Characterizing Vertebrate Rhodopsin Natural Variation in Evolution, Function, and Disease

CHARACTERIZING VERTEBRATE RHODOPSIN NATURAL VARIATION IN EVOLUTION, FUNCTION, AND DISEASE

Nihar Bhattacharya

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Cell and Systems Biology University of Toronto

CHARACTERIZING VERTEBRATE RHODOPSIN NATURAL VARIATION

IN EVOLUTION, FUNCTION, AND DISEASE

Nihar Bhattacharya

Doctor of Philosophy

Department of Cell and Systems Biology

University of Toronto

2018

ABSTRACT

Vertebrate dim light vision is mediated by the rod visual pigment, rhodopsin, a member of the G protein-coupled receptor (GPCR) superfamily of proteins. In the dark, rhodopsin is covalently bound to a vitamin A-derived 11-cis chromophore, which acts as an inverse agonist to stabilize the inactive state of rhodopsin. When exposed to light of a maximal wavelength (λmax), the 11-cis retinal chromophore isomerizes to an all-trans conformation, initiating a series of structural shifts to the light-activated state of rhodopsin. This results in a signalling cascade within the rod photoreceptor cell and, ultimately, the perception of light.

The goal of this thesis is to investigate natural variation in rhodopsin function in the context of evolutionary adaptation, chromophore usage, and disease mutations. Following a general introduction, in Chapter II, I characterize the visual system of the diurnal colubrid snake

Pituophis melanoleucus using immunohistochemistry of retinal sections and spectroscopy of

purified visual pigments expressed in vitro, revealing an unusual rhodopsin with cone opsin properties found in cone-like rod photoreceptors. In Chapter III, I investigate the effects of the rare vertebrate chromophore, 11-cis 3,4 dehydroretinal (A2), on the spectral and non- spectral properties of rhodopsin. In Chapters IV and V, I study the effects of pathogenic mutations in rhodopsin that cause the retinal degenerative disease retinitis pigmentosa (RP).

In Chapter IV of my thesis, I identify the phenotype of RP mutations found in the extracellular loop 2 of rhodopsin and assess the effects of functional rescue using two different approaches. Finally, in Chapter V, I characterize three novel RP mutations to investigate the relationship between the in vitro and clinical disease phenotypes. The investigations in this thesis expand our understanding of snake retinal evolution, the role of the chromophore in rhodopsin function, and the effect of pathogenic mutations on rhodopsin structure and function. This thesis combines data from non-model organisms, non- mammalian chromophores, and non-wildtype pathogenic mutations to significantly increase our understanding of the scope of rhodopsin functionality.

iii

ACKNOWLEDGMENTS

Over the last 6 years, I have grown, matured, and changed so much that when I think of myself pre-Toronto, I can hardly believe that was me. There are so many people I have to acknowledge and thank for this transformation. I would like to begin by thanking my supervisor, Belinda Chang. The experiences I had in her lab have been formative and helped shape me as a researcher and scientist. I thank her for indulging my biochem-y/wet lab ways while also hammering me into something resembling a coherent writer. I thank her for letting me try out new things and for allowing me to gain experience in so many other facets of research on top of those that are typical for a graduate student.

I would like to thank my supervisory committee: Vince Tropepe and Jane Mitchell.

They’ve made every single one of my committee meetings an absolute delight and were a source of great feedback and advice. I’d also like to thank Dr. David Hunt for agreeing to be my external, and Jennifer Mitchell for agreeing to be my internal examiner. I would also like to acknowledge Les Buck for his feedback and comments during my appraisal exam. I’d like to thank Vince again for being such a wonderful collaborator on the snake project. Also,

Massimo Olivucci and Akimori Wada for their patience and guidance during our collaboration on the A2 project. I would also like to thank Elise Heon for her contribution to the RP project. I thank Henry Hong and the Imaging Facility in Ramsay Wright for his help with everything microscopy related. Also, my thanks to Tamar, Ian and Jim for all the assistance and conversations over the years, all three of you made my PhD a more pleasant experience. I would also like to thank the Vision Science Research Program and Janet Wong for funding me and for exposing me to the larger world of ophthalmology.

