice After High Intensity Illumination ......................86
Apoptosis in hP23H Mouse Retinas After High Intensity Illumination..............89
D iscu ssio n ...................... .. .. ......... .. .. ..................................................7 0









5 RED FILTERED LIGHT FOR INJECTIONS PROTECTS AGAINST LIGHT
DAMAGE IN THE mrho+/-; hT17M MOUSE ................................................. 94

In tro du ctio n ...................................... ................................................ 9 4
M materials and M methods ....................................................................... ..................96
Creation of 600nm Filters .............. ..... ......... ... ............... 96
R etinal Illum nation ......... ............................................................ ... .... ....... 97
E lectroretinography .............................. ........................ .. ........ .... ............97
H isto lo g y ............... .................................................................9 7
TUNEL Visualization of Apoptosis ............. ......... ......... ............... 97
Test Injections Using 600nm Filtered Light....................................................97
R e su lts ....... ......... .............. ...... ......................... ............... 9 8
Spectrophotometric Analysis of 600nm Filters........................ ............... 98
ERG Analysis After 600nm Retinal Illumination............................................98
Apoptosis in Retinas Exposed to 600nm Illumination.......... ...............100
Test Injections Using 600nm Filtered Light.................................................101
D isc u ssio n ............ .. ............ ................. ................................ 1 0 4

6 DISCUSSION .............................. ......... ..... .. .......... .... 107

Summary ............. ............ .................................107
Mechanism of Light-Induced Photoreceptor Apoptosis............... .................. 110
C clinical Im pact .......................................................................................................... 117

L IST O F R EFER EN CE S ......... .................................... ........................ ............... 121

BIOGRAPHICAL SKETCH ...................................... ..................................138
















LIST OF TABLES


Table p

1-1 Tissue tropism and site of isolation of the various AAV serotypes. ......................28

2-1 ONL averages taken from mrho+/- and mrho+/-; hT17M mice .............................45

4-1 Percent reduction in eyes illuminated 10,000 lux white light ..................................88

5-1 Percent reduction in eyes illuminated with red or white light.............................101
















LIST OF FIGURES


Figure page

1-1 Funduscopic presentation of Retinitis Pigmentosa. ..........................................2

1-2 A rod photoreceptor cell.......................... ................... ................................... 4

1-3 Isomerization of 11-cis to all-trans retinal. ....................................................5

1-4 The phototransduction cascade. ........................................... .......................... 6

1-5 Illustration of rhodopsin as it is inserted into the outer segment .............................8

1-6 Generic structure of a hammerhead ribozyme ............................... ............... .22

1-7 Structure of a specific hairpin ribozyme .............. .................................... ........ 23

1-8 Ribozyme cleavage in trans ............................... ................. ... .............. 24

2-1 Agarose gel electrophoresis of PCR reactions .................................................... 35

2-2 Electroretinographic apparatus...................................................... ..... .......... 37

2-3 A n exam ple of an ERG tracing ........................................ .......................... 37

2-4 10dB intensity ERG a-wave natural history...........................................41

2-5 10dB intensity ERG b-wave natural history ................. ............... .............. 41

2-6 10 dB intensity a-wave responses charted as a percentage ....................................42

2-7 10 dB intensity b-wave responses charted as a percentage................................ 42

2-8 Fundus photographs and respective 10 dB intensity ERG tracings .......................43

2-9 Representative sections of mrho+/- mice (A-D) and mrho+/-; hT17M siblings......44

2-10 Tile-field mapped image of a mouse retina......................................................45

3-1 pXX-GS-HP MOPS 500 rAAV packaging plasmid .............................................57

3-2 Dissecting scope and fiber optic light used during subretinal injections ...............59









3-3 Cartoon depicting the subretinal injection ............. .............................................. 59

3-4 Primary structure of ribozymes HRzl and HRz3................. ............... ..............60

3-5 Representative Phosphorlmager scan of a time course assay ................................62

3-6 Time course of HRzl and HRz3 cleavage .................................... ............... 62

3-7 Relative ERG responses of a- and b-waves .................................. ............... 64

3-8 10dB intensity ERG tracings from mice with one eye injected............. ...............66

3-9 10dB intensity ERG tracings from mice with both eyes injected ......................... 67

4-1 A-wave ERG responses after high intensity illumination.................... ........ 78

4-2 B-wave ERG responses after high intensity illumination .....................................78

4-3 A-wave ERG responses after low intensity illumination...................................79

4-4 B-wave ERG responses after low intensity illumination ......................................79

4-5 TUNEL stained retinal sections from low intensity-illuminated mice ....................81

4-6 Tile-field mapped image of the TUNEL stained retina ................. .... ............82

4-7 A and b-wave ERG responses from mice with funds photography .....................84

4-8 Fundus pictures of mrho+/- and mrho+/-; hT17M mice at three and six weeks......85

4-9 TUNEL labeling of retinal sections from mice with right eye funduscopy.............86

4-10 10dB intensity a- and b-wave ERG responses of illuminated P23H mice ..............90

4-11 TUNEL labeling of retinal sections from P23H mice that were illuminated...........91

5-1 A bsorbance spectrum of rhodopsin.................................. ..................................... 95

5-2 600nm red light filters ....................... ...... .............. ..................................... 96

5-3 ERG amplitudes after right eye illumination with 600nm filtered light ................99

5-4 TUNEL labeled right and left eye sections ...... ......................................102

5-5 ERG responses measured two weeks after subretinal test injections.....................103

6-1 Eye from a human who inherited the T17M rhodopsin mutation..........................119


















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

INCREASED LIGHT SENSITIVITY IN MICE EXPRESSING A MUTANT HUMAN
RHODOPSIN TRANSGENE


By

Alan White

December 2005

Chair: Alfred Lewin
Major Department: Medical Sciences--Genetics

Retinitis Pigmentosa (RP) is a heterogeneous class of retinal disorders

characterized by an initial loss of peripheral and night vision, followed by loss of central

and daylight vision, affecting around 1.5 million people worldwide. Many RP patients

develop the disease because of mutations in a retinal protein called rhodopsin. Normal

rhodopsin is a vital component of the phototransduction cascade that allows the retina to

detect light, while mutant versions of this protein cause the cells of the retina to become

sick and die, leading to blindness.

Mouse models of the disease that are genetically engineered to express mutant

rhodopsin proteins are vital for studying the progression of RP and developing treatments

for the disease. My work describes one of these models, which expresses a human

rhodopsin transgene with a tyrosine to methionine mutation at the 17th amino acid of the

protein (hT17M). We bred our line to express the T17M human transgene on a









background that was hemizygous null for mouse rhodopsin (mrho+/-), which closely

modeled RP mutations in human patients in expressing one copy of mutant rhodopsin and

one copy of wild-type rhodopsin. We performed electroretinographic analysis (ERG) to

show that this line loses its visual responses over time. Histology confirmed that ERG

attenuation was accompanied by a loss of rod photoreceptors in the retina.

Unsuccessful attempts to treat the hT17M; mrho+/- mice with subretinal injections

of rAAV-expressed ribozymes led to the discovery of an hT17M-specific light sensitivity

that caused severe loss of a- and b-wave ERG responses. Histological analysis showed a

concomitant loss of photoreceptors, and TUNEL labeling of fragmented DNA in rod

photoreceptor cells demonstrated that the damage was occurring via an apoptotic

pathway. Attempts to reproduce this light damage phenotype in another mouse model of

retinal disease that expressed a human rhodopsin transgene with a proline to histidine

mutation at the 23rd amino acid were unsuccessful, leading us to conclude that this light

sensitivity was not common to all rhodopsin mutations. Finally, filters were developed

that removed the wavelengths of light responsible for the retinal damage, allowing for

non-damaging subretinal injections of the hT17M mice.














CHAPTER 1
INTRODUCTION

Retinitis Pigmentosa

History and Pathology

Retinitis pigmentosa (RP) is the common name for a group of retinal disorders

characterized by progressive photoreceptor degeneration that culminates in loss of vision

(Flannery et al., 1989; Hims et al., 2003; Farrar et al., 2002). The disease affects from

50,000 to 100,000 people in the United States and around 1.5 million people worldwide.

Initial symptoms include night blindness and loss of peripheral vision, usually occurring

during the late teens or early twenties. As the photoreceptors continue to degrade the

visual impairment progresses towards the center of the retina, eventually affecting the

cone photoreceptors that are responsible for central vision in bright light conditions, and

resulting in the manifestation of "tunnel vision", in which the visual fields of the patient

constrict to less than 200.

The loss of vision is accompanied by visual pigment depositions in the retina for

which the disease is named (Figure 1-1) (Stricker et al., 2005; Farber et al., 1987). In

most cases this is accompanied by a waxy pallor to the optic nerve, attenuated retinal

vasculature, and thinning of the retinal pigmented epithelium. In many cases, abnormal

ERG responses predict the development of retinitis pigmentosa in affected individuals

well before other symptoms present. For most patients the visual fields will continue to

constrict until they are completely blind (Humphries et al., 1992).









Causes of Retinitis Pigmentosa

Retinitis pigmentosa shows a high degree of genetic heterogeneity. Causative

mutations can be subdivided into several genetic categories: autosomal dominant

(ADRP), autosomal recessive (ARRP), X-linked (XLRP), or syndromic (Hims et al.,

2003; Farrar et al., 2002). The percent contribution of each type of disease varies among

different populations, but it is generally agreed that ADRP accounts for around 25% of

all cases, ARRP for 20%, X-linked and syndromic for around 8%, with the rest of the

cases believed to be the result of spontaneous mutations arising in the affected individual

(Diager et al., 2005). The majority of causative genes identified to date lead to the

autosomal dominant form of RP.


Figure 1-1. Funduscopic presentation of Retinitis Pigmentosa. This series shows a
normal retina (A) and a retina exhibiting evidence of Retinitis Pigmentosa (B)
(Rosenfeld and Dryja, 1995).

Retinitis pigmentosa can be caused by mutations in a wide variety of genes. The

first of these candidate genes to be discovered was rhodopsin, which is expressed in rod

photoreceptor cells and is an integral part of the phototransduction cascade.

Subsequently a mutation in the peripherin RDS gene, which is a structural protein

involved in maintaining rod photoreceptor outer segment disc morphology, was also


A








Normal Retitia


RP Rekia









shown to be linked to the disease. More recently, mutations in phototransduction

proteins including the a and 3 subunits of rod cGMP phosphodiesterase, the a subunit of

the rod cGMP-gated channel, and arrestin have been shown to lead to RP. RPE65, which

is expressed in the retinal pigmented epithelium and is involved in visual pigment

regeneration, as well as ROM 1, which has a structural role related to peripherin RDS,

have also been identified as RP candidate genes. In all, 44 different loci have been

identified as being associated with RP, of which 35 have been cloned (Humphries et al.,

1992; Diager et al., 2005; Kennan et al., 2005). It is of interest that some of the genes

associated with RP encode proteins needed in all cells for functions such as RNA splicing

or purine metabolism. Why mutations in these genes lead to retinal degeneration and not

other phenotypes is unknown (Bowne et al., 2002; Kennan et al., 2002; Martinez-Gimeno

et al., 2003; Maita et al., 2005).

Rhodopsin

Opsin is a seven-transmembrane G-coupled receptor protein found in the disc

membranes of the outer segments of rod and cone photoreceptor cells (Figure 1-2). The

protein is oriented in the rod outer segment (ROS) membrane such that its amino

terminus is located on the inside of the disc in the intralumenal space, while the carboxyl

terminus is found on the outside of the disc in the cytoplasmic space. The amino

terminus of the protein contains two glycosolation sites, at Asn2 and Asn 15, and there is

extensive association between the intralumenal portions of opsin and these carbohydrate

moieties (Stenkamp et al., 2005; Hargrave and Mcdowell, 1992). Opsin is incredibly

abundant, accounting for around 90% of the total protein in the rod outer segments

(Daiger et al., 1995).










Rhodopsin and the Visual Cycle

In its active form, the protein binds the visual pigment 11-cis retinal to form

rhodopsin. The term "rhodopsin" will be used to refer to this protein-chromophore

complex for the remainder of this dissertation. The 1 1-cis retinal is covalently bound to

rhodopsin via a protonated Schiff-base at Lys296 (Bownds, 1967). The opsin protein

itself does not absorb visible



Rod Photoreceptor Cell

------ ..hiW-'"CF Rhodopin
F'!> C }- DISC
Outer
Segment



,.,VInardsca Surfac
inner
Segment 17


Nucleus ,;
Cytplasrmic Surface
Fiber
Synaptic
Ending

Rhodopsin
Molecule





Figure 1-2 A rod photoreceptor cell, with expanded views showing outer segment disc
morphology and the orientation or rhodopsin within the outer disc membrane.
The chromophore 11-cis retinal is shown in a cutaway view to be bound to the
interior of the rhodopsin protein. Figure adapted from Hargrave and
McDowell, 1992.

light, but when it is bound to 11-cis retinal to form rhodopsin, the resultant molecule has

a broad absorption band with a peak at around 500nm. Photons impacting upon

rhodopsin provide energy that is able to temporarily convert the cis-double bond between

C-11 and C-12 of 11-cis retinal into a single bond, allowing the molecule to rotate










through 180 to an all-trans configuration (Figure 1-4). This isomerization leads to

conformational changes in rhodopsin that allow it to interact with downstream molecules

in the phototransduction cascade.

After light activation, rhodopsin shifts between two conformations, termed

metarhodopsin I and metarhodopsin II. Metarhodopsin II is able to transiently bind to the

next protein in the cascade, transducin, which is a heterotrimeric G-protein consisting of

cis linkage



26 10 12
fl, CH3 H CH3 H /


3 H H H
CH3 14 CH3
0 15
11-cis Retinal c 3 H trans linkage

CH3 CH3 H CH3 H/ CH3 H

1 C-
| H H Hi H
CH3
all-trans Retinal

Figure 1-3. Isomerization of 11-cis to all-trans retinal. The C-11 to C-12 linkage about
which the rotation occurs is highlighted in red.

three subunits, c, 3, and y. Each molecule of light-activated rhodopsin is able to

interact with hundreds of transducin molecules, resulting in the first of a series of signal

amplifications. The a subunit of transducin in its inactive state is bound to GDP.

Interaction with metarhodopsin II causes the a subunit to exchange its GDP for a GTP

molecule, causing it to dissociate from the other subunits and allowing it to bind to cGMP

phosphodiesterase, the next player in the cascade. cGMP phosphodiesterase is composed

of catalytic a and 3 subunits that are bound to and inhibited by two y subunits.









Interaction between activated transducin and one of the y subunits releases the ao and 3

subunits to hydrolyze cGMP to 5'GMP.

The resultant drop in the intracellular cGMP concentration results in the closing of

a few hundred to a few thousand cGMP-gated Ca2+ channels, resulting in a decrease in

the intracellular concentration of Ca2+, which causes a hyperpolarization of the ROS

plasma membrane. This signal is propagated along the plasma membrane to the synaptic

terminus of the rod photoreceptor (Figure 1-2), resulting in a reduction in the release of

glutamate. This decreases the activity of nearby bipolar cell glutamate receptors, which

in turn decreases the activation of a G-coupled receptor protein and




fT= lowIl 11I) .... ..

^ e rhim is" n OCAP ,


raw* d
eanin ucins




m,, .., "" "Ei r"



,oa m .
7.11


recomin



-, -,iii ii ^ !!:.. ':i
L.. I H :
......................

Figure 1-4. The phototransduction cascade. Figure courtesy of Dr. Helga Kolb (Kolb et
al., 2005).









leads to an increase in the cGMP concentration in the bipolar cell. The increase in cGMP

results in the opening of large numbers of cationic channels, resulting in bipolar cell

depolarization and the generation of an action potential (Figure 1-4) (Hargrave and

Mcdowell, 1992; Daiger et al., 1995; Maple and Wu, 1996).

Inactivation of this cascade is initiated by rhodopsin kinase, which binds to and

phosphorylates metarhodopisn II. Phosphorylated metarhodopsin II is then able to

interact with arrestin, which prevents interactions with transducin until metarhodopsin

releases the all-trans retinal. Release of all trans retinal is thought to inhibit the re-

opening of the cGMP-gated ion channels. Eleven-cis-retinal is eventually recycled

through the retinoid cycle that has steps in both photoreceptors and the retinal pigmented

epithelium (McCabe et al., 2004). In the meantime, cGMP is regenerated from GMP by

the protein guanylate cyclase, and the cGMP is able to bind to and eventually reopen the

cGMP-gated ion channels. This causes an influx of Ca2+ that restores the resting

potential of the ROS, stimulating the release of glutamate at the synaptic terminus and

terminating the light-induced signal.

