Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of autophagy in retinal rod photoreceptor cells in Xenopus laevis Wen, Runxia 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2018_september_wen_runxia.pdf [ 5.07MB ]
Metadata
JSON: 24-1.0369290.json
JSON-LD: 24-1.0369290-ld.json
RDF/XML (Pretty): 24-1.0369290-rdf.xml
RDF/JSON: 24-1.0369290-rdf.json
Turtle: 24-1.0369290-turtle.txt
N-Triples: 24-1.0369290-rdf-ntriples.txt
Original Record: 24-1.0369290-source.json
Full Text
24-1.0369290-fulltext.txt
Citation
24-1.0369290.ris

Full Text

iCHARACTERIZATION OF AUTOPHAGY IN RETINAL RODPHOTORECEPTOR CELLS IN XENOPUS LAEVISbyRunxia WenA THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinThe Faculty of Graduate and Postdoctoral Studies(Cell and Developmental Biology)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)AUGUST, 2018© Runxia Wen 2018iiThe following individuals certify that they have read, and recommend to the Faculty ofGraduate and Postdoctoral Studies for acceptance, the dissertation entitled:Charaterization of autophagy in retinal rod photoreceptor cells in Xenopus laevissubmitted by Runxia Wen in partial fulfillment of the requirements forthe degree of Doctor of Philosophyin Cell and Developmental BiologyExamining Committee:Orson Moritz, Ophthalmology and Visual ScienceSupervisorRobert Molday, BiochemistrySupervisory Committee MemberCheryl Gregory-Evans, Ophthalmology and Visual ScienceSupervisory Committee MemberCalvin Yip, BiochemistryUniversity ExaminerJoanne Matsubara, Ophthalmology and Visual ScienceUniversity ExaminerAdditional Supervisory Committee Member:Sharon Gorski, Medical GeneticsSupervisory Committee MemberiiiAbstractRetinitis pigmentosa (RP) is a genetic neurodegenerative disorder that causesprogressive cell death of the rod and cone photoreceptors, eventually leading toblindness. The light-sensitive protein in rods, rhodopsin, is composed of thechromophore 11-cis retinal and the protein rod opsin. Mutations in the rhodopsin geneare common causes of RP. Autophagy is a lysosomal-turnover pathway for degradingdysfunctional proteins, organelles or other cellular components that is necessary formaintaining cellular homeostasis. We observed an increase of autophagy structures inrods expressing the misfolding-prone rhodopsin mutant P23H (Bogéa et al. 2015).However, the role autophagy plays in RP is not clearly understood. To examine the roleof autophagy in normal and diseased rods, I generated transgenic Xenopus laevistadpoles expressing the autophagy reporter mRFP-eGFP-LC3.My results demonstrate that the autophagy process lasts for about 34 h in normal rods.Early autophagic structures persist for 6 to 8 h before fusing with lysosomes andacidification; acidified autolysosomes persist for about 28 h before complete digestion.Autophagy in normal rods is diurnally regulated, with more autophagic structuresgenerated in light and fewer in darkness; this regulation is non-circadian. Autophagyalso increased in rods co-expressing P23H rhodopsin.The rhodopsin chromophore, a pharmacological chaperone for rhodopsin, absorbsphotons to initiate phototransduction, and is consumed in this process; it also promotesivproper rhodopsin folding. To determine whether increased autophagy in light-exposednormal rods is caused by increased misfolding of wildtype rhodopsin due to lack ofchromophore, I used CRISPR/Cas9 to knock out the gene RPE65, which is essential forchromophore biosynthesis. I observed that eliminating chromophore does not promoteautophagy in dark-reared rods, but prevents induction of autophagy in light-exposedrods. This combination of outcomes suggests that, although rhodopsin misfolding caninduce autophagy, light-induced autophagy is not due to misfolding of rhodopsin, butrather due to phototransduction.Further, I found that a group of compounds called histone deacetylase (HDAC)inhibitors, valproic acid (VPA), sodium butyrate (NaBu) and CI-994, consistentlypromote autophagy in rods; these compounds were previously demonstrated toameliorate retinal degeneration associated with P23H rhodopsin (Vent-schmidt et al.2017).vLay SummaryRetinitis pigmentosa (RP) is a group of inherited eye disorders caused by progressivecell death in the retina, ultimately leading to blindness. Rhodopsin is a light-sensitiveprotein found in rod photoreceptors, the dim light-detecting cells of the retina. Mutationsin the rhodopsin gene frequently cause RP. Autophagy is a recycling process in the cellfor breaking down dysfunctional cellular components, which is necessary formaintaining a healthy cellular environment. We found autophagic structures wereincreased in diseased rods carrying a rhodopsin mutation (Bogéa et al. 2015), but therole that autophagy plays in these diseased rods is not clearly understood. I studiedautophagy in normal and diseased rods by using genetically modified frog models inwhich the autophagy process can be directly visualized. My research has allowedgreater understanding of autophagy in normal cell biology and the mechanisms of RP.viPrefaceIn this thesis, I set out to characterize autophagy, a lysosomal degradation process, inrod photoreceptors of Xenopus laevis, including the levels and time-course ofautophagy in both wildtype and degenerating rod photoreceptors. For this, I establishedtransgenic X. laevis using the transgenes XOP-mRFP-EGFP-LC3 and HSP70-mRFP-EGFP-LC3 to study autophagy in rod photoreceptors in vivo.By using the HSP70-mRFP-EGFP-LC3 transgenic animals, I characterized the time-course of autophagy in normal rod photoreceptors. By using the XOP-mRFP-EGFP-LC3transgenic animals, I characterized the levels of autophagy in normal rods, observed adiurnal regulation of autophagy in normal rod photoreceptor cells, and determined thatthis regulation is light-dependent and non-circadian. I also characterized the levels ofautophagy in degenerating rods expressing a misfolding-prone P23H mutant rhodopsin.I designed, performed these experiments and analyzed the data.To understand further the mechanism underlying light-dependent autophagy, I designedan experiment to generate transgenic XOP-mRFP-EGFP-LC3 X. laevis tadpoles thatare deficient in chromophore production, in order to determine whether light-dependentautophagy is present in rods without chromophore, which lack the ability to carry outphototransduction. This was done by CRISPR-cas9 mediated knockout of the geneRPE65, which is essential for chromophore production. My colleague Paloma Stanarviidesigned the guide RNA targeting the RPE65 gene. We performed the experimentstogether. I collected and analyzed the data.I also quantified the influence of a widely-used autophagy inducer rapamycin, anautophagy inhibitor chloroquine and a class of histone deacetylase (HDAC) inhibitors(including VPA, NaBu and CI-994) on autophagy by using the transgenic XOP-mRFP-EGFP-LC3 animals. I designed, performed these experiments and analyzed the data.Finally, to determine whether regulation of autophagy levels can affect the extent ofretinal degeneration, tadpoles with retinal degeneration were treated with either anautophagy inducer (rapamycin) or an autophagy inhibitor (chloroquine) and the levels ofretinal degeneration were measured after the treatment. I designed and performed theexperiments. Data were collected with help from my colleague Ruanne Vent-Schmidt,and I analyzed the data.A version of Figure 20 has been published in “Light induces ultrastructural changes inrod outer and inner segments, including autophagy, in a transgenic Xenopus laevisP23H rhodopsin model of retinitis pigmentosa”, Tami H. Bogéa, Runxia H. Wen, andOrson L. Moritz. Invest Ophthalmol Vis Sci. 2015 Dec; 56(13): 7947–7955. Publishedonline 2015 Dec 17.doi:10.1167/iovs.15-16799. PMCID: PMC4684193.viiiA version of Figure 21 has been published in “Opposing effects of valproic acidtreatment mediated by histone deacetylase inhibitor activity in four transgenic X. laevismodels of retinitis pigmentosa”. Ruanne Y.J. Vent-Schmidt, Runxia H. Wen, ZushengZong, Colette N. Chiu, Christopher G. May, Beatrice M. Tam and Orson L. Moritz. (2016)Journal of Neuroscience 9 December 2016,1647-16; DOI:https://doi.org/10.1523/JNEUROSCI.1647-16.2016All the animal experiments were performed in accordance with animal care certificates(A14-0276) issued by the UBC Animal Care Committee (ACC).ixTable of ContentsAbstract........................................................................................................................................................ iiiLay Summary............................................................................................................................................... vPreface..........................................................................................................................................................viTable of Contents........................................................................................................................................ ixList of Tables...............................................................................................................................................xiiList of Figures............................................................................................................................................ xiiiList of Abbreviations..................................................................................................................................xvAcknowledgements.....................................................................................................................................xxDedication..................................................................................................................................................xxii1. Introduction..............................................................................................................................................11.1 The Retina:......................................................................................................................................... 11.2 Photoreceptors:.................................................................................................................................. 51.3 Rhodopsin:........................................................................................................................................121.4 Phototransduction:.......................................................................................................................... 151.5 Retinitis Pigmentosa:.......................................................................................................................201.6 P23H rhodopsin:.............................................................................................................................. 241.7 How does P23H rhodopsin cause cell death in RP?......................................................................261.7.1 Gain of function:.......................................................................................................................271.7.2 Proteasome overload:............................................................................................................... 291.7.3 Dominant negative effect:........................................................................................................ 291.8 Autophagy:....................................................................................................................................... 311.8.1 The subtypes of autophagy:..................................................................................................... 321.8.2 Autophagy pathway:.................................................................................................................351.8.3 Molecular mechanism of autophagy:......................................................................................361.8.4 Regulation of autophagy:......................................................................................................... 411.9 Autophagy and retinal degeneration:............................................................................................ 431.9.1 Autophagic degradation of P23H rhodopsin in cell cultures:...............................................43x1.9.2 Autophagy in animal models of RP:....................................................................................... 441.9.3 Autophagy in RP caused by the P23H rhodopsin mutation:................................................451.10 Goals of this thesis:........................................................................................................................ 501.11 Significance:................................................................................................................................... 512. Methodology...........................................................................................................................................522.1 Generation of two transgene constructs:.......................................................................................522.2 Generation and rearing of transgenic X. laevis:............................................................................532.3 Drug treatments of transgenic X. laevis:........................................................................................552.4 Immunohistochemistry and confocal microscopy:.......................................................................552.5 Extraction of genomic DNA and qPCR genotyping:....................................................................582.6 Transmission electron microscopy (TEM):...................................................................................592.7 Generation of RPE65 knock out X. laevis tadpoles using CRISPR-cas9 technology:............... 592.8 Dot blot analyses:.............................................................................................................................612.9 Statistical Analysis........................................................................................................................... 613. Results.....................................................................................................................................................633.1 Development of transgenic X. laevis reporter lines.......................................................................633.2 Initial characterization of fluorescence in transgenic reporter lines.......................................... 633.3 Time-course of autophagy in rods:.................................................................................................693.4 Diurnal variation in autophagy:.....................................................................................................713.5 Diurnal variation in WT animals:..................................................................................................753.6 Diurnal variation of autophagy is due to light illumination, not circadian rhythm..................773.7 Autophagy induced by the misfolding rhodopsin mutant P23H.................................................793.8 Autophagy regulation by chromophore availability.....................................................................833.9 Autophagy levels increased in X. laevis rods expressing P23H rhodopsin................................. 883.10 HDAC inhibitors induced autophagic structures in wildtype X. laevis rods............................913.11 Rapamycin treatment up-regulated autophagic flux in X. laevis rods......................................953.12 Chloroquine treatment inhibited autophagic flux in X. laevis rods.......................................... 983.13 VPA treatment up-regulated autophagic flux in X. laevis rods...............................................1023.14 NaBu treatment up-regulated autophagic flux in X. laevis rods............................................. 1043.15 CI-994 treatment up-regulated autophagic flux in X. laevis rods........................................... 1073.16 Effect of rapamycin on P23H retinal degeneration..................................................................1093.17 Effect of chloroquine on P23H retinal degeneration................................................................113xi4. Discussion............................................................................................................................................. 1164.1 Expression:..................................................................................................................................... 1174.2 Large red puncta:........................................................................................................................1184.3 Light-dependent autophagy:.........................................................................................................1194.4 Light and P23H regulation of autophagy:...................................................................................1204.5 Autophagy protects against various neurodegenerations:.........................................................1224.6 The roles that autophagy plays in retinal degeneration associated with P23H rhodopsin:....1244.7 Controversial roles that autophagy plays in retinal degeneration:...........................................125Bibliography.............................................................................................................................................127Appendices................................................................................................................................................147A Introduction......................................................................................................................................147A.1 Photoreceptor outer segment disk renewal:...........................................................................147A.2 Peripherin-2:............................................................................................................................. 148A.3 Goal of this experiment:...........................................................................................................150BMethodology..................................................................................................................................... 150B.1 Generation of HSP70-peripherin-eGFP transgenic X. laevis tadpoles:............................... 150B.2 Immunohistochemistry and confocal microscopy:................................................................ 151C Results...............................................................................................................................................152D Conclusions and discussions:.......................................................................................................... 158xiiList of TablesTable 1 The average number of autophagic puncta per rod in non-treated andrapamycin-treated X. laevis tadpoles...............................................................97Table 2 The average number of autophagic puncta per rod in non-treated andchloroquine-treated X. laevis tadpoles.......................................................... 100Table 3 The average number of autophagic puncta per rod in non-treated andVPA-treated X. laevis tadpoles....................................................................... 103Table 4 The average number of autophagic puncta per rod in non-treated andNaBu-treated X. laevis tadpoles..................................................................... 105Table 5 The average number of autophagic puncta per rod in non-treated and CI-994-treated X. laevis tadpoles........................................................................ 108xiiiList of FiguresFigure 1 Structure of the retina. .........................................................................................2Figure 2 The mean absorbance spectra of human rods and cones.............................7Figure 3 Schematic diagram of vertebrate rods and cones........................................... 9Figure 4 Structure of rhodopsin........................................................................................ 14Figure 5 Schematic view of major proteins involved in phototransduction................17Figure 6 The mammalian visual cycle. ........................................................................... 19Figure 7 The comparison of a normal retina to a retina affected by retinitispigmentosa (RP). ............................................................................................... 21Figure 8 Different types of autophagy............................................................................. 34Figure 9 The process of autophagy. ............................................................................... 36Figure 10 The molecular mechanism of autophagy in mammals..................................40Figure 11 Light exposure induced autophagic structures in rod inner segments ofbP23H-expressing X. laevis tadpoles.............................................................. 46Figure 12 Plasmid Hsp70-2DI-GFP map.......................................................................... 53Figure 13 Establishment and initial characterization of transgenic reporter lines ..... 67Figure 14 Time-course of autophagy in rods. ..................................................................70Figure 15 Diurnal variation of autophagy in X. laevis rods expressing XOP-mRFP-eGFP-LC3. .......................................................................................................... 73Figure 16 Diurnal variation of autophagy in wildtype X. laevis rods............................. 76Figure 17 Diurnal variation of autophagy is due to light illumination, not circadianrhythm................................................................................................................... 78xivFigure 18 Increased autophagy in bP23H rhodopsin-expressing rods........................ 82Figure 19 Autophagy regulation by chromophore availability........................................86Figure 20 LC3B immunolabeling confirmed the presence of increased numbers ofautophagic compartments in rod inner segments of tadpoles expressingbP23H rhodopsin................................................................................................ 90Figure 21 HDAC inhibitors induced autophagic structures in wildtype X. laevis rods................................................................................................................................ 93Figure 22 Rapamycin treatment up-regulated autophagic flux in X. laevis rods........ 97Figure 23 Chloroquine treatment inhibited autophagic flux in X. laevis rods............ 101Figure 24 VPA treatment up-regulated autophagic flux in X. laevis rods.................. 103Figure 25 NaBu treatment up-regulated autophagic flux in X. laevis rods................ 106Figure 26 CI-994 treatment up-regulated autophagic flux in X. laevis rods.............. 108Figure 27 Effect of rapamycin on P23H retinal degeneration......................................112Figure 28 Effect of chloroquine on P23H retinal degeneration....................................114Figure 29 Peripherin-2 localization to the rim region of photoreceptor disks............ 149Figure 30 Measuring the disk displacement rate of X. laevis rods..............................154Figure 31 Measuring the disk displacement rate of X. laevis cones...........................157xvList of Abbreviationsautosomal-dominant Retinitis Pigmentosa (adRP)Alzheimer’s disease (AD)The AMP-activated protein kinase (AMPK)The activating transcription factor 4 (ATF4)The activating transcription factor (ATF6)Autophagy-related genes (Atg)Macro-autophagy (autophagy)Binding immunoglobulin protein (Bip)Connecting cilium (CC)cyclic guanosine monophosphate (cGMP)Transcriptional factor C/EBP homologous protein (CHOP)Chaperone-mediated autophagy (CMA)Cyclic-nucleotide gated ion channel (CNG)Cone photoreceptor cell (Cone)The cone outer segment (COS)Dimethyl sulfoxide (DMSO)day post-fertilization (dpf)The eukaryotic translation-initiaion factor 2(eIF2)The ectopic P-Granules autophagy protein 5 homolog (EPG5)Electron Microscopy (EM)Endoplasmic reticulum (ER)xviER associated degradation (ERAD)Gamma-aminobutyric acid receptor-associate protein (GABARAP)Gamma-aminobutyric acid receptor-associate protein-like protein 1 (GABARAPL1)Gamma-aminobutyric acid receptor-associate protein-like protein 2 (GABARAPL2)Guanylate cyclase (GC)Guanylyl cyclase activating proteins (GCAP)Retinal ganglion cell layer (GCL)G-protein coupled receptor (GPCR)The green fluorescent protein (GPF)G-protein receptor kinase (GRK)Hour (h)Huntington’s disease (HD)Histone deacetylase (HDAC)Human embryonic kidney cell (HEK cell)The homotypic fusion and protein sorting (HOPS)heat shock cognate protein of 70kDa (hsc70)hsp70 heat shock promoter (hsp70)Inner nuclear layer (INL)Inner plexiform layer (IPL)intrinsically-photosensitive ganglion cells (ipRGCs)The inositol-requiring enzyme1(IRE1)Inner segment (IS)Lysosome-associated membrane protein type 2A (LAMP-2A)xviiMicrotubule-associated protein 1A/1B light chain 3 (LC3)Minute (min)Long-wave cones (L-cones)Lecithin retinol acyltransferase (LRAT)Middle-wave cones (M-cones)Microtubule organizing center (MTOC)The mammalian target of rapamycin (mTOR)Sodium butyrate (NaBu)Optimal cutting temperature medium (O.C.T.)Outer nuclear layer (ONL)Outer plexiform layer (OPL)Outer segment (OS)p62 (nucleoporin 62)Parkinson’s disease (PD)Phospherdiesterase (PDE)The phosphodiesterase 6b gene (PDE 6b gene)Phosphatidylethanolamine (PE)The protein kinase R-like ER kinase (PERK)The class III phosphoinositide 3 kinase (PI3K)Phosphatidylinositol 3-phosphate (PI3P)The pleckstrin homology domain-containing family M member 1 (PLEKHM1)meta-rhodopsin II (R*)The ras-related protein rab-7 (Rab7)xviiiRetinol dehydrogenase (RDH)The retinal degeneration 10 mouse (rd10 mouse)Restriction enzyme mediated integration (REMI)The red fluorescent protein (RFP)Regulator of G-protein signalling 9 (RGS9)The rhodopsin gene (RHO)Rod photoreceptor cell (Rod)Room temperature (RT)The rod outer segment (ROS)Retinitis Pigmentosa (RP)Retinitis pigmentosa 1 (RP1)Retinal pigment epithelium (RPE)Retinal pigment epithelium-specific 65 kDa protein (RPE65)Retinitis pigmentosa GTPase regulator (RPGR)Short-wave cones (S-cones)small guide RNA (sgRNA)Synaptosome associated protein 29 (SNAP29)The autophagosomal soluble N-ethylmaleimide-sensitive factor (syntaxin 17)Transmission electron microscopy (TEM)The transcription factor EB (TFEB)unc-51 like autophagy activating kinase 1 (ULK1)The unfolded protein response (UPR)Vesicle-associated membrane protein 8 (VAMP8)xixValproic acid (VPA)Wheat germ agglutinin (WGA)The WD repeat domain phosphoinositide-interacting protein (WIPI)The X-box binding protein 1 (XBP1)Xenopus laevis (X. laevis)X. laevis rod opsin promoter (XOP)3 Dimentional (3D)3-methyladenine (3-MA)xxAcknowledgementsI first want to thank my supervisor Dr. Orson Moritz for his enormous patience andgenerous help during my studies in the Moritz Lab in the past five years. When I firstentered the Moritz Lab, I did not speak English fluently. Dr. Moritz taught me how to doexperiments himself in the first year and gave me lots of confidence during my studies.He is a real gentleman with lots of good personality traits, I feel so lucky to have him asmy supervisor.I want to thank my committee member Dr. Robert Molday, Dr. Cheryl Gregory-Evansand Dr. Sharon Gorski. Bob is a world famous scientist; it is my pleasure to have him onmy supervisory committee. He always brought up questions beyond what I was thinking,such as questions related to the future of visual science. He shines light on my researchroad. I also have two female scientists on my committee, Cheryl and Sharon. Cheryl isvery elegant, and she always replies to my emails. She edited my research proposaland thesis very carefully. Sharon is friendly, considerate and warm-hearted. She gaveme lots of help on autophagy study. They set good examples to me of how elegant andprofessional a scientist should be. I want to thank all my committee members again forcontributing time reading and editing my thesis, they brought up lots of good questionsand suggestions to me. Thanks for all your help.Special thanks to my colleague Dr. Brittany J. Carr who helped me edit my thesis.Although I did not overlap with Brit for very long, I can tell she is a warm-hearted person.xxiThanks to my colleague Colette Chiu who takes care of our frogs in the lab and helpedme practice my English in the first year of my study. She is a warm-hearted person andshe treats everyone so well. Thanks to my colleagues Paloma Stanar and RuanneVent-Schmidt who helped me with some of the experiments in this thesis. Thanks to myother colleague Beatrice M Tam, and my former colleagues Tami H. Bogea and JoannaM. Feehan. Beatrice is the lab manager, she is willing to share her broad knowledgewith everyone else in the lab. Tami is warm-hearted, she always answers my questionspatiently. Special thanks to Joanna, she gave me so much confidence and we had somuch fun together during our studies in Moritz lab.I want to thank my family members, especially my parents who always support andback up me in every stage of my life. It was initially not easy for me to separate fromthem and begin my study in another country where I did not know anyone. Many timesvideo calls to the other side of the ocean helped me through those difficult times. Alsothanks to