I’d also like to thank Melody, Chris, Ken, Tanja, Nyla and the rest of the teaching staff for giving me the opportunity to develop as a teacher and for giving me a space to retreat to in the building. I’d like to thank Virlana Shchuka for her friendship and all of the lovely and fascinating conversations we’ve had over the years. To all the Chang lab members, past and present, thank you for your friendship, your assistance, and your collaboration. I would specifically like to thank Amir for helping me so much with stats,

Alex for his patient help with my figures and A2 brainstorming, Ryan for dealing with my naïveté with evolutionary biology, and Ben for being my RP and wet lab comrade in arms. I would especially like to thank Frances, Sarah, Eduardo and Gianni for keeping me laughing, keeping me well fed, keeping me nerdy, and for keeping me sane with constant supply of cute animal/Makwa/Tabi/Gia/Tess pics. I could write an entire paragraph of thanks for each of you. My thanks to you all, and I hope we keep in touch. To my extended dojo family, thank you for giving me a different perspective on things and for the extremely fun nights out. Specific thanks to Sacha, Elizabeth and Greg, Darby and Ece, and Patrice for keeping me limber.

Finally, to my family: thank you for your patience with me and thank you for loving me. Choto Dida and Dadu and the Roy family, thank you for taking such good care of me in

Toronto and Atlanta. To Dada and Boudi, thank you for flying me out to SF so many times and for being so supportive, for nurturing my artistic side and for being a source of calm and sanity for me. And to Ma and Baba, I don’t even know where to start. You’ve both helped me so much during my PhD. Thank you for all the trips, for all the nature retreats, for still taking care of me, for worrying about me. I love you all so much

TABLE OF CONTENTS ACKNOWLEDGMENTS ...... IV TABLE OF CONTENTS ...... VI LIST OF TABLES ...... VIII LIST OF FIGURES ...... IX LIST OF ABBREVIATIONS ...... XI CHAPTER I: GENERAL INTRODUCTION...... 1 1.1 - EYE ...... 1 1.2 - RETINA ...... 6 1.3 - PHOTORECEPTORS ...... 9 1.3.1 - ROD AND CONE PHOTORECEPTORS ...... 10 1.3.2 - PHOTORECEPTOR TRANSMUTATION ...... 12 1.4 - PHOTOTRANSDUCTION ...... 16 1.5 - OPSINS ...... 20 1.6 - RHODOPSIN ...... 22 1.6.1 - RHODOPSIN STRUCTURE ...... 23 1.6.2 - RHODOPSIN ACTIVATION ...... 25 1.6.3 - RHODOPSIN VARIATION ...... 28 1.7 - VERTEBRATE CHROMOPHORE ...... 31 1.7.1 - THE A2 CHROMOPHORE ...... 31 1.7.2 - RETINOID CYCLE ...... 34 1.8 - RETINAL DISEASE ...... 37 1.8.1 - RETINITIS PIGMENTOSA ...... 37 1.8.2 - DISEASE MUTATIONS IN RHODOPSIN ...... 39 1.8.3 - RESCUE AND TREATMENT METHODS OF RETINITIS PIGMENTOSA ...... 44 1.8.4 - RETINITIS PIGMENTOSA MUTATIONS SITES IN RHODOPSIN...... 47 1.9 - THESIS OBJECTIVES ...... 50 1.10 - THESIS OVERVIEW ...... 52 1.11 – REFERENCES ...... 56 1.12 – COPYRIGHT ...... 79 CHAPTER II: CONE-LIKE RHODOPSIN EXPRESSED IN THE ALL CONE RETINA OF THE COLUBRID PINE SNAKE AS A POTENTIAL ADAPTATION TO DIURNALITY ...... 80 2.1 – ABSTRACT ...... 80 2.2 - INTRODUCTION ...... 82 2.3 - MATERIALS AND METHODS ...... 85 2.4 - RESULTS ...... 89 2.5 - DISCUSSION ...... 97 2.6 - REFERENCES ...... 105 2.7 - SUPPLEMENTAL INFORMATION ...... 117 CHAPTER III: INVESTIGATING THE 11-CIS 3,4 DEHYDRORETINAL (A2) CHROMOPHORE IN DIVERSE VERTEBRATE RHODOPSINS REVEALS INSIGHTS INTO THE SPECTRAL AND NON- SPECTRAL ROLES OF THE CHROMOPHORE IN RHODOPSIN FUNCTION ...... 124 3.1 - ABSTRACT ...... 124 3.2 - INTRODUCTION ...... 126 3.3 - MATERIALS AND METHODS ...... 131