Rhodopsin and Retinitis Pigmentosa

The first ADRP gene was identified and localized to the long arm of chromosome 3

in 1989 by researchers investigating the pedigree of a large Irish family with over fifty

individuals reporting symptoms consistent with retinitis pigmentosa (Bradley et al., 1989;

McWilliam et al., 1989). As RP was known to affect rod photoreceptors, and the gene

encoding the ROS protein rhodopsin had recently also been localized to the long arm of

chromosome 3, the race was on to identify a point mutation in rhodopsin that was

associated with ADRP. In 1990, the first ADRP-causing rhodopsin mutation was









reported, consisting of a DNA mutation (CCC to CAC) that caused histidine to be

substituted for proline (P23H) at the 23rd amino acid of the protein (Dryja et al., 1990).

Rhodopsin has subsequently become the most extensively characterized gene

associated with retinitis pigmentosa. Mutations in the rhodopsin gene account for around

10% of all reported cases of RP (Rivolta et al., 2002). Since discovery of the P23H

mutation, around 150 different mutations of rhodopsin have been shown to cause the

disease. A vast majority of these mutations lead to autosomal dominant RP, and most of

these mutations are thought to cause retinal degeneration by either a toxic gain of

function or a dominant negative fashion (Wilson and Wensel, 2003). Many of these

mutations are illustrated in Figure 1-5.

iddMI HOOC MI S
Rhodopsin oo .Z il i'
-T
Ac^ 9
Z0 .O ,^' k t

i ioplasa i p t ic





205.-, '-,^, '
S. .... -,, K C ,-



F, (F 0 1 ,) '
) )-." p i '-




T TV: xa Rina bNndg -i
I-un F'if' i 7
IPEi ,9T








Figure 1-5. Illustration of rhodopsin as it is inserted into the outer segment disc
membrane. The image depicts some secondary structure and also illustrates
key amino acids in which mutations lead to retinitis pigmentosa (Diager et al.,
2005).
2005).









Mutations affecting the Lys295 residue prevent rhodopsin from binding to 11-cis

retinal, and cause the protein to be constitutively activated, leading to retinitis

pigmentosa, possibly because this protein is able to continuously bind to and sequester

arrestin (Berson, 1996). It is also known that mutations at the Tyr4 and Tyrl7 residues

abolish the glycosylation sites at the N-terminus of rhodopsin, leading to its aberrant

trafficking in cell culture models, and in the case of the T17M mutation, inefficient

regeneration of the protein with 11-cis retinal (Kaushal et al., 1994). Addtionally, the

cystine residues at positions 110 and 187, which form a conformationally vital disulfide

bond, and the glutamate residue at codon 114, which provides the counter ion for the

Schiff s base retinal linkage at codon 296, are structurally important residues that, when

mutated, can lead to RP (Kamik et al., 1988; Kamik and Khorana, 1990; Daiger et al.,

1995). However, the mechanisms by which the majority of rhodopsin mutations lead to

RP are not well understood.

Several groups have attempted to categorize rhodopsin mutations by expressing

rhodopsin genes containing them in cultured cells and then analyzing the resultant

proteins with respect to a variety of factors. One such analysis utilized rhodopsin cDNAs

engineered to contain thirty four mutations known to cause ADRP in patients and

expressed in H293S cells. These experiments lead to two categories of rhodopsin

mutations, Class I and Class II. Class I mutations were less numerous, accounting for

only six of the 34 mutations studied. Class I mutants were similar to wild-type rhodopsin

expressed in the same system in terms of yield, subcellular localization (the plasma

membrane), and regeneration with 11-cis retinal, and tended to cluster in both the first

transmembrane and carboxyl terminal domains of the protein. Class II mutations were









more numerous, and were found clustered in the transmembrane and loop domains of

rhodopsin. Class II mutants accumulated to significantly lower than wild-type levels,

regenerated poorly with 11-cis retinal, were predominately mislocalized to the

endoplasmic reticulum, and were shown to form intracellular aggregates (Sung et al.,

1991; Sung et al., 1993).

A concurrent study involved introducing 35 mutations into a synthetic bovine

rhodopsin gene and expressing them in COS cells (Kaushal and Khorana, 1994). The

proteins expressed in this study were classified into three classes. Class I, like those of

the studies of Sung and coworkers, consisted of proteins that showed expression levels,

subcellular localization (to the plasma membrane) and chromophore regeneration that

were similar to wild-type rhodopsin. Class II consisted of proteins that showed folding

defects, were mislocalized to the endoplasmic reticulum, and were not able to reconstitute

with 11-cis retinal. Class III mutations also showed folding and localization defects, but

were able to partially regenerate with 11-cis retinal. Taken together, these two studies

provide evidence for at least two types of rhodopsin defect that can lead to ADRP.

Recently, Mendes and coworkers have proposed classifying rhodopsin mutations into

five groups based on additional characteristics such as whether they affect endocytosis or

whether they remain constituively activated (Mendes et al., 2005).

There are several theories concerning the mechanism of retinal degenerations

caused by rhodopsin mutations. Sung and coworkers note that their class II rhodopsin

mutations show retention/mislocalization to the endoplasmic reticulum that is similar to

that seen in other disease-causing mutant proteins such as Class II low density lipoprotein

receptors and cystic fibrosis transmembrane conductance regulator proteins. Indeed,









there is evidence for a variety of neurodegenerative diseases that share an accumulation

of aggregated, ubiquitinated mutant proteins, suggesting that these proteins are targeted

for destruction, possibly due to toxicity resulting from their expression (Wilson and

Wensel, 2003; Rajan et al., 2001; Bence et al., 2001; Illing et al., 2002). It is also thought

that overloading the cellular machinery responsible for the removal of misfolded and

toxic proteins, termed the unfolded-protein or ER-stress response, can lead to

programmed cell death (Mendes et al., 2005; Rutkowski and Kaufman, 2004).

In the case of mutations affecting the tyrosine residues at codons 4 and 17, the

pathogenic mechanism may in part be explained by the abolishment of glycosylation at

nearby residues, which has been shown to cause defective ROS membrane

morphogenesis in Xenopus laevis retinas (Fliesler et al., 1985). Defects in outer segment

morphogenesis are thought to lead to photoreceptor cell death, and can result from

mutations in rhodopsin that prevent it from being properly transported to the OS disc

membranes. It is known that the ROS sheds -10% of its outer segment discs each day

(Young and Bok, 1969). These discs are phagocytosed by the retinal pigmented

epithelium, and more must be synthesized at the base of the ROS each day to replace

them (Young and Bok, 1969; Papermaster et al., 1986; Steinberg et al., 1980). Studies of

carboxy terminal rhodopsin mutations have shown that this region of the protein is

important for dynein binding and cellular transport. Mutations in this region have been

shown to lead to a mislocalization of rhodopsin in the rod photoreceptors, causing

abnormal ROS disc morphogenesis, which would help to explain the pathogenesis of

Class I rhodopsin mutations (Tai et al., 1999; Sung et al., 1994). Additionally,

overwhelming synthesis of aberrantly folded rhodopsin may interfere with the subcellular









trafficking of non-mutant rhodopsin, again resulting in a hindering or prevention of disc

morphogenesis and causing a progressive shortening of the ROS that eventually leads to

cell death (Mendes et al., 2005; Besharse and Wetzel, 1995).

While the exact mechanisms by which the various rhodopsin mutations lead to RP

are still uncertain, it is clear that the ultimate fate of the affected photoreceptors is

programmed cell death, or apoptosis. Apoptosis is a thoroughly documented

phenomenon by which cells initiate a specific program of self-destruction in response to

intrinsic or extrinsic factors (Wenzel et al., 2005). These can include mechanical

damage, toxic chemical exposure, bacterial or viral infection, and various forms of

irradiation. Apoptosis is also responsible for the programmed deaths of cells for

developmental reasons, deaths of auto-reactive cells of the immune system, and deaths of

cells that have lost growth inhibition and could become cancerous. Two major apoptotic

pathways have been described, an extrinsic pathway involving CD95 and CD95 ligand

and an intrinsic pathway involving mitochondrial damage. Activation of the aspartyl

proteases, termed caspases, is common to both mechanisms, as is the ultimate release of

mitochondrial cytochrome c, which leads to full apoptotic activation. While the extrinsic

pathway plays an important role in shutting down the immune response, the intrinsic

pathway is thought to be more important in pathologic apoptosis such as that occurring in

RP (Vermeulen et al., 2005; Mohamad et al., 2005).

Additionally, caspase-independent pathways have also been described that involve

effector molecules including cathepsins, calpains, granzyme A and B, and serine

proteases such as AP24. Two major players in the caspase independent pathway are

apoptosis inducing factor (AIF) and PARP-1. AIF is an oxioreductase found in the









mitochondria, and decreased levels of AIF have been correlated with an increased

sensitivity to oxidative stress. Activation of the caspase-independent apoptotic pathway

leads to mitochondrial secretion of AIF, which is ultimately involved chromatin

condensation and recruits the endonuclease EndoG to effect chromatin degradation.

PARP-1 is involved in DNA repair, and overstimulation as a result of DNA damage is

thought to lead to cell death through metabolic depletion. Interestingly, PARP-1 is

actually a target for caspase cleavage in caspase-dependent apoptosis this is thought to

be a cell strategy to reduce metabolic depletion and increase the energy available to effect

orderly apoptosis through the caspase-dependent pathway (Wenzel et al., 2005).

Hallmarks of apoptosis include reduced cell size and the appearance of bubble-like

blebss" on the surface of the plasma membrane. The nuclear chromatin breaks down,

leading to a diagnostic "DNA laddering" morphology, and the mitochondria begin to lose

integrity, releasing cytochrome c into the cytoplasm. Eventually the entire cell breaks

down into small, membranous vesicles, which are finally engulfed by macrophages and

dendritic cells that recognize the apoptotic cells (Vaux, 1993; Reme et al., 1998).

The link between apoptosis and RP-related photoreceptor cell death has been well

established (Reme et al., 1998). In 1994, Portera-Cailliau and coworkers investigated the

mechanism of cell death in the three mouse models of retinitis pigmentosa: the rd mouse,

which contains a defect in the rod cGMP phosphodiesterase gene, the rds mouse, which

contains a defect in the structural gene peripherin, and in mice containing a Q334

termination mutation in the rhodopsin gene (Portera-Cailliau et al., 1994). In each

mouse, DNA fragmentation, a hallmark of apoptosis was seen in the photoreceptors. In

concurrent studies, researchers showed similar evidence for apoptotic cell death in RCS









rats, which carry defects in the ability of the RPE to phagocytose ROS discs (Tso et al.,

1994). In 1996, another study showed apoptotic cell death as the ultimate fate of rod

photoreceptors in mice engineered to contain a P23H rhodopsin mutation, and a

correlation was noted between increasing amounts of apoptosis and decreased ERG

findings in the same animals (Naash et al., 1996).

A hallmark of retinal degeneration resulting from RP is that after a certain number

of rod photoreceptors have died, normal rods and cones that do not express mutant opsins

begin to die as well. In 1993, chimeras between normal mice and mice carrying a P347S

mutant rhodopsin transgene were created in order to address this issue. In situ

hybridization assays confirmed that the chimeric retinas were made up of mixtures of

adjacent mutant and non-mutant rod photoreceptor cells, while histological examination

revealed that cell death was occurring simultaneously in both mutant and non-mutant

sections of the retina (Huang et al., 1993). This, taken together with the observations of

cone cell death resulting from a rod defect, has led to the belief that photoreceptor cells

secrete "survival factors" necessary to the survival of their neighbors, and thus the deaths

of mutant rod photoreceptors can have deleterious effects upon neighboring cones and

non-mutant rods (Rosenfeld and Dryja, 1995; Bredesen et al., 2005). This theory is

supported by experiments in which various cell survival factors (e.g. BDNF, CNTF, and

NGF) were injected intravitreally into both naturally occurring and artificial models of

retinal degeneration. These cellular factors slowed the degenerations in several of the

mouse lines (LaVail et al., 1998).

Animal Models of Retinitis Pigmentosa

The development of animal models has been vital to the study of retinitis

pigmentosa. Initial lines consisted of naturally occurring mutants that showed early onset









photoreceptor degeneration, the first of which was the rd mouse, identified in 1966,

which is currently known to involve a spontaneous mutation in the beta subunit of rod

cGMP phosphodiesterase (Keeler, 1966). Many other naturally occurring mouse mutants

have since been isolated including, among others, mouse models of Purkinje cell

degeneration, the rds mouse, which as mentioned before contains defects in its peripherin

gene, and models of Leber Congenital Amaurosis and cone photoreceptor function loss

(Chang et al., 2002; Pang et al., 2005). In addition to the naturally occurring mutations,

targeted gene disruption techniques have enabled researchers to create and study mouse

lines which lack other integral phototransduction proteins, such as arrestin and the Rpe-

65 protein.

With mutations in the rhodopsin gene accounting for over 10% of all incidences of

RP, it is not surprising that a large number of mouse models involving target rhodopsin

deletions and mutations have been designed. In 1992, a mouse model was created to

express a either a wild-type human rhodopsin transgene or a human rhodopsin transgene

carrying the P23H mutation. Three lines expressing the mutant transgene were created,

each of which expressed the mutant rhodopsin at a different level, and although all three

lines exhibited retinal degeneration, it was observed that the rate of retinal degeneration

was directly proportional to the expression level of the mutant transgene. This model

was also important in that it demonstrated that ADRP could arise from a single point

mutation in a single gene (Olsson et al., 1992).

Two lines expressing the normal human rhodopsin transgene were also created in

the studies by Olsson and coworkers, one of which expressed rhodopsin at levels

comparable to non-transgenic mice, and one that expressed five times as much rhodopsin.









Intriguingly, the line that exhibited overexpression of the wild-type human transgene also

showed a retinal degeneration, even though there was no mutation involved, which

suggests that overexpression of rhodopsin can lead to similar trafficking and aggregation

problems as expression of mutant rhodopsins at normal levels. Subsequent studies on

this mouse showed accumulation of mutant rhodopsin at abnormal sites in the rod

photoreceptor cells (Roof et al., 1994).

Another mouse model involved germline insertion of a mouse rhodopsin transgene

that was mutated to contain two silent RFLPs and three amino acid substitutions, one of

which was the P23H mutation, and the other two being nearby non ADRP-associated

amino acid substitutions that were included to provide an epitope tag. This mouse model,

termed the "VPP" model because of the three amino acid substitutions that were

introduced, showed normal expression levels of a mixture of wild-type and mutant

rhodopsin, and exhibited a slow photoreceptor degeneration that mimicked that seen in

human patients. Study of this mouse line revealed that dark-reared mutant animals had

significantly reduced rates of photoreceptor degeneration, with a threefold decrease in the

appearance of apoptotic photoreceptor cells when compared to mutant siblings raised in

twelve hour cyclic light. This suggests that light activation of mutant rhodopsin is a key

causative agent in rhodopsin-mediated RP (Naash et al., 1993; Naash et al., 1996; Goto et

al., 1995; Goto et al., 1996). The exacerbation of RP symptoms by intense light exposure

is a well-documented phenomenon that is crucial to this work and will be discussed in

more detail in future chapters. Further work with the VPP mouse line involving

immunohistochemical tracking of the mutant rhodopsin molecules revealed that the

mutant opsin was correctly synthesized and localized, that there was normal outer









segment disc shedding, but that there were defective and disorganized basal discs at the

connecting cilium, the site of outer segment disc morphogenesis, providing further

evidence of a trafficking/disc morphogenesis defect associated with some rhodopsin

mutations (Wu et al., 1998; Liu et al., 1997).