my grandma and my boyfriend Yue Huang, thanks for their supports.Finally, I want to thank Lin’s Chinese restaurant on Broadway Vancouver, which is veryclose to the Moritz lab. I want to thank this restaurant as it offered over 200 lunches tome during the last 5 year’s study.Thanks again to all the above mentioned people, thank you so much.xxiiDedicationThis thesis is dedicated to my dad and mom, without them I would not be here.11. Introduction1.1 The Retina:The retina is a thin layer of light-sensitive tissue located at the back of the eye thatconverts light into electrical signals. These signals are sent to the visual cortex of thebrain through the optic nerve where they are processed into visual perception. Retinalstructures and functions are highly conserved among vertebrates. The retina iscomposed of diverse neuronal cell types, including photoreceptor cells, three classes ofinterneurons (horizontal cells, amacrine cells and bipolar cells), ganglion cells and twotypes of supporting cells: Müller glia and retinal pigment epithelium (RPE) cells. Thesecells are arranged in an orderly fashion in the retina forming seven retinal layers: theretinal pigment epithelium (RPE) cell layer, the photoreceptor cell layer, the outernuclear layer (ONL), the outer plexiform layer (OPL), the inner nuclear layer (INL), theinner plexiform layer (IPL) and the retinal ganglion cell layer (GCL) (Figure 1A). Lightenters the retina from the GCL and passes through the other intermediate retinal layersbefore reaching the photoreceptor cell layer (Masland 2012).2Figure 1: Structure of the retina. (A): A confocal micrograph of a X. laevis retina sectionshowing different retinal layers. (B): Diagram of the neural circuit of the retina, showingthe six neuronal cell types and the two supporting cell types (Müller glia and retinalpigmented epithelium). In A, the scale bar =50 μm. Modified from Gramage, Li, &Hitchcock, 2014.Retinal circuitry is commonly broken into “vertical” and “lateral” visual processingpathways. The vertical pathway is photoreceptor cells-bipolar cells-ganglion cells; lateralpathways are the horizontal cells and the amacrine cells, which fine-tune signals comingfrom the photoreceptor cells and the bipolar cells before they reach the ganglion cells(Figure 1B).Photoreceptor cells are specialized retinal cells responsible for detecting light as the firstevent in vision and converting light into electrical signals. There are two types ofphotoreceptor cells in the retina, the rod photoreceptor cells (rods) and the cone3photoreceptor cells (cones), responsible for scotopic and photopic vision respectively(Molday & Moritz 2015). RPE cells are the pigmented layer of the retina immediatelyadjacent to the photoreceptor cells and provide nutrients to photoreceptor cells; they arevital for photoreceptor cell survival. RPE cells also have many other functions, such asabsorption of scattered light to diminish photo-oxidative stress and improve acuity,transportation of ions and fluid, conversion and storage of retinoid, phagocytosis ofphotoreceptor outer segments, and contributing to the blood-retina barrier (Sparrow etal. 2010). Horizontal cells belong to the lateral visual processing pathways, with theircell bodies located in the outermost third of the INL and dendrites spreading laterallyacross the OPL where they connect with the synapses of the photoreceptor cells. Theyprovide inhibitory feedback to photoreceptor cells and help adjust the output signalsfrom the photoreceptor cells, making our eyes see well under both bright and dim lightconditions, and contribute to the perception of contrast (Masland 2012). Bipolar cells arelocated between photoreceptor cells and ganglion cells, and transfer signals fromphotoreceptor cells directly or indirectly (through amacrine cells) to ganglion cells(Masland 2012). Amacrine cells, (meaning “short fiber” cells in Greek), belong to thelateral visual processing pathways, with their cell bodies located in the innermost third ofthe INL and dendrites spreading laterally across the IPL where they connected withbipolar cells and/or ganglion cells. They affect the output signals from bipolar cells, andcan send signals received from bipolar cells to ganglion cells (Masland 2012). Retinalganglion cells collect visual signals from bipolar cells or amacrine cells and send thesignals to the brain through the optic nerve as action potentials. They are the onlyretinal neurons that generate action potentials; other retinal neurons generate graded4potentials. Ganglion cells have very long axons which extend into the brain, comprisingthe optic nerve. Most ganglion cells function in collecting and transferring visual signalsfrom photoreceptors that can form visual perception. However, a subset of ganglioncells called intrinsically-photosensitive ganglion cells (ipRGCs) contain a photosensitivepigment called melanopsin. ipRGCs are not thought to contribute significantly to visualperception, but rather to contribute to entraining circadian rhythm by communicationwith the suprachaismatic nucleus in the hypothalamus, and to the pupillary light reflex(Yu et al. 2013; Sanes & Masland 2015). Müller cells, also called Müller glia, are theprinciple glial cells of the retina and are responsible for maintaining the homeostasis ofthe neural extracellular environment (Reichenbach & Bringmann 2013). Müller cells arethe largest cells in the retina and span across the entire inner retina. Besides Müllercells, there are also other glial cells in the retina, such as microglia.These different types of retinal neurons and other cells cooperate in forming visualperception (detecting light, transferring light to electrical signals, and conducting signalsto the brain) and in maintaining a healthy and stable retinal environment. I will focus onphotoreceptor cells in this thesis.51.2 Photoreceptors:The human retina contains approximately 120 million rods and 6 million cones, the ratioof the number of rods to cones is about 20 to 1. Rods are located mainly in the outer/peripheral retina with reducing density in the central retina known as the macula; thereare no rods in the centre of the macula, known as the fovea. Rods are extremelysensitive to light and can generate an electrical response to a single photon, allowingvisual responses under dim light conditions, thus rods are responsible for night vision.However, their electrical responses saturate under bright light conditions. There is onlyone type of rod in the human retina, containing a pigment called rhodopsin, a photo-sensitive protein responsible for detecting light in rods (Molday & Moritz 2015). Rodsrespond to a broad spectrum of light (from 400 nm to 600 nm), with a peak sensitivity at498nm, and are not sensitive to light with wavelength longer than 640 nm (Bowmaker &Dartnall 1980).In the human retina, cones are predominantly located in the central retina or macula;they are less efficient at transmitting single-photon responses. Unlike rods, which canbe saturated by intense light, cones do not saturate. Thus bright daylight saturates rodsand daytime vision is mostly produced by cones. Cones also respond to light faster thanrods, making them more effective at detecting movement (Mustafi et al. 2009). Conesare also responsible for color vision. There are three different types of cones in thehuman retina: the short-wave cones (S-cones), the middle-wave cones (M-cones) andthe long-wave cones (L-cones), each with different pigments called cone opsins. Conepigments differ from the rod rhodopsin pigment in that they respond to different6wavelengths of light. S-cones, also called blue cones, respond to short wavelengths ofvisible light, with a peak absorbance at 420 nm. M-cones, also called green cones,respond to middle wavelengths of visible light, with a peak absorbance at 534 nm. L-cones, also called red cones, respond to long wavelengths of visible light, with a peakabsorbance near 564 nm (Figure 2) (Bowmaker & Dartnall 1980). Due to their origin in arecent gene duplication event, the amino acid sequences of the human red and greencone pigments share high homology with changes in only a few amino acids, whichcause the differences in light absorption spectrum (J. Nathans et al. 1986). Thedifferences in signal intensities received from the three types of cones allow theperception of color vision (Kalloniatis & Luu 2005; Collin & Trezise 2004). For example,blue color is perceived when S-cones are predominantly stimulated, red color isperceived when L-cones are predominantly stimulated and yellow color is perceivedwhen L-cones are stimulated slightly more than M-cones. Mutations in genesresponsible for encoding cone pigments cause color deficiency or color blindness (JNathans et al. 1986; Deeb & Kohl 2003). There are also inner-retinal and corticalpathways responsible for color-specific visual processing; I will focus on photoreceptorcells in this thesis.7Figure 2: The mean absorbance spectra of human rods and cones. Rods respond to abroad spectrum of light (from 400 nm to 600 nm), with a peak absorbance at 498 nm. S-cones, also called blue cones, respond to short wavelengths of visible light, with a peakabsorbance at 420 nm. M-cones, also called green cones, respond to middlewavelengths of visible light, with a peak absorbance at 534 nm. L-cones, also called redcones, respond to long wavelengths of visible light, with a peak absorbance at 564 nm(Bowmaker & Dartnall 1980).Rods have low spatial resolution, while cones contribute to high visual acuity. In humanor other simian primates, a structure called fovea in the central retina contains a highdensity of cones only and mediates high acuity and color vision. Both rods and conesare connected to their secondary retinal neurons, the bipolar cells. There areapproximately 12 types of cone bipolar cells, but only one type of rod bipolar cell.Although there are many more rods than cones in the retina, cone bipolar cells8outnumber rod bipolar cells (Strettoi et al. 2010). As a consequence, cone signals haveto diverge to access the cone bipolar cells. For example in the primate fovea, each coneis connected to 3 or 4 cone bipolar cells (Ahmad et al. 2003), so that responses fromdifferent cones can be distinguished easily. Multiple rods converge on each rod bipolarcell, making responses from different rods harder to distinguish.Rods and cones have similar structures; they can be compartmentalized into fiveregions: the outer segment (OS), the connecting cilium (CC), the inner segment (IS), thenuclear region, and the synaptic region (Figure 3) (Molday & Moritz 2015). The OS is ahighly modified non-motile cilium responsible for detecting light and converting light intoelectrical signals. The rod OS is cylindrical in shape and consists of a plasmamembrane and thousands of stacked membranous disks; each disk is closed andseparated from other disks and the plasma membrane. In contrast, the cone OS isconical in shape. Overall, the structures are similar to those in rods, but cone disks arecontinuous with each other and with the plasma membrane (Steinberg et al. 1980;Mustafi et al. 2009). The mammalian rod OS is typically 20-30 µm long and 1.2-2.0 µmin diameter (Molday & Moritz 2015; Nickell et al. 2007). The frog (bullfrog) rod OS isabout 60 μm in length and 6-8 μm in diameter (Bownds & Brodie 1975), which is about3 times the size of mammalian rod OS. The OSs of rods and cones are continuouslyrenewed by adding newly generated disks to the basal OS and shedding off the distalOS which are phagocytosed by the RPE cells (Bm M Kevany & Palczewski 2010). TheIS contains the cell’s metabolic and biosynthetic machinery, housing multiple subcellularorganelles including mitochondria, endoplasmic reticulum (ER), Golgi complex,9ribosomes and lysosomes. The CC connects the OS and the IS, and proteinssynthesized in the IS are trafficked to OS through the CC. The nuclear region housesthe nucleus and is connected with the IS. The synaptic region contains a ribbonsynapse, and synaptic vesicles that release the neurotransmitter glutamate to stimulatebipolar and other secondary neurons in the retina (Molday & Moritz 2015).Figure 3: Schematic diagram of vertebrate rods and cones. Rods and cones havesimilar structures; they can be compartmentalized into five regions: the outer segment10(OS), the connecting cilium (CC), the inner segment (IS), the nucleus, and the synapticregion (Cote R.H., 2006).X. laevis is a useful model system for visual research, as they have large photoreceptorcells compared to those found in mammalian retinas (Moritz et al. 1999), allowingvisualization of details of some subcellular structures or photoreceptor proteinlocalizations under a confocal microscope. There are two types of rods in the Xenopusretina, the green rods and the red rods, named after their color observed axially under abrightfield microscope. The red rods are maximally sensitive to green light, accountingfor about 97% to 98% of the total rod population. The green rods are rare in theXenopus retina, and they are thinner and sensitive to blue light. There are three types ofcones in the Xenopus retina, the majority of them are sensitive to red light, with a smallproportion of them sensitive to blue light; the third type of cones in Xenopus are UV-sensitive cones, which are the least abundant (Röhlich & Szél 2000). The structures ofXenopus rods and cones are generally similar to that of human, with distinctly shapedouter segments comprised of membranous discs. In the human retina, the ratio of thenumber of rods to cones is about 20 to 1 (as mentioned above in the beginning of thissection), although this ratio varies dramatically across the retina. X. laevis has a muchhigher ratio of cones, making it a good system for studying cone photoreceptors(Röhlich & Szél 2000). Many of the genes involved in phototransduction andphotoreceptor structure are conserved between human and X. laevis. X. laevis arerelatively easy to genetically manipulate, so that we can model human retinal diseases11using X. laevis, such as retinitis pigmentosa (RP) (Tam & Moritz 2007a). There are alsosome differences between the human and X. laevis retina. For example, the humanretina has a structure called the fovea which is composed of high density of conephotoreceptors, but X. laevis does not have a fovea (Röhlich & Szél 2000). Anotherdifference is that the human retina has multiple rows of nuclei in the outer nuclear layer(ONL), while Xenopus only has only a single row of nuclei in the ONL (Röhlich & Szél2000). In this thesis, all the experiments were performed on X. leavis.121.3 Rhodopsin:Rod OS disk membranes contain very high levels of the photosensitive protein rodopsin, which is covalently coupled to a chromophore, 11-cis retinal; the complex isreferred to as rhodopsin. Rhodopsin is responsible for detecting light and convertinglight into electrical signals. Rhodopsin comprises over 85% of the disk membraneproteins with a density of over 30,000 rhodopsin molecules per square micrometer(Fotiadis, Liang, Slawomir Filipek, et al. 2003). Atomic force microscopy studies ofmouse photoreceptors suggest that rhodopsin is arranged in higher order oligomericstate in the native disk membranes, forming structural dimers that are organized inparacrystalline arrays (Fotiadis, Liang, Sławomir Filipek, et al. 2003; Liang et al. 2003).Recent cryoelectron tomography studies demonstrated that, in mouse rodphotoreceptors, rhodopsin forms dimers and at least 10 dimers form a row; rows formtracks (pairs) that are aligned parallel to the disk incisures (Gunkel et al. 2015). Rod OSare continuously renewed, necessitating continuous production of large quantities ofdisk membranes and rhodopsin (Bm M Kevany & Palczewski 2010).Rhodopsin belongs to the G-protein coupled receptor (GPCR) family which transducesextracellular stimuli into intracellular signals by activation of a coupled G protein. Thehuman GPCR family, including more than 800 members, is involved in a variety ofphysiological processes including vision, olfaction, and hormonal signaling. They shareconserved structural features, with seven transmembrane alpha helices connected bysix loops of varying lengths (Fredriksson 2003; Kobilka 2007; Zhou et al. 2012). The N-13terminal region of rhodopsin is located extracellularly and the C-terminal region iscytoplasmic (Palczewski 2006). Rhodopsin is glycosylated at Asn-2 and Asn-15 at its N-terminus (Fukuda et al. 1979). A conserved disulfide bond is formed between Cys-110and Cys-187 to connect helices IV and V in the extracellular region (Filipek et al. 2003).The chromophore is attached to the protein opsin via lysine residue 296 (Palczewski2006).Rhodopsin was the first GPCR with its three-dimensional (3D) structure to be solved byX-ray crystallography. The crystal structure of ground-state bovine rhodopsin(containing the chromophore 11-cis retinal) was first determined at a resolution of 2.8angstroms by Palczewski in 2000 (Figure 4), showing the major molecular structuralfeatures of GPCRs. The seven transmembrane helices are aligned roughlyperpendicular to the membrane plane. The ground-state chromophore 11-cis retinal isburied in the transmembrane region of the protein opsin (Palczewski et al. 2000); laterwith new crystallization conditions, the resolution of bovine rhodopsin was improved to2.2 angstroms and more details of the chromophore binding site were provided. Thechromophore covalently binds to the seventh transmembrane domain of opsin through aprotonated Schiff base linkage at lysine residue 296 (Okada et al. 2004). Thus far, thecrystal structures of activated rhodopsin and opsin, with and without the binding ofcoupled G protein, have all been published (Park et al. 2008; Scheerer et al. 2008;Choe et al. 2011).14Figure 4: Structure of rhodopsin. (A) Crystal structure of bovine rhodopsin covalentlylinked with 11-cis retinal. (B) “Beads on a string” diagram for bovine rhodopsin. EC:extracellular. IC: intracellular (Brogi et al. 2014).151.4 Phototransduction:Phototransduction, the process by which absorption of a photon is converted into anelectrical signal, is generally similar in rods and cones, although there are cone-specificisoforms of many of the proteins involved; in this section, details of thephototransduction cascade in rods are reviewed.In the dark, rods have high levels of cyclic guanosine monophosphate (cGMP), whichkeeps cyclic-nucleotide-gated (cGMP-gated) Na+ channels located in the plasmamembrane open, allowing an inward current of Na+ that depolarizes the rods. Thedepolarization of the cell membrane opens voltage-gated calcium channels allowinginflux of Ca2+. Increased Ca2+ concentration facilitates the fusion of glutamate-containingvesicles to the synaptic membranes, causing the release of the neurotransmitterglutamate, which inhibits the excitation of other secondary neurons.Upon light illumination, the rhodopsin chromophore 11-cis retinal absorbs a photon andundergoes isomerization to all-trans retinal. The isomerization of chromophore triggersa conformational change of the opsin protein, resulting in a photoactivated rhodopsin ormeta-rhodopsin II (R*). R* activates its coupled heterotrimeric G protein calledtransducin, which is composed of alpha, beta and gamma subunits. The activation oftransducin facilitates its bound GDP being exchanged by GTP. The alpha subunit oftransducin together with its bound GTP then dissociate from the beta and gammasubunits and activate phosphodiesterase (PDE), which cleaves cGMP into 5’-GMP.16Reduced cGMP levels cause the closure of cGMP-gated sodium channels andhyperpolarize the rods (Figure 5). The hyperpolarization of the cell membrane leads tothe closure of voltage-gated Ca2+ channels and a decrease in Ca2+ concentration, whichcauses reduced release of glutamate from rods and reduces inhibition of other retinalneurons (Arshavsky et al. 2002).Each step of phototransduction is then terminated: 1) photoactivated rhodopsin isphosphorylated by rhodopsin kinase (a G-protein coupled receptor kinase). 2)phosphorylation of photoactivated rhodopsin facilitates its binding to arrestin, whichblocks further G protein-mediated signaling (Figure 5). Additionally, R* is unstable, andthe bound all-trans retinal rapidly dissociates from the rod opsin (Arshavsky 2002; Burns& Pugh 2010). Transducin is inactivated by hydrolysis of the bound GTP to GDP, aprocess facilitated by the transducin GTPase activating protein regulator of G-proteinsignaling 9 (RGS9), leading to loss of phosphodiesterase binding and activation. TheRGS9 is in complex with G protein 5 and RGS9 associated protein (R9AP) (Veleri et al.2015).17Figure 5: Schematic view of major proteins involved in phototransduction.During phototransduction (black arrows), the rhodopsin chromophore 11-cis retinalabsorbs a photon, resulting in a photoactivated rhodopsin (R*). R* activates its coupledG protein called transducin, which further activates phosphodiesterase (PDE), cleavingcGMP into 5’-GMP. Reduced cGMP levels cause the closure of cGMP-gated sodiumchannels (CNG) and hyperpolarize the rods. At low Ca2+ levels, guanylyl cyclaseactivating proteins (GCAP) activates guanylate cyclase (GC) to restore cGMP levels,leading to re-opening of CNG (red arrows). Termination of phototransduction (redarrows and T bars) is mediated by phosphorylation of activated rhodopsin by rhodopsinkinase (GRK), facilitating the binding of arrestin to rhodopsin. In the dark, rod cellmembranes are depolarized, allowing the influx of Ca2+. At high intracellular Ca2++levels, recoverin inhibits GRK. Please see main text for details (Veleri et al. 2015).All-trans retinal is not photoactive and therefore it must be converted back to 11-cisretinal to regenerate rhodopsin. This process is known as the visual cycle. The all-trans18retinal is first reduced to all-trans retinol by an enzyme called all-trans retinoldehydrogenase (RDH 12) in rods. Retinol dehydrogenase (RDH) belongs to theoxidoreductase family, which catalyzes the transfer of electrons from reductant to theoxidant, and normally uses NAD+ as a cofactor. RDH can also be called “retinalreductase”. All-trans retinol is transferred to RPE cells and esterified by an enzymecalled lecithin retinol acyltransferase (LRAT) to all-trans retinyl ester, which is thenconverted to 11-cis retinol by a retinoid isomerohydrolase called RPE65 (retinal pigmentepithelium-specific 65 kDa protein). 11-cis retinol is oxidized to 11-cis retinal by 11-cisretinol dehydrogenase (RDH 5) and travels back to rod OS where it is re-conjugated toopsin (Figure 6) (Kefalov 2012).19Figure 6: The mammalian visual cycle. Upon light illumination, the rhodopsinchromophore 11-cis retinal absorbs a photon and undergoes isomerization to all-transretinal. The all-trans retinal is first reduced to all-trans retinol by an enzyme called all-trans retinol dehydrogenase (RDH 12) in rods. All-trans retinol is transferred to RPEcells and esterified by an enzyme called lecithin retinol acyltransferase (LRAT) to all-trans retinyl ester, which is then converted to 11-cis retinol by a retinoidisomerohydrolase called RPE65 (retinal pigment epithelium-specific 65 kDa protein).11-cis retinol is oxidized to 11-cis retinal by 11-cis retinol dehydrogenase (RDH 5) andtravels back to rod OS where it is re-conjugated to opsin (Xue et al. 2004).201.5 Retinitis Pigmentosa:Retinal degenerative disorders are a heterogeneous group of inherited degenerations ofthe retina, with more than 100 genes implicated so far. Retinitis pigmentosa (RP) is asubset of inherited retinal degenerations characterized by progressive loss of retinalrods and then cones. Patients with RP typically lose night vision first, due to rod death,and then progressively lose peripheral and central vision due to rod death followed bythe secondary death of cones (Figure 7). Many patients are severely impaired or blindbefore midlife. Other symptoms may include tunnel vision and defects in colordiscrimination. The worldwide prevalence of RP is about 1 in 4000, with more than 1million people affected; currently there is no cure for RP (Hartong et al. 2006).21Figure 7: The comparison of a normal retina to a retina affected by retinitis pigmentosa(RP). The confocal microscopy image on the left showed a normal retina section, theimage on the right showed a human retina section from an individual with RP. Red:Nuclei. Green: rods (micrograph provided by Dr. Robert N. Fariss).RP exhibits several types of heterogeneity, including genetic heterogeneity, allelicheterogeneity, phenotypic heterogeneity and clinical heterogeneity. RP exhibits (1)genetic heterogeneity as many different genes can cause RP (e.g. RHO, GNAT1,RPE65); (2) allelic heterogeneity as many different mutations in the same gene cancause RP (e.g. P23H, T4K, T17M, Q344Ter rhodopsin); (3) phenotypic heterogeneity as22different mutations in the same gene may cause different diseases; (4) clinicalheterogeneity as the same mutation in different individuals (or even among members ofthe same family) may show different clinical consequences (Daiger et al. 2013).RP has four inheritance patterns: autosomal-dominant, autosomal-recessive, X-linkedand mitochondrial inheritance. RP can occur in non-syndromic forms, or in syndromicforms, such as Usher syndrome which is RP with congenital or early onset deafness.Mitochondrial RP is also syndromic. Thus far, mutations in 56 genes are known tocause non-syndromic RP, with nearly 3100 disease-causing mutations reported (Daigeret al. 2013). These mutations can be substitutions, deletions or insertions that causemissense mutations or truncations. There are also mutations in other genes causingsyndromic RP. In this thesis, only non-syndromic RP (hereafter called RP) will bediscussed further. Mutations in 23 genes are reported to cause autosomal-dominant RP(adRP), 36 genes cause recessive RP and 3 genes cause X-linked RP (Daiger et al.2013). The currently known genes account for about 40% of all patients; the remaindermay have defects in unidentified genes (Maubaret & Hamel 2005).Most identified genes account for only a small proportion of RP cases, with the RHO(rhodopsin), RP1 (retinitis pigmentosa 1) and RPGR (retinitis pigmentosa GTPaseregulator) genes contributing to the greatest number of known RP-causing mutations(Wang et al. 2005). Mutations in the rhodopsin (or RHO) gene are the most commoncause of RP, and account for about 30% of all adRP (Hartong et al. 2006). The majorityof these mutations cause destabilization of the rhodopsin molecule and defects in23biosynthesis (Chen et al. 2014; Tam & Moritz 2006; Tam & Moritz 2009). Rhodopsinmutations have been characterized into two classes, Class I mutations and Class IImutations, based on their biochemical and cellular properties. Class I mutationsresemble wildtype rhodopsin in their expression levels (high yield) and can regeneratewith the chromophore 11-cis retinal; examples include G51V, V345M and P347S. ClassI mutations are normally correctly folded and accumulate in the plasma membranewhen expressed in cell culture. Class II mutations are usually misfolded and havedefects in chromophore assembly. They are produced at a lower levels and accumulatein the ER; examples include P23H (Sung et al. 1993). In addition, more complexclassification schemes have been proposed that more specifically categorize theunderlying biochemical defects (Athanasiou et al. 2017).241.6 P23H rhodopsin:The most commonly occurring cause of RP in North America is the rhodopsin mutationP23H, a single amino acid-substitution of proline to histidine at codon 23 of therhodopsin gene; it was also the first RP mutation to be identified (Dryja et al. 1990;Hartong et al. 2006). P23H rhodopsin causes retinal degeneration when expressed inmultiple species, including mouse (Mus musculus) (Jane E. Olsson et al. 1992), rat(Rattus norvegicus) (Machida et al. 2000), frog (Xenopus laevis) (Tam & Moritz 2006)and fruit fly (Drosophila melanogaster) (Galy et al. 2005). Researchers have developedvarious animal models to study the phenotypes and disease-causing mechanisms of RPassociated with P23H rhodopsin. There are P23H rhodopsin transgenic mouse models(J E Olsson et al. 1992; Roof et al. 1994; Goto et al. 1995), a P23H rhodopsin knock inmouse model (Sakami et al. 2011) a P23H-rhodopsin-GFP knock-in mouse model(Price et al. 2011), a transgenic P23H rhodopsin rat model (Machida et al. 2000) andtransgenic P23H rhodopsin frog models (Tam & Moritz 2006). Photoreceptordegenerations with shortened OS and thinned ONL were observed in P23H rhodopsinknock-in mice and P23H rhodopsin transgenic rats (Sakami et al. 2011; Machida et al.2000). Transgenic X. laevis with a P23H rhodopsin mutation also exhibit shortenedphotoreceptor OS, condensed/ pyknotic nuclei in the ONL, and gaps in thephotoreceptor cell layers indicating death of rods (Tam & Moritz 2006).It is generally thought that P23H rhodopsin causes misfolding of rhodopsin molecules(Illing et al. 2002; Kaushal & Khorana 1994; Tam & Moritz 2006). Interestingly,25biosynthesis of this rhodopsin mutant is significantly enhanced in the presence of the11-cis retinal chromophore, a pharmocological chaperone for the rod opsin molecule(Noorwez et al. 2004). Patients with this mutation often exhibit a phenotype calledsector RP in which the lower retina (which generally receives more light from overheadsources) is more affected (Lanum 1978). Animal models, including X. laevis models,exhibit retinal degeneration that is exacerbated by light (Wang et al. 1997; Tam et al.2010). These phenotypes are likely due to reduced availability of chromophore in lightexposed retinas, in