3.4 - RESULTS ...... 134 3.5 - DISCUSSION ...... 148 3.6 - REFERENCES ...... 158 3.7 – SUPPLEMENTAL ...... 170 CHAPTER IV: CHARACTERIZING THE CLINICAL AND IN VITRO PHENOTYPE OF TWO RETINITIS PIGMENTOSA MUTATIONS, P180L AND G182V, IN THE EXTRACELLULAR LOOP 2 OF RHODOPSIN ...... 171 4.1 - ABSTRACT ...... 171 4.2 - INTRODUCTION ...... 173 4.3 - MATERIALS AND METHODS ...... 178 4.4 - RESULTS ...... 182 4.5 - DISCUSSION ...... 198 4.6 - REFERENCES ...... 206 4.7 – SUPPLEMENTAL INFORMATION...... 216 CHAPTER V: COMPARING RETINITIS PIGMENTOSA RHODOPSIN MUTATIONS IN VITRO TO CLINICAL PHENOTYPES ...... 217 5.1 - ABSTRACT ...... 217 5.2 - INTRODUCTION ...... 219 5.3 – MATERIALS AND METHODS...... 222 5.4 - RESULTS ...... 225 5.5 - DISCUSSION ...... 236 5.6 - REFERENCES ...... 242 5.7 – SUPPLEMENTAL INFORMATION...... 250 CHAPTER VI: GENERAL DISCUSSION...... 252 6.1 – GENERAL SUMMARY ...... 252 6.2 – GENERAL DISCUSSION ...... 253 6.3 - CONCLUSION ...... 263 6.3 - REFERENCES ...... 264

vii

LIST OF TABLES

TABLE 1.1 - ALL KNOWN AND PREDICTED RETINITIS PIGMENTOSA MUTATIONS IN RHODOPSIN . 41 TABLE S2.1. DEGENERATE PRIMERS FOR SEQUENCING OPSIN GENES FROM GENOMIC DNA .. 117 TABLE S2.2. LIST OF SEQUENCES USED IN THE PHYLOGENETIC ANALYSES OF OPSIN GENES ... 119 TABLE 3.1 - MEASURED ΛMAX VALUES OF A1 AND A2 PAIRS ...... 146 TABLE 3.2 - ARRHENIUS PLOT DATA...... 147 TABLE 4.1 - EFFECTS OF PHARMACOLOGICAL AND STRUCTURAL RESCUE OF EL2 RP MUTANTS...... 197 TABLE S1 – CLINICAL PATIENT SUMMARIES...... 216 TABLE S5.1. SITE-DIRECTED MUTAGENESIS PRIMERS...... 250 TABLE S5.2 PHENOTYPE SUMMARY...... 251