An interesting and very useful model of ADRP was created in 1997 via a targeted

disruption of the entire rhodopsin gene. This model contained a Pol2: neomycin insertion

in exon II of the endogenous mouse rhodopsin gene. Subsequent mouse lines were bred

to contain either one disrupted rhodopsin gene hemizygouss null, or mRho+/- mice), or

two disrupted rhodopsin genes homozygouss null, or mRho-/- mice). Hemizygous null

mice showed some subcellular disorganization of the ROS, as well as shortening of the

outer segments in older mice when compared to wild-type mice, but little ERG reduction

or other signs of disease. Homozygous null mice, however, which completely lacked

expression of mouse rhodopsin, never formed rod outer segments, showed no ERG

response at eight weeks of age, and showed loss of cone photoreceptors by three months

of age (Humphries et al., 1997). These mice, along with a separate rhodopsin knockout

model that was subsequently developed (Lem et al., 1999), were important for several

reasons. Chief among these was their usefulness at providing a genetic background that

allowed researchers to breed hemizygous null (mrho+/-) rhodopsin lines upon which they

could express rhodopsin transgenes at an allelic ratio identical to that seen in human

patients with ADRP (i.e., one mutant copy and one wild-type copy).

Further mouse lines expressing rhodopsin mutations were created in the 1990s,

including one containing a Q344Ter mutation which caused synthesis of a rhodopsin

protein with an abnormally short carboxyl terminus. This mutant line exhibited









rhodopsin accumulation in the plasma membranes of rod photoreceptor cells,

demonstrating, as mentioned before, that the carboxy terminal of rhodopsin is necessary

for efficient and proper trafficking of rhodopsin (Sung et al., 1994). In 1998, Li et al.

reported creation of mouse models of ADRP expressing human rhodopsin transgenes

with either a T17M mutation (a class I mutation that abolishes the glycosylation site at

Argl5), or a P347S mutation (a class II mutation). Both lines exhibited progressive loss

of ERG response and decreasing thickness of the ONL. Interestingly, treatment with

Vitamin A supplementation led to partial rescue of the T17M-medaited RP, but not of the

P347S animals, providing further evidence that the separate classes of rhodopsin

mutations lead to retinitis pigmentosa via different biochemical or morphological

pathways (Li et al., 1998).

Gene Therapy for Retinitis Pigmentosa

It has long been the goal of medical scientists to effect therapies that act in targeted

cells at the level of the gene. In cases where disease symptoms are created by missing or

non-functional gene products, gene replacement strategies could be used to introduce

healthy gene products into tissues of interest. In cases where there is a toxic gain of

function caused by a mutant protein, therapeutic techniques designed to specifically

abolish expression of the mutant gene encoding that protein would be warranted. Both

types of gene therapy have been successfully demonstrated in animal models of retinitis

pigmentosa (Hauswirth and Lewin, 2000; Hauswirth et al., 2004).

The first efforts at gene replacement involved creating transgenic mice expressing

the corrective gene and crossing these with mice showing a retinal defect. In the case of

the naturally occurring rd mice, which undergo retinal degeneration due to a defect in the

beta subunit of the rod cGMP phosphodiesterase gene (PPDE), the introduction of a









functional human gene replacement through this method restored normal photoreceptor

morphology and function (Lem et al., 1992). In another study, these transgenic

techniques were used to express a human rhodopsin transgene on the homozygous null

rhodopsin (mRho-/-) mouse model of retinal degeneration. Expression of the functional

human rhodopsin in this case resulted in rescue of photoreceptor ultrastructure and ERG

response, and demonstrated the ability of the group's targeting construct to express the

transgene at therapeutic levels (McNally et al., 1999).

Following up on the work of Lem and colleagues with the rd mouse, two groups

have reported rescue of the photoreceptors through viral-mediated delivery of functional

PPDE genes. In 1996, researchers used replication-deficient adenoviral vectors to deliver

a murine cDNA expressing the wild-type 3PDE gene into the subretinal space of rd mice

(Bennett et al., 1996). This therapy resulted in expression of functional 3PDE that

resulted in a six week delay in photoreceptor degeneration. A similar strategy that

utilized intravitreal injection an adeno-associated viral vector delivering a wild-type

PPDE gene resulted in increased survival of photoreceptors and an increased ERG

response in rd mice receiving the therapy (Jomary et al., 1997). Gene replacement has

also been effective in treating the retinal degeneration seen in the naturally occurring rds

mouse, which as mentioned before suffers from defects in the ROS structural protein,

peripherin. Subretinal injection an adeno-associated viral vector engineered to express a

wild-type peripherin gene resulted in restoration of ROS ultrastructure and function in

these mice though the result was temporary (Ali et al., 2000). Long term rescue was also

seen with gene replacement designed to deliver a functional RPE65 gene to the retinas of









the rdl2 naturally occurring mouse model of the recessive retinal disease, Leber

Congenital Amaurosis (Pang et al., 2005).

Gene therapy designed to alleviate dominant retinal disease by suppressing

expression of mutant genes has also shown efficacy in mouse models. One such study

involved a rat model of retinal disease that was engineered to express a P23H mutant

rhodopsin transgene. Subretinal delivery of adeno-associated viral vectors engineered to

deliver catalytic RNAs, or ribozymes, designed to selectively degrade the mutant

transgene, while sparing the endogenous wild-type rat rhodopsin, was shown to result in

substantial, long-term rescue of photoreceptor structure and function (Lewin et al., 1998) .

This rescue, which was shown to be effective to eight months age in these animals, as

well as in treating animals that had already entered late stages of the disease (LaVail et

al., 2000), shows promise for the treatment of other forms of autosomal dominant

disease.

Ribozymes

Riboymes are RNA molecules with the ability to catalyze the cleavage and joining

of RNA. They were initially discovered by Altman and Cech, who described the

catalytic activity of the RNA component of RNaseP and the group I introns, respectively

(Cech, 1988a; Cech, 1988b; Guerrier-Takada et al., 1983; Guerrier-Takada and Altman,

1984). Initial Group I intron catalysis was seen to occur in cis, but it was soon

discovered that one could liberate the catalytic structure of the RNA from its substrate to

generate a ribozyme with the ability to cleave target molecules in trans.

Shortly after the work of Altman and Cech, smaller catalytic structures were

discovered in the sequences of certain plant pathogens which undergo site-specific, self-

catalyzed RNA cleavage as a part of their replicative process (Buzayan et al., 1988;









Haseloff and Gerlach, 1988). These are the hammerhead and hairpin ribozymes, which

are receiving much of the present attention for applications in gene therapy (Cech, 1988b;

Sigurdsson and Eckstein, 1995; Phylactou et al., 1998; Citti and Rainaldi, 2005).

Hammerhead ribozymes consist of three base paired stems surrounding a central catalytic

core of fifteen conserved nucleotides, eleven of which are necessary for catalytic activity.

The crystal structure of the hammerhead ribozyme reveals noncanonical base pairing

within the catalytic core and a magnesium binding site that is distal to the site of catalysis

(Scott et al., 1995). The helices are designated I, II, and III; the first and last helices (I

and III) form via base-pairing with the target molecule while the middle helix (II) is

responsible for stabilizing the catalytic core structure (Figure 1-4) (Sigurdsson and

Eckstein, 1995; Pierce and Ruffner, 1998). Hairpin ribozymes have two 5' stretches of

sequence which base pair with their RNA targets, forming helices I and II, followed by

two downstream helices (III and IV), which interact with one another and the substrate to

comprise the catalytic core (Figure 1-5) (Joseph et al., 1993; Berzal-Herranz et al., 1993).

The hairpin ribozyme has been crystallized either as a protein complex or as a pure RNA,

revealing internal base pairing within and essential interactions between the loop regions

of the secondary structure (Rupert and Ferre-D'Amare, 2001; Grum-Tokars et al., 2003).

No magnesium is seen at the catalytic site.

Mechanistic Description

Ribozymes achieve their cleavage by binding to their RNA target via

complementary base pairing with sequences flanking the cleavage site, folding into a

specific catalytic conformation, catalyzing the hydrolysis of the 5'3' phosphodiester bond

at the cleavage site, and dissociating from the resultant cleavage products (Figure 1-6).

The reaction generates two RNA fragments with either 5' hydroxyl or 2'3' cyclic









phosphate moieties. Divalent cations, such as magnesium, are thought to aid in the

folding of the ribozyme into the proper catalytic structure.

These cleavage reactions are sequence specific. The sequences flanking the

catalytic core of the ribozyme must be able to form helices with sequences flanking the

target cleavage site, and hydrolysis will only occur at a site containing certain

combinations of nucleotides. Hammerhead ribozymes will cleave 3' of a triplet sequence

of NUX, where N is any nucleotide, U is a uridine, and X is any nucleotide but a



cleavage


5'- NNNNNNNUXNNNNNNN -3'
3'- N N N N N NNA N N N N N NN -5'

Helix III A UG Helix I

A A
G AGU
N-N
N-N
N-N
N-N
N N Helix II
NN

Figure 1-6. Generic structure of a hammerhead ribozyme. The requisite NUX cleavage
triplet is shown in red, with the position of cleavage indicated by an arrow.
Green nucleotides indicate conserved ribozyme sequences necessary for
catalysis.











I u A. II
c a
3'- g c u u c a g g c u -5'
5'- C GAAGT CC GA
A A A-U -3'
G A -G
C-G
A-U
G-C
III c C
G A
A u
U
A
A U
A
C U
A G
C-G
A-U
C-G
G-C
IV c-G
U-A
C-G
G A
U A

Figure 1-7. Structure of a specific hairpin ribozyme. Cleavage is indicated by an arrow.
Blue nucleotides indicate conserved ribozyme sequences necessary for
catalysis

guanosine. The hairpin ribozymes show greater sequence constraint, and can be

engineered to cleave 5' of the G in the sequence 5'-NBNGUC, where N is any nucleotide

and B is G, C, or U. This theoretically allows a wide range of cleavage possibilities for

therapeutic applications of ribozymes.

In addition, recent experiments using large combinatorial libraries of ribozyme

candidate molecules have shown that it is possible to achieve cleavage of almost any

RNA target (Nieuwlandt, 2000; Piganeau et al., 2001). Most importantly, targets that

deviate from a ribozyme's optimal cleavage sequence, especially in the area of the

catalytic site, undergo reduced cleavage or none at all; even a single nucleotide difference

in critical areas is enough to completely abolish cleavage activity (Werner and









Uhlenbeck, 1995). This allows researchers to design ribozymes that are able to cleave

mutant RNA transcripts while leaving wild-type messages able to integrate into the target

cell genome, which makes them attractive as therapeutic effectors for the treatment of

autosomal dominant disease.




mRNA



Sannealing
Hammerhead
Ribozyme \


cleavage/






J J




Figure 1-8. Ribozyme cleavage in trans. A hammerhead ribozyme (black) is shown
annealing to its target mRNA (red) through complementary base-pairing,
cleaving it, and releasing the 5' and 3' cleavage products. Ribozymes are
truly catalytic, able to catalyze the cleavage of many successive target
molecules. Figure courtesy of Dr. Lynn C. Shaw.

AAV

Recombinant adeno-associated viral vectors offer a number of features ideal for

gene therapy, including the ability to infect a wide variety of both non-dividing and

dividing cell types (Loiler et al., 2003; Flotte, 2005; Fisher et al., 1997), the lack of

pathogenicity absence of immune or inflammatory response in transduced cells (Bennett,

2003; Hernandez et al., 1999; Conrad et al., 1996), and the ability to achieve long-term









expression of therapeutic genes (Guy et al., 1999). Adeno-associated viruses (AAV) are

human parvoviruses composed of a 4.7 kb single-stranded linear DNA genome packaged

in a capsid composed of three structural proteins. The genome is a model of efficiency,

consisting of overlapping open reading frames (orfs) that utilize alternative splicing and

variation in translation initiation sites to express several proteins from a relatively small

genome. The genes encoded by the orfs (and their gene products) are termed Rep (for

replication-associated proteins) and Cap (for capsid-associated proteins). The Rep genes

produce four protein products, termed Rep78, Rep68, Rep52, and Rep40, with the

numbers indicating the respective sizes in kilodaltons of each protein. The Cap genes

produce three capsid-forming gene products of 87, 73, and 62 kilodaltons in size, termed

VP1, VP2 and VP3, respectively, which are present in the viral capsid at a ratio of 1:1:10

(VP1 :VP2:VP3). These genes are present in the genome between two inverted terminal

repeats (TRs) consisting of 145 bases, which contain the viral origin of replication and

are necessary for viral packaging and integration. The entire genome is packaged within

a nonenveloped, icosahedral capsid that is around 20nm in diameter.

AAV belong to the class of viruses known as dependoviruses, because they require

co-infection of a helper virus to produce productive infection of their own. In the latent

phase of the virus it is found preferentially integrated into a particular site on human

chromosome 19 (Kotin et al., 1990). In this phase, the products of the Rep gene have

actually been found to be inhibitory to AAV replication. The latent phase of infection

will persist until the infected cell is rendered permissive, which can result from certain

cytotoxic exposures (including heat shock, irradiation, cycloheximide treatment), or until

the cell receives a secondary helper virus infection, usually with adenovirus, herpesvirus,









cytomegalovirus, or poxvirus. This co-infection is accompanied by the expression of

helper gene products from the new virus that help AAV achieve productive, lytic

infection. In the case of secondary adenoviral infection, expression of the early set of

genes (Ela, Elb, E2a, and E4) have been shown to provide helper functions such as

transcriptional transactivation and assistance in AAV mRNA accumulation and transport.

The ability of AAV to achieve stable, long term integration into target genomes,

coupled with a lack of association with insertional oncogenesis, led to the initial interest

in its use as a gene therapy vector. It has since become understood that integration of the

virus involves interactions between the chromosomal target site, the AAV TR structures,

and the Rep gene products (McLaughlin et al., 1988; Kotin et al., 1990; Surosky et al.,

1997; Weitzman et al., 1994). Since the recombinant AAV used for gene therapy lack

the Rep gene, integration seen with these vectors is random when it is seen at all. This

finding is of concern, as it raises the possibility of insertional activation of a proto-

oncogene, but these concerns are somewhat alleviated by a lack of reproducible

tumorigenesis in animals treated with rAAV therapy vectors (Donsante et al., 2001) and

by the fact that the rAAV genomes are generally thought to persist within the target cells

in an episomal fashion (Schnepp et al., 2005).

Production of AAV Vectors

To make recombinant AAV vectors, therapeutic sequences are inserted into a

packaging plasmid containing a cloning site and regulatory sequences flanked by the

AAV TR sequences that are necessary and sufficient for packaging as recombinant AAV

(McLaughlin et al., 1988). This construct is then co-transfected into HK 293 cells along

with a helper plasmid that encodes both the Rep and Cap AAV genes and the adenovirus

helper gene products. By this method it is possible to generate replication deficient,









recombinant AAV containing up to 4.5 kb of therapeutic DNA sequence bounded by the

AAV TRs (Hermonat et al., 1997; Zhou and Muzyczka, 1998). Providing helper gene

products on a plasmid rather than by adenoviral co-infection eliminates the possibility of

adenoviral contamination in the final viral preparation. These recombinant AAV (rAAV)

vectors are able to infect target cells and, with the inclusion of strong cis-acting promoter

sequences, generate good levels of expression of the therapeutic gene products of interest

for extended periods of time. Continued advances in the production and purification of

rAAV, including the use of high-volume cell factories and efficient, high-throughput

column purification methods, have led to consistently high yields of highly purified viral

vectors (Kapturczak et al., 2001; Potter et al., 2002; Blouin et al., 2004). Drawbacks

associated with first-generation therapeutic rAAV vectors included a delay in the

expression of therapeutic genes for as much as two weeks to a month, and limited

transduction of certain cell types. Both issues have been greatly overcome to large extent

through the use of pseudotyped rAAV vectors.

Different serotypes of adeno-associated virus can show great diversity in the amino

acid composition of their capsid proteins (Grimm and Kay, 2003). Although the original

viral delivery vectors were developed from serotype 2, many more serotypes have been

found to exist(Gao et al., 2002; Gao et al., 2005). The capsid divergence exhibited by

these serotypes cause them to have profound differences in the efficiency and speed with

which they are able to transduce various cell types (Grimm and Kay, 2003; Auricchio et

al., 2001; Hildinger et al., 2001; Rabinowitz et al., 2002). By using the AAV type 2

packaging vector in combination with helper plasmids providing capsid proteins from the

various other serotypes it is possible to create rAAV pseudotype vectors containing the









gene of interest and its ancillary sequences flanked by AAV2 TRs and packaged into

capsids derived from one of the other serotypes (Grimm, 2002; Burger et al., 2004; Choi

et al., 2005; Gao et al., 2005). Vectors created in this manner are termed pseudotype 2/*,

where the first number indicates that the packaging vector was derived from AAV

serotype 2 while the asterisk represents the serotype number of the capsid proteins (thus,

serotype 2 sequences packaged into a capsid derived from AAV serotype 4 would be

referred to as pseudotype 2/4). Table 1-1 summarizes the cell tropisms of the various

coat protein serotypes.