which rod outer segments act as a catalyst for destruction of 11-cisretinal, thereby promoting misfolding of P23H rod opsin (Tam et al. 2010).261.7 How does P23H rhodopsin cause cell death in RP?Early studies in transgenic mice expressing different mutant rhodopsin genes that causeRP in humans suggest that the loss of photoreceptor cells is due to apoptosis, aprocess of programmed cell death in which nuclear DNA fragmentation is a key feature(Chang et al. 1993; Portera-Cailliau et al. 1994). Apoptosis is a highly controlled celldeath process, where the dying cells are fragmented and form structures calledapoptotic bodies that can be cleared by phagocytic cells, thus preventing the leaking ofcellular contents to the surrounding environment and reducing damage to other cells.More recently, cell signaling studies in P23H rhodopsin-transgenic rats suggest thatmitochondria-induced apoptosis is involved in the cell death mechanism of RP (Sizovaet al. 2014); another study using a fly model for P23H rhodopsin suggests that stress-induced apoptosis is involved in photoreceptor cell death (Galy et al. 2005).There are other studies suggesting that cell death in RP is mediated by a non-apoptoticmechanism. Instead, the key features of this proposed mechanism include theactivation of calpain and poly ADP-ribose polymerase (PARP), but how they govern thecell death is unclear (Arango-Gonzalez et al. 2014; Kaur et al. 2011; Viringipurampeeret al. 2016). A recent study showed that rod and cone cell death were governed bydifferent signaling pathways in P23H transgenic rat (Viringipurampeer, 2016). Rod celldeath occurred via necroptosis, a caspase-independent regulated form of necrosis, withup-regulation of RIP1/RIP3/DRP1 in rods. Cone cell death occured via pyroptosis, ahighly-inflammatory form of necrosis, with up-regulation of NLRP3 and caspase-1 in27cones and expression of mature IL-1β and IL-18 inducing inflammation in the retina(Viringipurampeer, 2016).The exact mechanism by which P23H rhodopsin causes cell death in RP is not clearlyunderstood, and several proposed mechanisms are reviewed below.1.7.1 Gain of function:P23H rhodopsin is a Class II misfolded mutant and is proposed to cause RP via gain-of-function mechanism (Illing et al. 2002). Various cell culture and transgenic animal modelstudies suggest that P23H rhodopsin is misfolded and prone to form dimers and/orhigher molecular weight oligomers, which are retained within ER (Kaushal & Khorana1994; Illing et al. 2002; Tam & Moritz 2006; Chiang et al. 2012). Aggresomes wereobserved in cell cultures but not reported in transgenic animal models (Illing et al. 2002;Tam & Moritz 2006). Aggregation is a gain of function for P23H rhodopsin (Illing et al.2002). The retention of mutant rhodopsin within ER is likely due to ER protein qualitycontrol mechanisms (Saliba et al. 2002), which may induce ER stress and activate theunfolded protein response (UPR).The UPR is a set of signaling pathways induced by ER stress, which initially alleviatesER stress and promotes cell survival by reducing gene translation, enhancing proteinfolding and increasing the degradation of misfolded proteins; when cells are undergoingprolonged ER stress, the UPR can also trigger apoptotic cell death. The UPR ismediated by three ER stress sensors located on ER membranes. They are the inositol-requiring enzyme1 (IRE1), the activating transcription factor (ATF6) and the protein28kinase R-like ER kinase (PERK). Under ER stress, the ER chaperone protein BiP(binding immunoglobulin protein) is titrated from its binding sites on the ER stresssensor proteins IRE1, ATF6 and PERK by competitive binding to misfolded proteins,resulting in the activation of these stress sensors. Activated IRE1 catalyzes the splicingof the X-box binding protein 1 (XBP1) mRNA, resulting in production of functionaltranscription factor XBP1, which upregulates the expression of genes encoding ERchaperones and components of ER associated degradation (ERAD). ATF6 is cleavedand forms an active transcriptional factor cATF6, which translocates into the nucleusand enhances transcription of several ER chaperones, which are helpful for proteinfolding, degradation and ER expansion. PERK inhibits the eukaryotic translation-initiation factor 2 (eIF2) by phosphorylation and reduces global protein translation,except for certain selected mRNAs, such as the activating transcription factor 4 (ATF4),which induce the expression of pro-apoptotic factor CHOP (transcriptional factor C/EBPhomologous protein). CHOP can induce the expression of genes supporting cell death(Sano & Reed 2013).Mammalian HEK293 cell culture studies suggest that the IRE1 branch of UPRselectively promotes the degradation of P23H rhodopsin, but not wildtype rhodopsin,through both proteasomal and lysosomal pathways (Chiang et al. 2012). Animal studiessuggest that the UPR is activated in transgenic rats expressing a P23H rhodopsinmutation, and that the level of the pro-apoptotic factor CHOP is elevated underprolonged ER-stress, which may induce photoreceptor cell death (Lin et al. 2007).Other studies suggest that over-expression of the ER chaperone protein BiP promotes29photoreceptor cell survival and preserves retinal function in P23H transgenic rats, andthese effects are likely through the modulation of the UPR (Gorbatyuk et al. 2012).1.7.2 Proteasome overload:The very high levels of rhodopsin expression results in gain of function phenotypesassociated with rhodopsin misfolding mutants that likely cause cell death via overload ofcellular proteostasis mechanisms (Lobanova et al. 2013). The ubiquitin-proteasomesystem and autophagy-lysosome pathway are the two major mechanisms for degradingcellular proteins. It is well established that P23H rhodopsin is a substrate of ubiquitin-proteasome pathway (Illing et al. 2002; Saliba et al. 2002). However cell culture studiessuggest that the function of ubiquitin-proteasome system is impaired by P23H rhodopsinaggregation (Illing et al. 2002). Proteasome overload and consequent insufficientcapacity to degrade misfolded proteins likely cause cell death in transgenic mice with aP23H rhodopsin mutation (on the Rho+/− background) (Lobanova et al. 2013).1.7.3 Dominant negative effect:Besides toxic gain-of-function mechanisms, a dominant negative effect has also beenproposed in RP, supported by both cell culture and transgenic animal studies. Co-transfection of normal and mutant P23H rhodopsin in COS-7 cell cultures leads to therecruitment of the normal rhodopsin to aggresomes containing mutant rhodopsin (Salibaet al. 2002). In a transgenic mouse model with a P23H rhodopsin mutation,30photoreceptor degeneration was ameliorated upon overexpression of additional wildtyperhodopsin (Mao et al. 2011; Price et al. 2012), which suggests a dominant negativeeffect is involved. A mixture of both toxic gain-of-function mechanism and dominantnegative effects has also been suggested as the mechanisms underlying photoreceptordeath in RP (Mendes & Cheetham 2008).311.8 Autophagy:The term “autophagy” is derived from the Greek, and means “self-eating”. It was firstcoined by Christian de Duve in 1967 when he observed mitochondria and other intra-cellular structures were degraded within lysosomes of rat liver cells (Deter & De Duve1967). Since then, many discoveries have been made, and the study of autophagy hasdeveloped into an entire field. Autophagy is a lysosomal turnover process involved inthe degradation of cytoplasmic proteins, damaged organelles and other dysfunctionalcellular components, and releasing small molecules such as amino acids for recycling(Mizushima 2007). It is a normal and conserved process in eukaryotic cells that playsessential roles in maintaining cellular homeostasis. All cells have a basal autophagylevel that plays a role in tissue development and cellular house-keeping, such as theremoval of damaged organelles. Autophagy can also be induced by nutrient starvationand various forms of cellular stress, such as hypoxia, reactive oxygen species, DNAdamage, damaged organelles, and protein aggregates (Mizushima 2007). Nitrogenstarvation is the most potent autophagy stimulus in yeast, and lack of other essentialnutrients, such as carbon, amino acids and nucleic acids can also induce autophagy,but less efficiently. Nitrogen or carbon starvation can also induce autophagy in plantcells. Amino acid and insulin/growth factor starvation are common triggers of autophagyin mammalian cell cultures (Mizushima 2007; Glick et al. 2010). During starvation,autophagy breaks down large cellular molecules to maintain cellular energy levels andpromote cell survival.32Autophagy is implicated in many physiological and pathophysiological processes,including tissue development, starvation adaption, clearance of damaged proteins ororganelles, elimination of pathogens or microorganisms, cell death, anti-aging andtumor suppression. It also plays important roles in multiple human diseases, such ascancers, neurodegenerative diseases, diabetes, infectious and inflammatory diseases(Mizushima 2007; Yang et al. 2013).1.8.1 The subtypes of autophagy:There are several subtypes of autophagy, which are defined based on how the cargosare selected and delivered to lysosomes for degradation. The major three types includemacroautophagy, microautophagy and chaperone-mediated autophagy (Boya et al.2013).Macroautophagy is mediated by the formation of a unique double-membrane vesiclecalled an autophagosome, which engulfs a portion of cytoplasm and is transportedalong microtubules to fuse with a lysosome (Figure 8) (Boya et al. 2013). It wasoriginally thought that macroautophagy is non-selective as it engulfs a portion ofcytoplasm into the autophagosomes. But now many examples demonstrate thatmacroautophagy can be selective on substrates via adaptor proteins (Johansen &Lamark 2011). Substrates bind to the inner surface of the phagophore through adaptorsassociated with both the substrate and LC3 proteins (Zaffagnini & Martens 2016). Forexample, mitophagy, a selective form of macroautophagy, is the selective degradationof damaged or defective mitochondria by lysosomes (Lemasters 2005). Normally theterm “autophagy” refers to “macroautophagy”, unless otherwise specified.33Microautophagy is mediated by direct lysosomal membrane invagination to engulfcytoplasmic contents (Figure 8) (Boya et al. 2013). It is generally a non-selectivepathway. Chaperone-mediated autophagy (CMA) is a selective form of autophagymediated by a chaperone called heat shock cognate protein of 70 kDa (hsc70). Solublecytosolic proteins with a KFERQ motif are recognized by hsc70 and targeted to thelysosome surface, where the cargo protein-chaperone complex binds to a receptorcalled lysosome-associated membrane protein type 2A (LAMP-2A). LAMP-2A thenguides the cargo protein to unfold and translocate directly across the lysosomalmembrane for degradation (Figure 8). There is no formation of additional vesicles duringthis procedure (Boya et al. 2013). In this thesis, I mainly focus on macroautophagy.34Figure 8: Different types of autophagy. Macroautophagy is mediated by the formation ofa unique double-membrane vesicle called an autophagosome, which engulfs a portionof cytoplasm and is transported along microtubules to fuse with a lysosome.Microautophagy is mediated by direct lysosomal membrane invagination to engulfcytoplasmic contents. Chaperone-mediated autophagy (CMA) is mediated by achaperone hsc70. Soluble cytosolic proteins with a KFERQ motif are recognized byhsc70 and targeted to the lysosome surface, where LAMP-2A guides the cargo proteinto unfold and translocate directly across the lysosomal membrane for degradation(Mizushima & Komatsu 2011).351.8.2 Autophagy pathway:Macroautophagy (hereafter called autophagy) consists of 3 general steps, cargosequestration, autophagosome formation and fusion with lysosomes, and cargodegradation (Figure 9). Upon induction of autophagy, an isolation membrane (alsocalled a phagophore) is formed around a region of cytoplasm or selected substrates.The membrane origin of the phagophore is not clearly understood, but it is likely derivedfrom the lipid bilayers of the ER, the trans-Golgi network and/or endosomes (Axe et al.2008; Simonsen & Tooze 2009). A more recent study suggests that the outermembrane of the mitochondria is the main membrane source, and ER also contributesto the formation of isolation membrane (McEwan & Dikic 2010). A phagophoreelongates and turns in on itself to form a double-membrane vesicle called anautophagosome; bulk cytoplasm or other cargo is engulfed inside the autophagosome.The diameters of autophagosomes are typically between 500-900 nm in yeast(Takeshige et al. 1992), and 500-1500 nm in mammals (Mizushima et al. 2002) , and300-500 nm in X. laevis (based on my own observations). Autophagosomes aretransported along the microtubule tracks in a retrograde way to lysosome-enrichedareas, such as microtubule organizing centres (MTOCs), where they fuse withlysosomes and form structures called autolysosomes (Kimura et al. 2008). The cargosfrom autophagosomes are degraded by lysosomal acid hydrolases, yielding basicmetabolites, such as amino acids, into the cytoplasm through permeases onautolysosomal membrane for new synthesis or as a source of energy. In addition,autophagosomes can fuse with endosomes forming structures called amphisomes,which then fuse with lysosomes (Mizushima 2007; Glick et al. 2010). Amphisomes are36generally considered as autophagic precursor structures, and whether they have othercellular roles is not clear (Sanchez-Wandelmer & Reggiori 2013).Figure 9: The process of autophagy. Upon induction of autophagy, an isolationmembrane (also called a phagophore) is formed around a region of cytoplasm orselected substrates. Completion of the double membrane forms an autophagosomewhich is able to fuse with an endosome to form an amphisome. Fusion of anautophagosome or amphisome with a lysosome forms an autolysosome where thecargos are degraded by acid hydrolases from the lysosome (Hannigan & Gorski 2009).1.8.3 Molecular mechanism of autophagy:37Although autophagy has been extensively studied, most studies focuses on themorphology or functional analysis, little was known about its molecular mechanismsuntil Ohsumi’s group discovered the autophagy-related genes (Atg genes) (Tsukada &Ohsumi 1993). Since then, lots of discoveries were made. Autophagy is a dynamicprocess controlled by multiple genes. Genetic screening in yeast identified 32 differentAtg genes that are important for autophagy, and many of these genes are conserved inplants, insects and mammals (Nakatogawa et al. 2009). Among them, 18 Atg proteinsare involved in autophagosome formation, including Atg 1-10, Atg 12-14, Atg16-18,Atg29 and Atg31 (Mizushima 2007). In 2016, the importance of this discovery wasacknowledged by awarding of the Nobel prize in physiology to Y. Ohsumi.Microtubule-associated protein 1A/1B light chain 3 (MAP1LC3 or LC3), a mammalianhomologue of yeast Atg8, is stably and specifically associated with autophagicmembranes during the entire autophagy process, and thus it is widely used as a markerof autophagy. It is to date the only known protein present on both the outer and innermembranes of autophagosomes (Kabeya 2000). By tagging with fluorescent proteins,such as green fluorescent protein (GFP) or red fluorescent protein (RFP), LC3 has beenused extensively to monitor autophagy in cell culture or animal studies. Yeast Atg8 hassix homologs in mammals; They are divided into two subfamilies, the LC3 subfamily(including LC3A, LC3B and LC3C) and the GABARAP (Gamma-aminobutyric acidreceptor-associate protein) subfamily, including GABARAP, GABARAPL1 andGABARAPL2 (GABARAP-like protein 1/2) (Schaaf et al. 2016). Studies in LNCaPprostate cancer cells suggest that GABARAPs rather than LC3 subfamily are more38essential for autophagic bulk sequestration of cytosolic cargos (Szalai et al. 2015). Helacell culture studies suggest that LC3 and GABARAP family members are not requiredfor autophagosome formation. The GABARAP subfamily members are required forautophagosome–lysosome fusion, whereas the LC3 subfamily members are notnecessary (Nguyen et al. 2016). More of their functions and relationships are underinvestigation.Autophagy in mammals can be induced by nutrient starvation. Low energy or nutrientlevels inhibit a nutrient signaling regulator called the mammalian target of rapamycin(mTOR) and reduce its inhibition of unc-51 like autophagy activating kinase 1 (ULK1),which is in complex with Atg13, Atg101 and FIP 200. The ULK1 complex then activatesthe class III phosphoinositide 3 kinase (PI3K) complex (including VPS34, VPS15,Atg14L and Beclin1) through ULK1-dependent phosphorylation. These two proteincomplexes function in the initiation of autophagy and are essential for the nucleation ofthe isolation membrane/-phagophore. The class III PI3K complex generatesphosphatidylinositol 3-phosphate (PI3P) at the site of isolation membrane, resulting inrecruitment of PI3P-binding proteins: the WD repeat domain phosphoinositide-interacting protein (WIPI) family members, including WIPI-1 and WIPI-2. WIPI proteinshave a conserved 7-bladed beta-propeller which functions as a PI3P scaffold and helpsto recruit other subsequent proteins to the isolation membrane (Proikas-Cezanne et al.2015), including two protein complexes, the Atg5-Atg12-Atg16L complex and LC3 II-phosphatidylethanolamine (PE). These two protein complexes function in the elongationand closure of isolation membranes/ phagophores, resulting in the formation of39autophagosomes. WIPI-2 recruits the complex Atg5-Atg12-Atg16L to isolationmembranes through direct binding with Atg16L (Proikas-Cezanne et al. 2015), the Atg5-Atg12-Atg16L complex then catalyzes the lipidation of LC3. ProLC3 is soluble in thecytoplasm and is first cleaved by Atg4 and becomes LC3 I. LC3 I is conjugated to PE onthe isolation membrane in the presence of Atg7 (E1 like ligase), Atg3 (E2 like ligase)and the complex Atg5-Atg12-Atg16L (E3 like ligase), resulting in LC3 II-PE, whichcontributes to phagophore/ isolation membrane elongation and closure (Green & Levine2014). The closed autophagosomes fuse with lysosomes to be degraded (Figure 10).Although the mechanisms of autophagosome-lysosome fusion are not fully understood,a variety of proteins are involved in this process, including the autophagosomal solubleN-ethylmaleimide-sensitive factor syntaxin 17, synaptosome associated protein 29(SNAP29), vesicle-associated membrane protein 8 (VAMP8) and the adaptor proteinpleckstrin homology domain-containing family M member 1 (PLEKHM1) and itsassociating protein the homotypic fusion and protein sorting (HOPS) complex andectopic P-Granules autophagy protein 5 homolog (EPG5), Atg14L, and rab-7 (ras-related protein 7) (Nakamura & Yoshimori 2017; McEwan et al. 2015).40Figure 10: The molecular mechanism of autophagy in mammals.Autophagy in mammals can be induced by nutrient starvation, energy depletion, orhypoxia. The autophagy process begins with the nucleation of the isolation membrane.The two protein complexes ATG16L1-ATG5-ATG12 and LC3-II participate in theexpansion and the closure of the isolation membrane. The complete double-membraned structure is called an autophagosome, which fuses with an lysosome. Thecargos from autophagosomes are degraded by lysosomal acid hydrolases, yieldingbasic metabolites, such as amino acids, into the cytoplasm for new synthesis or as asource of energy (Karakaş & Gözüaçik 2014).411.8.4 Regulation of autophagy:Autophagy induction can be regulated through ULK1 phosphorylation by two conservednutrient signaling pathways, the mTOR and AMP-activated protein kinase (AMPK)signaling pathways.Autophagy is negatively regulated by mTOR, which is a master regulator of cell growthin response to nutrients, growth factors and cellular energy levels. Under nutrientsufficiency, mTOR is activated and inhibits the ULK1 complex by phosphorylating theULK1 Ser 757 (Kim et al. 2011), thus suppressing autophagy induction. Duringstarvation, the low concentration of amino acids inhibits mTOR complex1 activity,resulting in reduced phosphorylation of its substrate transcription factor EB (TFEB),which then translocates from cytosol into the nucleus and enhances the transcription ofgenes supporting increased autophagy (Martina et al. 2012). Treatment with mTORinhibitors, such as rapamycin, can be used to induce autophagy in yeast and animals(Noda & Ohsumi 1998; Ravikumar, Vacher, Berger, Janet E. Davies, et al. 2004).Autophagy can also be induced by AMPK activation. AMPK is an energy sensor whichplays important roles in maintaining cellular energy homeostasis. AMPK is activatedduring energy starvation; it then activates ULK1 through phosphorylation at Ser 317 andSer 777 and initiates autophagy (Kim et al. 2011). AMPK can also induce autophagythrough indirect mTOR inactivation (Gwinn et al. 2008).There are also studies proposing that light can regulate autophagy levels inphotoreceptor cells. The number of autophagic structures/vacuoles increased42significantly in albino rats when the intensity of light was increased from 3 lux to 200 lux(Remé et al. 1999). Another study in wildtype C57BL/6 mice showed that the levels ofautophagy in photoreceptors exhibited a bimodal pattern in cyclic light with light-dependent increase of autophagy (Yao et al. 2014).431.9 Autophagy and retinal degeneration:1.9.1 Autophagic degradation of P23H rhodopsin in cell cultures:Autophagy is one of two major protein degradation pathways in eukaryotic cells togetherwith the ubiquitin-proteasome pathway. Misfolded proteins can be resulted from geneticmutations, abnormal protein modifications, incorrect protein folding or protein complexassembly (Yao 2010). Misfolded proteins are aggregated-prone and can interfere withnormal cellular functions and/or cause cellular stress, thus they are closely monitored byprotein quality control mechanisms. The ubiquitin-proteasome system specificallyrecognizes ubiquitinated proteins for degradation. It is generally thought that misfoldedproteins are polyubiquitinated and targeted to proteasomes for degradation (Yao 2010).Growing evidence shows that autophagy can also mediate selective degradation ofmisfolded proteins or protein aggregates, where ubiquitinated misfolded proteins arerecruited into autophagosomes via adaptor protein p62 (nucleoporin 62, an ubiquitin-binding protein) (Yao 2010).Mammalian cell culture studies using human embryonic kidney (HEK) 293 cellsdemonstrate that P23H rhodopsin is a substrate of autophagy, and that pharmacologicalregulation of autophagy levels significantly affect the degradation of misfolded P23Hrhodopsin (Kaushal 2006). Upregulation of autophagy by rapamycin enhanced thedegradation of P23H rhodopsin in HEK293 cell cultures, while inhibition of autophagy by3-methyladenine (3-MA) resulted in higher levels of P23H rhodopsin, suggesting a44decreased degradation of P23H rhodopsin under autophagy inhibition (Kaushal 2006).3-MA is an autophagic inhibitor, and it inhibits autophagy by suppressing class III PI3Kactivity, which is essential for the induction of autophagy (Wu et al. 2010).1.9.2 Autophagy in animal models of RP:Although cell culture studies demonstrated that autophagy can degrade misfolded P23Hrhodopsin, the role that autophagy plays in retinal degeneration still requires furtherinvestigation. As mentioned previously, RP is a heterogeneous group of retinaldegeneration which can be caused by multiple different mutations in the same ordifferent genes. Even mutations within the same gene can cause RP via differentmechanisms, for example the rhodopsin mutation P23H causes misfolding of rhodopsinmolecules (Tam & Moritz 2006), while T4K causes destabilization of photoactivatedrhodopsin molecules (Tam et al. 2014), and Q344Ter causes mislocalization ofrhodopsin molecules (Concepcion & Chen 2010). Thus, it may be quite difficult orimpossible to define a common role that autophagy plays in all forms of RP. Autophagymay even play different (or no) roles in RP caused by different mutations within thesame gene, depending on the underlying disease-causing mechanisms.Studies in Drosophila revealed that knocking-down of autophagy essential componentsAtg7 or Atg8 resulted in accumulation of rhodopsin and subsequent retinal degeneration,suggesting a protective role of autophagy in preventing retinal degeneration(Midorikawa et al. 2010). On the other hand, autophagy exacerbates photoreceptor45degeneration in the retinal degeneration 10 (rd10) mouse, which is an autosomalrecessive retinal degeneration model. The rd10 mouse carries a missense mutation inthe phosphodiesterase6β (Pde6β) gene, encoding the beta subunit of cGMP-PDE thatfunctions in the phototransduction cascade (Chang et al. 2002), leading to rod and conedegeneration (Gargini et al. 2007). Autophagy level was reduced in the retinas of rd10mice before the onset of photoreceptor degeneration, and activation of autophagy byrapamycin increased photoreceptor apoptotic cell death (Rodríguez-Muela et al. 2015).1.9.3 Autophagy in RP caused by the P23H rhodopsin mutation:We formerly observed and reported increased number of autophagic structures byelectron microscopy in the inner segments of X. laevis rods expressing a misfolding-prone P23H rhodopsin mutant, including both human and bovine P23H rhodopsin(Vent-schmidt et al. 2017; Bogéa et al. 2015). In transgenic X. laevis, retinaldegeneration caused by bovine P23H rhodopsin can be prevented by constantdarkness, and induced by light exposure (Tam & Moritz 2007b). We observed that thenumber of autophagic structures was increased by light exposure (panels B, C, E ofFigure 11) in transgenic tadpoles, but was low in wild-type tadpoles kept either inconstant darkness (not shown) or exposed to light (panel D of Figure 11). At later timepoints (panel C of Figure 11), massive vacuolization occurred in rods of bP23H-expressing tadpoles. These results suggest that autophagy is associated with retinaldegenerations caused by P23H rhodopsin, either acting as a protective response tolarge amounts of misfolding rhodopsin or in executing rod cell death.46Figure 11: Light exposure induced autophagic structures in rod inner segments ofbP23H-expressing X. laevis tadpoles. Note the presence of compartments withprominent contents (arrowheads), consistent with autophagosomes (arrowheads) orautolysosomes (beveled arrowheads). Possible phagophore formation (arrow) andnumerous vesicular structures without identifiable contents (asterisks) are also present.(A) Dark-reared bP23H-expressing tadpole. (B) Dark-reared bP23H-expressing tadpoleexposed to 12 h of light. (C) bP23H-expressing tadpole kept in 1.5 light cycles (12L:12D:12L). (D) wildtype tadpole reared in constant darkness and exposed to 12 h of light.