viii

LIST OF FIGURES

FIGURE 1.1 – GROSS ANATOMY OF THE EYE...... 5 FIGURE 1.2 – LIGHT IMAGE AND SIMPLIFIED DIAGRAM OF RETINAL LAYERS...... 8 FIGURE 1.3 – GENERAL ROD AND CONE PHOTORECEPTOR ANATOMY ...... 10 FIGURE 1.4 – PHOTORECEPTOR TRANSMUTATION AS HYPOTHESIZED BY WALLS (1942)...... 15 FIGURE 1.5 – PHOTOTRANSDUCTION CASCADE IN THE PHOTORECEPTOR CELL...... 19 FIGURE 1.6 – SCHEMATIC OF VERTEBRATE COMPLEMENT OF PHOTORECEPTORS ...... 21 FIGURE 1.7 – SNAKE PLOT OF HUMAN RHODOPSIN ...... 22 FIGURE 1.8 – RHODOPSIN STRUCTURAL MOTIFS...... 25 FIGURE 1.9 – SPECTRAL SENSITIVITY SHIFTING WITH ENVIRONMENT VISUAL SPECTRUM...... 30 FIGURE 1.10 - MOLECULAR STRUCTURES OF THE TWO VERTEBRATE CHROMOPHORES ...... 31 FIGURE 1.11 - THE RETINAL PIGMENT EPITHELIUM (RPE) VISUAL CYCLE...... 36 FIGURE 1.12 – RP MUTATIONS IN RHODOPSIN ...... 39 FIGURE 2.1: IMMUNOHISTOCHEMICAL STAINING ...... 94 FIGURE 2.2: UV-VISIBLE DARK ABSORPTION SPECTRA OF PINE SNAKE OPSINS...... 95 FIGURE 2.3: HYDROXYLAMINE REACTIVITY OF PINE SNAKE ...... 96 FIGURE S2.1. RHODOPSIN GENE TREE...... 121 FIGURE S2.2. SWS1 GENE TREE...... 122 FIGURE S2.3. LWS GENE TREE ...... 123 FIGURE 3.1 - MOLECULAR DIAGRAM OF THE (A) 11-CIS RETINAL (A1) CHROMOPHORE AND THE (B) 11-CIS 3,4-DEHYDRORETINAL (A2) CHROMOPHORE...... 138 FIGURE 3.2 - NORMALIZED UV-VISIBLE ABSORBANCE SPECTRA OF VERTEBRATE RHODOPSIN WITH A1 (BLACK) AND A2 (RED) CHROMOPHORES...... 139 FIGURE 3.3 - (A) A1 AND A2 ΛMAX PAIRS FROM 11 VERTEBRATE RHODOPSINS ...... 141 FIGURE 3.4 - LIGHT-ACTIVATED RETINAL RELEASE OF A1 AND A2 BOVINE RHODOPSIN ...... 142 FIGURE 3.5 - LIGHT ACTIVATED RETINAL RELEASE OF A1/A2 CICHLID AND ZEBRAFISH ...... 143 FIGURE 3.6 - HOMOLOGY MODELLING OF ZEBRAFISH AND BOWERBIRD RHODOPSIN ...... 144 FIGURE 3.7 - CHROMOPHORE CHARGE SURFACE ...... 145 FIGURE S3.1 – NMR SPECTRA OF 11-CIS A2 CHROMOPHORE ...... 170 FIGURE 4.1 - STRUCTURE OF WILDTYPE RHODOPSIN ...... 189 FIGURE 4.2 - PHYSICAL MAP OF THE PGFP EXPRESSION VECTOR ...... 190

FIGURE 4.3 - OCULAR PHENOTYPE OF CASE 1 (CARRYING THE P180L VARIANT) ...... 191 FIGURE 4.4 - RHODOPSIN RP MUTANTS ...... 192 FIGURE 4.5 - THE STABILITY MUTANT ...... 193 FIGURE 4.6 - UV-VISIBLE DARK ABSORPTION SPECTRA OF WILDTYPE AND RP RHODOPSIN MUTANTS ...... 194 FIGURE 4.7 - COMPARISON OF HYDROXYLAMINE REACTIVITY OF RP ...... 195 FIGURE 4.8: SCHEMATIC DIAGRAM ILLUSTRATING A MODEL FOR PHARMACOLOGICAL RESCUE AND STABILITY MUTANT RESCUE IN EL2 RP RHODOPSIN MUTATIONS...... 196 FIGURE 5.1: 2D AND 3D VISUALIZATIONS OF DARK-STATE RHODOPSIN STRUCTURE...... 229 FIGURE 5.2: CONFOCAL FLUORESCENT MICROSCOPY IMAGES OF SK-N-SH CELLS ...... 230 FIGURE 5.3: UV-VISIBLE ABSORBANCE DIFFERENCE SPECTRA ...... 232 FIGURE 5.4: CLINICAL ASSESSMENT OF PATIENTS WITH RHODOPSIN MUTATIONS ...... 233 FIGURE 5.5: HOMOLOGY MODELLING...... 235