Table 1-1. Tissue tropism and site of isolation of the various AAV serotypes. Bold type
indicates high levels of expression, while bold italic type denotes highest
expression of all AAV serotypes. Adapted from Hildinger and Aurricchio,
2004.

Serotype Isolated In Tissue/Cell Tropsim

AAV1 Cell line Muscle, eye, liver, lung

AAV2 Cell line Muscle, brain, liver, eye

AAV3 Cell line Not determined

AAV4 Cell line Brain

AAV5 Human lesion Brain, muscle, liver, lung,
eye
AAV6 Cell line Muscle, eye, liver, lung

AAV7 Monkey Muscle, liver

AAV8 Monkey Liver


The usefulness of pseudotyped delivery vectors becomes apparent when examining

the different transduction efficiencies of AAV 2/1, AAV 2/2 and AAV 2/5 when used to

deliver a green fluorescent marker protein (GFP) to the retina. GFP expression is seen in

both the retinal pigmented epithelium and in photoreceptor cells of retinas receiving

subretinal injection of either AAV 2/2 or AAV 2/5. However, the use of AAV 2/5









resulted in a 400-fold increase in the number of transgene-expressing cells, as compared

to transduction by AAV 2/2, and the number of viral genome copies per eye was thirty

times higher. Additionally, AAV2/5 showed a faster onset of transgene expression, and

was shown to achieve higher levels of transgene expression. AAV2/1, in contrast,

transduces cells of the retinal pigmented epithelium almost exclusively, being fifteen

times more efficient than AAV2 and achieving higher levels of transgene expression

(Yang et al., 2002). Clearly the selection of the proper rAAV pseudotype is of great

importance in designing rAAV-mediated gene therapies (Dinculescu et al., 2005;

Auricchio and Rolling, 2005). Recent advances in this area have focused on directly

altering individual capsid epitopes to further enhance and refine the selective tropism of

these vector systems (Opie et al., 2003; Warrington, Jr. et al., 2004; Gigout et al., 2005;

Muzyczka and Warrington, Jr., 2005).

AAV and Retinal Gene Therapy

Adeno-associated vectors are ideally suited to deliver therapeutic DNA sequences

to retinal cells. These recombinant viruses are able to achieve long term expression in

retinal photoreceptor cells, retinal ganglion cells, and cells of the retinal pigmented

epithelium (Dinculescu et al., 2005; Guy et al., 1999; Flannery et al., 1997).

Transduction of retinal cells by rAAV is achieved with little or no toxic or immunogenic

side-effects (Bennett, 2003). Finally, the use of the various adeno-associated viral

pseudotypes with their selective tropism allows researchers to selectively target specific

retinal cell types for transduction.

Additionally, there are many characteristics of the retina itself that make it

attractive for rAAV-mediated gene therapy. Injections into the subretinal space can be

routinely performed with great speed and precision, and when properly performed lead to









little or damage of the injected tissues (Timmers et al., 2001). Also, the eye itself is held

to a certain extent to be an immune privileged site, a characteristic which enhances the

lack of an immune response to rAAV and the therapeutic proteins or sequences that it

delivers (Streilein et al., 1992; Sonoda and Streilein, 1992). Finally, the presence of two

eyes in animal models ensures that there is always a built-in control available to the

researcher in every experimental animal, as one eye can simply remain uninjected and

then be compared to the contralateral, treated eye.

As mentioned before, rAAV gene delivery vectors have been used to treat wide

variety of retinal disease, including the rd, rds, rdl2 mouse models of retinal

degeneration, and mouse and rat models of P23H mutant rhodopsin-mediated ADRP.

These vectors have also been successful components of therapies designed to treat the

lysosomal storage defects in the retinal pigmented epithelium seen in the MPVII (Bosch

et al., 2000; Hennig et al., 2004) and MP IIIB (Fu et al., 2002) mice, MerTK deficiency

in the Royal College of Surgeons (RCS) rat (Smith et al., 2003), retinal degeneration in

the naturally occurring RPE65-/- Briard dog analogue of the rdl2 mouse (Acland et al.,

2005; Acland et al., 2001), and retinal degeneration in a mouse model of X-linked

juvenile retinoschisis (Min et al., 2005). rAAV-delivered neurotrophic factors such as

GDNF (Wu et al., 2004), CNTF (Adamus et al., 2003; Liang et al., 2001), and FGF (Lau

et al., 2000; Lau and Flannery, 2003) have also been used to alleviate retinal degeneration

in various animal models of retinal disease. Finally, these vectors have been used to

deliver anti-angiogenic molecules such as PEDF and angiostatin for the treatment of

ocular neovascularization in animal models of diabetic retinopathy, retinopathy of

prmaturity, and the wet form of age-related macular degeneration (Raisler et al., 2002;









Auricchio et al., 2002). Although rAAV has been shown to be effective at delivering

therapy to a wide variety of organs and tissues, it is clearly the vector of choice for

transduction of retinal cells.

Project

The following chapters describe experiments designed to characterize and treat a

mouse model of retinitis pigmentosa that expresses a human rhodopsin transgene with a

tyrosine to methionine mutation at the 17th amino acid of the protein. I will discuss

breeding designed to express this transgene on a hemizygous null mouse rhodopsin

background to more closely imitate the genotype found in human RP patients. The

progression of retinitis pigmentosa in this mouse line will be documented, as will the

development of a ribozyme-mediated gene therapy strategy to treat the mice. Finally I

will discuss the results of a pilot study involving subretinal injection of rAAV expressing

these therapeutic ribozymes that led to some interesting observations concerning the light

sensitivity of the hT17M transgenic line, and strategies to circumvent this issue.














CHAPTER 2
CHARACTERIZATION OF A NOVEL MOUSE MODEL OF RETINITIS
PIGMENTOSA

Introduction

Human patients with the T17M rhodopsin mutation exhibit classic symptoms of

autosomal dominant retinitis pigmentosa. Affected individuals report loss of peripheral

and vision and night blindness, accompanied by decreased ERG response and eventually

culminating in loss of central vision. Postmortem examination of eyes from affected

patients reveals heavy deposits of bone spicule-like pigmentation in the inferior retina

that are accompanied by severe loss of photoreceptors (Li et al., 1994). Interestingly,

photoreceptors of the superior retina are relatively well-preserved, a feature that is

uncommon in RP patients. These mutants are intriguing in that they reduce or abolish

glycosylation at position 15 of the gene (Kaushal et al., 1994). As inhibition of

glycosylation by tunicamycin causes defects in ROS morphology in frogs (Fliesler et al.,

1985), this raises the possibility that T17M rhodopsin may not be correctly incorporated

into the outer segment discs.

In 1998, Li et al. described the creation of a mouse model of T17M rhodopsin-

mediated ADRP. These transgenic mice were created using a 17 kilobase human

geneome fragment that included the rhodopsin gene flanked by 4.8 kilobases of upstream

and 6.2 kilobases of downstream sequence. The gene contained a single nucleotide

substitution to change codon 17 from threonine to methionine. ERG analysis of these

mutant animals demonstrated that they underwent complete photoreceptor degeneration









by eight months of age. Subsequent studies with this animal model demonstrated a

therapeutic effect of vitamin A supplementation, which reduced the rate of decline in a-

and b-wave ERG amplitudes. This partial rescue was accompanied by an increase in

photoreceptor survival in the outer nuclear layer. Parallel experiments involving a mouse

line containing a P347S rhodopsin mutation did not show vitamin A rescue, providing

evidence for different paths of retinal degeneration between the two mutants (Li et al.,

1998).

We obtained this line from Dr. Li with the goal of developing a ribozyme therapy

for T17M-mediated ADRP. It was impossible to create mice containing two copies of

the human transgene (the cross led to embryonic lethality), with the result that mice

containing a 1:1 genotypic ratio of mutant to wild-type rhodopsin could not be created on

a wild-type rhodopsin background. In order to establish a mouse model that underwent a

more rapid degeneration, as well as one in which the genotype of the model more closely

mimicked the naturally occurring disease phenotype (i.e. one mutant allele and one wild-

type allele), we decided to breed our hT17M animals onto a mouse rhodopsin knockout

(mrho-/-) background (Lem et al., 1999). The resultant animals, which would contain

one copy of mutant hT17M human rhodopsin transgene and no copies of wild-type

mouse rhodopsin, could then be crossed to wild-type (mrho+/+) mice to produce animals

that are heterozygous null at the mouse rhodopsin locus (mrho+/-), of which half would

also contain the hT17M mutant rhodopsin transgene.

In this chapter I describe the creation and analysis of an mrho+/-; hT17M mouse

model of retinal disease. I will discuss the breeding and PCR genotyping of these









animals, and will document the visual degeneration in this line as assayed by ERG,

funduscopy, and histological examination.

Materials and Methods

DNA Oligonucleotides

DNA oligonucleotides were ordered from Invitrogen (Palo Alto, CA), at a 50

nmole scale of synthesis. Oligonucleotides were desalted, but otherwise unpurified by

the manufacturer. The sequences were as follows:

hExon 2 sense primer: 5'-GAGTGCACCCTCCTTAGGCA-3'

hExon 2 antisense primer: 5'-TCCTGACTGGAGGACCCTAC-3'

mRHO Exon 1 sense primer: 5'-CCAAGCAGCCTTGGTCTCTGTCTA-3'

mRHO Exon 1 antisense primer: 5'-TGTGCGCAGCTTCTTGTGGCT-3'

Neo sense primer: 5'-AGGATCTCCTGTCATCTCACCTTGCTCCTG-3'

Neo antisense primer: 5'-AAGAACTCGTCAAGAAGGCGATAGAAGGCG-3'



Isolation of Genomic DNA

Genomic DNA was isolated from 0.5 cm tail snips from candidate mice using the

Quiagen DNeasy Kit (Quaigen Inc, Valencia, CA), as per the manufacturer's instructions.

In brief, the tails were digested overnight at 500C in digestion buffer supplemented with

Proteinase K. The resultant suspension was diluted, bound to the DNeasy column,

washed twice, and eluted twice with 100 microliters of elution buffer, for a final volume

of 200 microliters of genomic DNA suspension. Three microliters of this genomic DNA

were used for PCR analysis, the rest was stored at either -20 or 4C.









PCR Analysis of Genomic DNA

Endogenous mouse rhodopsin genotypes were determined using mRHO Exon 1

and Neo primers. Genomic DNA from mice that are wild-type at the mouse rhodopsin

locus (mrho+/+) produce products with the mRHO Exon 1 primers, which amplify a 270

bp fragment, while the Neo primers will produce no product, as there is no Neo knock-

out cassette to amplify.








490bp
270bp
ZQbp



A B C D E F G H I J

Figure 2-1. Agarose gel electrophoresis of PCR reactions performed to determine the
mouse rhodopsin genotype (left gel) or confirm the presence or absence of the
hT17M transgene (right gel). Lane B shows amplification of the 490 bp Neo
and 270 bp mRho PCR products from a mRho+/- mouse. Lanes D and J show
amplification of the 290 bp human rhodopsin PCR product. Lanes E I are
negative for human rhodopsin. Lanes A and C are size standards.

Conversely, DNA from mice that are homozygous for a knockout at the mouse

rhodopsin locus (mrho-/-) will produce a 490 bp fragment with the Neo primers, while

producing no product when amplified with the mRHO Exon 1 primers. DNA from mice

that are heterozygous null at the mouse rhodopsin locus (mrho+/-) are able to produce

both fragments. Presence of the human T17M rhodopsin transgene was determined by

PCR amplification of genomic DNA with the hExon 2 primers, which amplify a 290 bp

fragment in the presence of the transgene (Figure 2-1).









PCR reactions were set up in 50 [tl volumes as follows: 3[tl (- ~1g) genomic

DNA, 5[tl 10X PCR Buffer (Sigma, St. Louis, MO), 0.5 [tl 100mM dNTP mix (Sigma),

0.25[tl 100mM sense primer, 0.25[tl 100mM antisense primer, 0.5[tl (1 Unit) Taq

Polymerase (Promega, San Luis, CA), and 40.5 ml dH20. PCR was then carried out as

follows: 1) 95C for 10 minutes; 2) 95 C for 45 seconds; 3) 54 C for 45 seconds; 4) 72

C for 60 seconds; 5) repeat steps 2-4 for 28 cycles; 6) 72 C for 10 minutes. PCR

amplifications were performed on a Gene Amp PCR System 2700 thermocycler (Applied

Biosystems, Foster City, CA). The presence or absence of the PCR product of interest

was verified by agarose gel electrophoresis.

Electroretinography

Mice were dark adapted overnight. All subsequent ERG procedures were

performed under dim red light (wavelength >600nm), which does not activate rhodopsin.

Mice were anesthetized with IP injections of xylazine (13mg/kg) and ketamine

(87mg/kg) (Phoenix Pharmaceuticals, St. Joseph, MO). The mouse corneas were

anesthetized with a drop of 0.5% proparacaine HC1 (Akom, Buffalo Grove, IL), and

dilated with a drop of 2.5% phenylephrine HC1 (Akorn). Measurement electrodes tipped

with gold wire loops were placed upon both corneas with a drop of 2.5% hypromellose

(Akom) to maintain electrode contact and corneal hydration. A reference electrode was

placed subcutaneously in the center of the lower scalp of the mouse, and a ground

electrode was placed subcutaneously in the hind leg. The mouse rested on a homemade

sliding platform that kept the animal at a constant temperature of 370 C. The animal was

positioned so that its entire head rested inside of the Ganzfeld (full-field) illumination

dome of a UTAS-E 2000 Visual Electrodiagnostic System (LKC Technologies, Inc.,









Gaithersburg, MD), as shown in Figure 2-2. Full-field scotopic ERGs were measured by

10 msec flashes at an intensity of 0.9 and and 1.9 log cd m-2 at 1 minute intervals.


Figure 2-2. Electroretinographic apparatus. The anesthetized animal rests on a warming
tray that can slide in and out of a Ganzfeld illumination dome. Electrodes are
held in place with homemade articulable plastic arms.


b


Figure 2-3. An example of an ERG tracing. The X axis of this trace represents the
elapsed time of the signal, while the Y axis shows the intensity of the
response. The amplitudes used to calculate a and b-waves are shown .









Responses were amplified at a gain of 4,000, filtered between 0.3 to 500Hz and digitized

at a rate of 2,000 Hz on two channels. Five responses were averaged at each intensity.

The wave traces analyzed using UTAS-E 2000 software package (LKC Technologies,

Inc.). A-waves were measured from the baseline to the peak in the cornea-negative

direction; b-waves were measured from the cornea-negative peak to the major cornea-

positive peak (Figure 2-3).

Funduscopy

Mice were anesthetized, and their corneas anesthetized and dilated as described

above. Fundus photography was performed by Dr. Quihong Li with a Kowa Genesis

hand held funds camera (Kowa Company, Ltd., Tokyo, Japan) focused through a Volk

Super 66 Stereo Fundus Lens (Keeler, Berkshire, England). Two pictures of each eye

were generally taken to ensure a properly focused image.

Histology

Mice were euthanized by an overdose of Isoflurane (Abbot, North Chicago, IL)

followed by cervical dislocation. Eyes were quickly removed and fixed overnight in 4%

paraformaldehyde and transferred to a solution of phosbate buffered saline (PBS)

(137mM NaC1, 10mM PO4, 2.7Mm KC1, pH 7.4). Histological sectioning and

subsequent H&E staining was performed by UF Histology core technicians. This

program of sectioning resulted in twelve serial sections through the entire eye. Sections

that contained the optic nerve were then photographed at 20X power using a Zeiss

Axiophot Z microscope equipped with a Sony DXC-970MD 3CCD Color Vid Camera

and an MCID Elite Stage, utilizing MCID (Imaging Research, Inc., Ontario, Canada)

Analysis Software (Imaging Research, Inc.) that stitched individual images together to

create a tile-field composite image of the entire retina. The images were viewed with









Adobe Photoshop, and a radial template overlay was used to define six equivalent and

equally spaced regions of the retina. From each of these areas, the mean value from three

separate ONL counts was determined, and these regional counts were then averaged to

generate a value that represented the ONL thickness of the retina. Statistical comparisons

between the transgenic and non-transgenic values were performed to generate P values

using the paired, one-tailed Student's t-test feature of Exel spreadsheet software

(Microsoft, Redmond, WA).