(E) Juxtaposed rod (left) and cone (right) inner segments from a transgenic tadpole keptin one complete light cycle. CN, cone nucleus; P, cone paraboloid. Scale bars: 2 μm(Bogéa et al. 2015).47A controversial study of a cohort of seven patients suggested that valproic acid (VPA)shows promise as an RP treatment (Clemson et al. 2011), but was criticized for lack ofsupporting research demonstrating a therapeutic mechanism (Sandberg et al. 2011;Van Schooneveld et al. 2011). Follow-up reports suggested detrimental effects onacuity (Sisk 2012; Bhalla et al. 2013). A more recent study indicated opposing effects ofVPA in mice with Pde6β mutations (Mitton et al. 2014).VPA has anticonvulsant and mood stabilizing activities that are used to treat epilepsyand bipolar disorder. Its mechanisms of action are unclear due to complexpharmacology; it is a HDAC inhibitor (Phiel et al. 2001), a GABA transaminase inhibitor,and a sodium channel blocker (Johannessen 2000; Löscher 2002; Owens & Nemeroff2003). Its mood stabilizing effects are due to inositol depletion (Williams et al. 2002). Itcan reduce ER stress via glycogen synthase kinase inhibition (Chen et al. 1999; Bownet al. 2002) and activate autophagy via PI3K (Sarkar et al. 2005; Williams et al. 2008;Renna et al. 2010). VPA is under investigation as a treatment for neurodegenerativediseases, including Huntington's (Chiu et al. 2011), Parkinson's (Monti et al. 2010) andAlzheimer's (Loy & Tariot 2002; Tariot & Aisen 2009) diseases, and is reported toenhance clearance of protein aggregates by autophagy (Renna et al. 2010; Fleming etal. 2011).We formerly investigated the therapeutic potential of VPA using four transgenic X. laevismodels of ADRP-expressing human rhodopsin mutants, including P23H rhodopsin, T4K48rhodopsin, T17M rhodopsin, and Q344Ter rhodopsin (Vent-schmidt et al. 2017). Thefirst three models show marked light dependence of RD (Tam & Moritz 2007b; Tam &Moritz 2009). We found that VPA may ameliorate or exacerbate RD depending on theunderlying mutation. VPA ameliorated RD and vision deficits caused by P23H rhodopsin,but dramatically exacerbated RD caused by T17M rhodopsin in the presence of light.Effects in other models were negative and less pronounced (Vent-schmidt et al. 2017).Similar results were obtained with three other HDAC inhibitors, but not with otherantipsychotics, VPA structural analogs, or chemical chaperones, indicating that both thepositive and negative effects were due to HDAC inhibitor activity (Vent-schmidt et al.2017).We found that HDAC inhibitors, including VPA and sodium butyrate (NaBu), inducedautophagy in wild type X. laevis rods as supported by electron microscopy (Vent-schmidt et al. 2017). Thus, upregulation of autophagy, possibly leading to enhancedclearance of P23H rhodopsin, is a possible mechanism by which HDAC inhibitors exerttheir beneficial effects in RP caused by P23H rhodopsin.We also found that in models of ADRP caused by mutations that do not promotemisfolding of rhodopsin, including T17M, T4K and Q344Ter, the effects of VPA weredetrimental. The mechanisms underlying these degenerations do not involve proteinmisfolding during biosynthesis, but rather involve photoactivation of the mutantrhodopsin, possibly resulting in instability and membrane disruptions in outer segments49(T17M and T4K) or rhodopsin mis-localization (Q344ter). The mechanism underlyingthis detrimental effect is not clear, but it was also associated with the HDAC inhibitoractivity of VPA, because it was similarly reproduced by treatment with NaBu and CI-994.These findings together suggest that autophagy is associated with retinal degenerations,and may represent a potential therapeutic target. However, the role that autophagyplays in retinal degenerations is unclear.501.10 Goals of this thesis:1) To develop transgenic lines expressing fluorescent markers of autophagy thatallow visualization of autophagy and autophagic flux in photoreceptors.2) To characterize autophagy in normal rod photoreceptor cells, including the time-course of autophagy, and the effects of light on the level of autophagy in rods.3) To study autophagy in diseased rods and determine whether it is altered indisease states.4) To determine the role that autophagy plays in RP caused by P23H rhodopsinmutation.511.11 Significance:In this thesis, I developed transgenic X. laevis expressing a fluorescent autophagyreporter mRFP-eGFP-LC3 with expression driven by either a heat-shock promoter(HSP70) or the Xenopus rod opsin promoter (XOP). By using these animals, Icharacterized the time-course of autophagy in normal rods, and determined a non-circadian diurnal variation of autophagy levels in normal rods. I also characterizedautophagy levels in rods expressing a misfolding-prone P23H rhodopsin mutation andfound that autophagy levels were increased in rods expressing P23H rhodopsincompared to that of rods expressing wildtype rhodopsin. I further determined that theincrease of autophagy in light-exposed normal rods was not caused by increasedmisfolding of wildtype rhodopsin due to lack of chromophore, but rather due to theactivation of phototransduction.My study systematically characterized autophagy in the rods of X. laevis, including bothwildtype and degenerating rods, and significantly advanced our knowledge in this field.It also allows us to better understand the mechanisms underlying RP caused bymisfolding-prone rhodopsin and suggests that autophagy may play a role in this form ofRP, and possibly other forms of RP caused by class II rhodopsin mutations. Additionally,my study also raises interesting questions for future investigations, such as what are thepathways by which light promotes autophagy in normal rods, and what is the purpose ofthis pathway?522. Methodology2.1 Generation of two transgene constructs:The original mRFP-eGFP-LC3B sequence was amplified from the plasmid ptfLC3(Plasmid #21074 from Addgene) using primers flanked by EcoRI and NotI sites. Togenerate the hsp70- mRFP-eGFP-LC3 construct, this sequence was cloned into theEcoRI/NotI sites of our plasmid Hsp70-2DI-GFP which contains a heat shock promotersequence (plasmid map in Figure 12), replacing the sequence of the GFP cDNA. Thisvector was made by Dr. Damian Lee (unpublished), who obtained the hsp70 promoterby PCR from Xenopus laevis genomic DNA using the published sequence (Bienz 1984).53Figure 12: Plasmid Hsp70-2DI-GFP map. DI: chicken beta-globin double insulatorsequence. Xhsp70: X. laevis hsp70 promoter. MCS: multiple cloning site. GFP: GFPcDNA. SV40 pA: SV40 polyadenylation signal. Neo: neomycin (G418) resistance.HSV TK pA: HSV thymidine kinase polyadenylation sequence. pUC ori: pUC origin ofreplication. Figure by Dr. Damian Lee, unpublished.To generate the XOP-mRFP-eGFP-LC3 constructs, the mRFP-eGFP-LC3 sequencewas similarly cloned into the same two sites of our XOP-eGFP-N1 vector, replacing theoriginal GFP fragment. This vector has an identical construction to that shown in FigureL, except that the Hsp70 sequence is replaced by an 800 bp fragment of the Xenopusrod opsin promoter (XOP).2.2 Generation and rearing of transgenic X. laevis:Primary transgenic X. laevis tadpoles were generated through restriction enzymemediated integration (REMI) method of transgenesis (Ishibashi et al. 2012; Tam et al.2013). Briefly, transgene constructs were first linearized using the restriction enzymeFseI, and diluted to 75 µg/ml. X. laevis sperm nuclei were decondensed by incubationwith high speed egg extract and incubated for 5 minutes (min) with the linearizedtransgene construct followed by 10 min of incubation with dilute restriction enzyme FseI.The integrated nuclei (with the transgene) were then diluted to a concentration of 3nuclei per 20 nl, and transplanted by microinjection into dejellied unfertilized wildtype X.laevis eggs. Correctly dividing fertilized embryos were identified visually at the 4 to 8 cellstage, and transferred into 0.1x Marc's modified Ringer (MMR) solution (100 mM NaCl,542 mM KCl, 2 mM CaCl2, 0.1 mM ethylene- diaminetetraacetic acid, 1 mM MgCl2, 5 mMHEPES-KOH at pH 7.8) with 6% ficoll and 10 mg/ml gentamycin. The following day, theembryos were transferred into regular 1x tadpole ringer solution (10 mM NaCl, 0.15 mMKCl, 0.2 mM CaCl2, and 0.1 mM MgCl2) from 1 day post-fertilization (dpf). The primarytransgenic tadpoles were raised to adult frogs in clear tanks containing 1x tadpole ringer,requiring between 8 months to 2 years.F1 tadpoles were generated by mating the primary transgenic frogs to wildtype frogs.Fertilized embryos were collected in a petri-dish and dejellied using 2% cysteine in 1xMMR solution. Dejellied embryos were then transferred into a 4 L clear tank containingregular 1x tadpole ringer solution and raised in standard lighting conditions comprising alight cycle of 12 h light and 12 h dark (12L: 12D) in an 18°C incubator. Standard lightlevels were 1700 lux. Tadpoles were fed daily with powdered NASCO frog brittle.For visualizing the diurnal variation of autophagy, a transgenic frog carrying the XOP-mRFP-eGFP-LC3 transgene (female1) was mated to a wildtype male frog. Tadpoleswere sacrificed every 3 h for 24 h from 12 dpf to 13 dpf.To determine the time-course of autophagy, a transgenic frog carrying the hsp70-mRFP-eGFP-LC3 (male2) was mated to a wildtype female frog. Tadpoles were eithersacrificed without the heat-shock on 6 dpf or sacrificed at different time points afterreceiving a 1 h heat-shock at 33°C from 6 dpf to 14 dpf. For heat-shock protocols,tadpoles were raised at 18°C prior to the heat-shock. A beaker containing 100 ml of 1xtadpole ringer solution was incubated in 37 °C incubator overnight (without cover). The55actual temperature of the ringer solution in the beaker was 33 °C as measured by athermometer. Tadpoles were then transferred to the beaker using a plastic transferpipette. The beaker containing tadpoles was then put into the 37 °C incubator for 1h.After heat-shock, tadpoles were first transferred to 400 ml of room-temperature ringersolution for 5 min, then transferred back into normal 18 °C incubator.2.3 Drug treatments of transgenic X. laevis:A transgenic frog expressing XOP-mRFP-eGFP-LC3 (female1, carrying the transgeneat a single integration site) was mated to a wildtype male frog; offspring were raised inan 18°C incubator on a 12 h light/dark cycle. At 2 dpf (corresponding to developmentalstage 23), animals of either sex were divided into two groups of n = 13. One group ofanimals was raised in 1× tadpole ringer solution with 10 μM VPA (Sigma-Aldrich); theother group of animals was raised in 1× tadpole ringer solution without the drug. Freshdrug solution was prepared daily and renewed for 4 days (from 2 dpf to 5 dpf). Animalswere sacrificed on 6 dpf after receiving 5 h of light illumination.Other two drugs, sodium butyrate (NaBu) (Sigma-Aldrich) and CI-994 (CaymanChemical) were also tested on tadpoles expressing XOP-mRFP-eGFP-LC3. We used300 μM NaBu and 10 µg/ml (or 37.2 µM) CI-994 respectively. Drug therapy protocols forNaBu and CI-994 were almost identical to VPA, except that each group had 10 animals.2.4 Immunohistochemistry and confocal microscopy:Tadpoles were sacrificed by pithing. Eyes were enucleated and fixed in 4%paraformaldehyde in 0.1 M sodium phosphate buffer pH 7.4 overnight. Fixed eyes were56embedded in O.C.T. (optimal cutting temperature) medium (Tissue Tek) andcryosectioned (12 µm) as previously described (Tam 2006). Frozen sections werestained with Alexa-647-conjugated wheat germ agglutinin (WGA) (Thermo Fisher) at a1:10,000 dilution of 1 mg/ml solution and Hoechst 33342 (Sigma-Aldrich) (1: 1000) asdescribed previously (Moritz et al. 1999). Sections were imaged by using a Zeiss 510laser scanning confocal microscope with a 10× (air, NA 0.3) or 40× (water-immersion,NA 1.2) objectives. Three Dimensional (3D) Z-stack images were collected using the40X objective with a spacing of 0.5 microns between consecutive Z-planes. Withinindividual experiments, all images were taken at the same gain and contrast settings forthe GFP and mRFP channels. Imaging processing was performed using AdobePhotoshop. For construction of figures, the fluorescent signals derived from GFP andmRFP channels were adjusted linearly to best demonstrate the numbers of fluorescentpuncta. Signals derived from Alexa-647-WGA and Hoechst 33342 staining wereadjusted nonlinearly to best demonstrate retinal architecture.For counting the number of fluorescent puncta, only rods in which soluble green signalwas present were used. Rods with no soluble green signal (or no green signal) butpunctate red signal were frequently observed, and were attributed to transgenesilencing (Moritz et al. 2001). In general, older tadpoles contained fewer rodphotoreceptors with fluorescent transgene signal, indicating that both lines of transgenicanimals (XOP-mRFP-eGFP-LC3 female1 and female 2) exhibited progressivetransgene shutoff over time. The numbers of green and red puncta were counted fromGFP and mRFP channels of 3D renderings of the Z-stack images respectively. The57counts were not blinded. The brightness/intensity of GFP or mRFP signals were freelyadjusted during counting of puncta (Zen2009 software) such that both bright and dimpuncta could be visualized by the experimenter. All green or red fluorescent punctawere counted regardless of the intensity and size of the fluorescent puncta. Colocalizedgreen and red puncta with similar shape and size when the GFP and mRFP channelswere superimposed were counted as yellow puncta, regardless of whether green or redsignal predominated.For visualizing the RPE cell layer, frozen retinal sections were labelled with primaryrabbit polyclonal anti-RPE65 antibody (custom-made by Yenzyme) YZS437 at a 1:3,000dilution, and then stained with the secondary anti-rabbit antibody conjugated to Cy5(1:750) and counterstained with Hoechst 33342 (Sigma-Aldrich) (1: 1000).To determine whether the large red puncta contained mitochondria, some retinalsections were labelled with cytochrome C (anti-mouse monoclonal) antibody (Novusbiologicals) at 1:100 dilution. Antigen retrieval was performed before the primaryantibody labelling following the procedure described in Daigo Inoue 2011(Inoue &Wittbrodt 2011). Eyes were fixed in 4% paraformaldehyde in 0.1 M sodium phosphatebuffer pH7.4 overnight and were immersed in 30% sucrose overnight at 4℃. Each eyewas incubated with 100 μl 150 mM Tris-HCl at pH 9.0 for 5 min at RT then heated at70℃ for 15 min. The eyes were transferred into 30% sucrose overnight at 4℃. Eyeswere then cryosectioned and labelled with the primary cytochrome C antibody at a581:100 dilution, and then labelled with the secondary Cy5 anti-mouse antibody (1:750)and counterstained with Hoechst 33342 (Sigma-Aldrich) (1:1000).2.5 Extraction of genomic DNA and qPCR genotyping:Clippings of tail tissue were obtained from sacrificed tadpoles. The clipped tadpole tailswere immersed in 100 µl of genomic DNA extraction buffer (50 mM Tris pH8.5; 0.5%Tween-20; 200 µg/ml proteinase K) and incubated at 55°C for 2h, 95°C for 10 min.Genomic DNA was amplified by PCR using PCR primers specific for the target genes.Amplification was carried out in the presence of a FAM-labelled probe, and fluorescencederived from the probe was detected using an MJ Research Opticon 2 Q-PCR system.Primers and probes were custom-synthesized by Integrated DNA Technologies.To determine whether tadpoles carried the XOP-mRFP-eGFP-LC3, genomic DNA wasamplified using the primers “5’-TGA ACT GCA AGA CGA GGC -3’ ” and “5’-CAA GGTGAG ATG ACA GGA GAT C-3’ ”, which amplify a sequence within the neomycinresistance cassette present in the transgene vector (see Figure 12). Amplification wascarried out in the presence of the FAM-labelled probe “5’-/56-FAM/CAC TGAAGC/ZEN/GGG AAG GGA C”.To determine whether tadpoles carried the bovine P23H rhodopsin transgene, genomicDNA was amplified using the primers “5’-TCA CAG TCC AGC ACA AGA AG-3’” and “5’-TAC CCA TGG AGA GAG GTG TAG-3’”, which amplify a sequence within the bovine59rhodopsin cDNA. Amplification was carried out in the presence of the FAM-labelledprobe “5’-/56-FAM/CAC ACC GCT /ZEN/CAA CTA CAT CC.2.6 Transmission electron microscopy (TEM):Procedures for electron microscopy were performed essentially as described previously(Tam et al. 2015). Briefly, tadpoles expressing XOP-mRFP-eGFP-LC3 were sacrificeddirectly at either 12 h following light onset (receiving 12 h of light exposure) on 12 dpf or0 h prior to light onset (receiving 12 h of dark treatment) on 13 dpf. Eyes were fixed in4% paraformaldehyde 1% glutaraldehyde in 0.1M sodium phosphate buffer, pH 7.4 andinfiltrated with 2.3 M sucrose in 0.1M phosphate, embedded in O.C.T. medium(TissueTek) and cryosectioned (20 µm). Cryosections were osmicated, dehydrated,infiltrated with Eponate 12 resin, and embedded in Eponate 12 resin. Sections 70 nmthick were cut using a diamond blade, mounted on formvar film, stained with uranylacetate and lead citrate, and examined in a Hitachi H7600 transmission electronmicroscope.2.7 Generation of RPE65 knock out X. laevis tadpoles usingCRISPR-cas9 technology:Procedures for CRISPR-cas9 mediated gene knock out were performed essentially asdescribed previously (Feehan et al. 2017). Briefly, a small guide RNA (sgRNA) wasdesigned to target exon 7 (of 14 exons total) in the two rpe65 genes located on both60chromosome 4S and 4L (rpe65.L sequence:http://gbrowse.xenbase.org/fgb2/gene_model_details/xl7_1?feature_id=9193997,rpe65.S sequence:http://gbrowse.xenbase.org/fgb2/gene_model_details/xl9_1?feature_id=619378). Thetargeting sequence was AGGTTGTTGTTCAGTTCC, with the cut site positionedbetween AGG and TTGTTGTTCAGTTCC. Oligonucleotides corresponding to thissgRNA targeting sequence were cloned into the BbsI site of pDR274 (A gift from KeithJoung - Addgene plasmid #42250). The resulting plasmid was linearized and used as atemplate to generate sgRNA via in vitro transcription using the MAXIscript in vitrotranscription kit (Ambion). The sgRNA was subsequently purified using miRNeasy kit(Qiagen). S. Pyogenes Cas9 mRNA and eGFP mRNA were also in vitro transcribed andsubsequently purified. RNA products were evaluated for size by agarose gelelectrophoresis (1.5% agarose at 120 mV for 20 min) and quantified by absorbance at260/280 nm with a NanoDrop 2000c spectrophotometer (Thermo Scientific). The Cas9mRNA, eGFP mRNA and sgRNA were combined and microinjected into one-cell stagedejellied in vitro fertilized X. laevis eggs. Healthy embryos were picked into 0.1x MMRsolution with 6% Ficoll and 10 µg/mL gentamicin and stored in an 18 C incubatorovernight. At 2 dpf, embryos were screened for eGFP fluorescence using anepifluorescence-equipped Leica MZ16F dissecting microscope. Embryos with high,uniform eGFP expression were transferred to 0.1x MMR solution. Starting from 3 dpf,the embryos were raised in 1x tadpole ringer solution. At 12 h following light onset(receiving 12 h of light exposure) on 12 dpf or 0 h prior to light onset (receiving 12 h ofdark treatment) on 13 dpf, tadpoles were sacrificed and tails were clipped. Genomic61DNA was extracted from tail clippings as described above. For analysis of insertionsand deletions (indels) introduced in the genomic DNA by this procedure, regions aroundthe target sequences of the two RPE65 genes were amplified by PCR, with differentprimer pairs used for the RPE65.L and RPE65.S genes. For RPE65.L, the forwardprimer was 5’-CCTGCCTTTATGATACCCAG-3’ and the reverse primer was 5’-GTATAG CAG AGA TCC CCA AC-3’ (total amplicon length is 411 bp). For RPE65.S, theforward primer was 5’-CCT CAA TAT CCC ACC TTT GTG-3’and the reverse primerwas 5’-CTG GCT TGA GGT TTC TAC AG-3’ (total amplicon length is 436 bp). PCRproducts were treated with exonuclease I and shrimp alkaline phosphatase to removeexcess primers and dNTPs, and sequenced directly using the forward primers using acommercial sequencing service (Genewiz, South Plainfield, NJ, USA).2.8 Dot blot analyses:Dot blots of X. laevis eye extracts were performed as described previously (Tam 2006).Briefly, intact eyes were solubilized in 100 μl of a 1:1 mixture of PBS and SDS-PAGEloading buffer containing 1 mm EDTA and 100 μg/ml PMSF in Eppendorf tubes using aplastic pestle (Kontes). Blots were probed with primary monoclonal antibody B630N(Adamus et al. 1991) at a 1:15 dilution of tissue culture supernatant, followed by IR-dye800-conjugated goat anti-mouse antibody at 1:10,000 of 1 mg/ml solution(Rockland). Each dot blot included standards containing 100% X. laevis rod opsin(derived from age matched WT retinas).2.9 Statistical Analysis.62The experiment comparing the number of autophagic structures between XOP-mRFP-eGFP-LC3 line1 and line2 described in section 3.2 was analyzed by student’s t-test.The experiment involving measurements of the number of autophagic structures atdifferent time points of the light cycle described in section 3.4 was analyzed by 1-wayANOVA and student’s t-test.The experiment demonstrating that light-dependent autophagy is non-circadiandescribed in section 3.6 was analyzed by student’s t-test.The experiment comparing the number of autophagic structures between bP23Hrhodopsin-expressing rods and rods expressing only endogenous rhodopsin describedin section 3.7 was analyzed by student’s t-test.The experiment involving RPE65 gene knockout and measurements of the number ofautophagic structures described in section 3.8 was analyzed by 2-way ANOVA.Experiments involving drug treatments and measurements of the number of autophagicstructures (including rapamycin, chloroquine, VPA, NaBu and CI994 treatments onXOP-mRFP-eGFP-LC3 tadpoles) described in sections 3.11 - 3.15 were analyzed bystudent’s t-test.Statistical analyses of dot blot data was performed on log-transformed integratedintensity values. Experiments involving drug treatments (either rapamycin orchloroquine) and measurements of total rod opsin levels per retina described in sections3.16 - 3.17 were analyzed by two-way ANOVA. Statistical tests were performed usingSPSS software.633. Results3.1 Development of transgenic X. laevis reporter linesTo study autophagy in rod photoreceptors, I established transgenic X. laevis linesexpressing the dually fluorescent autophagy marker mRFP-eGFP-LC3 (Kimura et al.2007). LC3, the microtubule-associated protein 1A/1B light chain 3, is an ubiquitin-likeprotein conjugated to the membranes of autophagic vacuoles, and associates withautophagic membranes throughout the autophagy process (Mizushima 2007). Twotransgene constructs containing the mRFP-eGFP-LC3 cDNA linked to either the heatshock inducible hsp70 promoter (hsp70) or the Xenopus laevis rod opsin promoter(XOP) were generated, and referred to as hsp70-mRFP-eGFP-LC3 and XOP-mRFP-eGFP-LC3 respectively. X. laevis tadpoles carrying either the hsp70-mRFP-eGFP-LC3or XOP-mRFP-eGFP-LC3 transgenes were generated via the restriction-enzyme-mediated integration method (Amaya & Kroll 2010). Tadpoles were screened forfluorescence, raised to maturity, and bred with wildtype X. laevis. Two transgenicfounders were identified for XOP-mRFP-eGFP-LC3, and one founder was identified forhsp70-mRFP-eGFP-LC3.3.2 Initial characterization of fluorescence in transgenicreporter linesUpon induction of autophagy, the dually fluorescent LC3 marker mRFP-eGFP-LC3 isincorporated into early autophagic structures (phagophores and autophagosomes)64appearing as dually fluorescent puncta in micrographs (Kimura et al. 2007). In latestages, autophagosomes fuse with lysosomes and acid hydrolases destroy the acid-labile GFP signal, while the acid-resistant mRFP signal remains, allowing tracking of thedynamic process of autophagy (Kimura et al. 2007).I examined retinas of animals constitutively expressing dual fluorescent LC3 in rods.Fluorescent LC3 was observed only in X. leavis red rods (Figure 13 A), which comprisethe majority of X. laevis rods. The offspring of both founders showed mosaic expressionof the transgene in a subset of rods (Moritz et al. 2001), such that individual transgene-expressing photoreceptors were easily imaged. I observed abundant green, yellow andred puncta in rod inner segments (Figure 13 B), with a fainter diffuse green fluorescentbackground that also extended into the outer segment along the microtubules of thecilliary axoneme. No diffuse red mRFP signal was observed. The three categories ofpuncta were consistent with the previously reported slower maturation time of mRFPrelative to GFP (Campbell et al. 2002), which has been used in similar duallyfluorescent proteins as a timer (Khmelinskii et al. 2012), and is likely fortuitouslyexaggerated by the relatively low (18°C) temperatures of our normal X. laevis housing(Campbell et al. 2002). In some photoreceptors, orderly lines of puncta were observedthat appeared to follow defined tracks, possibly microtubules and/or contours of the ER.The diffuse green signal and puncta were consistent with LC3 transitioning from asoluble form to a membrane-associated form during autophagosome formation, whilethe absence of diffuse mRFP signal suggests that in rods LC3 is rapidly turned over byautophagic activity.65The number of autophagic structures per rod were measured and compared betweenthe offspring of the two founders. The number of structures did not differ significantlybetween the two lines (Figure 13 E). On average, 27 structures were observed per rodfor animals sacrificed after 4 h of light exposure.In transgenic lines expressing mRFP-eGFP-LC3, I consistently observed largefluorescent structures in rod inner segments, which were about 4 to 7 times larger indiameter than the predominant smaller autophagic structures (1.3 to 1.7 µm vs 0.3 to0.5 µm). These were almost exclusively red in confocal micrographs, but very rarelygreen (Figure 13 F) or yellow examples were seen. Each rod typically contained asingle large red structure, typically located immediately below the connecting cilium andbelow the mitochondria, adjacent to the microtubule organizing center, but occasionallyother localizations or more than one structure were observed. To further determine thecontents of these structures, I labelled them with anti-cytochrome C antibody, astandard marker for mitochondria, and found some minor colocalization withcytochrome C, suggesting that at least some of these large structures may beconnected to the mitophagy pathway (Lemasters 2005). However, as the nature ofthese structures was unclear, and as they comprised a small minority of the total puncta,they were excluded from all subsequent analyses (except for the experiment describedin section 3.3: the time-course of autophagy in rods). Because of their low abundance,this had negligible effects on the total numbers of structures counted.66In hsp70-mRFP-GFP-LC3 transgenic animals, expression of the dually fluorescentautophagy marker was inducible by heat shock. Prior to heat shock there was noobservable expression of dually fluorescent LC3 in the retina (Figure 13 C); after heatshock, dually fluorescent LC3 was observed in the entire tadpole body (not shown), andabundant fluorescent LC3 expression was observed in all layers of retina (Figure 13 D).67Figure 13: Establishment and initial characterization of transgenic reporter lines. Twotransgenic reporter lines were established, the XOP-mRFP-eGFP-LC3 and hsp70-68mRFP-eGFP-LC3. (A) A retina section from a X. laevis tadpole expressing XOP-mRFP-eGFP-LC3 which was sacrificed at 6 dpf. Scale bars: 100 μm. (B) Rods of tadpolesexpressing XOP-mRFP-eGFP-LC3. Note the presence of green puncta (greenarrowhead), yellow puncta (yellow arrowhead), red puncta (red arrowhead) and thelarge red structure (white arrowhead) in the inner segment of rod. A fainter diffuse greenfluorescent background also extended into the outer segment along the microtubules ofthe cilliary axoneme (white arrows). Rod outer segments were labeled with WGA-Alexa647, represented by the magenta color in the confocal micrograph. Scale bars: 5 μm. (C)A retina section from a hsp70-mRFP-eGFP-LC3 trasngenic tadpole sacrificed withoutheat-shock at 6 dpf. Scale bars: 100 μm. (D) A retina section from a hsp70-mRFP-eGFP-LC3 trasngenic tadpole which received a 1 h heat-shock at 6 dpf and wassacrificed at 24 h after the heat-shock. Scale bars: 100 μm. (E) Quantification of thenumber of autophagic structures per rod of XOP-mRFP-eGFP-LC3 line1 and line2. Thenumbers of green, yellow, red and total puncta per rod were quantified respectively.Error Bars= S.E.M. The number of autophagic structures did not differ significantlybetween the two transgenic lines. (T-test for green, yellow, red, total puncta, p = 0.53,0.67, 0.34, 0.54 respectively). (F) A rod expressing XOP-mRFP-eGFP-LC3 containing alarge green autophagic structure (white arrowhead) in the inner segment. Scale bars: 5μm. (G) Colocalization of a large red structure with cytochrome C labeling. A XOP-mRFP-eGFP-LC3-expressing rod containing a large red autophagic structure that waslabeled by cytochrome C antibody (represented by cyan color). Scale bars: 5 μm. (G’)The cytochrome C labeling in graph G, without showing the mRFP channel. Scale bars:5 μm. CA, cilliary axoneme; IS, inner segment; RPE: retinal pigmented epithelium; PR:69photoreceptor; ONL: outer nuclear layer; INL: inner nuclear layer; GC: ganglion cell.Magenta: WGA-Alexa 647; Blue: Hoechst 33342; Cyan: Cytochrome C.3.3 Time-course of autophagy in rods:I examined the time-course of autophagy in rods using the inducible hsp70-mRFP-eGFP-LC3 line. F1 offspring were raised in cyclic light, and sacrificed at regularintervals; eyes were fixed and examined by confocal microscopy. The number ofautophagic structures per rod were counted and plotted for each time point (Figure 14).Diffuse green signal and green puncta first appeared at 2 h post heat-shock, indicatingthe minimum time required for LC3 biosynthesis, acquisition of green fluorescence, andincorporation into autophagosomes. Yellow (green and red simultaneously) punctaappeared at 8 h, indicating mRFP maturation required a 6 h delay relative to GFP, anda few red puncta were also observed, indicating some autophagosomes fused withlysosomes and became acidified as early as 8 h after heat shock (Figure 14). Thus,autophagosomes persisted for 6 h before starting to fuse with lysosomes andacidification, resulting in red-only puncta. Although the number of these acidifiedautolysosomes began to decline by 28 h, some red puncta persisted for a considerableperiod of time. In contrast, the number of large red structures peaked at 48 h after heat-shock, indicating that they form relatively slowly. Finally, the lack of soluble redfluorescent signal (Figure 13) in combination with the time course data (Figure 14)suggests that the lifetime of soluble LC3 in rods is less than 2 h on average.7071Figure 14: Time-course of autophagy in rods. (A) Quantification of the number ofautophagic structures per rod of hsp70-mRFP-eGFP-LC3 transgenic tadpoles, thatreceived a 1 h heat-shock at time 0 and were sacrificed at different intervals after theheat-shock. The numbers of green, yellow, red and large (or giant) puncta per rod werequantified respectively. Error Bars= S.E.M. N=5 animals per group, at least 3 rods wereimaged per animal. (B-G) Superimposed 3D-zstack confocal micrographs showing rodsof hsp70-mRFP-eGFP-LC3 transgenic tadpoles sacrificed either without heat-shock orat 4, 24, 48, 74 and 174 h after the heat-shock respectively. HS, heat-shock; OS, outersegment; IS, inner segment; N, nucleus. Magenta: WGA-Alexa 647; Blue: Hoechst33342. Scale bar: 5 μm.3.4 Diurnal variation in autophagy:Previous reports suggest an increased number of autophagic vacuoles under lightillumination in frog photoreceptor cells in vitro (Reme & Knop 1980). To examine howautophagy levels in X. laevis rods vary in vivo throughout the day, F1 XOP-mRFP-eGFP-LC3 offspring were raised on a 12L: 12D light cycle and sacrificed every 3 h for24 h. Eyes were fixed and imaged by confocal microscopy.The number of autophagic structures in rods showed a diurnal variation with moreautophagic structures generated in the light and fewer in the dark. The number ofautophagic structures started to increase at 1 h of light exposure, peaked at 10 h, andthen started to decrease after 1 h of darkness. The average number of autophagicstructures per rod increased from 18.6 ± 3.3 (S.E.M.) to 49.5 ± 1.9 (S.E.M.), up to 1.672fold, after 10 h of 1,700 lux light illumination (Figure 15). The composition of