LIST OF ABBREVIATIONS Abbreviation Definition A1 Retinal A2 3,4 dehydroretinal A280 Absorbance at 280 nm adRP Autosomal dominant retinitis pigmentosa arRP Autosomal recessive retinitis pigmentosa ATF6 Activating transcription factor 6 Aλmax Absorbance at λmax BiP Binding immunoglobulin protein BSI Blue-shifted intermediate cDNA Complimentary deoxyribonucleic acid cGMP Cyclic guanosine monophosphate CNG Cyclic nucleotide gated cation CSNB Congenital stationary night blindness CYP27c1 Cytochrome P450 family protein DHA Docosahexaenoic acid DHA docosahexaenoic acid DM N-dodecyl-b-D-maltoside DNA Deoxyribonucleic acid DOPE Discrete optimized protein energy Ea Activation energy EL2 Extracellular loop 2 EM Electron microscopy ER Endoplasmic reticulum ERG Electroretinogram GCAP Guanylate cyclase activating proteins gDNA Genomic deoxyribonucleic acid GDP Guanosine diphosphate GFP Green fluorescent protein GPCR G protein-coupled receptor GRK G protein-coupled receptor kinase GRK1 Rhodopsin kinase GRK7 Cone opsin kinase GTP Guanosine triphosphate GVF Goldmann visual fields Human recombinant green fluorescent hrGFP II protein IRBP Interphotoreceptor retinoid binding protein IRE1 Inositol requiring enzyme 1 LRAT Lecithin:retinol acyltransferase Lumi-I Lumirhodopsin I Lumi-II Lumirhodopsin II LWS Long wavelength-sensitive opsin MCS Multiple cloning site

MI Metarhodopsin I MII Metarhodopsin II MIII Metarhodopsin III min Minute mRNA Messenger ribonucleic acid MSP Microspectrophotometry nm Nanometer PBS Phosphate buffered solution PBT Phosphate buffered solution with Tween-20 PCR Polymerase chain reaction PDE Phosphodiesterase Phosphate buffered solution with Tween-20 PDT and DMSO Double-stranded RNA-activated protein PERK kinase PKR-like ER kinase PP2A Phosphatase 2A RGS9 Regulator of G protein signalling RH1 Rhodopsin RH2 Middle wavelength-sensitive opsin RNA Ribonucleic acid RP Retinitis pigmentosa RPE Retinal pigment epithelium RPE65 Retinoid isomerohydrolase SWS1 Short wavelength-sensitive opsin 1 SWS2 Short wavelength-sensitive opsin 2 t1/2 Half-life TM Transmembrane helix UPR Unfolded protein response UPS Ubiquitin proteasome system UV Ultraviolet UV-Vis Ultraviolet-visible VA Visual acuity λmax Maximal wavelength of absorption

xii CHAPTER 1: INTRODUCTION 1

CHAPTER I: GENERAL INTRODUCTION

Visual systems enable organisms to interact with and sense their surroundings by translating information about their environment contained in visible light into a biological signal. The complexity of visual anatomy varies greatly across organisms. The most simplistic systems are present in organisms that can detect only light, but a complex eye is required for animals to possess image-forming vision (Land, 2005). Here I will summarize the major components of the visual system and retinal disease in vertebrates.

1.1 - EYE

The basic vertebrate eye includes the sclera, choroid, cornea, iris with a pupil, ciliary body, lens and a retina with retinal pigment epithelium (Land, 2005; Walls, 1942) (Figure

1.1A). The outer surface of the eye consists of a clear cornea and a white, opaque sclera. The fibrous sclera maintains the overall spherical shape of the eye (Walls, 1942). Light enters the eyeball through the cornea located on the anterior surface of the eye, where it acts as a structural barrier protecting the interior of the eye (DelMonte and Kim, 2011; Walls, 1942).

The cornea refracts the incoming light which continues through aqueous humor-filled anterior chamber until it encounters the iris. The iris is typically a circular structure that controls the size and shape of the pupil (Walls, 1942). The pupil is the opening through which light passes to the lens and into the interior of the eye (Kolb, 2007). Adjustments in pupil size compensate for different light intensities in the organisms environment (Walls,

1942). Once light has entered the eye through the pupil, it is refracted by the clear lens and an inverted image is projected onto the light-sensitive retina at the posterior of the eye (Kolb,

2007; Walls, 1942). The lens is adjusted by the ciliary body to focus on objects at variable distances. Depending on the species, the ciliary body does this by either adjusting the shape of the lens or by adjusting the distance of the lens from the retina (Kolb, 2007; Walls, 1942)