Results

Breeding Founder and Experimental Mice

A breeding pair of mrho+/- mice containing the hT17M transgene was kindly

provided by Dr. Tiansen Li in 2002. However, this line had subsequently undergone a

mating crisis due to poor fecundity and had become poorly characterized by the time

these studies were initiated. It was uncertain which members of the colony actually

contained the hT17M transgene, and of what respective mouse rhodopsin genotype those

animals were. It was necessary to screen the entire colony by PCR for the hT17M

transgene and then work from there. Several mice were found that contained the hT17M

transgene; these animals were bred to mrho-/- mice, and their progeny were screened by

PCR for the hT17M transgene, as well as for their respective genotype at the mouse

rhodopsin locus. Eventually, mice were obtained that were heterozygous null for mouse

rhodopsin (mrho-/-) and also contained the hT17M mutant rhodopsin transgene. Two

mrho-/-; hT17M males were crossed to wild-type C57B16 female mice to breed the

mrho+/-; hT17M animals used in these experiments. Concurrently, mrho-/-; hT17M

females were crossed to mrho-/- males to maintain the hT17M transgene on a mouse

rhodopsin null background.









ERG Natural History

Once these lines were breeding reliably, an ERG natural history study was

conducted on the mrho+/-; hT17M mice and non-transgenic littermates. Two litters were

genotyped for this study, producing eight mrho+/- animals and eight mrho+/-; hT17M

siblings. These animals were subjected to ERG analysis every two weeks for the next six

and a half months. Right and left eye amplitudes were averaged for each animal, and the

averages of each group were plotted for each time point (Figures 2-4 and 2-5).

Additionally, the ratio of the mrho+/- response to the mrho+/-; hT17M response for each

time point was plotted (Figures 2-6 and 2-7).

There was a definite loss of both a- and b- wave responses over the life of the

hT17M transgenic animals compared to their non-hT17M littermates. The a-wave

response showed particularly early degradation, as the mutant response was only half that

of the non-mutant siblings at one month of age. Mutant a-wave responses remained at

around 50% of those of the non-transgenic animals for the next 2.5 months, when they

again underwent a significant drop relative to their non-mutant siblings, to around 30% of

non-mutant a-wave amplitude. The b-wave amplitudes of the transgenic animals were

more robust, starting out at around 70 to 80% percent of the non-transgenic littermates

and remaining so for the next 2.5 months, when they began to undergo steady a steady

decrease in amplitude at around the time that the a-waves exhibited their drop to 30%.

Funduscopy

In order to see if the degeneration indicated by the ERG natural history study

would be apparent upon funduscopic examination of the retina, two mrho+/- siblings, one

of which contained the hT17M transgene, underwent ERG analysis at 6 months of age.

The following day, funds photographs were taken to visualize the retinas of these











Mutant T17M and Non-Mutant Littermate AWave
Amplitudes Over Time


* Rho+/-
SRho+/-; T17M


,0 4, 4, 1, , ,
N C1, ,. .

Age of Animal



Figure 2-4. 10 dB intensity ERG a-wave natural history of mrho+/- and mrho+/-; hT17M
mice.


Mutant T17M and Non-mutant Littermate B Wave
Amplitudes Over Time


* Rho +/-
0 Rho +/-:T17M


Age of Animal



Figure 2-5. 10 dB intensity ERG b-wave natural history of mrho+/- and mrho+/-; hT17M
mice. ERGs were performed at the 10dB light intensity.


400
350
300
>
250
S200

E
< 100
50
0
I1 lil


900
800
- 700
Z 600
. 500
. 400
- 300
< 200
100
0










AWave Responses: Percentage of Mutant to Non-Mutant
Littermates

S100
90
80
E 70
60
z 50
., 40
o 30
20
I 10
ll ^. .
S 0-



Age of Animal


Figure 2-6. 10 dB intensity a-wave responses charted as a percentage of the average
mrho+/-; hT17M response to the average response of the mrho+/- siblings at
each time point.


B Wave Responses: Percentage of Mutant to Non-
Mutant Littermates

S 100
S 90
S 80
E 70
60
zo 50
"" 40
C 30
0 20
M 10
0



Age of Animal


Figure 2-7. 10 dB intensity b-wave responses charted as a percentage of the average
mrho+/-; hT17M response to the average response of the mrho+/- siblings at
each time point.






43














Co i



I .. E-



Time in Milliseconds Time in Milliseconds


Figure 2-8. Fundus photographs and respective 10 dB intensity ERG tracings taken at 6
months of age. Photos and tracings from a mrho+/- animal are shown on the
left (A), and those of its mrho+/-; hT 17M littermate are shown on the right
(B).

animals. The resultant images, matched with their respective ERG tracings, are shown in

Figure 2-8. The funds of the mrho+/- mouse looked normal, exhibiting relatively even

pigmentation and healthy-appearing retinal morphology, and ERG analysis of this mouse

showed robust a- and b-wave amplitudes. The hT17M sibling displayed a markedly

depressed ERG tracing, as well as shadowy spotting in its funds picture, indicating that

the retina of this mutant animal had thinned as its photoreceptors degenerated.



Histology

Two of the animals (hT17M mutant and non-mutant sibling) from the natural

history were sacrificed at 4 months of age to obtain an early look at the histological









results of the hT17M -mediated retinal degeneration. The rest of the mice were all

sacrificed at 6.5 months, as the a- and b-wave loss seemed to plateau and it was feared

that waiting longer would result in retinas that were too degraded to provide good

sections. Representative sections from these animals showed severe, progressive

degeneration of the outer nuclear layer (ONL) that is consistent with the loss of ERG

response, both of which are hallmarks of retinitis pigmentosa. (Figure 2-9).

























Figure 2-9. Representative sections of mrho+/- mice (A-D) and mrho+/-; hT17M siblings
(E-H). Sections A and E are from animals sacrificed at 4 months of age, all
other sections are from animals sacrificed at 6.5 months of age, at the
completion of the natural history study. All hT17M animals (E-H) show
severe degeneration of the outer nuclear layer (ONL). Arrows indicate ONL.

In order to obtain a more concrete idea of the extent of the ONL degeneration, tile

field mapped images of the hT17m and non-hT17M retinas were examined. A semi-

transparent template (Figure 2-10) was used to pinpoint eight areas of each retina from

which ONL thickness could be measured (in numbers of nuclei) and averaged. The mean









of the counts taken from the six points was determined, and this number represents the

average thickness of the ONL for that eye. For the animals sacrificed at 6.5 months of

age, the ONL values from both eyes were averaged together, while only one eye from


Figure 2-10. Tile-field mapped image of a mouse retina, with a translucent overlay that
defines eight evenly-spaced sections of the retina from which outer nuclei
counts can be obtained.


Table 2-1. ONL averages taken from mrho+/- and mrho+/-; hT17M mice sacrificed and
analyzed at 4 and 6.5 months of age. ** indicates significant difference
between mutant and non-mutant values with a P value of less than .001.

mRho+/- mice mrho+/-; hT17M mice

ONL Thickness at 4 months 8.8 Rows 5.9 Rows

ONL Thickness at 6.5 months 7.7 Rows** 2.4 Rows**


each animal at the 4 month time point was able to be analyzed. These results are

summarized on Table 2-1. By this measure, hT17M mice exhibited significant thinning









of the outer nuclear layer when compared to their non-hT17M littermates sacrificed at the

6.5 month time point. hT17M mice sacrificed and analyzed at the 4 month time point

also showed a reduction in the ONL thickness.

Discussion

The original mrho+/+; hT17M mouse line obtained from Dr. Tiansen Li provided a

useful animal model for the study of retinal disease. These animals showed progressive

photoreceptor loss that was accompanied by a loss of a- and b-wave ERG amplitudes,

and experiments involving treatment of this mouse with vitamin A supplementation

produced some compelling results. However, by the time we began these experiments,

the line had become uncharacterized for reasons already described. A goal of the

preceding experiments was to develop efficient methods for genotyping these mice and to

breed the remaining hT17M positive animals to a final mrho-/- background. We then

wanted to use these mrho-/-; hT17M mice both for maintenance of the transgenic line and

for creation of a mrho+/-; hT17M transgenic line that would be useful in developing

therapy for autosomal dominant retinal disease.

The PCR assays that we have developed to characterize these lines are simple and

accurate. Use of the Quiagen DNeasy kit allows a researcher to easily purify substantial

amounts of genomic DNA from small pieces of mouse tail in a short period of time. Two

separate PCR reactions allow for determination of the endogenous mouse rhodopsin

genotype (mrho+/+, mrho+/-, or mrho-/-), and to screen for the presence or absence of the

T17M human mutant rhodopsin transgene. The line is now maintained on both the mrho-

/- and mrho+/+ genetic background.

The retinal degeneration displayed by mrho+/-; hT17M transgenic line is an

excellent model of the vision loss observed in human patients suffering from ADRP. Bi-









monthly ERG recordings demonstrated progressive loss of rod photoreceptors in the

hT17M transgenic animals, culminating in a severe loss of visual response by six months

of age, while the non-transgenic littermates retained normal visual function. The loss of

ERG was accompanied by a progressive thinning of the ONL as determined from retinal

sections, with mrho+/-; hT17M mice displaying a loss of 40% of their ONL by 4 months

of age and a loss of 70% of their ONL by 6.5 months of age, as compared to their non-

transgenic littermates. Funduscopic analysis of the transgenic animals revealed marked

retinal thinning by 6 months of age, with non-transgenic littermates again displaying

normal retinal morphology.

It is unfortunate that it has so far been impossible to breed the hT17M transgene to

homozygosity. This is probably due to insertional inactivation of an unknown gene that

is embryonic lethal if not present in at least one copy. The lines have been outbred for

several years now since their arrival, and future attempts to breed the transgene to

homozygosity could well meet with success. At present, our breeder mice are limited to

one copy of the hT17M transgene, with the result being that, at most, around 75% hT17M

transgenic mice can be produced from a breeding of two hT17M transgenic mice. As our

breeders to date have consisted of one mrho-/-; hT17M transgenic mouse crossed with a

wild-type C57BL6 mate, we have been producing litters composed of around 50%

transgenic and 50% non-transgenic pups on an mrho+/- background. This is a drawback

in terms of requiring twice as many mouse pups to provide sufficient numbers of

transgenic animals for testing therapy for ADRP. However, this drawback has proven

useful in that most experimental litters contain several non-transgenic control mice. In









fact, these non-transgenic littermates were essential in discovering an hT17M rhodopsin-

mediated light sensitivity that will be discussed in future chapters.

Not all mouse models of retinal disease are practical for developing therapy. For

example, one of the mouse lines created by Olsson et al. to express a human rhodopsin

transgene containing the P23H mutation showed nearly complete photoreceptor

degeneration by 20 days of age (Olsson et al., 1992). This is too rapid a degeneration to

provide an effective test subject for all but the most rapidly acting therapies. On the other

hand, an animal model that undergoes degeneration too slowly is also problematic

because experiments utilizing the line can take a long time to provide useful data.

The hT17M human transgene expressed on an mrho+/+ background underwent a

progressive loss of photoreceptors culminating in total loss of vision by around eight

months of age. We decided to breed these animals to a mrho+/- background with the

goal of both creating a mouse line that contained one copy each of mutant human and

wild-type rhodopsin genes, and of developing a model that would undergo a more rapid

retinal degeneration while still providing a therapeutic window for treatment. The

resultant hT17M; mrho+/- animals did undergo a more rapid retinal degeneration, with

almost complete loss of functional rod photoreceptors by around 6.5 months of age.

Additionally, this line displayed a-wave and b-wave ERG responses of 100 [LV and 400

[LV, respectively, at 2.5 months of age. By 6.5 months of age these responses had

degraded to an a-wave response of less than 50 [LV and a b-wave response of less than

200 [LV. Studies have shown that AAV pseudotype 2/5 can achieve significant transgene

expression of mouse photoreceptor cells by twelve days following subretinal injection

(Yang et al., 2002; Rabinowitz et al., 2002). Assuming that our hT17M transgenic mice









receive treatment at weaning (21 days of age), the line should retain ample photoreceptor

function by the time one can expect expression of rAAV-delivered therapy designed to

prolong retinal function. This therapeutic window should also be wide enough to be

amenable to pharmacological treatment.

In conclusion, the hT17M transgene has been maintained and expressed on

mrho+/+, mrho+/-, and mrho-/- backgrounds. This line undergoes a progressive

photoreceptor degeneration that can be monitored by ERG measurements and that

correlates with thinning of the ONL as visualized by funduscopy and by retinal histology.

The degeneration of the mrho+/-; hT17M line is practically complete by 6.5 months; yet

these mice retain sufficient retinal function and photoreceptor survival at early ages to

make them amenable to therapeutic intervention.















CHAPTER 3
AAV-MEDIATED RIBOZYME TREATMENT OF MRHO+/-; hT17M MICE

Introduction

The hT17M mutant human rhodopsin transgene causes the autosomal dominant

form of retinitis pigmentosa, meaning that expression of the mutant allele is responsible

for the disease. One way to treat an autosomal dominant disorder is to selectively destroy

the mRNA encoding the mutant allele in theory this should abolish expression of the

mutant protein and rescue the disease (Hauswirth et al., 2000). Mice bred to be

hemizygous null at the rhodopsin allele (mrho+/-) show only slightly reduced ERG

responses when compared to mice that contain two wild type copies of the gene

(Humphries et al., 1997). Thus in our mrho+/-; hT17M line, removal of the mutant

human mRNA while leaving the wild-type mouse message intact should protect against

the vision loss associated with the mutant rhodopsin gene product.

Ribozymes are ideally suited for the treatment of autosomal dominant disease. As

discussed before, the sequence specificity of ribozyme cleavage is stringent enough that a

ribozyme can often be designed to discriminate between mutant and wild-type target

RNAs that differ by a single nucleotide. This makes it feasible to treat certain dominant

diseases arising from point mutations with ribozymes that selectively degrade the mutant

RNA, thus selectively reducing or abolishing expression of the mutant protein. It was

recently shown that the use of such a ribozyme that targeted the reduction of a P23H

mutant rhodopsin message resulted in rescue of vision in a rat model of ADRP (Lewin et









al., 1998; LaVail et al., 2000; Drenser et al., 1998). We wanted to explore the efficacy of

a similar ribozyme-mediated therapy for the threonine to methionine at position 17

(T17M) mutation carried in our rho+/-; hT17M mouse line.

To this end we designed ribozymes to selectively cleave human rhodopsin mRNAs.

In most forms of ribozyme gene therapy for autosomal dominant disease, the ribozymes

are designed to select between mutant and non-mutant messages because of sequence

differences at the site of the causative mutation. This means that the therapeutic

ribozyme must be specifically tailored to cleave the message at the mutation site, and

often this target site is less than ideal for ribozyme-mediated cleavage, if it is susceptible

to cleavage at all. In the hT17M transgenic mouse model, expression of a mutant human

gene on a mouse genetic background is responsible for the disease, so we were able to

design ribozymes that were targeted to ideal cleavage sites in the human rhodopsin gene

that differed from the endogenous, wild-type mouse sequence, rather than having to

design a ribozyme to a target site that was restricted to the site of the mutation. Using

this strategy we were able to create two highly active ribozymes that cleaved the human

rhodopsin message but should theoretically leave the mouse message intact.