theautophagosome population also varied. An immediate increase in the number of greenpuncta at light onset indicated increased autophagic activity; In contrast, the number ofred puncta gradually increased throughout the day, consistent with their delayedproduction and relatively long lifespan. The results suggest an increase in the rate ofgeneration of autophagic structures during light exposure, with minimal compensatingincrease in the rate of autophagosome processing, leading to an accumulation ofstructures throughout the day, until a higher equilibrium number of structures is reached.The observation of diurnal variation in the numbers of red puncta is also consistent withthe observed lifetime of about 28 h (Figure 14) (i.e. if the lifetime was substantiallylonger it would mask diurnal variation in production). The large red structures wereexcluded from this analysis, as they did not occur in sufficient numbers for rigorousanalysis, although they appeared to be relatively invariant, consistent with their slowrate of formation and long lifetime.7374Figure 15: Diurnal variation of autophagy in X. laevis rods expressing XOP-mRFP-eGFP-LC3. (A) Quantification of the number of autophagic structures per rod of XOP-mRFP-eGFP-LC3 transgenic tadpoles that were sacrificed at different time points of theday. The time that light was turned on in the incubator is considered as time 0, tadpoleswere kept in the light during 0-12 h and in the dark during 12-24 h. The numbers ofgreen, yellow, red and total puncta per rod were quantified and indicated as green,yellow, red and black bars respectively in the bar chart. Error Bars= S.E.M. N≥ 5animals per time-point, at least 3 rods were imaged per animal. (B,C) Confocalmicrographs showing rods of XOP-mRFP-eGFP-LC3 transgenic tadpoles which weresacrificed after 1 and 10 h of 1,700 lux light illumination respectively. N≥ 5 animals pertime-point, at least 3 rods were imaged per animal. (D,E) Confocal micrographsshowing rods of XOP-mRFP-eGFP-LC3 transgenic tadpoles which were sacrificed at 16h (12 h of light illumination and 4 h of darkness) and 22 h (12 h of light illumination and10 h of darkness) respectively. N≥ 5 animals per time-point, at least 3 rods were imagedper animal. Magenta: WGA-Alexa 647; Blue: Hoechst 33342. Scale bar: 5 μm. Thenumbers of autophagic structures in rods, including green, yellow, red and total puncta,showed a diurnal variation with more generated in the light and fewer in the dark. (1-way ANOVA, for green, yellow, red, total puncta, P = 0.31, 2.8 X 10-4, 2.6 X 10-8, 1.2 X10-6 respectively; T-test: number of green puncta in the light (any time point) v.s. in thedark (any time point), P = 2.1 X 10-2).753.5 Diurnal variation in WT animals:To determine whether autophagy in transgenic X. laevis expressing XOP-mRFP-eGFP-LC3 behaved similarly to autophagy in non-transgenic siblings, a female XOP-mRFP-eGFP-LC3 transgenic X. laevis (line 1) was mated to a non-transgenic male. Fiftypercent of the resulting offspring carried the transgene. The offspring were raised in 12L:12D light cycles. At 12-13 dpf, tadpoles were sacrificed after 12 h of light or 12 h ofdarkness. Eyes were fixed and processed for electron microscopy (EM).The resulting electron micrographs demonstrate that transgenic X. laevis expressingXOP-mRFP-eGFP-LC3 behaved similarly to non-transgenic siblings; both genotypesshowed more autophagic structures after 12 h of light exposure relative to 12 h ofdarkness (Figure 16). Consistent with published observations that LC3 is not involved ininitiation of autophagy (Nguyen et al. 2016), there was no evidence that expression ofXOP-mRFP-eGFP-LC3 significantly enhanced autophagy in the transgenic animals.76Figure 16: Diurnal variation of autophagy in wildtype X. laevis rods. Electronmicrographs of the inner segments of rods from (A) a wildtype tadpole that wassacrificed after 12 h of 1,700 lux light illumination, (B) a wildtype tadpole that wassacrificed after 12 h of darkness, (C) a XOP-mRFP-eGFP-LC3 transgenic tadpole thatwas sacrificed after 12 h of 1,700 lux light illumination, (D) a XOP-mRFP-eGFP-LC3transgenic tadpole that was sacrificed after 12 h of darkness. N = 2 animals per group.Note the presence of autophagic structures are indicated by white arrowheads. Scalebar: 500 nm.773.6 Diurnal variation of autophagy is due to lightillumination, not circadian rhythm.To determine whether the diurnal variation of autophagy observed in Figure 15 wasregulated via circadian rhythm, XOP-mRFP-eGFP-LC3 line 1 F1 offspring were raisedin 12L: 12D cyclic light incubator from fertilization onwards. On 12 dpf, a subset oftadpoles were transferred into constant darkness prior to light onset (indicated by blueline in Figure 17A), and a subset were transferred to constant light prior to light offset(indicated by green line in Figure 17A), and another subset were exposed to 10 h oflight then transferred to darkness (with 0-10 h indicated by green line and 10-34 hindicated by red line in Figure 17A). Tadpoles were sacrificed at different time points,and the numbers of autophagic structures were compared. The large red puncta wereexcluded from the analysis. In contrast to animals exposed to cyclic light, animals inextended dark or light conditions showed little variation in autophagy levels (Figure 17),indicating that the diurnal variation of autophagy in rods is not controlled by circadianrhythm. Moreover, tadpoles kept in extended light exposure (TO - T22, green line)maintained a high level of autophagy; tadpoles kept in extended dark (T22 - T34, redline) maintained a low level of autophagy, with a slightly decrease. These resultssuggest that autophagy level in rods is controlled exclusively by the level of illumination.Neither the induction of autophagy in the light nor the decline in autophagy in the dark isinfluenced by circadian rhythm.7879Figure 17: Diurnal variation of autophagy is due to light illumination, not circadianrhythm. (A) Quantification of the number of autophagic structures per rod of transgenicXOP-mRFP-eGFP-LC3 tadpoles sacrificed at different time points. Error Bars= S.E.M.(B-J) Confocal micrographs showing rods of XOP-mRFP-eGFP-LC3 transgenictadpoles that were sacrificed at different lightening conditions: (B) time 0, before lightonset, (C) 1 h of darkness, (D) 10 h of darkness, (E) 22 h of darkness, (F) 1 h of of1,700 lux light illumination, (G) 10 h of light illumination, (H) 22 h of light illumination, (I)10 h of light illumination plus 12 h of darkness, (J) 10 h of light illumination plus 24 h ofdarkness. N ≥ 4 animals per group, at least 2 rods were imaged per animal. After 10 hof light illumination, the number of autophagic structures in tadpole rods wassignificantly higher than that of rods of tadpoles that remained in the dark (t-test, *P =1.1 X 10-3); the number of autophagic structures of tadpole rods maintained in the lightdid not vary significantly with time (T10 v.s. T22) (t-test, P = 0.48); the number ofautophagic structures of tadpole rods maintained in the dark did not vary significantlywith time (T10 v.s. T22) (t-test, P = 0.22). Magenta: WGA-Alexa 647; Blue: Hoechst33342. Scale bar: 5 μm.3.7 Autophagy induced by the misfolding rhodopsin mutantP23H.To determine whether misfolding of rhodopsin can cause increased autophagy, Igenerated tadpoles co-expressing the RP-associated destabilized rhodopsin mutantP23H and a dually fluorescent LC3 autophagy marker. The P23H mutation promotes80instability and misfolding of rhodopsin, and can cause retinal degeneration in manyspecies (as mentioned above in section 1.6). In transgenic X. laevis, retinaldegeneration caused by bovine P23H rhodopsin (bP23H) can be fully prevented byconstant darkness, and induced by light exposure (Tam & Moritz 2007b).A XOP-mRFP-eGFP-LC3 female frog (line 2) was mated to a heterozygous bP23Hmale frog of a previously described lineage (Tam & Moritz 2007b). Embryos werecollected, and raised in constant darkness. At 6 dpf, tadpoles expressing the fluorescentautophagy marker were identified by mRFP/GFP fluorescence and returned to the dark.At 10 dpf, a group of tadpoles were exposed to 12 h of 1700 lux light and sacrificed(T12); another group of tadpoles were exposed to one complete light cycle (12 h of1700 lux light + 12 h of darkness) and sacrificed (T24). Genotypes were determined byQ-PCR of tail tissue using bovine rhodopsin-specific primers and a TaqMan probe.Autophagic structures were counted and compared between tadpoles expressing onlyendogenous rhodopsin and tadpoles expressing bP23H rhodopsin.Rods expressing bP23H rhodopsin had higher autophagy levels than rods expressingonly endogenous rhodopsin after 12 h of light exposure (Figure 18A). The averagenumber of total autophagic structures per rod is 49.6 ± 2.5 (S.E.M) (N= 6 animals) inwild-type rhodopsin expressing-rods and is 68.8 ± 5.5 (S.E.M) (N=6 animals) in bP23Hrhodopsin-expressing rods, with about 39% increase, indicating that bP23H rhodopsinexpression can upregulate autophagy in rods.After one complete light cycle (T24:12 h of light + 12 h of darkness), extensive retinaldegeneration had occurred; in combination with the mosaic transgene expression, this81greatly limited the number of rods expressing mRFP-eGFP-LC3 available forexamination at this and further time points; therefore similar counting of structures wasnot attempted. The few remaining rods frequently showed abnormal nuclearmorphologies indicating cell death was in progress (Figure 18 D’&E’). Qualitatively, incomparison to other samples observed throughout this study, the autophagic structuresin these rods appeared atypical, with a higher proportion of green structures andabnormally large morphologies (Figure 18 D&E), suggesting either enhanced initiationof autophagy, or failure of autophagosomes to mature. These large structures wereconsistent with our previous EM studies (Figure 11) (Bogéa et al. 2015). The resultssuggest that in light exposed retinas, rhodopsin misfolding may place demands on theautophagy pathway that exceed the cells’ capacity, resulting in retinal degeneration.8283Figure 18: Increased autophagy in bP23H rhodopsin-expressing rods. (A)Quantification of the number of autophagic structures per rod of XOP-mRFP-eGFP-LC3transgenic tadpoles either co-expressing wildtype rhodopsin or bP23H rhodopsin, bothgroups of animals were exposed to 12 h of 1,700 lux light illumination. The black andgrey bars indicate the number of autophagic structures per rod of wildtype rhodopsin-expressing tadpoles and bP23H-expressing tadpoles respectively. The numbers ofgreen, yellow, red and total puncta per rod were quantified respectively. Error Bars=S.E.M. N=6 animals per group, at least 3 rods were imaged per animal. Rodsexpressing bP23H rhodopsin had higher autophagy levels than rods expressing onlyendogenous rhodopsin (T-test, for green, yellow, red and total puncta, P = 2.3 X 10-2,6.4 X 10-2, 6.8 X 10-2, 1.6 X 10-2 respectively). * represents P < 0.05. (B, C) Confocalmicrographs showing rods of either a wildtype rhodopsin-expressing tadpole or abP23H-expressing tadpole after 12 h of light illumination. (D, E) Confocal micrographsof bP23H-expressing rods exposed to one light cycle (12L + 12D). Notice the presenceof condensed nuclei are marked by white arrowheads. (D’, E’) The Hoechst channels ofgraph D and E showing the condensed nuclei. Magenta: WGA-Alexa 647; Blue:Hoechst 33342. Scale bar: 5 μm3.8 Autophagy regulation by chromophore availability.Rhodopsin is composed of the protein moiety opsin and the chromophore 11-cis retinal,which is responsible for absorbing photons and initiating the phototransduction, and 11-cis retinal is converted to all-trans retinal in this process (Molday & Moritz 2015). The84rhodopsin chromophore can act as a pharmacological chaperone to promote the foldingof P23H and wildtype opsins (Noorwez et al. 2004). I hypothesized that the increases inautophagy caused by bP23H rhodopsin and light exposure may be related, becauselight exposure may promote misfolding of wildtype rhodopsin by limiting chromophoreavailability during biosynthesis. To test this hypothesis I knocked out the genesencoding the enzyme RPE65, the isomerohydrolase that is essential for chromophoreregeneration (see introduction section 1.4 phototransduction). X. laevis have twoRPE65 homologues (RPE65.L and RPE65.S). I injected cas9 mRNA and a single guideRNA (sgRNA) targeting both RPE65 genes into single-cell embryos derived from XOP-mRFP-eGFP-LC3 line 1. Embryos without Cas9 and sgRNA injection served as controls.Both the knock-out and control groups were raised in 12L: 12D light cycles for 12 daysand sacrificed on either day 12 after 12 h of light exposure or day 13 after 12 h ofdarkness. Genomic DNA was isolated from tail tissue, and eyes were fixed,cryosectioned and examined by confocal microscopy.I examined editing of the REP65.L and RPE65.S genes by amplifying regionscontaining the predicted cut site by PCR, and analyzing PCR products by Sangersequencing. Sequencing traces showed markedly decreased sequencing fidelity justprior to the predicted cut site in samples injected with Cas9/sgRNA (Figure 19B),indicating the presence of multiple PCR products differing in size by a small number ofbases. These results were consistent with our previous use of CRISPR/Cas9 in X.laevis (Feehan et al. 2017). Control retinal sections showed abundant labelling in theRPE cell layer using an anti-RPE65 antibody, while the majority of animals in thesgRNA-injected group showed much fainter or no RPE65 antibody labelling (Figure8519A), although knock-out efficiency varied between samples. I then selected samples(N = 7 animals per group) with no or very minimal RPE65 antibody labelling, indicatinghigh knock-out efficiency, for quantification of autophagic structures.Consistent with experimental results shown in Figure 17, rods of the control groupshowed light-dependent autophagy. I observed abundant autophagic structures in rodinner segments of tadpoles sacrificed after 12 h of light exposure, and very fewautophagic structures in rods of tadpoles sacrificed after 12 h of darkness. In the knock-out group, animals sacrificed after 12 h of light exposure showed dramaticallydecreased autophagy compared to the control group. The number of autophagicstructures found in knock-out animals sacrificed after 12 h of darkness did not show asignificant change compared to the control group (Figure 19C). The results weresignificant by two-way ANOVA with P = 3.7 X 10-10 (N=7 animals per group). Theeffects of lighting (P = 2.2 X 10-7) and knock-out (P = 2.5 X 10-7) were both significant,as was the interaction between lighting and knockout (P = 2.1 X 10-6), indicating thatRPE65 knock-out specifically reduces autophagy in light-exposed animals. Thiscombination of results suggests that increased autophagy in light-exposed rods is notcaused by misfolding of wild-type rhodopsin due to lack of chromophore, and insteadsuggests that autophagy in photoreceptors is regulated by phototransduction.8687Figure 19:Autophagy regulation by chromophore availability. (A) Confocalmicrographs showing retinal sections labeled with anti-RPE65 antibody (representedwith magenta color) derived from either the control animals (top panel) or RPE65 knock-out animals (bottom panel) either exposed to 12 h of light or reared in 12 h of dark.N=10 animals per group. The Hoechst dye labeled the nuclei of RPE cells. Scale bars:20 μm. (B) Regions containing the predicted cut site of the REP65.L and RPE65.Sgenes were amplified by PCR and sequenced. The top and bottom panels show thesequencing traces from a control animal and a RPE65-knock out animal respectively. (C)Confocal micrographs of rods from either a control tadpole or an RPE65-kncok outtadpole either exposed to 12 h of light or kept in 12 h of dark. N=7 animals per group, atleast 3 rods were imaged per animal. The bar chart shows the quantification of thenumber of autophagic structures per rod, with the black and gray bars indicating thenumber of autophagic structures in rods of control animals and RPE65-knock outtadpoles respectively. The results were significant by two-way ANOVA with P = 3.7 X10-10. The effects of lighting (P = 2.2 X 10-7) and knock-out (P = 2.5 X 10-7) were bothsignificant, as was the interaction between lighting and knockout (P = 2.1 X 10-6),indicating that RPE65 knock-out specifically reduces autophagy in light-exposedanimals. Magenta: WGA-Alexa 647; Blue: Hoechst 33342. Scale bar: 5 μm883.9 Autophagy levels increased in X. laevis rods expressingP23H rhodopsinWe have previously reported that there were increased numbers of autophagicstructures in X. laevis rods expressing a misfolding-prone P23H rhodopsin mutant(Bogéa et al. 2015). In transgenic X. laevis, retinal degeneration caused by bovineP23H rhodopsin can be fully prevented by constant darkness, and induced by lightexposure (Tam & Moritz 2007b). Increased numbers of autophagic structures wereobserved in bP23H-expressing rods exposed to light and in hP23H-expressing rods byelectron microscopy (Bogéa et al. 2015; Vent-schmidt et al. 2017); Larger vesicularstructures were apparent in these samples, consistent with the induction of autophagy.To verify the presence of autophagosomes in these cells by an alternate method, aheterozygous frog expressing bovine P23H rhodopsin was mated to a wild-type frog.Offspring were reared without light exposure until 14 dpf. Under these conditions, retinaldegeneration is completely prevented in these animals (Tam & Moritz 2007b). A groupof 30 tadpoles were sacrificed in the dark (TO) on 14 dpf, the remaining 30 tadpoleswere transferred to cyclic light and sacrificed after 12 h of light illumination (T12).Tadpole eyes were labeled with anti-light chain 3 (LC3) A/B rabbit polyclonal antibody(product code Ab128025; AbCam, Cambridge, UK) at 1:1000 dilution, which associateswith autophagosomes and autolysosomes. The antibody recognizes both LC3I (soluble)and LC3II (cleaved, lipidated, and associated with autophagosomes) forms. PunctateLC3 labeling confirmed the presence of autophagic compartments in wild-type and89bP23H-expressing retinas of tadpoles kept in the dark or exposed to light for 12 h.Figures 20A through 20E show that these compartments were located in rod innersegments, but they were also observed in all other retinal cell layers, and in the outersegment layer. Relative to WT siblings, I observed increased numbers of puncta in light-exposed bP23H photoreceptors (Figure 20F), confirming results obtained by electronmicroscopy. LC3–positive puncta located in the OS layer were likely associated withinterdigitating processes of the RPE (Figures 20G–I), and therefore were not includedin counts shown in Figure 20F.9091Figure 20: LC3B immunolabeling confirmed the presence of increased number ofautophagic compartments in rod inner segments of tadpoles expressing bP23H. Imagesare derived from dark-reared WT tadpole (A); WT tadpole exposed to 12 h of light (B);dark-reared bP23H-expressing tadpole (C); and bP23H-expressing tadpole exposed to12 h of light (D). Arrowheads indicate LC3 puncta. (E) wildtype tadpole-control labelingwithout primary antibody. (F) Quantification of LC3-positive puncta in rod innersegments. Error bars = S.E.M. n ≥ 4 animals per group. *P = 0.041. (G–I) LC3-positivepuncta present in the OS layer are likely associated with interdigitating processes of theRPE. Electron micrographs show no autophagic structures within rod and cone OS (G),but double-membrane compartments are present in interdigitating processes of the RPE(arrowheads). (H, I) Further magnification of the boxed regions in (G). Scale bars: (A)10 μm; (G) 2 μm; (H, I) 500 nm (Bogéa et al. 2015).3.10 HDAC inhibitors induced autophagic structures inwildtype X. laevis rods.We formerly investigated the therapeutic potential of VPA using four transgenic X. laevismodels of adRP-expressing human rhodopsin mutants, including P23H rhodopsin, T4Krhodopsin, T17M rhodopsin, and Q344Ter rhodopsin (Vent-schmidt et al. 2017). Thefirst three models show marked light dependence of RD (Tam & Moritz 2007b; Tam &Moritz 2009). We found that VPA may ameliorate or exacerbate RD depending on theunderlying mutation. VPA ameliorated RD and vision deficits caused by P23H rhodopsin,but dramatically exacerbated RD caused by T17M rhodopsin in the presence of light.92Effects in other models were negative and less pronounced (Vent-schmidt et al. 2017).Similar results were obtained with three other HDAC inhibitors, but not with otherantipsychotics, VPA structural analogs, or chemical chaperones, indicating that both thepositive and negative effects that we observed were due to HDAC inhibitor activity(Vent-schmidt et al. 2017).We also found that VPA treatment decreased the burden of misfolded P23H rhodopsin(Vent-schmidt et al. 2017) in X. laevis expressing P23H rhodopsin. VPA has beenreported to be an activator of autophagy (Sarkar et al. 2005). To investigate whetherVPA increased autophagy in rods, I examined rods in WT retinas treated with VPA byEM and found that their inner segments contained increased numbers of autophagicstructures, identifiable as vesicular structures with resolvable membranous orcytoplasmic contents, sometimes surrounded by a double membrane (Figure 21 C&D).NaBu has also been reported to increase autophagy (Seong Lee & Lee 2012) and Iobserved similar effects in NaBu-treated retinas (21 E,E′). In addition, I found signs ofincreased autophagy in untreated retinas expressing human P23H rhodopsin (Figure 21F,F′), similar to results obtained previously with bovine P23H rhodopsin (Bogéa et al.2015), suggesting that RD caused by human P23H rhodopsin is associated withincreased autophagy, potentially as a protective response to large quantities ofmisfolded rod opsin. However, whether the beneficial effect of HDAC inhibitors on P23Hretinal degeneration is mediated by activation of autophagy is unclear.9394Figure 21: HDAC inhibitors induced autophagic structures in wildtype X. laevis rods. A,B, Untreated WT. C, D, VPA-treated WT. Structures indicated by arrowheads areconsistent with autophagosomes or autolysosomes. Structures indicated by arrows areconsistent with newly forming autophagosomes (phagophores). Small vesicularstructures morphologically consistent with autophagosomes and autolysosomesincreased with VPA treatment consistent with an increase in autophagy. E, Vesicularstructures also increased in rods treated with NaBu. F, Vesicular structures wereincreased in photoreceptors expressing P23H rhodopsin, consistent with previousstudies suggesting induction of autophagy during retinal degeneration. Larger vesicularstructures were also apparent in these samples. C′, E′, F′, Boxed structures from C, E,and F shown at higher magnification. N= 3 animals per group. M, Mitochondria; ER,endoplasmic reticulum; N, nucleus. Scale bar, 500 nm (Vent-schmidt et al. 2017).The above mentioned results (section 3.9 and 3.10) showed increased numbers ofautophagic structures in P23H rhodopsin-expressing rods by immunohistochemistry andwildtype rods treated with HDAC inhibitors by EM. These results showed the transientautophagy levels at the moments when animals were sacrificed, but not the autophagicflux. To determine whether HDAC inhibitors can increase autophagic flux in X. laevisrods, tadpoles expressing XOP-mRFP-EGFP-LC3 were treated with three differentHDAC inhibitors, including VPA, NaBu and CI-994. The autophagy levels of tadpolestreated with HDAC inhibitors were quantified and compared to that of non-treatedtadpoles. The XOP-mRFP-EGFP-LC3 reporter system allows me to track the dynamicchanges of green, yellow, red and total autophagic puncta.953.11 Rapamycin treatment up-regulated autophagic flux in X.laevis rods.I first tested the XOP-mRFP-EGFP-LC3 reporter system with a commonly usedautophagic inducer rapamycin which can cross the blood-brain barrier (Chi et al. 2017).By suppressing the mTOR signaling pathway, it has been widely reported to induceautophagy both in vivo and in vitro (Yang et al. 2013). It has also been used as ananti-rejection drug in organ transplantation (Saunders et al. 2001). To assess theeffects of rapamycin on autophagy in X. laevis, I first examined its toxicity. I testeddifferent concentrations of rapamycin (Sigma-Aldrich 37094, CAS number 53123-88-9)on wildtype X. laevis larvae. Three groups of larvae were treated with 0.04 μM (or 0.04μg/ml), 0.22 μM (or 0.2 μg/ml) and 1.10 μM (or 1μg/ml) rapamycin in 0.2% dimethylsulfoxide (DMSO) ringer solution respectively, each group had 10 tadpoles; anothergroup of 5 tadpoles were treated with 5.48 μM (or 5 μg/ml) rapamycin in 0.2% DMSOringer solution; a group of 10 tadpoles were raised in 0.2% DMSO ringer solution andserved as controls. Each tadpole was treated with 2ml of tadpole ringer solution with orwithout rapamycin in a 24-well plate. Treatment started from 2 dpf to 7 dpf, with drugsrenewed daily. Tadpoles treated with rapamycin were all alive after 6 days of treatmentregardless of the concentrations of rapamycin used, they all looked healthy and weresimilar to tadpoles in the non-treated control group. I then used the highestconcentration tested (5.48 μM) in the future experiments that required rapamycintreatments.96To determine whether rapamycin can enhance autophagy in X. laevis rods, XOP-mRFP-EGFP-LC3 female 1 (carrying the transgene at a single integration site) wasmated with a wildtype male frog. Half of the offspring expressed the XOP-mRFP-EGFP-LC3 transgene and the other half were wildtype. The offspring were divided into twogroups; one group of 30 animals were treated with 5.48 μM rapamycin (CaymanChemical ,item number 13346, CAS number 53123-88-9 ) in 0.2% DMSO ringersolution from 2 dpf to 5 dpf , with drugs renewed daily, while the other group of 30animals were raised in 1x tadpole ringer with 0.2% DMSO and served as controls. Bothgroups of animals were raised in 12L:12D cyclic light cycles at 18°C. Rapamycinsolution was renewed at the beginning of dark cycles daily (rapamycin is light sensitive).At 5 dpf, tadpoles were screened for eGFP and mRFP fluorescence by using anepifluorescence-equipped Leica MZ16F dissecting microscope; only transgenictadpoles expressing XOP-mRFP-EGFP-LC3 were raised and analyzed. At 6 dpf, bothgroups of tadpoles were sacrificed after receiving 5 h of light illumination. Eyes werefixed, cryosectioned and examined by confocal microscope. 