Once an effective ribozyme has been created, the next step in developing a therapy

for retinal disease is to deliver the ribozymes to the pertinent retinal cells. For this

purpose we used recombinant adeno-associated virus (AAV) delivery vectors. As

previously discussed, AAV is extremely well suited for retinal gene transfer (Flannery et

al., 1997). For our experiments we decided to use AAV pseudotype 5, as this pseudotype

has demonstrated preferentially high levels oftransduction in photoreceptor cells, as well

as a rapid onset of transgene expression (Yang et al., 2002). To drive the expression of









our human rhodopsin-specific ribozymes we chose the MOPS500 promoter, which has

been shown to achieve impressive levels of transgene expression in the rod

photoreceptors of rats and mice (Flannery et al., 1997). The combination of AAV

pseudotype 5 vectors delivering ribozymes under the control of the MOPS500 promoter

helps ensure high levels of ribozyme expression that are specific to the rod

photoreceptors of treated animals.

In this chapter, I describe the creation and in vitro testing of two ribozymes

designed to specifically cleave the human rhodopsin transgene. I will discuss how these

ribozymes were cloned in specialized plasmids for packaging as recombinant AAV.

Finally, I will detail the subretinal injection technique and the results of delivering this

virus to the subretinal space of mrho+/-; hT17M transgenic animals.

Materials and Methods

RNA Oligonucleotides

RNA nucleotides were ordered from Dharmacon Research Inc. (Boulder, CO), at

the 50 t[molar scale. The sequences were as follows:

Rzl: 5'-CCGAACUGAUGAGCCGUUCGCGGCGAAACGAAG-3'

Rz3: 5' -GUGAACUGAUGAGCCGUUCGCGGCGAAACGAGC-3'

Targetl: 5'-CUUCGUCUUCGG-3'

Target 3: 5'-GCUCGUCUUCAC-3'

DNA Oligonucleotides

DNA oligonucleotides were obtained from Invitrogen at the 40 nmolar scale of

preparation. Oligonucleotides used for cloning had 5' phosphate groups chemically

added by the manufacturer, and were purified by desalting. Their sequences were as

follows:











Rzl Cloning Sense:

5'-AGCTTCCGAACTGATGAGCCGTTCGCGGCGAAACGAAGATGCA-3'

Rzl Cloning Antisense

5'-TCTTCGTTTCGCCGCGAACGGCTCATCAGTTCGGA-3'

Rz3 Cloning Sense

5'-AGCTTGTGAACTGATGAGCCGTTCGCGGCGAAACGAGCATGCA-3'

Rz3 Cloning Antisense

5'-TGCTCGTTTCGCCGCGAACGGCTCATCAGTTCACA-3'

Red nucleotides indicate restriction sites for HindIII (AAGCTT) and Nsil

(ATGCAT).

Preparation of Synthetic RNA Ribozymes and Substrates

All hammerhead ribozymes as well as the substrate RNAs were purchased from

Dharmacon Research, Inc. (Boulder, CO). The RNA oligonucleotides were chemically

synthesized with an acid-labile orthoester protecting group (to reduce ribonuclease

degradation) on the 2'-hydroxyl (2'-ACE) that must be deprotected by incubation at pH

3.8 at 60C according to the manufacturer's protocol prior to use. Deprotected RNA

oligos were suspended at a final concentration of 300 picomoles per microliter.

5' End-labeling of Deprotected Target RNAs

Prior to in vitro kinetic analysis, the target RNAs were 5' end-labeled with [y-32P]-

ATP (ICN, Irvine, CA) using T4 Polynucleotide Kinase (T4 PNK) (Promega, Madison,

WI). A typical reaction contained: 2 [tl target RNA oligo (20 picomoles total), 1 dtl of

10X PNK Buffer [700 mM Tris-HCl (pH 7.6 at 250C), 100 mM MgC12], 1 tl RNasin









(Promega), 1 [tl 0.1M DTT, 3 tl H20, 1 tl [y-32P]-ATP, and 1 tl T4 PNK. Reactions

were incubated at 370C for 30 minutes, and then 90 [tl of H20 was added. The mixture

was then incubated at 65 for 5 minutes to inactivate the T4 PNK. Two

Phenol:Chloroform:Isoamyl Alcohol extractions were then performed, and 90 [tl of the

aqueous phase was purified over a Sephadex G-25 Spin Column (Pharmacia, Piscataway,

NJ) to separate the labeled target molecule from unincorporated radionucleotides. The

resulting labeled target solution was at a final concentration of 0.2 picomoles per

microliter.

In Vitro Ribozyme Time Course Analysis

Separate ribozyme and target mixes were made. The ribozyme mix consisted of:

13 [l 400 mM Tris-HCl (pH 7.45), 1 [tl ribozyme (diluted to 2 pmol/tl), and 80 [tl H20.

The target mix consisted of 1 tl radiolabeled target RNA, (diluted to 0.2 pmol/[tl), 1 [l

cold target RNA (diluted to 20 pmol/|tl), and 8 [tl of H20. After these two mixes were

created, the ribozyme mix was heated to 65C for two minutes, and then cooled to room

temperature for at least ten minutes. 13 [tl of a 1:10 RNasin:0.1M DTT mix and 13 [tl of

50 mM MgCl2 (for a final reaction concentration of 5 mM Mg Cl2) were then added, and

the solution was equilibrated at 370C for ten minutes.

For the reaction, the 10 [tl target mix was added to the ribozyme mix, the solution

was mixed thoroughly by vigorous pipetting, and 10 [tl of the reaction mix was

immediately added to 10 ul formamide stop buffer (90% formamide, 50 mM EDTA,

0.05% bromophenol blue, and 0.05% xylene cyanol), and placed on ice (this was the 0

minute time point), and the remaining reaction mix was incubated at 370C. At

subsequent intervals of 1, 2, 4, 8, 16, 32, 64, and 128 minutes after the start of the









reaction 10 [tl of the reaction mix was removed and likewise added to 10 [tl formamide

stop buffer and placed on ice.

The reaction/formamide stop buffer mix from each time point was then heat

denatured at 95C for five to ten minutes, then placed on ice for five minutes, and then

separated on a 10% polyacrylamide / 8M urea gel. The gel was fixed in 2 L of fixation

solution (40% methanol, 10% acetic acid, 3% glycerol) for thirty minutes, dried, exposed

to radioanalytic phosphorescent screens, and analyzed using a Molecular Dynamics

Phospholmager system and ImageQuant software (Molecular Dynamics, Sunnyvale,

CA). The percentage of substrate cleaved in each sample was determined from the ratio

of radioactivity in the 5'-end labeled cleavage product (P) to the sum of the radioactivity

in the 5'-end labeled cleavage product and the substrate band (S): % Cleavage = P/P+S.

Using Excel (Microsoft, Redmond, WA) the percentage substrate cleaved was then

plotted as a function of time to generate a graphical representation of the cleavage time

course.

Ligating Ribozyme Sequences into rAAV Packaging Vectors

Complementary DNA oligonucleotides encoding the sense and antisense strands of

HRzl and HRz3 were ordered. In addition to the ribozyme sequence these

oligonucleotides contained sequences (shown above in red) appended to their 5' and 3'

ends so that when they annealed they formed the sticky overhangs corresponding to

HindIII at the 5' end and Nsil at the 3' ends. This allowed the oligonucleotides to be

ligated into the rAAV packaging vector, pXX-GS-HP-MOPS500 (Figure 3-1) at a

multiple cloning site containing a HindIII restriction site upstream of an Nsil restriction









site. The oligonucleotides came from the supplier with 5'-P04 groups already attached to

allow immediate ligation into the packaging vector.

To linearize pXX-GS-HP-MOPS 500, 5[tg of plasmid DNA was digested with

HindIII for three hours at 37C. The DNA was then ethanol precipitated, resuspended,

and digested with Nsil for three hours at 37C. After the two digestions were complete,

the resultant fragment was run on a 1% agarose gel, and the digested plasmid band was

visualized by ethidium bromide staining under UV illumination. The band was excised

from the gel and the DNA was purified using a "freeze squeeze" technique (Sugden et al.,

1975). In brief, the gel fragment was crushed and mixed in an equal volume of phenol in

a 1.5 mL Eppendorftube. The tube was then incubated at -80C for 2 hours. Next the

tube was spun at a speed of 13,000 rpm for 10 minutes, and the aqueous solution was

removed. The aqueous solution was then extracted with an equal volume of

phenol:chloroform:isoamyl alcohol, ethanol precipitated, and the resultant purified

plasmid pellet resuspended at a concentration of 0.5 [tg/[tl.

To ligate the ribozyme-encoding oligonucleotides with the linearized packaging

vector, the complementary oligonucleotides were mixed together for a final concentration

of 20 picomoles each in a volume of 4 microliters. To facilitate proper annealing, the

oligos were heated to 95C for five minutes, and then allowed to slowly cool to room

temperature. 0.5 |tg of linearized plasmid (1 [tl) was then added to the mixture along

with 5 [tl of 5X ligation buffer [250mM Tris (pH 7.5), 50 mM MgC12, 5mM ATP], 1 [tl

25 mM DTT, 3 [tl PEG 4000, 10 [tl H20, and 1 [tl T4 DNA ligase (Promega). Reactions

were incubated at 25C overnight, and then 2 [tl was transformed into 50 [tl of

electrocompetent E. coli by electroporation using a Bio-Rad Gene Pulser II









electroporation apparatus (Bio-Rad, Hercules, CA), utilizing 0.1 mm electroporation

cuvettes (USA Scientific, Ocala, FL). The E. coli was plated on LB plates containing

ampicillin, and the resultant transformants were picked, their DNA isolated, and

sequenced for the proper ribozyme insert.



RR MOPS 500 Promoter
SV40 SSA
Mrndll(8 17)
1 or IRz3

AnpRSi Nsl(86
IRES

cDNA
p>XX- HP-tz
81cr
SV40 pA)
ori PY I41 polmaenhancer


% GH pdIx
IIR


Figure 3-1. pXX-GS-HP MOPS 500 rAAV packaging plasmid. Human rhodopsin
specific ribozymes were cloned into the HindIII-NsiI restriction sites as
indicated.

Once each ribozyme was successfully cloned into pXX-GS-HP MOPS 500, 700ug

of DNA was produced from a 1 L E. coli culture, purified via cesium banding on an

ultracentrifuge, and finally packaged as recombinant AAV type 5 at the UF

Ophthalmology Packaging Core.









Subretinal Injection of rAAV Ribozyme Delivery Vectors

At weaning age (21-24 days) litters were removed from their parents and their right

eyes were dilated with 1% atropine sulfate solution (Bausch and Lomb, Tampa, FL). The

next morning, the right eyes were again dilated with 1% atropine sulfate, and again an

hour before the injection procedure, at which time the eyes also received a drop of 2.5%

phenylephrine HC1 and 0.5% proparacaine HC1. An hour after this final dilation, the

animals were anesthetized by ketamine/xylazine injection and again treated with a drop

each of 1% atropine sulfate, 2.5% phenylephrine HC1, and 0.5% proparacaine HC1. The

right eyes of these animals then received a drop of 2.5% hypromellose to aid in retinal

visualization and to help keep the retina hydrated. Injections were visualized with a

Nikon SM2800 (Nikon, Melville, NY) dissecting microscope, with illumination provided

by a Southern Micro Instruments 150 Watt fiber optic light source with Schott Fostec

fiber optic arms (Southern Micron Instruments, Marietta, GA) (Figure 3-2), which at full

power provided an intensity of illumination of around 10,000 lux. A hole was placed in

the inferior cornea of the eye with a 28 gauge needle. A blunt 32 gauge needle was then

inserted into the hole, the tip of the needle was rotated around the lens, and pushed

through the retina until it came to rest at the sclera, which could be visualized by the eye

sinking back into the socket. 0.5 [tl of rAAV suspension was then slowly delivered into

the subretinal space over a period of 20-30 seconds. This injection strategy is depicted in

Figure 3-3. VPP antibiotic ointment (Akorn) was placed upon both eyes to maintain

hydration and prevent infection in the injected eyes, and the animals were ear marked and

0.5 cm sections of tail tip were removed for genotyping as described previously. The

animals were then allowed to recover on a warming plate at 370C.





























Figure 3-2. Dissecting scope and fiber optic light used during subretinal injections and
experimental retinal illumination.


Retina

INL Rods and Cones
OPL
ONL

AI (
Choroid





/ Sclera


Figure 3-3. Cartoon depicting the subretinal injection. The blunt injection needle is
shown passing through the cornea, around the lens, and into the subretinal
space (right). Once positioned thus, the rAAV solution can be delivered to the
subretinal space (left), resulting in a localized retinal detachment that resolves
itself over time as the virus spreads laterally from the site of injection (red
arrow). Figure courtesy of Dr. Lynn C. Shaw.

Electroretinography

Electroretinographic analysis of ribozyme-treated animals was performed as


described above.










Results

Ribozyme Creation

Our therapeutic plan called for the creation of a ribozyme that would specifically

cleave the mutant human rhodopsin mRNA while leaving the endogenous, wild-type

mouse mRNA intact. A benefit of this strategy is that it allowed us to consider all

possible ribozyme cleavage sites in areas of the human gene that showed polymorphisms

with the mouse gene. We were also able to look for the specific cleavage site GUCUU.

It has been consistently demonstrated that hammerhead ribozymes cleave more

efficiently at the GUC target site than at any other (Shimayama et al., 1995). It has also

been reported that hammerhead ribozyme cleavage can be enhanced when the triplet

target sequence is followed by a UU or UA dinucleotide (Clouet-d'Orval and Uhlenbeck,

1997).

Mouse Target 5'-C U UC G U C U UUG G-3' Mouse Target 5'-G C UG U C U U C A C-3'
Human Target 5'-C U U C G U CU UG G-3' Human Target 5'-G C UCG U CUU C A C-3'
Ribozyme HRzl 3'-G A A G CA AA CC-5' Ribozyme HRz3 3'-C GA CA A A G U G-5'
A C A C
UG UG
A A
A U A U
G A G AG
C-G C-G
G-C G-C
G-C G-C
C-G C-G
G U G U
CU CU

Figure 3-4. Primary structure of ribozymes HRzl and HRz3 (green), shown paired with
their human (black) and mouse (red) target sequences. Polymorphisms between the
mouse and human rhodopsin genes are boxed.

The human rhodopsin gene contains 19 GUC sites in its reading frame. Of these,

three contained a UU dinucleotide directly following the cleavage site, and two of these

GUCUU sites contained single nucleotide polymorphisms between the endogenous

mouse gene and the mutant human transgene. The first of these sites is a GUCUU site

beginning at nucleotide 310 as measured from the start of the coding sequence for the









gene. The second of these begins at nucleotide 679. The predicted structure of these two

areas of the mRNA showed no serious energetically favorable secondary structure that

would inhibit proper ribozyme binding to their respective target sites. The two

ribozymes, named HRzl and HRz3, are illustrated in Figure 3-4.

In Vitro Time Course Analysis of HRzl and HRz3

In vitro cleavage analysis showed HRzl and HRz3 to be efficient at cleaving the

human rhodopsin mRNA. Reactions were performed under a condition of 10-fold excess

of substrate relative to ribozyme (10nM to InM). Both ribozymes achieved 20%

substrate cleavage in one minute in a reaction mixture containing 20mM MgCl2.

Magnesium is necessary in cell free reactions to promote the folding of the ribozyme, but

is not required in cells. Figure 3-5 shows a representative time course cleavage reaction

performed by incubating the HRzl RNA oligonucleotide with its 12 nucleotide, 5' end-

labeled target as described. These reactions generate two bands when separated on

polyacrylamide gels. The top band is the radioactively-labeled, uncut 12 nucleotide RNA

target molecule, while the bottom band is the 7 nucleotide 5' cleavage product (as only

the 5' end of the RNA target oligonucleotide was labeled, the 5 nucleotide 3' cleavage

product was not detectable by autoradiography). Phosphorimager analysis was used to

determine the relative intensity of these two bands, which in turn were used to calculate

the time course cleavage rates as described. Identical time course reactions were

performed using the HRz3 ribozyme/target combination (data not shown). Graphical

representation of the time course reactions of HRzl and HRz3, illustrated in Figure 3-6,

confirmed the activity of our ribozyme selections and prompted us to initiate efforts to

treat the hT17M mice with rAAV expressing these ribozymes.










HRzl


O' 1' 2'


e


4' 8' 16' 32' 64' 128'



PSu b s t r a t e




SProduct


-AEA .


Figure 3-5. Representative Phosphorlmager scan of a time course assay showing HRzl
cleavage of a 12 nucleotide synthetic human rhodopsin target RNA.