3D Z-stack imaging wasperformed on intact individual rod that expressed fluorescent LC3 as described above(section 2.4). The number of autophagic structures per rod was then quantified basedon the 3D images. At least 3 rods were imaged per animal, and 10 animals in both non-treated and rapamycin treated groups were included in the data analyses.Compared to the tadpoles in the non-treated control group, rapamycin treated tadpolesshowed more autophagic puncta in their rods (Table 1). The average number of totalautophagic structures per rod is 44.2 ± 2.4 (S.E.M) (n=10 animals) in non-treated rods97and is 55.6 ± 1.0 (S.E.M) (n=10 animals) in rapamycin-treated rods, with a 27%increase. Rapamycin treated tadpoles showed increases in all categories of green,yellow and red puncta, with a 10%, 8% and 36% increase respectively, indicating thatautophagic flux was up-regulated (Figure 22).Average number ofautophagic puncta per rodGreenpunctaYellowpunctaRedpunctaTotalpunctaNon-treated group 9.8± 0.7812.7± 2.221.7± 1.644.2± 2.4Rapamycin treated group 11.3± 1.314.0± 1.130.4± 1.755.6± 1.0Table 1: The average number of autophagic puncta per rod in non-treated andrapamycin-treated X. laevis tadpoles. Error=S.E.M.98Figure 22: Rapamycin treatment up-regulated autophagic flux in X. laevis rods. (A) Aconfocal micrograph of a rod from a non-treated X. laevis tadpole expressing XOP-mRFP-EGFP-LC3. (B) A confocal micrograph of a rod from a rapamycin-treated tadpoleexpressing XOP-mRFP-EGFP-LC3. (C) Quantification of the numbers of autophagicstructures per rod of non-treated and rapamycin treated tadpoles. The black barsindicate the number of autophagic structures in rods of non-treated animals and thegray bars indicate the number of autophagic structures in rods of rapamycin treatedtadpoles. The numbers of green, yellow, red and total puncta per rod were quantifiedrespectively. Error bar = S.E.M. N=10 animals per group, at least 3 rods were imagedper animal. (T-test, for green, yellow, red and total puncta, P = 0.37, 0.62, 1.5 X 10-3,9.7 X 10-4 respectively). * represents P < 0.05. NT, non-treated; Rapa: Rapamycin;magenta: WGA-Alexa 647; Blue: Hoechst 33342. Scale bars: 5 μm.3.12 Chloroquine treatment inhibited autophagic flux in X.laevis rods.I also tested the XOP-mRFP-EGFP-LC3 reporter system with a widely used autophagyinhibitor chloroquine (Yang et al. 2013). It is also a well-known medication used toprevent and treat malaria (Homewood et al. 1972). Upon treatment, chloroquineaccumulates inside acidic cellular compartments, such as endosomes and lysosomes,and inhibits their acidification (Steinman et al. 1983). By raising the lysosomal pH,chloroquine inhibits the fusion of autophagosomes with lysosomes, and also inhibitslysosomal protein degradation (Shintani & Klionsky 2004; Wibo & Poole 1974).99Chloroquine can cross the blood-brain barrier (Mielke et al. 1997). Interestingly,hydroxychloroquine was reported to cause retinal degeneration in humans, such aspatients with autoimmune diseases who received hydroxychloroquine treatment(Motarjemizadeh et al. 2016; Geamănu Pancă et al. 2014).To determine whether chloroquine can inhibit autophagy in rods of X. laevis, the frogXOP-mRFP-EGFP-LC3 female 1 was again mated with a wildtype male frog. I firstexamined its toxicity and no concentration below 1 mM was found to be toxic. Offspringwere collected and screened for GFP/mRFP as described under section 3.11 above. At5 dpf, the transgenic tadpoles were divided into two groups. One group of 15 tadpoleswas treated with 1 mM chloroquine for 24 h starting at 5 h following light onset on 5 dpf,while the other group of 15 tadpoles were raised in 1x tadpole ringer and served ascontrols. Both groups of tadpoles were sacrificed on 6 dpf and imaged the same way asdescribed under section 3.11.Compared to the non-treated control group, tadpoles treated with chloroquine had 30%fewer green autophagic puncta and 91% more yellow puncta per rod (Figure 23),indicating autophagosome accumulation. This was consistent with inhibition ofacidification, which would be expected to preserve green fluorescence inautolysosomes, and reduced fusion of autophagosomes with lysosomes. Tadpolestreated with chloroquine also had 25% more red autophagic puncta per rod (Figure 23),indicating autolysosome accumulation, consistent with the ability of chloroquine toinhibit lysosomal protein degradation. These data together suggest that chloroquine100treatment partially inhibited autophagic flux in rods. The reduction in green puncta (t-test,P = 7.5 X 10-4) may indicate the presence of a feedback mechanism leading todecreased initiation of autophagy on inhibition of autophagic flux, although I am notaware of any reports of such a mechanism.By treating the XOP-mRFP-eGFP-LC3 reporter system with a commonly usedautophagy activator rapamycin and an autophagy inhibitor chloroquine, I determinedthat the XOP-mRFP-eGFP-LC3 reporter system can report changes of autophagy fluxin rods.Average number ofautophagic puncta per rodGreenpunctaYellowpunctaRedpunctaTotalpunctaNon-treated control group 10.3± 0.7210.7± 0.8616.3± 1.337.3± 2.0chloroquine treated group 6.7± 0.4920.5± 1.520.3± 2.047.5± 2.5Table 2: The average number of autophagic puncta per rod in non-treated andchloroquine treated X. laevis tadpoles. Error=S.E.M.101Figure 23: Chloroquine treatment inhibited autophagic flux in X. laevis rods. (A) AConfocal micrograph of a rod from a non-treated X. laevis tadpole expressing XOP-mRFP-EGFP-LC3. (B) A confocal micrograph of a rod from a chloroquine-treated X.laevis tadpole expressing XOP-mRFP-EGFP-LC3. Tadpoles were raised in ringersolution before 5 dpf and were treated with 1mM chloroquine for 24 h before sacrificedon 6 dpf. (C) Quantification of the numbers of autophagic structures per rod of non-treated and chloroquine treated tadpoles. The black bars indicate the number ofautophagic structures in rods of non-treated animals and the gray bars indicate thenumber of autophagic structures in rods of chloroquine treated tadpoles. The numbersof green, yellow, red and total puncta per rod were quantified respectively. Error bar=S.E.M. N=10 animals per group, at least 3 rods were imaged per animal. (T-test, forgreen, yellow, red and total puncta, P = 7.5 X 10-4, 5.2 X 10-5, 0.11, 5.3 X 10-3respectively). * represents P < 0.05. NT, non-treated; CQ: chloroquine; magenta: WGA-Alexa 647; Blue: Hoechst 33342. Scale bars: 5 μm.1023.13 VPA treatment up-regulated autophagic flux in X. laevisrods.VPA has been reported to be an activator of autophagy (Sarkar et al. 2005). It is also anHDAC inhibitor (Vent-schmidt et al. 2017). We previously determined that VPAameliorated retinal degeneration associated with P23H rhodopsin and increasedautophagic structures in rod inner segments (Vent-schmidt et al. 2017). To determinewhether VPA can increase autophagic flux in X. laevis rods, XOP-mRFP-EGFP-LC3female 1 was mated with a wildtype male frog. The offspring were separated into twogroups, either treated with 10 µM VPA in 1x Ringer solution or raised in 1x Ringersolution without drug (For details of drug therapy procedures, please see section 2.3).On 6 dpf, animals were sacrificed and imaged the same way as described above insection 3.11.Compared to the tadpoles in the non-treated control group, tadpoles treated with VPAhad more fluorescent autophagic puncta in their rods. The average number of totalautophagic structures per rod is 44.5 ± 2.2 (S.E.M) (n=13 animals) in non-treated rodsand is 54.8 ± 3.1 (S.E.M) (n=13 animals) in VPA-treated rods, with a 22% increase(Table 3). After treatment with VPA, the number of green puncta, indicating newlysynthesized autophagic structures, increased by 7%; the number of yellow punctaincreased by 50%; the number of red puncta, indicating late stage autolysosomes,increased by 22% (Figure 24). These data together suggest that VPA treatmentincreased autophagic flux in rods.103average number ofautophagic structures per rodGreenpunctaYellowpunctaRedpunctaTotalpunctaNon-treated control group 14.6± 0.686.2± 0.6822.9± 1.944.5± 2.2VPA treated group 16.4± 0.869.2± 0.9828.2± 2.654.8± 3.1Table 3: The average number of autophagic puncta per rod in non-treated and VPAtreated X. laevis tadpoles. Error=S.E.M.Figure 24: VPA treatment up-regulated autophagic flux in X. laevis rods. (A) A confocalmicrograph of a rod from a non-treated X. laevis tadpole expressing XOP-mRFP-EGFP-LC3. (B) A confocal micrograph of a rod from a VPA-treated tadpole expressing XOP-mRFP-EGFP-LC3. (C) Quantification of the numbers of autophagic structures per rod of104non-treated and VPA treated tadpoles. The black bars indicate the number ofautophagic structures in rods of non-treated animals and the gray bars indicate thenumber of autophagic structures in rods of VPA treated tadpoles. The numbers of green,yellow, red and total puncta per rod were quantified respectively. Error bars = S.E.M.N=13 animals per group, at least 3 rods were imaged per animal. For green, yellow, redand total puncta, P = 0.11, 1.8 X 10-2, 0.12, 1.2 X 10-2 respectively. * represents P <0.05. NT, non-treated; magenta: WGA-Alexa 647; Blue: Hoechst 33342. Scale bars: 5μm.3.14 NaBu treatment up-regulated autophagic flux in X.laevis rods.NaBu has been reported to increase autophagy in Chinese hamster ovary cells (SeongLee & Lee 2012) and I also observed increased autophagic structures in NaBu-treatedrods by EM (Figure 21E). NaBu is also an HDAC inhibitor (Vent-schmidt et al. 2017). Todetermine whether NaBu can increase autophagic flux, XOP-mRFP-EGFP-LC3 female1 was mated with a wildtype male frog. The offspring (n=30 animals) were separatedinto two groups, either treated with 300 μM NaBu in 1 X Ringer solution or raised in 1xRinger solution without the drug (For details of drug therapy procedures, please seesection 2.3). The concentrations known to alter retinal degeneration were previouslydetermined (Vent-schmidt et al. 2017). Animals were sacrificed and imaged asdescribed above in section 3.11.105Compared to the tadpoles in non-treated control group, tadpoles treated with NaBuhave more fluorescent autophagic puncta in their rods. The average number of totalautophagic structures per rod is 45.5 ± 1.6 (S.E.M) (n=10 animals) in non-treated rodsand is 58.1 ± 2.3 (S.E.M) (n=10 animals) in NaBu-treated rods, with a 26% increase(Table 4). After treatment with NaBu, the number of green puncta, indicating newlysynthesized autophagic structures, increased by 9%; the number of yellow punctaincreased by 10%; the number of red puncta, indicating late stage autolysosomes,increased by 40% (Figure 25). These data together suggest that NaBu treatmentincreased autophagic flux in rods.average number ofautophagic structures per rodGreenpunctaYellowpunctaRedpunctaTotalpunctaNon-treated control group 11.0± 0.989.9± 0.9324.6± 2.045.5± 1.6NaBu treated group 12.4± 0.5210.6± 1.235.1± 1.858.1± 2.3Table 4: The average number of autophagic puncta per rod in non-treated and NaButreated X. laevis tadpoles. Error=S.E.M.Note: Three (out of 10) NaBu-treated retinas did not appear healthy. Some of the rodsin these three retinas have almost no fluorescent LC3 puncta, only with a solublefluorescent background. These“abnormal”retinas were excluded from the analyses.106Figure 25: NaBu treatment up-regulated autophagic flux in X. laevis rods. (A) Aconfocal micrograph of a rod from a non-treated X. laevis tadpole expressing XOP-mRFP-EGFP-LC3. (B) A confocal micrograph of a rod from a NaBu-treated tadpoleexpressing XOP-mRFP-EGFP-LC3. (C) Quantification of the numbers of autophagicstructures per rod of non-treated and NaBu treated tadpoles. The black bars indicatethe number of autophagic structures in rods of non-treated animals and the gray barsindicate the number of autophagic structures in rods of NaBu-treated tadpoles. Thenumbers of green, yellow, red and total puncta per rod were quantified respectively.Error bars= S.E.M. N=10 animals per group, at least 3 rods were imaged per animal.For green, yellow, red and total puncta, P = 0.22, 0.68, 9.4 X 10-4, 3.3 X 10-4respectively. * represents P < 0.05. NT, non-treated; magenta: WGA-Alexa 647; Blue:Hoechst 33342. Scale bars: 5 μm.1073.15 CI-994 treatment up-regulated autophagic flux in X.laevis rods.CI-994 is an HDAC inhibitor and we previously determined that it can ameliorate retinaldegeneration associated with P23H rhodopsin and induce autophagic structures in rods(Vent-schmidt et al. 2017). To determine whether CI-994 can increase autophagic fluxin rods, XOP-mRFP-EGFP-LC3 female 1 was mated with a wildtype male frog. Theoffspring were separated into two groups, either treated with 10 µg/ml (or 37.2 µM) CI-994 or non-treated (For details of drug therapy procedures, please see section 2.3). Theconcentrations known to alter retinal degeneration were previously determined (Vent-schmidt et al. 2017). Animals were sacrificed and imaged the same way as describedabove in section 3.11.Compared to the tadpoles in non-treated control group, tadpoles treated with CI-994have more fluorescent autophagic puncta in their rods. The average number of totalautophagic structures per rod is 42.2 ± 2.4 (S.E.M) (n=10 animals) in non-treated rodsand is 57.6 ± 3.1(S.E.M) (n=10 animals) in CI-994-treated rods, with a 38% increase(Table 5). After treatment with CI-994, the number of green puncta, indicating newlysynthesized autophagic structures, increased by 22%; the number of yellow punctadidn’t change significantly; the number of red puncta, indicating late stageautolysosomes, increased by 78% (Figure 26). These data together suggested that CI-994 treatment increased autophagic flux in rods.108average number ofautophagic structures per rodGreenpunctaYellowpunctaRedpunctaTotalpunctaNon-treated control group 9.2± 0.7315.4± 2.017.6± 2.742.2± 2.4CI-994 treated group 10.8± 1.315.1± 1.531.7± 2.457.6± 3.1Table 5: The average number of autophagic puncta per rod in non-treated and CI-994treated X. laevis tadpoles. Error=S.E.M.Figure 26: CI-994 treatment up-regulated autophagic flux in X. laevis rods. (A) Aconfocal micrograph of a rod from a non-treated X. laevis tadpole expressing XOP-mRFP-EGFP-LC3. (B) A confocal micrograph of a rod from a CI-994-treated tadpoleexpressing XOP-mRFP-EGFP-LC3. (C) Quantification of the numbers of autophagic109structures per rod of non-treated and CI-994 treated tadpoles. N=10 animals per group,at least 3 rods were imaged per animal. The black bars indicate the number ofautophagic structures in rods of non-treated animals and the gray bars indicate thenumber of autophagic structures in rods of CI-994 treated tadpoles. The numbers ofgreen, yellow, red and total puncta per rod were quantified respectively. Error bars =S.E.M. N ≥ 10 per group. For green, yellow, red and total puncta, P = 0.28, 0.91, 9.8 X10 -4, 1.1 X 10-3 respectively. * represents P < 0.05. NT, non-treated; magenta: WGA-Alexa 647; Blue: Hoechst 33342. Scale bars: 5 μm.3.16 Effect of rapamycin on P23H retinal degeneration.In the former sections (3.13-3.15), I determined that HDAC inhibitors, including VPA,NaBu and CI-994, consistently promoted autophagy flux in X. laevis rods. Whether thebeneficial effect of HDAC inhibitors on P23H retinal degeneration (as mentioned abovein section 3.10) is mediated by activation of autophagy is not clear. I hypothesized thatautophagy can degrade misfolded rhodopsin proteins, and rescue retinal degenerationby reducing the toxic effects of misfolded rhodopsin. To test this hypothesis, I usedrapamycin to enhance autophagy in rods expressing a misfolding-prone P23Hrhodopsin and measured the extent of retinal degeneration.A heterozygous female frog expressing human P23H rhodopsin (f1-f2-f1) was matedwith a wildtype male frog. The offspring were collected and separated into four groups110(N=22 animals per group). Two groups were raised in cyclic light either non-treated ortreated with 5.5 μM rapamycin; the other two groups were raised in constant dark eithernon-treated or treated with 5.5 μM rapamycin. Treatment started at 2 dpf, with drugsrenewed daily. On 8 dpf, all the tadpoles were sacrificed with eyes solubilized. Total rodopsin levels for each eye were measured by dot blot. The levels of the total rod opsinindicate the extent of retinal degeneration, with higher rod opsin levels indicatinghealthier retinas and lower rod opsin levels indicating retinas with more severe retinaldegeneration.In the dark group, rapamycin-treated P23H rhodopsin-expressing retinas had 41%higher average rod opsin levels compared to the non-treated P23H rhodopsin-expressing retinas (24.4 compared to 17.3, all the data were normalized to the rod opsinlevels of non-treated wildtype tadpoles in the light; for interaction of genotype andrapamycin treatment, p = 7.4 X 10-4), suggesting that rapamycin-treated retinas in thedark had less retinal degeneration. In the cyclic light group, rapamycin treated P23Hrhodopsin-expressing retinas had 41% lower average rod opsin levels compared to thenon-treated P23H rhodopsin-expressing retinas (9.2 compared 15.5; not statisticallysignificant), suggesting that rapamycin treated retinas in the light had more severeretinal degeneration (Figure 27). I also noticed that rapamycin treatment resulted in anapproximately 50% decrease in the levels of rod opsin in wildtype tadpoles in both lightand dark; moreover, some of the rapamycin treated retinas showed abnormalmorphology, that was likely due to prolonged drug treatment (6 days) compared to the4-day treatment used in the former experiment.111These results indicate that, more complicated than my hypothesis, enhancingautophagy using rapamycin does not reduce retinal degeneration in the light, although itseems to reduce retinal degeneration in the dark. In contrast, the opposite relationshipwas previously seen for HDAC inhibitors (more beneficial in animals reared in the light,minimal benefit in the dark), suggesting the beneficial effect of HDAC inhibitors on P23Hanimals was not mediated by activation of autophagy.Interestingly, enhancing autophagy in the light vs. in the dark caused opposing effectson retinal degeneration; it caused more severe retinal degeneration in the light, while itreduced retinal degeneration in the dark. This is potentially due to an “add on” effect oflight-dependent autophagy. In light-exposed conditions, it is possible that rapamycinand light both increase the level of autophagy, resulting in a very high level ofautophagy which may exceed the cell’s capacity and damage the cell. In the dark(without rapamycin), autophagy in rods is maintained at a relative low level. Rapamycintreatment in the dark promotes autophagy to a higher (but not saturating) level, andpromotes the degradation of misfolded rhodopsin.Alternatively, it is possible that the mechanism of cell death differs significantly betweendark and light conditions, and that enhancing autophagy is only beneficial in the case ofthe mechanism that occurs in dark-reared animals.112Figure 27: Effect of rapamycin on P23H retinal degeneration. (A) The bar chart showsthe levels of total rod opsin (both the hP23H rod opsin and endogenous wildtype rodopsin) per retina from tadpoles raised in the light, either rapamycin-treated or not. (B)The bar chart shows the levels of total rod opsin per retina from tadpoles raised in dark,either rapamycin-treated or not. The total rod opsin level of each retina was measuredby dot blot using antibody B630N, which recognizes both the mammalian andamphibian rod opsins. All the data were normalized to the rod opsin levels of non-treated wildtype tadpoles in the light. The black and gray bars indicate wildtype tadpolesand hP23H rhodopsin-expressing tadpoles respectively. Error bars = S.E.M. N ≥ 6animals per group. The results in the light were analyzed by 2-way ANOVA, forgenotype, rapamycin treatment, interaction of genotype and rapamycin treatment, P=1.8 X 10-11, 2.0 X 10-3, 3.4 X 10-1 respectively; the results in the dark were alsoanalyzed by 2-way ANOVA, for genotype, rapamycin treatment, interaction of genotypeand rapamycin treatment, P = 3.0 X 10-8, 6.4 X 10-1, 7.4 X 10-4 respectively. Rapa:rapamycin.1133.17 Effect of chloroquine on P23H retinal degeneration.To test the hypothesis above, I also used an autophagic inhibitor, chloroquine, tosuppress autophagy in rods expressing hP23H rhodopsin and measured the extent ofretinal degeneration. The drug therapy was similar to the rapamycin treatment, exceptthat tadpoles were treated with 1 mM chloroquine for 5 days from 2 dpf to 7 dpf.In the dark group, chloroquine-treated P23H rhodopsin-expressing retinas had loweraverage rod opsin levels compared to the non-treated P23H rhodopsin-expressingretinas (19.0 compared to 27.2, all the data were normalized to the rod opsin levels ofnon-treated wildtype tadpoles in the light; not statistically significant), suggesting thatchloroquine-treated retinas in the dark had more severe retinal degeneration. In thecyclic light group, chloroquine treated P23H rhodopsin-expressing retinas had higheraverage rod opsin levels compared to the non-treated retinas expressing P23Hrhodopsin (16.6 compared to 9.6; not statistically significant), suggesting thatchloroquine treated retinas in the light had less retinal degeneration (Figure 28).Suppressing autophagy in the light vs. in the dark also caused opposing effects onretinal degeneration; it caused more severe retinal degeneration in the dark, while itreduced retinal degeneration in the light, but the effect was not statistically significant.In short, activators and inhibitors of autophagy have minimal effects on retinaldegeneration associated with P23H rhodopsin, or prevent retinal degeneration under114different conditions than HDAC inhibitors. Therefore, it is unlikely that the beneficialeffects previously observed with HDAC inhibitors are due to effects on autophagy.Figure 28: Effect of chloroquine on P23H retinal degeneration. (A) The bar chart showsthe levels of total rod opsin (both the hP23H rod opsin and endogenous wildtype rodopsin) per retina from tadpoles raised in the light, either chloroquine-treated or not. (B)The bar chart shows the levels of total rod opsin per retina from tadpoles raised in dark,either chloroquine-treated or not. The total rod opsin level of each retina was measuredby dot blot using antibody B630N. All the data were normalized to the rhodopsin levelsof non-treated wildtype tadpoles in the light. The black and gray bars indicate wildtypetadpoles and hP23H rhodopsin-expressing tadpoles respectively. Error bars= S.E.M. N≥ 6 animals per group. The results in the light were analyzed by 2-way ANOVA, forgenotype, chloroquine treatment, interaction of genotype and chloroquine treatment, P =2.7 X 10-11, 7.3 X 10-1, 7.4 X 10-2 respectively; the results in the dark were alsoanalyzed by 2-way ANOVA, for genotype, chloroquine treatment, interaction of115genotype and chloroquine treatment, P= 8.3 X 10-9, 7.9 X 10-1, 3.7 X 10-1 respectively.CQ: chloroquine.1164. DiscussionIn this study, I used expression of a fluorescent mRFP-GFP-LC3 fusion protein tocharacterize autophagy in X. laevis rods in terms of time-course (Figure 14), lightdependence (Figure 15), and regulatory mechanisms. I found that autophagy levelswere regulated by light and misfolded rhodopsin, with more autophagic structuresgenerated under bright light illumination and less in dark (Figure 15), and moreautophagic structures generated in cells expressing a misfolding mutant rhodopsinassociated with RP (Figure 18). Furthermore, light-dependent autophagy is notregulated by circadian rhythm (Figure 17), and is mechanistically distinct fromautophagy induced by misfolded rhodopsin.We have previously reported that a class of pharmacological drugs called HDACinhibitors, including VPA, NaBu and CI994, ameliorated retinal degeneration associatedwith misfolding-prone P23H rhodopsin (Vent-schmidt et al. 2017). In this study, I furtherdetermined that HDAC inhibitors consistently up-regulated autophagy levels in rods(Figure 24, 25, 26), which may represent a protective mechanism for RP caused bymisfolding-prone proteins.Finally, I used an autophagy enhancer rapamycin and an autophagy inhibitorchloroquine to treat transgenic tadpoles expressing P23H rhodopsin respectively, andfound that up-regulating autophagy caused more severe retinal degeneration in the light,but it ameliorated retinal degeneration in the dark; vice versa, suppressing autophagyameliorated retinal degeneration in the light, but it caused more severe retinal117degeneration in the dark (Figure 27, 28). These results suggest that the mechanism bywhich HDAC inhibitors promote or prevent retinal degeneration is not due to their effectson autophagy.4.1 Expression:I successfully expressed fluorescent mRFP-eGFP-LC3 in X. laevis rods under thecontrol of both heat shock (HSP) and rod opsin promoters (XOP) (Figure 13). Asexpected, the fluorescent proteins had a punctate distribution that allowed me to trackautophagic structures, and as previously documented (Kimura et al. 2007), matured tored-only puncta on acidification. An added benefit of the X. laevis system is the slowmaturation time of dsRed derivatives (mRFP), which allowed me to distinguish green-only puncta as an early stage of autophagy.By using the HSP70-mRFP-EGFP-LC3 line, I characterized the time-course ofautophagy in normal rods, about which very minimal information is known. I found thatthe entire autophagy process lasts about 34 h. Green and yellow puncta, indicating theearly and intermediate autophagic structures, last for about 2 and 4 h respectively; whilethe red puncta, indicating late/acidified autolyososomes, persist for about 28 h beforedeclining (Figure 14).1184.2 Large red puncta:In transgenic tadpoles expressing XOP-mRFP-EGFP-LC3, I observed abundant green,yellow and red fluorescent puncta in rod inner segments, indicating early, intermediateand late stages of autophagic structures, respectively. Large puncta were also observed,which were about 4 to 7 times the size of the predominant smaller autophagic structures(Figure 13). They were almost exclusively red in confocal micrographs, this probablycan be explained by the longer lifetime of the red puncta (Figure 14) which permitted alonger time-course, therefore larger chance to be observed. They were typically locatedimmediately below the connecting cilium and mitochondria, adjacent to the microtubuleorganizing centre where lysosomes are enriched. Autophagosomes are trafficked alongthe microtubule to fuse with lysosomes in lysosome- enriched areas, this may explainwhy I observed the large red puncta around the microtubule organizing centre.Autophagic structures may resemble lysosomes in some morphological aspects.Lysosomes vary in size by as much as 10-fold (Lüllmann-Rauch 2005). Lysosome sizecan also increase when accumulating non-degraded material (Lüllmann-Rauch 2005). Itmight be possible, then, that the autophagic structures can also grow in size, with thelarge structures representing enlarged or exaggerated normal autophagic structures. Ifthis is true, the large red puncta should initially be non-distinguishable from smallerones. Later on, as they accumulate contents that are hard to digest such as thefluorescent mRFP-EGFP-LC3 fused proteins, they might grow into the large puncta.This may explain why the large red puncta show a significant delay in accumulationrelative to the smaller red puncta in our experiments involving heat-shock inducedexpression (Figure 14). An alternative explanation is that the large red puncta do not119form by merging of smaller structures, but rather represent a rare subset of autophagicevents. This is supported by my occasional observation of large green puncta (Figure13F). However, it is not clear that all of the large red puncta are derived from largegreen puncta; and the large green structures could also represent yet another subset ofstructures. The long lifetimes and low abundance of large green and red structuresmade their synthesis and turnover much more difficult to study. The nature of the largered puncta is not clear. I examined rods by EM, and observed occasional largestructures consistent with large autophagosomes or lysosomes, potentially representingthe large structures observed in confocal micrographs. These structures containedmembranes, and some small double-membrane organelles. However, conclusive linkingof confocal and EM results requires more sophisticated procedures, such as immuno-EM or correlative light and EM.4.3 Light-dependent autophagy:I observed that autophagy levels were regulated in a diurnal pattern in normal rods, withmore autophagic structures generated under bright light illumination and less in the dark(Figure 15). This is in agreement with previous studies of amphibian retina in vitro, andmouse retina (Reme & Knop 1980; Yao et al. 2014). Moreover, I demonstrate that thisvariation is linked to light exposure, and not to circadian rhythm. This finding hasimportant implications, in that the time of day that animals are sacrificed can affect theautophagy levels in rods. Thus, experimenters need to be consistent with time points atwhich animals are sacrificed within an experiment or between similar experiments, toavoid the time- or light-dependent differences in autophagy levels.120According to Reme (Reme & Knop 1980), autophagy in cones can also be promoted bybright light illumination, so I have a reason to suspect that this diurnal regulation ofautophagy may also exist in cones. Additionally, previous studies by Reme suggest thatbright light can promote autophagy more than dim red light in frog visual cells in vitro(Reme & Knop 1980), indicating light intensity may also have an effect on photoreceptorautophagy levels. These factors were not addressed in this thesis, but could beexamined in future studies.4.4 Light and P23H regulation of autophagy:Using the transgenic X. laevis system, I was also able to confirm upregulation ofautophagy in rods expressing a misfolding-prone mutant rhodopsin that is associatedwith RP (Figure 18), in agreement with results previously obtained by EM (Bogéa et al.2015). This led me to question whether upregulation of autophagy by light andmisfolded rhodopsin share a common pathway. As it is known that the 11-cis retinalchromophore can act as a pharmacological chaperone for both wildtype and P23Hrhodopsin, I reasoned that light exposure may limit chromophore availability, promotingmisfolding of both wildtype and P23H rhodopsin, and producing diurnal variations inautophagy. However, my results using CRISPR/Cas9-mediated knockout of the RPE65genes that are essential for chromophore production refute this hypothesis, and insteadsuggest that photoactivation of rhodopsin is necessary for upregulation of autophagy(Figure 19), implicating the phototransduction pathway in regulation of autophagy inrods. Although light-dependent autophagy in normal rods was not caused by increasedmisfolding of wildtype rhodopsin due to lack of chromophore, misfolding-prone mutant121P23H rhodopsin increased autophagy by 39% in rods (Figure 18). These resultssuggest that the extent of misfolding of wildtype rhodopsin under daylight illuminationmay not be enough to trigger autophagy, while the expression level of misfoldedrhodopsin is much higher in P23H mutants, and is sufficient to induce autophagy in rods.An unresolved question is the role of regulated autophagy in rods. Increased autophagyin response to misfolded rhodopsin could be a route for removal of the toxic misfoldedprotein, or for removal of damaged biosynthetic membranes such as ER or Golgi. Bothpossibilities are suggested, as autophagy is capable of targeting rhodopsin (Kaushal2006) and electron-micrographs indicate the presence of ER membranes inautophagosomes (Figure 21). My results are also consistent with the hypothesis that anoverwhelming of the cellular machinery for protein quality control, including autophagy,may be a trigger for cell death caused by mutant rhodopsin, as I did not observesignificant increases in autophagic structures on exposure of bP23H-expressinganimals to light, but rather impaired autophagosome maturation and autophagicstructures with morphologies (Figure 18).The role of regulation of autophagy by light exposure is less clear. Since it is not linkedto an increase in need for rhodopsin quality control, it may be linked to the presence ofincreased biosynthetic membrane damage caused by higher rates of outer segmentdisk membrane synthesis in the light (J. Besharse et al. 1977), or increased reactiveoxygen species generated by light exposure (Kuse et al. 2014). Further characterizationof the light-dependent autophagy in terms of wavelength and light intensity maydifferentiate these possibilities. It is also possible that regulation of autophagy may122serve more exotic purposes, such as contributing to regulation