HRz1 Timecourse at 20mM MgCI2


0 20 40 60 80 100 120
Time in Minutes


HRz3 Timecourse at 20mM MgCI2


0 20 40 60 80 100 120
Time in Minutes


Figure 3-6. Time course of HRzl and HRz3 cleavage. The percent cleavage of each 12
nucleotide synthetic target by its respective ribozyme is plotted as a function
of time. Both ribozymes are able to achieve 20% cleavage in a minute or less
in the presence of 20mM MgCl2.

ERG Analysis of hT17M Transgenic Mice Treated With HRzl and HRz3

Several features of the pXX-MOPS-GS-HP recombinant AAV packaging vector

used in these studies merit attention (Figure 3-1). The plasmid contains a multiple

cloning site under the control of the MOPS 500 promoter, which has been shown









previously to induce high levels of transgene expression in mouse and rat photoreceptor

cells (Flannery et al., 1997). MOPS500 consists of 483 bp of the mouse opsin proximal

promoter, including 70 bp of the 5' untranslated region of the mRNA coding sequence.

Immediately following the promoter is an SV40 intron (SD/SA), which has also been

shown to increase expression of RNAs by promoting nuclear export via the spliceosomal

pathway (Bertrand et al., 1997). Next is the multiple cloning site, in which either HRzl

or HRz3 was inserted into the HindIII/Nsil junction as described. Following this is a

downstream hairpin ribozyme that generates well-defined 3' ends for the ribozyme

transcript, reducing the possibility of the ribozyme interacting with excess downstream

sequence in such a way as to cause it to fold into an inactive or inaccessible (to the target

mRNA) conformation (Altschuler et al., 1992).

The vector also contains an ampicillin antibiotic resistance gene to aid with

bacterial cloning, a neomycin resistance gene enabling selection in mammalian cells

using the antibiotic G418, and a GFP marker gene under the expression control of an

internal ribosomal entry site (IRES). The entire ribozyme expression, GFP marker, and

neomycin resistance cassettes are contained within inverted terminal repeat sequences

(ITRs) that are necessary for these various elements to be packaged as recombinant AAV

The resultant viruses were purified to a titer of 2x1013 genome copies/ml (HRzl) and

lxl013 genome copies/ml (HRz3).

Our initial treatment attempt involved subretinally injecting the right eyes of a litter

consisting of five mrho+/-; hT17M transgenic mice and two mrho+/- siblings with 0.5 [tl

of rAAV expressing HRzl. The injections were performed at 21 days of age, and the

animals underwent ERG analysis one and a half months later to assay rescue of the









injected eye. In all five hT17M mice, the injected eyes showed substantial ERG

reduction relative to the uninjected eyes. The non-transgenic, mrho+/- siblings were not

affected. These results are summarized in Figures 3-7 and 3-8. Injection of recombinant

AAV expressing the HRz3 ribozyme produced similar results: significant depression of

both a- and b-wave ERG responses in the injected eyes of the animals containing the

hT17M transgene, but not in their non-transgenic littermates that were also injected.


Relative ERG Responses of Mice Receiving
Subretinal HRZ1 Injections


120

0
C 100
e 80 -
S 0 *T17M Mice
S60 -
uw O non Rho+/- Mice
> 40 -
LiJ
20 -

0
20dB A 20dB B 10dB A 10dB B


Figure 3-7. Relative ERG responses of a- and b-waves at 20 and 10dB flash intensities.
Each bar represents the average of the ratio of right to left eye ERG responses
for all animals in that group. These mice were analyzed at one month after
subretinal injection of the right eye with 0.5 microliters of rAAV expressing
HRzl. Subretinal injection led to significantly greater damage in the hT17M
transgenic animals (dark green bars) than in their non-transgenic littermates
(light blue bars).

Data generated during the creation of another mouse model of retinitis pigmentosa

expressing a human rhodopsin transgene containing a P23H mutation demonstrated that

such mice expressing two copies of wild type rhodopsin (mrho+/+) degenerated more

slowly than those expressing only one copy (mrho+/-). This led us to attempt to treat the









hT17M model, which was expressed on a mrho+/- background, with recombinant AAV

expressing wild-type mouse rhodopsin on the premise that increasing the level of normal

rhodopsin might dilute the impact of the mutant transgene. These particular animals also

received contralateral injections of recombinant AAV expressing only the GFP marker

protein to control for any rescue that might result from an ocular response to injection

damage. These injections led to ERG reductions in the mutant mice (Figure 3-9) that

were similar to those seen with the ribozyme injections, only they were seen in both eyes

of the hT17M transgenic animals, as both eyes were injected.

Discussion

We developed ribozymes with the ability to efficiently cleave the mRNA

associated with a human rhodopsin transgene, which could be useful as therapeutic

reagents for treatment of the hT17M mouse model of RP. The ribozymes were designed

to allow them to discriminate between endogenous, wild-type mouse rhodopsin and the

mutant human transgene in such a way as to abolish expression of the mutant RNA while

leaving the wild-type RNA intact (Figure 3-4). HRz3 contained a mismatch with the

mouse target at the first nucleotide upstream of the cleavage triplet, a site where

mismatches between the ribozyme and its target sequence has been shown to abolish the

catalytic step of hammerhead ribozyme cleavage in vitro (Werner and Uhlenbeck, 1995).

HRzl contained a mismatch with the mouse target that is located three nucleotides

downstream of the cleavage triplet, and although sequence differences in this area are not

though to severely reduce the catalytic step of ribozyme cleavage, they are thought to

cause sufficient disruption of ribozyme binding to allow preferential cleavage of a

perfectly matched target sequence. In vitro analysis showed these ribozymes to be

catalytically efficient, with around 20% cleavage of a 12nt target RNA sequence

















-~-I-`-


5 Is per, Division


I-R /

2-~ -'_ xY~~


5 is per Division


59 2-1 1
ui/div


5 Ms per Division


5 ms per Division


D



-1I
t-ll^^ ~


5 Ms per Division


5








5s pe, vision



5 MS pOP Division


'-I rSJ*~-c--

~-2-


5 ms per Division


Figure 3-8. 10dB intensity ERG tracings from mice receiving subretinal injections of
their right eyes with rAAV expressing HRzl. mrho+/- mice (A and B) show
normal ERG responses in both left and right eyes, while the mrho+/-; hT17M
transgenic mice (C-G) show substantial reduction in the ERG response of the
injected eye. These tracings were used to generate the data in Figure 3-7.
















-2C i


5 ms per Division

C


uV/div -- --------_-



5 ns pen Division







50 2- .. -. .



5 Mo pep Division


I---i-
a-L


5 ns per Division

D










5 ms per Division

F


59i Iit--".-


5 is per Division


Figure 3-9. 10dB intensity ERG tracings from mice receiving subretinal injections of
both eyes. The right eyes were injected with rAAV delivering a wild type
mouse rhodopsin transgene, while the left eyes were injected with control
rAAV delivering the GFP marker gene. mro+/- mice (A and B) show normal
ERG responses in both left and right eyes, while the mrho+/-; hT17M
transgenic mice (C,D,E, and F) show substantial reduction in the ERG
response of the injected eyes.









observed at the one minute time point at a MgCl2 concentration of 20mM. This time

course cleavage rate was comparable to that demonstrated by other hammerhead

ribozymes in our laboratory that have been shown to be effective in animal models

(Lewin et al., 1998; Fritz et al., 2002; Liu et al., 2005; Gorbatyuk et al., 2005).

The ribozymes were cloned and packaged as recombinant AAV (rAAV), with

expression controlled by a rhodopsin-specific promoter sequence. Ancillary sequences

were included in these viral vectors that should allow for efficient expression of ribozyme

molecules in the target photoreceptor cells. Downstream, self-cleaving ribozyme

sequences were also included to generate precise 3' ends for the therapeutic ribozymes.

rAAV viral vectors expressing both ribozymes were purified to a high titer, and injections

were performed to deliver 0.5 [tl of the HRzl-expressing virus (1.0 x 1010 genome copies

per injection) to the subretinal space of mrho+/-; hT17M transgenic animals, along with

their non-transgenic littermates. Subsequent PCR genotyping was performed to

determine the mrho and hT17M genotype of each animal.

Unfortunately, these injections actually resulted in significant retinal damage

instead of the intended rescue. ERG analysis of mice containing the mutant human

transgene performed one month post-injection revealed severe attenuation of both a- and

b- wave ERG responses (Figure 3-8). The non-transgenic littermates were unaffected.

Repetition of these experiments with several subsequent litters using either HRzl or

HRz3 as the delivered ribozyme produced the same results severe damage to the retina

as determined by ERG analysis one month after injection. An attempt was also made to

achieve treatment by injecting vectors designed to produce a surplus of wild-type

rhodopsin in the retinas of the hT17M transgenic animals. This experiment utilized a









control injection of rAAV designed to deliver a GFP marker protein to the photoreceptors

of the contralateral eye. This paired injection experiment led to severe ERG attenuation

of BOTH eyes of the transgenic animals, while again the non-transgenic littermates were

unaffected (Figure 3-9).

These results were surprising. Although a slight amount of ERG attenuation is

often seen in animals following subretinal injection, it is usually neither as severe nor as

prolonged as the reduction observed in these experiments (Timmers et al., 2001).

Occasionally bad injections are accidentally performed that can cause severe retinal

damage, but these are rare when the technique is performed, as ours was, by experienced

personnel. Attempts to treat our hT17M transgenic mouse line invariably resulted in

grossly high failure rates, and it eventually became evident that the severe injection

damage was always seen in mice containing the hT17M transgene and not in their non-

transgenic littermates. This led us to suspect a defect in injection tolerance in the hT17M

mice and to design experiments to determine which aspects of the subretinal injection

technique were responsible for such a severe loss of ERG response.















CHAPTER 4
INCREASED LIGHT SENSITIVITY IN mRHO+/-; hT17M MICE

Introduction

Repeated attempts were made to treat hT17M-mediated retinal degeneration in

mice with subretinal injection of rAAV delivering either human rhodopsin-specific

ribozymes or wild-type mouse rhodopsin transgenes. Analysis of treated animals showed

severely reduced ERG responses in the injected eyes. At first this was thought to be the

result of injection damage, but it was soon observed that the ERG reduction was seen

only in mice carrying the hT17M mutant rhodopsin transgene. After this trend was

shown to repeat itself in several experimental groups, we hypothesized that the

mechanism of injection itself was detrimental to the visual response of the hT17M mutant

animals and not to their non-transgenic siblings. The subretinal injection technique had

two components that seemed likely candidates for the damage: the introduction of a

virus-containing solution into the subretinal space and the use of bright fiber optic light to

illuminate the extremely dilated eyes of these animals during the actual procedure.

Subretinal injection is known to create a retinal detachment that resolves itself over

time as the injected solution (in this case containing rAAV expressing the human

rhodopsin-specific ribozyme) is removed from the eye (Timmers et al., 2001). If the

technique is not performed properly, it is possible to damage the retina so severely that

the visual response is affected. It is also possible that mice expressing hT17M mutant

rhodopsin are for some reason unable to resolve their retinal detachments, leading to









decreased ERG response following what would normally have been a successful

subretinal injection. Injecting the transgenic mice with saline solution resulted in a

reduction in the ERG response similar to that observed following subretinal injections of

rAAV solutions, suggesting retinal damage (data not shown).

Light-mediated retinal damage (LMD) has been studied extensively since the first

reports of its occurrence in laboratory rats four decades ago (Noell et al., 1966). It has

since been documented in various other laboratory animals, including pigmented and

non-pigmented fish (Penn, 1985), mice (LaVail et al., 1987), rabbits (McKechnie and

Johnson, 1977), dogs (Cideciyan et al., 2005), and monkeys (Lawwill et al., 1980). In

many instances, simply maintaining these animals under continuous room-level

illumination can lead to severely attenuated ERG responses that correlate with a loss of

rod photoreceptor cells. Although the phenomenon has been extensively studied, the

exact mechanism by which light damage leads to photoreceptor cell death is still not fully

understood (Wenzel et al., 2005). All forms of light-induced retinal damage have two

things in common: rhodopsin is the initial effector molecule, and the ultimate fate of the

damaged photoreceptor cells is death by apoptosis.

There is significant evidence to demonstrate the involvement of rhodopsin in

LMD. Studies of the RPE65 knockout mouse model show that this mouse is completely

resistant to light damage. Since RPE65 is involved in regeneration of 11-cis retinal from

all-trans retinal, this observation led researchers to conclude that reconstituted rhodopsin

(i.e. opsin bound to the 11-cis retinal chromophore) is the initial mediator of LMD

(Grimm et al., 2000b). Protection against light damage can also be achieved by

inhibiting rhodopsin reconstitution through pharmacological means. Intravitreal









treatment with 11-cis retinoic acid, which presumably competes with 11-cis retinal

binding to RPE-65 and thus inhibits reconstitution of active rhodopsin, has been shown to

protect against light damage in rats (Sieving et al., 2001). Administration of halothane

anesthesia has also been shown to block regeneration of 11-cis retinal, and such treatment

prior to light damaging light exposure can prevent LMD in albino rats and mice (Keller et

al., 2001).

Apoptotic involvement in light mediated retinal damage was convincingly

demonstrated in albino rats through TUNEL labeling of fragmented DNA in the affected

retinas (Aonuma et al., 1999). In 1998, studies involving p53 knockout mice

demonstrated that they were not resistant to LMD, indicating that the gene was not

involved in the light-induced apoptotic pathway (Lansel et al., 1998). Subsequent studies

reported involvement of the apoptotic effector molecule c-fos, which is a member of the

API transcription factor complex, in light-mediated apoptosis, and elevated levels of API

have been demonstrated in several models of acute LMD (Reme et al., 1998; Wenzel et

al., 2002; Naash et al., 1996; Cideciyan et al., 2005).

Genetic mutations have been shown to increase light sensitivity in the retina. In

1987, LaVail and coworkers reported increased light sensitivity in Balb/c mice as

compared to C57BL6 mice (LaVail et al., 1987). Matings between the two strains

produced Fl progeny that showed an intermediate phenotype, demonstrating a

segregating genetic trait as the mediator of light sensitivity. It is now known that this is

the RPE65 gene, and that a polymorphism at codon 450 leucinee in Balb/c mice and

methionine in C57BL6 mice) leads to increased RPE65 activity in the leucine variant,

leading to accelerated rhodopsin regeneration which, as discussed above, leads to









increased susceptibility to light damage (Wenzel et al., 2001b; Danciger et al., 2004).

However, these results could not be duplicated in rat models of LMD (Beatrice et al.,

2003). Mice with mutations in the SOD 1 gene also show increased susceptibility to light

damage (Mittag et al., 1999), suggesting a role for reactive oxygen species as mediators

of damage.

As it has been demonstrated that rhodopsin is the initial mediator of retinal light

damage, it is perhaps unsurprising that mutations affecting this gene have been associated

with increased susceptibility to such damage. Both the P23H and S334ter rhodopsin

mutations have been associated with increased light sensitivity in rat and mouse models

of ADRP (Ranchon et al., 2003; Wang et al., 1997). Indeed, dark-reared mice containing

the P23H mutation show substantial reduction in the rate of retinal degeneration when

compared to P23H litters raised in normal cyclic light, suggesting that cell death as a

result of ADRP could share at least some apoptotic mechanisms with certain forms of

light-induced damage (Naash et al., 1996).

Recently a dog model of human retinitis pigmentosa containing a tyrosine to

arginine mutation at rhodopsin codon 4 has been associated with an extreme sensitivity to

light damage (Cideciyan et al., 2005). These animals were subjected to focused retinal

light exposures with light intensities that were 1500 to 6000 times less intense than those

typically used in animal models ofLMD. The researchers reported significant loss of

retinal thickness at the site of light exposure, as measured by optical coherence

tomography. This result is intriguing, as both T4R and T17M mutations have been

shown to affect glycosylation at the amino terminus of rhodopsin (Kaushal et al., 1994;

Zhu et al., 2004).









Given the association of rhodopsin mutations with LMD, and considering that the

subretinal injection technique utilized in our studies involved shining bright fiber optic

light into the severely dilated retinas of anesthetized animals, experiments were designed

to test whether this intense exposure to light could have a damaging effect on the retinas

of mice containing the hT17M mutant rhodopsin transgene. This chapter describes those

experiments.

Materials and Methods

Retinal Illumination

Breedings were arranged as described above to create mrho+/- litters, of which a

portion would also contain the hT17M mutant rhodopsin transgene. At weaning age (21-

24 days) these litters were removed from their parents and their right eyes were dilated

with 1% atropine. The next morning, the right eyes were again dilated with 1% atropine,

and again an hour before the illumination procedure, at which time the eyes also received

a drop of 2.5% phenylephrine and 0.5% proparacaine HC1. An hour after this final

dilation, the animals were anesthetized and again treated with a drop each of 1% atropine,

2.5% phenylephrine, and 0.5% proparacaine HC1. The right eyes of these animals then

received a drop of 2.5% hypomellose to aid in retinal visualization and to help keep the

retina hydrated, and were illuminated with a Southern Micro Instruments 150 Watt fiber

optic light source with Schott Fostec fiber optic arms at an intensity of 10,000 or 5,000

lux for a period of 2.5 minutes. Light intensities were measured with an Extech Data

Logging Light Meter (Extech, Waltham, MA). Retinas were visualized under a

dissecting microscope, as described previously for the subretinal injections, to ensure that

the pupils remained dilated and that the light remained focused on the retina throughout

the duration of the experiment. Both eyes of the animals then received smears of VPP









ointment, and the mice were allowed to recover on a warming tray. This experiment was

designed to closely mimic the subretinal injection protocol in all ways except for the

actual injection.

Genotyping

Tail snips were taken from the mice while they were anesthetized for retinal

illumination, funduscopy, or in the case of the animals used for histological examination,

after they were sacrificed. Genomic DNA isolation and PCR analysis was performed as

described above to identify animals containing the hT17M mutant human rhodopsin

transgene.

Electroretinography

Electroretinography was performed as described above. Statistical comparisons

between the illuminated and non-illumated eyes were performed to generate P values

using the paired, one-tailed Student's t-test feature of Exel spreadsheet software

(Microsoft, Redmond, WA).



Funduscopy

Funduscopy was performed as described above.

Histology

Animals were sacrificed by overdose of Isoflurane, followed by cervical

dislocation. Eyes were enucleated, and a small hole was placed in the cornea with an

insulin needle. They were then fixed overnight at 40C in freshly-made 4%

paraformaldehyde. The next day they were incubated in solutions of sucrose diluted in

phosphate buffer (pH 7.4) at concentrations of 7% (2 hours at 4C), 15% (2 hours at 4C),

and 30% (overnight at 4C), for cryoprotection. After the final incubation, the eyes were









suspended in 15xl5x5mm disposable base molds (Electron Microscopy Sciences, Ft.

Washington, PA) in Tissue Tek OCT Compound Embedding Medium (Sakura Finetek,

Torrance, CA) such that the cornea and optic nerve formed an axis parallel to the bottom

of the mold, with the cornea to the front. The blocks were then frozen in isopentane at a

temperature of-40C. Frozen eyes were stored at -800C. 12-14 micron retinal sections

were then obtained from these frozen eyes using a Microm H550 (Microm, Walldorf,

Germany) cryostat, with particular care taken to obtain sections around the optic nerve.

Fisherbrand (Fisher Scientific, Pittsburgh, PA) Superfrost Plus microscope slides of size

75x25x1.0mm were used to collect the sections, which were stored at -800C.

TUNEL Visualization of Apoptosis

DNA fragmenting, a characteristic of apoptosis, was detected using a terminal

deoxynucleotide-mediated nick end-labeling (TUNEL) assay. For these experiments, the

In Situ Cell Death Detection Kit, TMR red (Roche Applied Science, Mannheim

Germany) was used, as per the manufacturer's instructions. In brief, sections were

thawed for twenty minutes, and then washed twice for five minutes in room temperature

lX PBS. Next, the sections were permeablized in a solution of 0.1% sodium citrate and

0.1% Triton X-100 detergent for two minutes on ice. The sections were then washed

twice for ten minutes with IX PBS at room temperature. A hydrophobic slide marker

pen (Daido Sangyo Co., Ltd., Tokyo, Japan) was used to surround the retinal sections so

that the TUNEL reagents would not leak off. TUNEL label mix was subsequently added

to the slides as per the manufacturer's instructions. Cover slips (size 24x60mm, Fisher

Scientific) were added, and the sections incubated in the dark at 37 C in a humid

chamber. After an hour, the cover slips were gently removed, the sections were rinsed

three more times in room temperature 1X PBS for 10 minutes each wash, dried briefly,









mounted with Vectashield Mounting Medium with DAPI (Vector Laboratories,

Burlingame, CA), and re-covered. Cover slips were sealed to the slides with Sally

Hansen Double Duty Nail Polish (Del Laboratories, Inc., Farmingdale, NY). Apoptotic

cells fluoresced red when visualized and photographed on a Zeiss Axiokop 2 mot plus

microscope utilizing Axiovision 4 software (Zeiss International).

Results

High Intensity Illumination

The right eyes of an experimental group consisting of six mrho+/- mice and six

mrho+/-; hT17M mice, aged 21-23 days, were illuminated as described with light of

10,000 lux intensity for 2.5 minutes. A period of 2.5 minutes was chosen for the

illumination because it is the approximate time an animal is exposed during an actual

injection procedure. Electroretinographic analysis was then performed on the mice at

intervals of 1, 3, and 5 weeks, and the results plotted (Figures 4-1 and 4-2). The results

demonstrate a significant decrease of around 30% in both a- and b-wave responses at

each time point in the illuminated eyes of animals containing the hT17M transgene, but

not their non-transgenic siblings.

Low Intensity Illumination

One possible way to reduce the damage caused by retinal illumination during

subretinal injections would be to reduce the intensity of the fiber optic light used to

visualize the retina. To test whether a reduction in intensity could decrease or eliminate

light-induced retinal degeneration, nine mrho+/- and eight mrho+/-; hT17M mice were

subjected to illumination as described, with the intensity of illumination reduced to 5,000

lux. Electroretinographic analysis was performed on the mice at intervals of 1, 3, and

five weeks after illumination, and the results plotted (Figures 4-3 and 4-4). Low intensity










2.5 Min Dissecting Scope Light Exposure 10dB A
Waves


250

200
-

*- 150
(u
o 100



0


1 Week


3 Weeks


* T17M Left
O T17M Right
* Rho +/- Left
O Rho+/- Right


5 Weeks


Figure 4-1. A-wave ERG responses at one, three, and five weeks after 10,000 lux
intensity, 2.5 minute illumination of mrho+/- and mrho+/-;T17M mice.
Asterisks indicate a difference between the right eye and the left eye with a P
value of less than 0.05.

2.5 Min Dissecting Scope Light Exposure 10dB B
Waves

800
700
> 600 T17M Left
.E 500 E T17M Right
u 400 Rho +/- Left
O 300 0 Rho+/- Right
200
100
0 Ii
1 Week 3 Weeks 5 Weeks


Figure 4-2. B-wave ERG responses at one, three, and five weeks after 10,000 lux
intensity, 2.5 minute illumination of mrho+/- and mrho+/-;T17M mice.
Asterisks indicate a difference between the right eye and the left eye with a P
value of less than 0.05.


in r


-













3(

2i

S2



S 1

L 1


2.5 Min Dissecting Light Exposure 10dB A waves

00

50

ST17M Left
00
O T17M Right
50 Rho+/- Left
00 _O Rho +/- Right
00
c n


1 Week 3 Weeks 5 Weeks


Figure 4-3. A-wave ERG responses at one, three, and five weeks after 5,000 lux
intensity, 2.5 minute illumination of mrho+/- and mrho+/-;T17M mice.
Asterisks indicate a difference between the right eye and the left eye with a P
value of less than 0.05.


2.5 Min Dissecting Light Exposure 10dB B waves

800
700 -
> 6e00 3T1TM Left
500 []T1TM Right

i 400 Rhono lLeft
300 Rho -+- Right
(9 200
100
0
1 Week 3 Weeks 5 Weeks


Figure 4-4. B-wave ERG responses at one, three, and five weeks after 5,000 lux
intensity, 2.5 minute illumination of mrho+/- and mrho+/-;T17M mice.
Asterisks indicate a difference between the right eye and the left eye with a P
value of less than .05.









illumination led to a 20-30% decrease in both a- and b-wave amplitudes at all three time

points. These results closely resembled the results observed after the similar, high-

intensity illumination experiments.

Apoptosis in Retinas Damaged by Low Intenisty Illumination

Light damage to the retina has been extensively studied. It has been demonstrated

that light-induced retinal damage is caused by apoptosis of photoreceptor cells (Hao et

al., 2002; Farrar et al., 2002). It is this cell death that is responsible for the depressed

ERG responses that are seen in light-damaged animals. To test whether apoptotic cell

death was the cause of the ERG reduction that was seen in the retinal illumination

experiments, two mrho+/- and three mrho+/-; hT17M mice were subjected to 5000 lux

illumination as described previously. The animals were sacrificed after 24 hours, their

eyes were fixed and sectioned, and TUNEL labeling was used to visualize apoptosis. The

results, shown in Figures 4-5 and 4-6, show evidence of intense photoreceptor cell death

in the illuminated retinas of the animals containing the hT17M transgene, but not in their

non-transgenic littermates.

In order to document the extent of photoreceptor apoptosis, pan-retinal images were

obtained from the TUNEL stained sections using tile-field mapping of 20X images on a

Zeiss Axiophot Z microscope equipped with a Sony DXC-970MD 3CCD Color Vid

Camera and an MCID Elite Stage, utilizing MCID (Imaging Research, Inc., Ontario,

Canada) Analysis Software (Imaging Research, Inc.). The results, shown in Figure 4-6,

show that the photoreceptor damage is pan-retinal, rather than tightly localized.

Funduscopic Illumination

Funduscopic examination is one of the most common ophthalmologic procedures,

often performed either as part of a routine physical or complete eye examination to detect









Left Right



A







B







C







D







E





Figure 4-5. TUNEL stained retinal sections from mrho+/- and mrho+/-; hT17M mice
whose right eyes were illuminated with 5000 lux white light for 2.5 minutes.
Sections of mrho+/- mice (rows A and B) show no evidence of apoptosis,
while sections from mrho+/-; hT17M mice (rows C, D, and E) show extensive
apoptosis, as evidenced by large numbers of red-labeled photoreceptor nuclei.
In each row, left and right eye sections are from the same experimental
animal.



























Figure 4-6. Tile-field mapped image of the TUNEL stained retina of an eye from a
mrho+/-; hT17M animal that was illuminated for 2.5 minutes with 5,000 lux
light. Arrows indicate pan-retinal red fluorescence indicative of apoptotic
photoreceptors.

and evaluate symptoms of eye disease, such as glaucoma or retinal detachment, or if

diabetes, hypertension, or other vascular disease is suspected. The characteristic "bone

spicule" deposits associated with retinitis pigmentosa are among the indicators that a

patient presenting with reduced visual fields and impaired night vision is actually

suffering from RP. During this procedure, the back of the retina is visualized through the

dilated iris of the patient using a bright white light. If photographs of the retina are taken,

they too must utilize intense flashes of light to record their images. It has been

demonstrated that funduscopic examination is damaging to the retinas of dogs containing

a T4R rhodopsin mutation (Cideciyan et al., 2005).

In order to determine if funduscopic examination and photography of the retinas of

mice carrying the hT17M human mutant rhodopsin transgene was harmful, eight mrho+/-

mice, four with the hT17M transgene and four that were non-transgenic, had two funds

pictures taken of their right eyes at three and six weeks of age. One week after each set









of photographs, electroretinography was performed as described, and the results averaged

and plotted (Figure 4-7). These ERG recordings show clearly depressed a- and b-wave

amplitudes in the hT17M transgenic mice following both sets of funduscopic

photography, while their non-transgenic littermates were unaffected.

The first set of funds pictures show no evidence of retinitis pigmentosa in either

the mrho+/- mice or their mrho+/-; hT17M littermates. However the second set of funds

images, taken at six weeks of age, reveal punctate regions of the retina in the hT17M

transgenic animals, suggesting loss of pigment in the retina. The images of the non-

transgenic mice looked normal. These results are summarized in Figure 4-8.

Apoptosis in Retinas Damaged by Fundus Photography

It seemed reasonable to assume that the depression of ERG response seen in hT17M

transgenic mice following funds photography would be accompanied by photoreceptor

apoptosis, as was noted with the animals damaged by low-intensity fiber optic

illumination. In order to confirm this, two mrho+/- littermate mice, one containing the

hT17M mutant rhodopsin transgene and one that was non transgenic, were subjected to

funds photography of the right eye at 21 days of age. One day later, the animals were

sacrificed, and their eyes were enucleated, fixed and sectioned. Sections containing the

optic nerve were then TUNEL stained to visualize apoptotic cells, with DAPI

counterstain to reveal retinal morphology. The results, illustrated in Figure 4-9, show

that funds photography clearly induced apoptosis in the rod photoreceptor cells of

mrho+/-; hT17M mice, but not in their non-hT17M littermates.












250

200

150

100

50

0


AWave ERG Following R Eye Funduscopy




Rho +/-;T17M Left
o Rho +/-;T17M Right
m Rho+/- Left
[ Rho+/- Right



First Series Second Series


B Wave ERG Following R Eye Funduscopy

700
600 -
500 m Rho +/-;T17M Left
400 O Rho +/-;T17M Right
300 Rho+/- Left
200 O Rho+/- Right
100
0


First Series


Second Series


Figure 4-7. A and b-wave ERG responses of mrho+/- and mrho+/-; hT17M mice
subjected to funds photography at 3 weeks (first series) and 6 weeks (second
series) of age. indicates significant difference between right and left eyes
with a P value of less than 0.05.








Week 3 Week 6 Week 3 Week 6


mrho+/- mrho+/-;hT17M


Figure 4-8. Fundus pictures of mrho+/- (left sets) and mrho+/-; hT17M (right sets) mice
at three and six weeks of age. Evidence of retinal degeneration is seen in the
six week set of hT17M transgenic mice.








Left


Right


mrho+/-;
hT17M+





mrho+/-;
hT17M+






mrho+/-;


Figure 4-9. TUNEL labeling of retinal sections from mice whose right eyes underwent
funds photography at 21 days of age. Sections of mrho+/- mice (bottom
row) show no evidence of apoptosis, while sections from mrho+/-; hT17M
mice (top rows) show extensive apoptosis, as evidenced by large numbers of
red-labeled photoreceptor nuclei. Images in the top row were taken at 20X
magnification; those in the bottom two rows were taken at 40X magnification.
ERG Analysis of hP23H Mice After High Intensity Illumination
As has been discussed, rats and mice containing a human rhodopsin transgene that
contains a proline to histidine mutation at codon 23 have been extensively used to study
and model retinal disease. Mice that are bred to be hemizygous null at the mouse
rhodopsin locus and that also contain the hP23H transgene undergo rapid retinal









degeneration with concomitant photoreceptor loss, culminating with a 60% loss of ERG

response at three months of age. This model has been successfully treated with

ribozymes designed to specifically cleave the mutant human transgene (F. Fritz UF

doctoral dissertation and unpublished results).

Since procedures used for subretinal injection of both lines of mice are identical, it

was important to explore the effect of focused fiber optic light on the retinas of mice

containing the hP23H transgene. If the hP23H mice showed light sensitivity similar to

the hT17M transgenics, it could have a negative impact on the ultimate success of

treatment with subretinal ribozyme injections. It was also of interest to determine

whether these two closely located mutations, which lead to different rates of retinal

degeneration, would respond differently to intense retinal illumination.

Our initial experiments examined the effect of high intensity fiber optic light on the

ERG responses of mrho+/-;hP23H mice. Eight rhodopsin hemizygous mice containing

the hP23H transgene and their five non-transgenic siblings were exposed in their right

eyes to 2.5 minutes of 10,000 lux intensity light, as described above. The animals then

underwent ERG analysis at one, three, and five weeks post-illumination to determine the

effect of the light exposure upon their visual response. These data are summarized in

Figure 4-10.

Although there were significant depressions of a- and b- wave ERG amplitudes in

the hP23H transgenic mice after illumination, the differences averaged around a 10%

reduction of a- and b- wave responses at 1 week after illumination, as compared to

around 26% depression in the hT17M transgenic mice. Furthermore, the hT17M mice

continued to exhibit reduction in a- and b-wave ERG responses at five weeks post