of the size of the ER, orturnover of signal transduction components such as transducin, possibly influencingphotosensitivity, as suggested by other researchers (Yao et al. 2014; Schuck et al. 2014;Bernales et al. 2007). Notably, RPE65 mutations cause recessive Leber congenitalamaurosis, a blindness due to loss of photoreceptor light sensitivity coupled with a slowloss of photoreceptors (Cai et al. 2009). My studies suggest that dysregulation ofautophagy is a potential cause of the retinal degeneration in these individuals. Theexact mechanism that regulates the increase of rod autophagy levels under lightillumination is still unclear. It will be interesting to identify the pathway by which lightpromotes rod autophagy levels. This could be done by knocking out other genesinvolved in the phototransduction cascade, such as transducin, and determiningwhether these genes are also involved in the regulation of rod autophagy levels.4.5 Autophagy protects against various neurodegenerations:Previous cell culture studies demonstrate that P23H mutant rhodopsin is a substrate ofautophagy (Kaushal 2006), and pharmacological regulation of autophagy affects thedegradation of mutant rhodopsin in cell culture. Rapamycin, an autophagy activator,enhanced the degradation of P23H rhodopsin, while inhibition of autophagy resulted inincreased levels of P23H mutant rhodopsin (Kaushal 2006; Chiang et al. 2012),suggesting that autophagy may represent a promising protective mechanism for retinaldegeneration via augmenting the clearance of misfolded rhodopsin. However, there isalso significant evidence suggesting that rhodopsin is degraded via the proteasomepathway (Lobanova et al. 2013; Saliba et al. 2002). Therefore, the involvement of123autophagy in quality control of normal and mutant rhodopsin in photoreceptors isunclear, although it is likely that both pathways participate, particularly when demand forquality control is high.Autophagy can degrade various aggregate-prone disease-causing proteins inneurodegenerative diseases, including mutant huntingtin associated with Huntington’sdisease (HD) (Ravikumar et al. 2002), mutant alpha-synuclein associated withParkinson’s disease (PD) (Webb et al. 2003) and beta-amyloid peptides associated withAlzheimer’s disease (AD) (Yu et al. 2005). Furthermore, induction of autophagy byrapamycin enhanced the degradation of these misfolded or aggregated proteins,including mutant huntingtin (Ravikumar et al. 2002), alpha-synuclein mutants (Webb etal. 2003) and beta-amyloid peptides (Caccamo et al. 2010); while inhibition ofautophagy resulted in decreased mutant huntingtin clearance (Ravikumar et al. 2002).Induction of autophagy by rapamycin provided neuroprotection in various animal models,such as fly and mouse models of HD (Ravikumar, Vacher, Berger, Janet E Davies, et al.2004), mouse models of AD (Spilman et al. 2010), and PD (Dehay et al. 2010). Thesefindings suggest that autophagy is a common protective mechanism in variousneurodegenerative diseases, and may be a potential therapeutic target in the treatmentof RP. My transgenic animals provide a useful tool for identifying pharmacologicalagents that promote autophagy in amphibian rods, allowing assessment of theirtherapeutic potential in additional transgenic X. laevis models of RP, as well as furtherinvestigations of quality control in photoreceptors.1244.6 The roles that autophagy plays in retinal degenerationassociated with P23H rhodopsin:We formerly published that a class of pharmacological drugs called HDAC inhibitorsameliorated retinal degeneration associated with P23H rhodopsin and inducedautophagy in wildtype rods (Vent-schmidt et al. 2017). By treating transgenic XOP-mRFP-EGFP-LC3 tadpoles with three different HDAC inhibitors, including VPA, NaBuand CI-994, I found that HDAC inhibitors consistently promoted autophagy in rods, withthe most effective drug being CI-994. However, it was unclear whether HDAC inhibitorsameliorated retinal degeneration associated with P23H rhodopsin via triggeringautophagy and enhancing the degradation of misfolded P23H rhodopsin, or if someother mechanism is involved.To determine whether autophagy levels can affect the extent of retinal degenerationcaused by misfolding-prone P23H rhodopsin, tadpoles expressing P23H rhodopsinwere treated with either an autophagy inducer rapamycin or an autophagy inhibitorchloroquine, and the levels of retinal degeneration were measured after the treatment. Ifound that increasing autophagy caused more severe retinal degeneration in the light,while it has an opposing effect in the dark, that is amelioration of retinal degeneration(for interaction of genotype and rapamycin treatment, p = 7.4 X 10-4) (Figure 27).Interestingly, this effect was the opposite of that previously observed with HDACinhibitors (more beneficial in animals reared in the light, minimal benefit in the dark)(Vent-schmidt et al. 2017), suggesting the beneficial effects previously observed withHDAC inhibitors are due to effects on autophagy.125Similarly, inhibiting autophagy also resulted in opposing effects in the light vs. in thedark. Inhibiting autophagy reduced retinal degeneration in the light and caused moresevere retinal degeneration in the dark (not statistically significant) (Figure 28). Theseresults are very interesting and suggest that the mechanisms by which autophagyrescue retinal degeneration may be different in the light versus in the dark. Currently,we have minimal understanding of the underlying mechanisms. I hypothesize that rodshave a threshold for the numbers of autophagic structures; when the numbers ofautophagic structures are below this threshold, increasing autophagy can represent aprotective mechanism for retinal degeneration by enhancing the clearance of misfoldedrhodopsin. When the demand for autophagic structures exceeds this threshold,autophagy may be inadequate or impaired, resulting in more severe retinaldegeneration. This may explain why increasing autophagy in the light caused moresevere retinal degeneration, as autophagy levels in the light are much higher than thosein the dark. It may also explain the unusual morphology of autophagic structures afterlight exposure in P23H retinas (Figure 18 C, D, E).4.7 Controversial roles that autophagy plays in retinaldegeneration:Controversies exist in defining the roles that autophagy plays in retinal degeneration. Astudy in Drosophila revealed that, besides the endosomal pathway, activated rhodopsinwas also degraded by an autophagy pathway to prevent retinal degeneration(Midorikawa et al. 2010). On the other hand, activation of autophagy exacerbated126photoreceptor degeneration in a mouse model of RP, the rd10 mouse. The rd10 mouseis a commonly used mouse model of retinal degeneration that harbors a missensemutation in the phosphodiesterase6β (Pde6β) gene. Activation of autophagy byrapamycin increased photoreceptor cell death in the rd10 mice (Rodríguez-Muela et al.2015). The roles that autophagy might play in retinal degeneration may depend on thedisease-causing mechanisms.127BibliographyAdamus, G. et al., 1991. Anti-rhodopsin monoclonal antibodies of defined specificity:Characterization and application. Vision Research, 31(1), pp.17–31.Ahmad, K.M. et al., 2003. Cell density ratios in a foveal patch in macaque retina. VisualNeuroscience, 20(2), pp.189–209.Amaya, E. & Kroll, K., 2010. Production of transgenic Xenopus laevis by restrictionenzyme mediated integration and nuclear transplantation. Journal of visualizedexperiments : JoVE, 200(42), pp.2–5. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3156005&tool=pmcentrez&rendertype=abstract.Anderson, D.H., Fisher, S.K. & Steinberg, R.H., 1978. Mammalian cones: disc sheddingphagocytosis, and renewal. Investigative Ophthalmology and Visual Science, 17(2),pp.117–133.Arango-Gonzalez, B. et al., 2014. Identification of a common non-apoptotic cell deathmechanism in hereditary retinal degeneration. PloS one, 9(11), p.e112142.Available at:http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0112142 [AccessedAugust 27, 2015].Arikawa, K. et al., 1992. Localization of peripherin/rds in the disk membranes of coneand rod photoreceptors: Relationship to disk membrane morphogenesis and retinaldegeneration. Journal of Cell Biology, 116(3), pp.659–667.Arshavsky, V.Y., 2002. Rhodopsin phosphorylation: From terminating single photonresponses to photoreceptor dark adaptation. Trends in Neurosciences, 25(3),pp.124–126.Arshavsky, V.Y., Lamb, T.D. & Pugh, E.N., 2002. G Proteins and Phototransduction.Annual Review of Physiology, 64(1), pp.153–187. Available at:http://www.annualreviews.org/doi/10.1146/annurev.physiol.64.082701.102229.Athanasiou, D. et al., 2017. The molecular and cellular basis of rhodopsin retinitispigmentosa reveals potential strategies for therapy. Progress in Retinal and Eye128Research. Available at:http://linkinghub.elsevier.com/retrieve/pii/S1350946217300769.Axe, E.L. et al., 2008. Autophagosome formation from membrane compartmentsenriched in phosphatidylinositol 3-phosphate and dynamically connected to theendoplasmic reticulum. Journal of Cell Biology, 182(4), pp.685–701.Bernales, S., Schuck, S. & Walter, P., 2007. ER-phagy: Selective autophagy of theendoplasmic reticulum. Autophagy, 3(3), pp.285–287.Besharse, J., Hollyfield, J. & Rayborn, M., 1977. Photoreceptor outer segments:accelerated membrane renewal in rods after exposure to light. Science, 196(4289),pp.536–538. Available at:http://www.sciencemag.org/cgi/doi/10.1126/science.300504.Besharse, J.C., Hollyfield, J.G. & Rayborn, M.E., 1977. Turnover of rod photoreceptorouter segments: II. membrane addition and loss in relationship to light. Journal ofCell Biology, 75(2), pp.507–527.Bhalla, S. et al., 2013. Long-term follow-up for efficacy and safety of treatment ofretinitis pigmentosa with valproic acid. The British journal of ophthalmology, 97(7),pp.895–9. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23603755.Bienz, M., 1984. Xenopus hsp 70 genes are constitutively expressed in injected oocytes.The EMBO journal, 3(11), pp.2477–83. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=557715&tool=pmcentrez&rendertype=abstract [Accessed August 26, 2015].Bogéa, T.H., Wen, R.H. & Moritz, O.L., 2015. Light induces ultrastructural changes inrod outer and inner segments, including autophagy, in a transgenic Xenopus laevisP23H rhodopsin model of retinitis pigmentosa. Investigative Ophthalmology andVisual Science, 56(13), pp.7947–7955.Bowmaker, J.K. & Dartnall, H.J., 1980. Visual pigments of rods and cones in a humanretina. The Journal of Physiology, 298(1), pp.501–511.Bown, C.D. et al., 2002. Regulation of ER stress proteins by valproate: Therapeuticimplications. Bipolar Disorders, 4(2), pp.145–151.Bownds, D. & Brodie, a E., 1975. Light-sensitive swelling of isolated frog rod outersegments as an in vitro assay for visual transduction and dark adaptation. TheJournal of general physiology, 66(4), pp.407–25.129Boya, P., Reggiori, F. & Codogno, P., 2013. Emerging regulation and functions ofautophagy. Nature Cell Biology, 15(7), pp.713–720.Brogi, S. et al., 2014. Discovery of GPCR ligands for probing signal transductionpathways. Frontiers in Pharmacology, 5(NOV).Burns, M.E. & Pugh, E.N., 2010. Lessons from Photoreceptors: Turning Off G-ProteinSignaling in Living Cells. Physiology, 25(2), pp.72–84. Available at:http://physiologyonline.physiology.org/cgi/doi/10.1152/physiol.00001.2010.Caccamo, A. et al., 2010. Molecular interplay between mammalian target of rapamycin(mTOR), amyloid-beta, and Tau: effects on cognitive impairments. The Journal ofbiological chemistry, 285(17), pp.13107–20. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2857107&tool=pmcentrez&rendertype=abstract [Accessed June 4, 2015].Cai, X., Conley, S.M. & Naash, M.I., 2009. RPE65: Role in the visual cycle, humanretinal disease, and gene therapy. Ophthalmic Genetics, 30(2), pp.57–62.Campbell, R.E. et al., 2002. A monomeric red fluorescent protein. Proceedings of theNational Academy of Sciences of the United States of America, 99(12), pp.7877–82.Available at: http://www.ncbi.nlm.nih.gov/pubmed/12060735.Chang, B. et al., 2002. Retinal degeneration mutants in the mouse. Vision Research,42(4), pp.517–525.Chang, G.Q., Hao, Y. & Wong, F., 1993. Apoptosis: final common pathway ofphotoreceptor death in rd, rds, and rhodopsin mutant mice. Neuron, 11(4), pp.595–605. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8398150 [Accessed April 25,2015].Chen, G. et al., 1999. The mood-stabilizing agent valproate inhibits the activity ofglycogen synthase kinase-3. Journal of Neurochemistry, 72(3), pp.1327–1330.Chen, Y. et al., 2014. Inherent instability of the retinitis pigmentosa P23H mutant opsin.Journal of Biological Chemistry, 289(13), pp.9288–9303.Chi, O.Z. et al., 2017. Rapamycin decreased blood-brain barrier permeability in controlbut not in diabetic rats in early cerebral ischemia. Neuroscience Letters, 654,pp.17–22.130Chiang, W.-C., Messah, C. & Lin, J.H., 2012. IRE1 directs proteasomal and lysosomaldegradation of misfolded rhodopsin. Molecular biology of the cell, 23(5), pp.758–70.Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3290636&tool=pmcentrez&rendertype=abstract [Accessed June 3, 2015].Chiu, C.-T. et al., 2011. Combined treatment with the mood stabilizers lithium andvalproate produces multiple beneficial effects in transgenic mouse models ofHuntington’s disease. Neuropsychopharmacology : official publication of theAmerican College of Neuropsychopharmacology, 36(12), pp.2406–2421. Availableat:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3194069&tool=pmcentrez&rendertype=abstract.Choe, H.W. et al., 2011. Crystal structure of metarhodopsin II. Nature, 471(7340),pp.651–655.Clemson, C.M. et al., 2011. Therapeutic potential of valproic acid for retinitispigmentosa. The British journal of ophthalmology, 95(1), pp.89–93.Collin, S.P. & Trezise, A.E., 2004. The origins of colour vision in vertebrates. Clinicaland Experimental Optometry, 87(4-5), pp.217–223.Concepcion, F. & Chen, J., 2010. Q344ter mutation causes mislocalization of rhodopsinmolecules that are catalytically active: A mouse model of Q344ter-induced retinaldegeneration. PLoS ONE, 5(6).Connell, G. et al., 1991. Photoreceptor peripherin is the normal product of the generesponsible for retinal degeneration in the rds mouse. Proceedings of the NationalAcademy of Sciences of the United States of America, 88(3), pp.723–6. Availableat:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=50885&tool=pmcentrez&rendertype=abstract.Daiger, S.P., Sullivan, L.S. & Bowne, S.J., 2013. Genes and mutations causing retinitispigmentosa. Clinical Genetics, 84(2), pp.132–141.Deeb, S.S. & Kohl, S., 2003. Genetics of color vision deficiencies. Developments inOphthalmology, 37, pp.170–187. Available at:https://drive.google.com/file/d/0B4n4-9JAReFnYS1ZcGt4eDl4cVk/view\nhttp://www.ncbi.nlm.nih.gov/pubmed/12876837.131Dehay, B. et al., 2010. Pathogenic lysosomal depletion in Parkinson’s disease. TheJournal of neuroscience : the official journal of the Society for Neuroscience, 30(37),pp.12535–12544.Deter, R.L. & De Duve, C., 1967. Influence of glucagon, an inducer of cellularautophagy, on some physical properties of rat liver lysosomes. Journal of CellBiology, 33(2), pp.437–449.Dryja, T.P. et al., 1990. A point mutation of the rhodopsin gene in one form of retinitispigmentosa. Nature, 343(6256), pp.364–366. Available at:http://www.nature.com/doifinder/10.1038/343364a0.Farjo, R. et al., 2006. Retention of function without normal disc morphogenesis occursin cone but not rod photoreceptors. Journal of Cell Biology, 173(1), pp.59–68.Farjo, R., Fliesler, S.J. & Naash, M.I., 2007. Effect of Rds abundance on cone outersegment morphogenesis, photoreceptor gene expression, and outer limitingmembrane integrity. Journal of Comparative Neurology, 504(6), pp.619–630.Farjo, R. & Naash, M.I., 2006. The role of Rds in outer segment morphogenesis andhuman retinal disease. Ophthalmic Genetics, 27(4), pp.117–122.Feehan, J.M. et al., 2017. Modeling Dominant and Recessive Forms of RetinitisPigmentosa by Editing Three Rhodopsin-Encoding Genes in Xenopus Laevis UsingCrispr/Cas9. Scientific Reports, 7(1).Filipek, S. et al., 2003. G Protein-Coupled Receptor Rhodopsin: A Prospectus. AnnualReview of Physiology, 65(1), pp.851–879. Available at:http://www.annualreviews.org/doi/10.1146/annurev.physiol.65.092101.142611.Fleming, A. et al., 2011. Chemical modulators of autophagy as biological probes andpotential therapeutics. Nature Chemical Biology, 7(1), pp.9–17.Fotiadis, D., Liang, Y., Filipek, S., et al., 2003. Is Rhodopsin dimeric in native retinalrods? Commentary. Nature, 426(6962), p.31.Fotiadis, D., Liang, Y., Filipek, S., et al., 2003. Rhodopsin dimers in native discmembranes. Nature, 421(6919), pp.127–128.Fredriksson, R., 2003. The G-Protein-Coupled Receptors in the Human Genome FormFive Main Families. Phylogenetic Analysis, Paralogon Groups, and Fingerprints.132Molecular Pharmacology, 63(6), pp.1256–1272. Available at:http://molpharm.aspetjournals.org/cgi/doi/10.1124/mol.63.6.1256.Fukuda, M.N., Papermaster, D.S. & Hargrave, P.A., 1979. Rhodopsin carbohydrate.Structure of small oligosaccharides attached at two sites near the NH2 terminus.Journal of Biological Chemistry, 254(17), pp.8201–8207.Galy, A. et al., 2005. Rhodopsin maturation defects induce photoreceptor death byapoptosis: a fly model for RhodopsinPro23His human retinitis pigmentosa. Humanmolecular genetics, 14(17), pp.2547–57. Available at:http://www.ncbi.nlm.nih.gov/pubmed/16049034 [Accessed May 15, 2015].Gargini, C. et al., 2007. Retinal organization in the retinal degeneration 10 (rd10) mutantmouse: A morphological and ERG study. Journal of Comparative Neurology, 500(2),pp.222–238.Geamănu Pancă, A. et al., 2014. Retinal toxicity associated with chronic exposure tohydroxychloroquine and its ocular screening. Review. Journal of medicine and life,7(3), pp.322–6.Glick, D., Barth, S. & Macleod, K.F., 2010. Autophagy : cellular and molecularmechanisms. Journal of Pathology The, 221(1), pp.3–12.Gorbatyuk, M.S. et al., 2012. Functional rescue of P23H rhodopsin photoreceptors bygene delivery. In Advances in Experimental Medicine and Biology. pp. 191–197.Goto, Y. et al., 1995. Functional abnormalities in transgenic mice expressing a mutantrhodopsin gene. Investigative ophthalmology & visual science, 36(1), pp.62–71.Available at: http://www.ncbi.nlm.nih.gov/pubmed/7822160.Gramage, E., Li, J. & Hitchcock, P., 2014. The expression and function of midkine in thevertebrate retina. British Journal of Pharmacology, 171(4), pp.913–923.Green, D.R. & Levine, B., 2014. To be or not to be? How selective autophagy and celldeath govern cell fate. Cell, 157(1), pp.65–75.Gunkel, M. et al., 2015. Higher-order architecture of rhodopsin in intact photoreceptorsand its implication for phototransduction kinetics. Structure, 23(4), pp.628–638.Gwinn, D.M. et al., 2008. AMPK Phosphorylation of Raptor Mediates a MetabolicCheckpoint. Molecular Cell, 30(2), pp.214–226.133Han, Z., Anderson, D.W. & Papermaster, D.S., 2012. Prominin-1 localizes to the openrims of outer segment lamellae in xenopus laevis rod and cone photoreceptors.Investigative Ophthalmology and Visual Science, 53(1), pp.361–373.Hannigan, A.M. & Gorski, S.M., 2009. Macroautophagy: The key ingredient to a healthydiet? Autophagy, 5(2), pp.140–151.Hartong, D.T., Berson, E.L. & Dryja, T.P., 2006. Retinitis pigmentosa. Lancet,368(9549), pp.1795–809. Available at:http://www.ncbi.nlm.nih.gov/pubmed/17113430 [Accessed December 29, 2014].Homewood, C.A. et al., 1972. Lysosomes, pH and the anti-malarial action ofchloroquine [9]. Nature, 235(5332), pp.50–52.Illing, M.E. et al., 2002. A rhodopsin mutant linked to autosomal dominant retinitispigmentosa is prone to aggregate and interacts with the ubiquitin proteasomesystem. The Journal of biological chemistry, 277(37), pp.34150–60. Available at:http://www.ncbi.nlm.nih.gov/pubmed/12091393 [Accessed May 25, 2015].Inoue, D. & Wittbrodt, J., 2011. One for all-a highly efficient and versatile method forfluorescent immunostaining in fish embryos. PLoS ONE, 6(5).Ishibashi, S., Kroll, K.L. & Amaya, E., 2012. Generating transgenic frog embryos byrestriction enzyme mediated integration (REMI). Methods in molecular biology(Clifton, N.J.), 917, pp.185–203. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3818119&tool=pmcentrez&rendertype=abstract [Accessed August 18, 2015].Johannessen, C.U., 2000. Mechanisms of action of valproate: A commentatory.Neurochemistry International, 37(2-3), pp.103–110.Johansen, T. & Lamark, T., 2011. Selective autophagy mediated by autophagic adapterproteins. Autophagy, 7(3), pp.279–296.Kabeya, Y., 2000. LC3, a mammalian homologue of yeast Apg8p, is localized inautophagosome membranes after processing. The EMBO Journal, 19(21),pp.5720–5728. Available at:http://emboj.embopress.org/cgi/doi/10.1093/emboj/19.21.5720.Kalloniatis, M. & Luu, C., 2005. The perception of Color. In Webvision: The Organizationof the Retina and Visual System [Internet]. Available at:http://www.ncbi.nlm.nih.gov/books/NBK11538/.134Karakaş, H.E. & Gözüaçik, D., 2014. Autophagy and cancer. Turkish Journal of Biology,38(6), pp.720–739.Kaur, J. et al., 2011. Calpain and PARP activation during photoreceptor cell death inP23H and S334ter rhodopsin mutant rats. PloS one, 6(7), p.e22181. Available at:http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0022181 [AccessedAugust 27, 2015].Kaushal, S., 2006. Effect of rapamycin on the fate of P23H opsin associated withretinitis pigmentosa (an American Ophthalmological Society thesis). Transactionsof the American Ophthalmological Society, 104(Figure 1), pp.517–529.Kaushal, S. & Khorana, H.G., 1994. Structure and Function in Rhodopsin. 7. PointMutations Associated with Autosomal Dominant Retinitis Pigmentosa. Biochemistry,33(20), pp.6121–6128. Available at: http://dx.doi.org/10.1021/bi00186a011[Accessed June 3, 2015].Kefalov, V.J., 2012. Rod and cone visual pigments and phototransduction throughpharmacological, genetic, and physiological approaches. Journal of BiologicalChemistry, 287(3), pp.1635–1641.Kevany, B.M. & Palczewski, K., 2010. Phagocytosis of retinal rod and conephotoreceptors. Physiology (Bethesda), 25(1), pp.8–15. Available at:http://www.ncbi.nlm.nih.gov/pubmed/20134024\nhttp://physiologyonline.physiology.org/content/nips/25/1/8.full.pdf.Kevany, B.M. & Palczewski, K., 2010. Phagocytosis of retinal rod and conephotoreceptors. Physiology (Bethesda, Md.), 25(1), pp.8–15. Available at:http://www.ncbi.nlm.nih.gov/pubmed/20134024\nhttp://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC2839896.Khattree, N., Ritter, L.M. & Goldberg, A.F.X., 2013. Membrane curvature generation bya C-terminal amphipathic helix in peripherin-2/rds, a tetraspanin required forphotoreceptor sensory cilium morphogenesis. Journal of cell science, 126,pp.4659–70. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23886945.Khmelinskii, A. et al., 2012. Tandem fluorescent protein timers for in vivo analysis ofprotein dynamics. Nature Biotechnology, 30(7), pp.708–714.Kim, J. et al., 2011. AMPK and mTOR regulate autophagy through directphosphorylation of Ulk1. Nature Cell Biology, 13(2), pp.132–141.135Kimura, S., Noda, T. & Yoshimori, T., 2007. Dissection of the autophagosomematuration process by a novel reporter protein, tandem fluorescent-tagged LC3.Autophagy, 3(5), pp.452–460.Kimura, S., Noda, T. & Yoshimori, T., 2008. Dynein-dependent movement ofautophagosomes mediates efficient encounters with lysosomes. Cell structure andfunction, 33(1), pp.109–122.Kobilka, B.K., 2007. G protein coupled receptor structure and activation. Biochimica etBiophysica Acta (BBA) - Biomembranes, 1768(4), pp.794–807. Available at:http://linkinghub.elsevier.com/retrieve/pii/S0005273606003981.Kuse, Y. et al., 2014. Damage of photoreceptor-derived cells in culture induced by lightemitting diode-derived blue light. Scientific Reports, 4.Lanum, J., 1978. The damaging effects of light on the retina. Empirical findings,theoretical and practical implications. Survey of ophthalmology, 22(4), pp.221–249.Lemasters, J.J., 2005. Selective Mitochondrial Autophagy, or Mitophagy, as a TargetedDefense Against Oxidative Stress, Mitochondrial Dysfunction, and Aging.Rejuvenation Research, 8(1), pp.3–5. Available at:http://www.liebertonline.com/doi/abs/10.1089/rej.2005.8.3.Liang, Y. et al., 2003. Organization of the G protein-coupled receptors rhodopsin andopsin in native membranes. Journal of Biological Chemistry, 278(24), pp.21655–21662.Lin, J.H. et al., 2007. IRE1 signaling affects cell fate during the unfolded proteinresponse. Science (New York, N.Y.), 318(5852), pp.944–9. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3670588&tool=pmcentrez&rendertype=abstract [Accessed April 15, 2015].Lobanova, E.S. et al., 2013. Proteasome overload is a common stress factor in multipleforms of inherited retinal degeneration. Proceedings of the National Academy ofSciences of the United States of America, 110(24), pp.9986–91. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3683722&tool=pmcentrez&rendertype=abstract [Accessed June 3, 2015].Loewen, C.J. et al., 2003. The role of subunit assembly in peripherin-2 targeting to rodphotoreceptor disk membranes and retinitis pigmentosa. Mol Biol Cell, 14(8),pp.3400–3413. Available at:136http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=12925772.Löscher, W., 2002. Basic pharmacology of valproate: a review after 35 years of clinicaluse for the treatment of epilepsy. CNS drugs, 16(10), pp.669–694.Loy, R. & Tariot, P.N., 2002. Neuroprotective properties of valproate: potential benefitfor AD and tauopathies. Journal of Molecular Neuroscience, 19(3), pp.303–7.Lüllmann-Rauch, R., 2005. History and Morphology of the Lysosome. In Lysosomes. pp.1–16. Available at: http://link.springer.com/10.1007/0-387-28957-7_1.Machida, S. et al., 2000. P23H rhodopsin transgenic rat: correlation of retinal functionwith histopathology. Investigative ophthalmology & visual science, 41(10),pp.3200–9. Available at: http://www.ncbi.nlm.nih.gov/pubmed/10967084 [AccessedMay 15, 2015].Mao, H. et al., 2011. AAV delivery of wild-type rhodopsin preserves retinal function in amouse model of autosomal dominant retinitis pigmentosa. Human gene therapy,22(5), pp.567–75. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3131806&tool=pmcentrez&rendertype=abstract [Accessed August 14, 2015].Martina, J.A. et al., 2012. MTORC1 functions as a transcriptional regulator of autophagyby preventing nuclear transport of TFEB. Autophagy, 8(6), pp.903–914.Masland, R.H., 2012. The Neuronal Organization of the Retina. Neuron, 76(2), pp.266–280.Maubaret, C. & Hamel, C., 2005. [Genetics of retinitis pigmentosa: metabolicclassification and phenotype/genotype correlations]. J Fr Ophtalmol, 28(1), pp.71–92. Available at:http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=15767903.McEwan, D.G. et al., 2015. PLEKHM1 regulates autophagosome-lysosome fusionthrough HOPS complex and LC3/GABARAP proteins. Molecular Cell, 57(1), pp.39–54.McEwan, D.G. & Dikic, I., 2010. Not All Autophagy Membranes Are Created Equal. Cell,141(4), pp.564–566.137McMahon, H.T. & Boucrot, E., 2015. Membrane curvature at a glance. Journal of CellScience, 128(6), pp.1065–1070. Available at:http://jcs.biologists.org/cgi/doi/10.1242/jcs.114454.McMahon, H.T. & Gallop, J.L., 2005. Membrane curvature and mechanisms of dynamiccell membrane remodelling. Nature, 438(7068), pp.590–596.Mendes, H.F. & Cheetham, M.E., 2008. Pharmacological manipulation of gain-of-function and dominant-negative mechanisms in rhodopsin retinitis pigmentosa.Human molecular genetics, 17(19), pp.3043–54. Available at:http://www.ncbi.nlm.nih.gov/pubmed/18635576 [Accessed July 15, 2015].Midorikawa, R. et al., 2010. Autophagy-dependent rhodopsin degradation preventsretinal degeneration in Drosophila. The Journal of neuroscience : the official journalof the Society for Neuroscience, 30(32), pp.10703–19. Available at:http://www.ncbi.nlm.nih.gov/pubmed/20702701 [Accessed June 5, 2015].Mielke, J.G. et al., 1997. Chloroquine administration in mice increases β-amyloidimmunoreactivity and attenuates kainate-induced blood-brain barrier dysfunction.Neuroscience Letters, 227(3), pp.169–172.Mitton, K.P. et al., 2014. Different effects of valproic acid on photoreceptor loss in Rd1and Rd10 retinal degeneration mice. Molecular vision, 20(November), pp.1527–44.Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=4225157&tool=pmcentrez&rendertype=abstract.Mizushima, N., 2007. Autophagy: Process and function. Genes and Development, 21,pp.2861–2873.Mizushima, N. & Komatsu, M., 2011. Autophagy: Renovation of cells and tissues. Cell,147(4), pp.728–741.Mizushima, N., Ohsumi, Y. & Yoshimori, T., 2002. Autophagosome Formation inMammalian Cells. CELL STRUCTURE AND FUNCTION, 27, pp.421–429.Molday, R.S., Hicks, D. & Molday, L., 1987. Peripherin. A rim-specific membrane proteinof rod outer segment discs. Investigative Ophthalmology and Visual Science, 28(1),pp.50–61.138Molday, R.S. & Moritz, O.L., 2015. Photoreceptors at a glance. Journal of Cell Science,128(22), pp.4039–4045. Available at:http://jcs.biologists.org/cgi/doi/10.1242/jcs.175687.Monti, B. et al., 2010. Valproic acid is neuroprotective in the rotenone rat model ofParkinson’s disease: involvement of alpha-synuclein. Neurotoxicity research, 17(2),pp.130–41. Available at: http://www.ncbi.nlm.nih.gov/pubmed/19626387.Moritz, O.L. et al., 2001. A Functional Rhodopsin-Green Fluorescent Protein FusionProtein Localizes Correctly in Transgenic Xenopus laevis Retinal Rods and isExpressed in a Time-dependent Pattern. Journal of Biological Chemistry, 276(30),pp.28242–28251.Moritz, O.L. et al., 1999. Fluorescent photoreceptors of transgenic Xenopus laevisimaged in vivo by two microscopy techniques. Investigative Ophthalmology andVisual Science, 40(13), pp.3276–3280.Motarjemizadeh, Q., Aidenloo, N.S. & Abbaszadeh, M., 2016. Detection ofHydroxychloroquine Retinal Toxicity by Automated Perimetry in 60 RheumatoidArthritis Patients with Normal Fundoscopic Findings. Global journal of healthscience, 8(3), pp.59–64.Mustafi, D., Engel, A.H. & Palczewski, K., 2009. Structure of cone photoreceptors.Progress in Retinal and Eye Research, 28(4), pp.289–302.Nakamura, S. & Yoshimori, T., 2017. New insights into autophagosome–lysosomefusion. Journal of Cell Science, 130(7), pp.1209–1216. Available at:http://jcs.biologists.org/lookup/doi/10.1242/jcs.196352.Nakatogawa, H. et al., 2009. Dynamics and diversity in autophagy mechanisms:Lessons from yeast. Nature Reviews Molecular Cell Biology, 10(7), pp.458–467.Nathans, J. et al., 1986. Molecular genetics of inherited variation in human color vision.Science, 232(4747), pp.203–210. Available at:http://www.sciencemag.org/cgi/doi/10.1126/science.3485310.Nathans, J., Thomas, D. & Hogness, D., 1986. Molecular genetics of human color vision:the genes encoding blue, green, and red pigments. Science, 232(4747), pp.193–202.139Nguyen, T.N. et al., 2016. Atg8 family LC3 / GAB ​ ARAP proteins are crucial forautophagosome – lysosome fusion but not autophagosome formation during PINK1/ Parkin mitophagy and starvation. Journal of Cell Biology, 215(6), pp.1–18.Nguyen-Legros, J. & Hicks, D., 2000. Renewal of photoreceptor outer segments andtheir phagocytosis by the retinal pigment epithelium. Int Rev Cytol, 196, pp.245–313.Nickell, S. et al., 2007. Three-dimensional architecture of murine rod outer segmentsdetermined by cryoelectron tomography. Journal of Cell Biology, 177(5), pp.917–925.Noda, T. & Ohsumi, Y., 1998. Tor, a phosphatidylinositol kinase homologue, controlsautophagy in yeast. Journal of Biological Chemistry, 273(7), pp.3963–3966.Noorwez, S.M. et al., 2004. Retinoids Assist the Cellular Folding of the AutosomalDominant Retinitis Pigmentosa Opsin Mutant P23H. Journal of Biological Chemistry,279(16), pp.16278–16284.Okada, T. et al., 2004. The retinal conformation and its environment in rhodopsin in lightof a new 2.2 Å crystal structure. Journal of Molecular Biology, 342(2), pp.571–583.Olsson, J.E. et al., 1992. Transgenic mice with a rhodopsin mutation (Pro23His): amouse model of autosomal dominant retinitis pigmentosa. Neuron, 9(5), pp.815–30.Available at: http://www.ncbi.nlm.nih.gov/pubmed/1418997 [Accessed May 15,2015].Owens, M.J. & Nemeroff, C.B., 2003. Pharmacology of valproate. PsychopharmacolBull, 37 Suppl 2, pp.17–24. Available at:http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=14624230.Palczewski, K. et al., 2000. Crystal Structure of Rhodopsin: A G Protein-CoupledReceptor. Science (New York, NY), 289(5480), pp.739–745. Available at:http://www.sciencemag.org/content/289/5480/739.full\npapers3://publication/doi/10.1126/science.289.5480.739.Palczewski, K., 2006. G Protein–Coupled Receptor Rhodopsin. Annual Review ofBiochemistry, 75(1), pp.743–767. Available at:http://www.annualreviews.org/doi/10.1146/annurev.biochem.75.103004.142743.140Park, J.H. et al., 2008. Crystal structure of the ligand-free G-protein-coupled receptoropsin. Nature, 454(7201), pp.183–187.Phiel, C.J. et al., 2001. Histone Deacetylase is a Direct Target of Valproic Acid, a PotentAnticonvulsant, Mood Stabilizer, and Teratogen. Journal of Biological Chemistry,276(39), pp.36734–36741.Portera-Cailliau, C. et al., 1994. Apoptotic photoreceptor cell death in mouse models ofretinitis pigmentosa. Proceedings of the National Academy of Sciences of theUnited States of America, 91(3), pp.974–8. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=521436&tool=pmcentrez&rendertype=abstract.Price, B.A. et al., 2011. Mislocalization and degradation of human P23H-rhodopsin-GFPin a knockin mouse model of retinitis pigmentosa. Investigative ophthalmology &visual science, 52(13), pp.9728–36. Available at:http://iovs.arvojournals.org/article.aspx?articleid=2187412 [Accessed May 15,2015].Price, B.A. et al., 2012. Rhodopsin gene expression determines rod outer segment sizeand rod cell resistance to a dominant-negative neurodegeneration mutant. PloSone, 7(11), p.e49889. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3503812&tool=pmcentrez&rendertype=abstract [Accessed August 14, 2015].Proikas-Cezanne, T. et al., 2015. WIPI proteins: essential PtdIns3P effectors at thenascent autophagosome. Journal of Cell Science, 128(2), pp.207–217. Available at:http://jcs.biologists.org/cgi/doi/10.1242/jcs.146258.Ravikumar, B., Vacher, C., Berger, Z., Davies, J.E., et al., 2004. Inhibition of mTORinduces autophagy and reduces toxicity of polyglutamine expansions in fly andmouse models of Huntington disease. Nature Genetics, 36(6), pp.585–595.Ravikumar, B., Duden, R. & Rubinsztein, D.C., 2002. Aggregate-prone proteins withpolyglutamine and polyalanine expansions are degraded by autophagy. Humanmolecular genetics, 11(9), pp.1107–1117.Reichenbach, A. & Bringmann, A., 2013. New functions of muller cells. Glia, 61(5),pp.651–678.141Remé, C.E. et al., 1999. Photoreceptor autophagy: Effects of light history on numberand opsin content of degradative vacuoles. Investigative Ophthalmology and VisualScience, 40(10), pp.2398–2404.Reme, C.E. & Knop, M., 1980. Autophagy in frog visual cells in vitro. InvestigativeOphthalmology and Visual Science, 19(5), pp.439–456.Renna, M. et al., 2010. Chemical inducers of autophagy that enhance the clearance ofmutant proteins in neurodegenerative diseases. Journal of Biological Chemistry,285(15), pp.11061–11067.Rodríguez-Muela, N. et al., 2015. Lysosomal membrane permeabilization andautophagy blockade contribute to photoreceptor cell death in a mouse model ofretinitis pigmentosa. Cell death and differentiation, 22(3), pp.476–87. Available at:http://www.ncbi.nlm.nih.gov/pubmed/25501597 [Accessed August 17, 2015].Röhlich, P. & Szél, Á., 2000. Photoreceptor cells in the Xenopus retina. MicroscopyResearch and Technique, 50(5), pp.327–337.Roof, D.J., Adamian, M. & Hayes, A., 1994. Rhodopsin accumulation at abnormal sitesin retinas of mice with a human P23H rhodopsin transgene. Investigativeophthalmology & visual science, 35(12), pp.4049–62. Available at:http://www.ncbi.nlm.nih.gov/pubmed/7960587 [Accessed May 15, 2015].Sakami, S. et al., 2011. Probing mechanisms of photoreceptor degeneration in a newmouse model of the common form of autosomal dominant retinitis pigmentosa dueto P23H opsin mutations. The Journal of biological chemistry, 286(12), pp.10551–67. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3060508&tool=pmcentrez&rendertype=abstract [Accessed August 27, 2015].Saliba, R.S. et al., 2002. The cellular fate of mutant rhodopsin: quality control,degradation and aggresome formation. Journal of cell science, 115(Pt 14),pp.2907–18. Available at: http://www.ncbi.nlm.nih.gov/pubmed/12082151[Accessed May 25, 2015].Sanchez-Wandelmer, J. & Reggiori, F., 2013. Amphisomes: Out of the autophagosomeshadow? EMBO Journal, 32(24), pp.3116–3118.Sandberg, M.A. et al., 2011. Lack of scientific rationale for use of valproic acid forretinitis pigmentosa. British Journal of Ophthalmology, 95(5), p.744.142Sanes, J.R. & Masland, R.H., 2015. The Types of Retinal Ganglion Cells: CurrentStatus and Implications for Neuronal Classification. Annual Review ofNeuroscience, 38(1), pp.221–246. Available at:http://www.annualreviews.org/doi/10.1146/annurev-neuro-071714-034120.Sano, R. & Reed, J.C., 2013. ER stress-induced cell death mechanisms. Biochimica etBiophysica Acta - Molecular Cell Research, 1833(12), pp.3460–3470.Sarkar, S. et al., 2005. Lithium induces autophagy by inhibiting inositolmonophosphatase. Journal of Cell Biology, 170(7), pp.1101–1111.Saunders, R.N., Metcalfe, M.S. & Nicholson, M.L., 2001. Rapamycin in transplantation:a review of the evidence. Kidney Int., 59(0085-2538 (Print)), pp.3–16.Schaaf, M.B.E. et al., 2016. LC3/GABARAP family proteins: Autophagy-(un)relatedfunctions. FASEB Journal, 30(12), pp.3961–3978.Scheerer, P. et al., 2008. Crystal structure of opsin in its G-protein-interactingconformation. Nature, 455(7212), pp.497–502.Van Schooneveld, M.J. et al., 2011. The conclusions of Clemson et al concerningvalproic acid are premature. British Journal of Ophthalmology, 95(1), p.153.Schuck, S., Gallagher, C.M. & Walter, P., 2014. ER-phagy mediates selectivedegradation of endoplasmic reticulum independently of the core autophagymachinery. Journal of Cell Science, 127(18), pp.4078–4088. Available at:http://jcs.biologists.org/cgi/doi/10.1242/jcs.154716.Seong Lee, J. & Lee, G.M., 2012. Effect of sodium butyrate on autophagy andapoptosis in Chinese hamster ovary cells. Biotechnology Progress, 28(2), pp.349–357. Available at: http://doi.wiley.com/10.1002/btpr.1512.Shintani, T. & Klionsky, D.J., 2004. Autophagy in health and disease: A double-edgedsword. Science, 306(5698), pp.990–995.Simonsen, A. & Tooze, S.A., 2009. Coordination of membrane events during autophagyby multiple class III PI3-kinase complexes. Journal of Cell Biology, 186(6), pp.773–782.Sisk, R.A., 2012. Valproic acid treatment may be harmful in non-dominant forms ofretinitis pigmentosa. British Journal of Ophthalmology, 96(8), pp.1154–1155.143Sizova, O.S. et al., 2014. Modulation of cellular signaling pathways in P23H rhodopsinphotoreceptors. Cellular Signalling, 26(4), pp.665–672.Sparrow, J.R., Hicks, D. & Hamel, C.P., 2010. The retinal pigment epithelium in healthand disease. Current molecular medicine, 10(9), pp.802–23. Available at:http://www.ncbi.nlm.nih.gov/pubmed/21091424\nhttp://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC4120883.Spilman, P. et al., 2010. Inhibition of mTOR by Rapamycin Abolishes Cognitive Deficitsand Reduces Amyloid-β Levels in a Mouse Model of Alzheimer’s Disease P. F.Ferrari, ed. PLoS ONE, 5(4), p.e9979. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2848616&tool=pmcentrez&rendertype=abstract [Accessed March 22, 2015].Steinberg, R.H., Fisher, S.K. & Anderson, D.H., 1980. Disc morphogenesis in vertebratephotoreceptors. Journal of Comparative Neurology, 190(3), pp.501–518.Steinman, R.M. et al., 1983. Endocytosis and the recycling of plasma membrane.Journal of Cell Biology, 96(1), pp.1–27.Strettoi, E. et al., 2010. Complexity of retinal cone bipolar cells. Progress in Retinal andEye Research, 29(4), pp.272–283.Stuck, M.W., Conley, S.M. & Naash, M.I., 2016. PRPH2/RDS and ROM-1: Historicalcontext, current views and future considerations. Progress in Retinal and EyeResearch, 52, pp.47–63.Sung, C.H., Davenport, C.M. & Nathans, J., 1993. Rhodopsin mutations responsible forautosomal dominant retinitis pigmentosa. Clustering of functional classes along thepolypeptide chain. Journal of Biological Chemistry, 268(35), pp.26645–26649.Szalai, P. et al., 2015. Autophagic bulk sequestration of cytosolic cargo is independentof LC3, but requires GABARAPs. Experimental Cell Research, 333(1), pp.21–38.Takeshige, K. et al., 1992. Autophagy in yeast demonstrated with proteinase-deficientmutants and conditions for its induction. Journal of Cell Biology, 119(2), pp.301–312.Tam, B.M. et al., 2013. Generation of Transgenic X. laevis Models of RetinalDegeneration. Methods in Molecular Biology, 935, pp.113–125.144Tam, B.M., 2006. Mislocalized Rhodopsin Does Not Require Activation to Cause RetinalDegeneration and Neurite Outgrowth in Xenopus laevis. Journal of Neuroscience,26(1), pp.203–209. Available at:http://www.jneurosci.org/cgi/doi/10.1523/JNEUROSCI.3849-05.2006.Tam, B.M. et al., 2014. Photoactivation-Induced Instability of Rhodopsin Mutants T4Kand T17M in Rod Outer Segments Underlies Retinal Degeneration in X. laevisTransgenic Models of Retinitis Pigmentosa. Journal of Neuroscience, 34(40),pp.13336–13348. Available at:http://www.jneurosci.org/cgi/doi/10.1523/JNEUROSCI.1655-14.2014.Tam, B.M. et al., 2015. Preparation of Xenopus laevis retinal cryosections for electronmicroscopy. Experimental Eye Research, 136, pp.86–90.Tam, B.M. et al., 2010. The dependence of retinal degeneration caused by therhodopsin P23H mutation on light exposure and vitamin a deprivation. InvestigativeOphthalmology and Visual Science, 51(3), pp.1327–1334.Tam, B.M. & Moritz, O.L., 2006. Characterization of rhodopsin P23H-induced retinaldegeneration in a Xenopus laevis model of retinitis pigmentosa. Investigativeophthalmology & visual science, 47(8), pp.3234–41. Available at:http://iovs.arvojournals.org/article.aspx?articleid=2126677 [Accessed April 30,2015].Tam, B.M. & Moritz, O.L., 2007. Dark rearing rescues P23H rhodopsin-induced retinaldegeneration in a transgenic Xenopus laevis model of retinitis pigmentosa: achromophore-dependent mechanism characterized by production of N-terminallytruncated mutant rhodopsin. The Journal of neuroscience : the official journal of theSociety for Neuroscience, 27(34), pp.9043–53. Available at:http://www.jneurosci.org/content/27/34/9043.long [Accessed April 30, 2015].Tam, B.M. & Moritz, O.L., 2009. The Role of Rhodopsin Glycosylation in Protein Folding,Trafficking, and Light-Sensitive Retinal Degeneration. Journal of Neuroscience,29(48), pp.15145–15154. Available at:http://www.jneurosci.org/cgi/doi/10.1523/JNEUROSCI.4259-09.2009.Tariot, P.N. & Aisen, P.S., 2009. Can lithium or valproate untie tangles in Alzheimer’sdisease? Journal of Clinical Psychiatry, 70(6), pp.919–921.Tsukada, M. & Ohsumi, Y., 1993. Isolation and characterization of autophagy-defectivemutants of Saccharomyces cerevisiae. FEBS Letters, 333(1-2), pp.169–174.145Veleri, S. et al., 2015. Biology and therapy of inherited retinal degenerative disease:insights from mouse models. Disease Models and Mechanisms, 8(2), pp.109–129.Available at: http://dmm.biologists.org/content/8/2/109.abstract.Vent-schmidt, R.Y.J. et al., 2017. Opposing Effects of Valproic Acid Treatment Mediatedby Histone Deacetylase Inhibitor Activity in Four Transgenic X . laevis Models ofRetinitis Pigmentosa. Journal of Neuros, 37(4), pp.1039–1054.Viringipurampeer, I.A. et al., 2016. NLRP3 inflammasome activation drives bystandercone photoreceptor cell death in a P23H rhodopsin model of retinal degeneration.Human Molecular Genetics, 25(8), pp.1501–1516.Wang, D.Y. et al., 2005. Gene mutations in retinitis pigmentosa and their clinicalimplications. Clinica chimica acta; international journal of clinical chemistry, 351(1-2), pp.5–16. Available at:http://www.sciencedirect.com/science/article/pii/S0009898104004097.Wang, M. et al., 1997. Expression of a mutant opsin gene increases the susceptibility ofthe retina to light damage. Visual Neuroscience, 14(1), pp.55–62.Webb, J.L. et al., 2003. Alpha-Synuclein is degraded by both autophagy and theproteasome. The Journal of biological chemistry, 278(27), pp.25009–13. Availableat: http://www.ncbi.nlm.nih.gov/pubmed/12719433 [Accessed April 17, 2015].Wibo, M. & Poole, B., 1974. Protein degradation in cultured cells: II. the uptake ofchloroquine by rat fibroblasts and the inhibition of cellular protein degradation andcathepsin B1. Journal of Cell Biology, 63(2), pp.430–440.Williams, A. et al., 2008. Novel targets for Huntington’s disease in an mTOR-independent autophagy pathway. Nature Chemical Biology, 4(5), pp.295–305.Williams, R.S.B. et al., 2002. A common mechanism of action for three mood-stabilizingdrugs. Nature, 417(6886), pp.292–295.Wu, Y.T. et al., 2010. Dual role of 3-methyladenine in modulation of autophagy viadifferent temporal patterns of inhibition on class I and III phosphoinositide 3-kinase.Journal of Biological Chemistry, 285(14), pp.10850–10861.Xue, L. et al., 2004. A palmitoylation switch mechanism in the regulation of the visualcycle. Cell, 117(6), pp.761–771.146Yang, Y.P. et al., 2013. Application and interpretation of current autophagy inhibitorsand activators. Acta Pharmacologica Sinica, 34(5), pp.625–635.Yao, J. et al., 2014. Circadian and noncircadian modulation of autophagy inphotoreceptors and retinal pigment epithelium. Investigative Ophthalmology andVisual Science, 55(5), pp.3237–3246.Yao, T.P., 2010. The role of ubiquitin in autophagy-dependent protein aggregateprocessing. Genes and Cancer, 1(7), pp.779–786.Young, R.W., 1967. The renewal of photoreceptor cell outer segments. Journal of CellBiology, 33(1), pp.61–72.Yu, D.Y. et al., 2013. Retinal ganglion cells: Energetics, compartmentation, axonaltransport, cytoskeletons and vulnerability. Progress in Retinal and Eye Research,36, pp.217–246.Yu, W.H. et al., 2005. Macroautophagy--a novel Beta-amyloid peptide-generatingpathway activated in Alzheimer’s disease. The Journal of cell biology, 171(1),pp.87–98. Available at:http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2171227&tool=pmcentrez&rendertype=abstract [Accessed May 25, 2015].Zaffagnini, G. & Martens, S., 2016. Mechanisms of Selective Autophagy. Journal ofMolecular Biology, 428(9), pp.1714–1724.Zhou, X.E., Melcher, K. & Xu, H.E., 2012. Structure and activation of rhodopsin. ActaPharmacologica Sinica, 33(3), pp.291–299.147AppendicesA IntroductionDuring my studies of autophagy in X. laevis rods, I also developed a line of transgenic X.laevis that allowed me to measure disk displacement rates in X. laevis rods and conesusing a peripherin-2 molecule tagged with a fluorescent protein as a reporter.A.1 Photoreceptor outer segment disk renewal:The disks in rod and cone outer segments are continuously renewed, with newlysynthesized disks added to the base of the outer segments and the distal parts of theolder outer segments phagocytized by the RPE cells in the retina (Brian M Kevany &Palczewski 2010). The rod outer segment (ROS) is renewed in an ordered way: newmembranes are incorporated into the ROS via the connecting cilium and evaginate fromthe cilium at the base of the outer segment. These evaginations expand into the full-width outer segment disks and become isolated from the plasma membrane and fromeach other (Nguyen-Legros & Hicks 2000; Young 1967). The mechanism of cone discrenewal is not completely understood. However, studies in ground squirrel, monkey andhuman showed that distal disks of the cone outer segment (COS) are phagocytized bythe RPE cells, similar to rods (Anderson et al. 1978). It has been proposed that the conedisk renewal may be more complex than that of rods, as the cone disks areinterconnected with each other and with the plasma membrane, such that maintaining148outer segment integrity during the cone disk renewal is more difficult (Anderson et al.1978).The disk displacement rate of rods in X. laevis was first determined by autoradiographyalmost 30 years ago (Young 1967; J. C. Besharse et al. 1977). Tadpoles receivedintraperitoneal injections of radioactive amino acids and were sacrificed at different timepoints. Eyes were fixed and sectioned, and the sections were processed forautoradiography. The disk displacement rate of rods was measured as radioactive banddisplacement. However, the disk displacement rate of cone photoreceptor cells cannotbe determined by this way, because the radioactive amino acids diffused throughout theCOS. This is likely due to diffusion of cone opsin through the interconnected cone “opendisk” structure.A.2 Peripherin-2:Peripherin-2 is a 33 kDa transmembrane protein required for disk morphogenesis; itwas first discovered by Robert Molday’s group in 1987 (Molday et al. 1987). In humans,peripherin-2 is encoded by PRPH2 gene and is required for normal disk morphogenesis(Arikawa et al. 1992; Connell et al. 1991). The peripherin-2 protein is localized to thedisk rims and incisures of rods and the inner disk rims of cones (Molday et al. 1987).Deletion of peripherin-2 produced nonfunctional rod precursors that underwentapoptosis and cones with atypical outer segments and reduced phototransductionefficiency (Farjo & Naash 2006; Farjo et al. 2007; Farjo et al. 2006). The exact functionof peripherin-2 is not clearly understood; although it may function in shaping the disk rimcurvatures (McMahon & Boucrot 2015; McMahon & Gallop 2005; Khattree et al. 2013).149Mutations in PRPH2 genes can cause various retinal diseases, including autosomaldominant retinitis pigmentosa (adRP) (Loewen et al. 2003).I described a new method using a fluorescent protein tagged peripherin-2 reporter todetermine the disk displacement rates of X. laevis rods and cones.Figure 29: Peripherin-2 localization to the rim region of photoreceptor disks. Peripherin-2 distributes to the entire rim of closed mature disks in ROS and also the inner rim ofopen disks in ROS and COS (Stuck et al. 2016).150A.3 Goal of this experiment:To measure the disk displacement rates in X. laevis rods and cones.B MethodologyB.1 Generation of HSP70-peripherin-eGFP transgenic X. laevistadpoles:The original peripherin-eGFP sequence was amplified from a plasmid containing thefused “X. Laevis peripherin-2 (Xrds-38) -eGFP” sequence (Loewen et al. 2003). Togenerate the HSP70-peripherin-GFP constructs, the “peripherin-GFP” sequence wascloned into the EcoRI/NotI sites of our plasmid Hsp70-2DI-GFP, which contains a heatshock promoter sequence (as mentioned above in section 2.1), replacing the sequenceof the GFP cDNA. This was done using the InFusion cloning system (Clontech), theupstream PCR primer 5’-CTC AAG CTT CGA ATT CGC CAC CAT GGC CCT GATGAA AAC TAA-3’ and the down-stream PCR primer 5’-TCT AGA GTC GCG GCC GCT-3’. Transgenic tadpoles expressing HSP70-peripherin-eGFP were generated viarestriction enzyme mediated integration (REMI) method as described in section 2.2above.151B.2 Immunohistochemistry and confocal microscopy:For measuring rod disk displacement rate, frozen retinal sections were stained withWGA-Alexa594 (at a 1:100 dilution) and Hoechst 33342 (Sigma-Aldrich) (at a 1:1000dilution) as described in section 2.4 above.For measuring cone disk displacement rate, frozen retinal sections were labeled with aprimary polyclonal cone opsin antibody YZ3526 at a 1: 1000 dilution of the originalrabbit serum (Moritz lab, unpublished reagent) and the Cy3 conjugated secondary goatanti-rabbit Cy3 (at a 1:1000 dilution), and counterstained with WGA-Alexa 647 (ThermoFisher) (at a 1: 10,000 dilution) and Hoechst 33342 (Sigma-Aldrich) (at a 1: 1000dilution).For imaging axoneme cilia in outer segments, retinal sections were labeled with twoprimary antibodies: the monoclonal anti-acetylated tubulin antibody T7451 (Sigma-Aldrich) (at 1:15,000 dilution of 1mg/ml) and the polyclonal cone opsin antibody YZ3526(at a dilution of 1:1000 of serum concentration), and the corresponding secondaryantibodies Cy3 conjugated goat anti-mouse antibody (at a dilution of 1:750) and Cy5conjugated goat anti-rabbit antibody (at a dilution of 1:750), and counterstained withHoechst 33342 (Sigma-Aldrich).Sections were imaged by using a Zeiss 510 laser scanning confocal microscope with a10× (air, NA 0.3) or 40× (water-immersion, NA1.2) objectives. Within individual152experiments, all images were taken at the same gain and contrast settings for the GFPchannel. Imaging processing was performed using Adobe Photoshop. For constructionof figures, the fluorescent signals derived from GFP channel were adjusted linearly tobest demonstrate the fluorescent peripherin-eGFP band. Signals derived from WGA-Alexa 594, WGA-Alexa 647 and Hoechst 33342 staining were adjusted nonlinearly tobest demonstrate retinal architecture. Signals derived from anti-rabbit Cy3 antibody,anti-mouse Cy3 antibody, and anti-rabbit Cy5 antibody were also adjusted nonlinearly tobest demonstrate retinal architecture.The widths of fluorescent bands were measured as the distances from the base of theouter segments to the top of fluorescent bands using Image J software.C ResultsA heterozygous male frog expressing HSP70-peripherin-eGFP was mated with awildtype female frog; offspring were raised in 12 h light and 12 h dark light cycles in an18℃ incubator. On 6 dpf, tadpoles were screened for eGFP fluorescence using anepifluorescence-equipped Leica MZ16F dissecting microscope, and only tadpoles withobservable fluorescence were raised. On 14 dpf, 5 tadpoles were sacrificed withoutheat-shock, the rest of tadpoles received an 1 h heat-shock at 33℃ in the light.Tadpoles were sacrificed at 12h, 24 h, 48 h, 96 h and 168 h following the heat-shock,respectively, with 5 tadpoles in each group. Eyes were fixed, cryo-sectioned andexamined by confocal microscopy.153In rods, there was no fluorescence observed without heat-shock. Green fluorescencewas first observed as very thin bands at the base of ROS at 12 h following the 1 h heat-shock. The widths of fluorescent bands (the distances from the base of ROS to the topof fluorescent bands) gradually increased from 24 to 168 h following the heat-shock,suggesting new proteins were continuously synthesized and added to the base of theROS. Fluorescent bands were striated, consistent with the known localization ofperipherin-2 to disk incisures. At 168 h following the heat-shock, the widths offluorescent bands were almost half the lengths of ROS. Based on the quantification byImage J, I calculated that the rod disk displacement rate is about 0.10 μm/h (Figure 30).154Figure 30: Measuring the disk displacement rate of X. laevis rods. The line chart showsthe quantification of the width of fluorescent bands in rods at different time points155following the 1 h heat-shock. Figure (A-F), confocal micrographs show rods expressingHSP70-perihperin-eGFP that were imaged either without heat-shock or at 12, 24, 48, 96and 168 h after the heat-shock respectively. (G) The line chart shows the quantificationof the widths of fluorescent bands in rods at different time points following the 1 h heat-shock. The widths of fluorescent bands in rods (the distances from the base of ROS tothe top of fluorescent bands) were measured by Image J. Error bars = S.E.M. N ≥ 3animals per group. Red: WGA-Alexa594. Blue: Hoechst 33342. Scale bar 5μm. OS:outer segment. IS: inner segment. N: nucleus.In cones, fluorescent signals were only observed after heat-shock. Green fluorescencewas first observed at the base of COS at 12 h following the heat-shock, and thefluorescent signals were concentrated on one side of COS (Figure 31). By labelling withthe anti-acetylated tubulin antibody T7451 (Sigma-Aldrich), I determined that thefluorescent peripherin-2 signals were concentrated on the sides labelled by tubulinantibody (data not shown), suggesting that peripherin-2 was localized to the inner rimsof COS, consistent with findings of others (Han et al. 2012). The widths of thefluorescent bands (the distances from the base of COS to the top of fluorescent bands)gradually increased in COS from 24 to 168 h following the heat-shock (Figure 31),suggesting that the COS were continuously renewed.Fluorescent peripherin-eGFP proteins were mostly recruited to the newly synthesizeddisks at the base of COS, additionally, some of the fluorescent proteins were alsorecruited to the distal old/mature discs, suggesting that the old discs may undergo someform of disk remodeling. Based on the quantification from Image J, I calculated that the156cone disk displacement rate is around 0.066 μm/h (Figure 31), about 34% slowercompared to that of rods.157158Figure 31: Measuring the disk displacement rate of X. laevis cones. Figure (A-F),confocal micrographs show cones expressing HSP70-perihperin-eGFP that wereimaged either without heat-shock or at 12, 24, 48, 96 and 168 h after the heat-shockrespectively. Cones are identified by arrowheads. (G-H) Confocal micrograph shows acone expressing HSP70-perihperin-eGFP that were imaged at 96 h after the heat-shockand labeled with the cone opsin antibody (magenta). Figure H is a superimposed 3D Z-stack image. (I) The line chart shows the quantification of the widths of fluorescentbands in cones at different time points following the 1 h heat-shock. The widths offluorescent bands in cones (the distances from the base of COS to the top offluorescent bands) were quantified by Image J. Error bars= S.E.M. N ≥ 2 animals pergroup. Scale bar 5 μm. Red: WGA-Alexa 647, Magenta: cone opsin antibody YZ3526,Blue: Hoechst 33342. COS: cone outer segment. IS: inner segment. N: nucleus.D Conclusions and discussions:I demonstrate a method for examining disk renewal rates in X. laevis cones that allowsdirect comparison with rod disk renewal rates. Unlike cone opsin, the cone disk rimcomponent peripherin-2 cannot freely diffuse between adjacent disks, thus it can beused as a reporter to measure cone disk renewal rate. By using HSP70-peripherin-2-eGFP transgenic X. laevis, I determined the rod and cone disk rim displacement ratesand that cone disk rims are renewed more slowly than rod disk rims. It is likely that mynew method reports only on disk rim renewal rates, and not renewal of disk lamellaecomponents such as cone opsin.159There is some suggestion that although peripherin-2 is inserted primarily at the cone OSbase, small quantities are inserted more distally, or possibly move distally duringremodeling steps (Figure 31G, H). Cone disks are remodeled, in that they becomesmaller as they are displaced upwards, resulting in smaller circumferences. It is unclearhow the excess disk rim components are extracted or displaced during this process,although several models have been proposed. These transgenic animals may be usefulfor examining disk rim remodeling as disk mature, and for examining factors thatinfluence cone outer segment biosynthesis.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0369290/manifest

Comment

Related Items