(Figure 1.1F). The ciliary process, which provides nutrients to the anterior eye, is also found in the ciliary body (Walls, 1942). The ciliary process is formed by the invagination of the choroid which is the vascular layer between the sclera and the retinal epithelium layer of the retina (Kolb, 2007; Walls, 1942). The choroid is heavily pigmented and absorbs light to minimize reflections within the eye while also providing the majority of the eye with nutrients and blood supply (Kolb, 2007; Walls, 1942). The retina and the retinal pigment epithelium layer (RPE) line the inner surface of the eye. The RPE lies below the retina and, as the name suggests, is typically highly pigmented and absorbs excess light (Kolb, 2007;

Walls, 1942). The blood supply to the retina is provided by the RPE and it is where the retinoid cycle takes place (see section 1.7.2 (Kolb, 2007; Lamb and Pugh, 2004; Walls,

1942)). The retina is the layered neural tissue lining the back of the eye that is responsible for the detection of light and the propagation of the neurological signaling from the retina, through the optic nerve to the brain (see section 1.2 (Walls, 1942)). While the general structure of the vertebrate eye remains somewhat constant among species, there is variation in morphology between species and groups enabling visual specializations. I will briefly discuss some of the more common specializations in gross eye anatomy among vertebrates.

Degenerate eyes have been observed in multiple vertebrate groups, usually associated with ecological niches with low to no light. Degenerate eyes may feature no lens, an undifferentiated sclera and choroid, a disorganized retina, and occasionally epidermis enclosing the eye. Degenerate eyes have been observed in some deep sea teleosts, cave fish,

3 hagfish, blind snakes, and naked mole rats (Caprette, 2005; Hetling et al., 2005; Walls, 1942)

(Figure 1.1B).

Adaptations for aquatic environments are quite varied among all vertebrate groups, but universally, all aquatic eyes have to contend with the different refractive properties of water versus air. As the density of the cornea is close that of water, the refractive properties of the cornea are negated. Certain teleosts have yellowed corneas and lenses to act as both a light filter but also to detect camouflaged species (Fernald, 1988). Generally, all aquatic eyes are somewhat flattened and ellipsoid in nature for assist in streamlining, though this can limit mobility of the eye in the socket (Fernald, 1988; Mass and Supin, 2007; Walls, 1942).

Additionally, the sclera can vary drastically in thickness around the optic nerve, usually due to blood supply requirements of the eye (Fernald, 1988; Mass and Supin, 2007; Walls, 1942)

(Figure 1.1C).

Some adaptations in eye anatomy to dark environments usually result in a more tubular shaped eye as seen in owls and deep-sea fish (Fernald, 1988; Hall and Ross, 2006; Walls,

1942) (Figure 1.1D). More general are adaptations in size and retinal composition, with larger eyes (Hall and Ross, 2006), and all rod retinas or a high proportion of rod photoreceptors mediating dim light vision, being observed in some owls, nocturnal mammals, cetaceans, fishes and other aquatic species (Fernald, 1988; Hall and Ross, 2006;

Mass and Supin, 2007; Walls, 1942). Certain avian species, such as eagles, also show adaptation for high visual acuity and resolution. These species have a “globose” shaped eyeball with prominent concave region by the ciliary body and the border of the large cornea.

These species also have a thick choroid, thickest at the fundus, and a thick, highly ordered and organized retina (Walls, 1942) (Figure 1.1E).

Other general and universal eye trends can be seen in response to body size, with larger bodied vertebrates having thicker sclera (Walls, 1942). A cartilaginous eye cup is found in most vertebrates excluding some teleosts, salamanders, snakes and non-monotreme mammals

(Walls, 1942). Additionally, the size and shape of the anterior chamber can vary depending on how the lens accommodation is achieved in the species. Certain animals have very small anterior chambers which prevent lens accommodation, while larger anterior chambers are usually required for species which accommodate the lens via movement towards/away from the retina (Caprette, 2005; Fernald, 1988; Walls, 1942) (Figure 1.1F).

Figure 1.1 – Gross anatomy of the eye. (A) the typical vertebrate eye with sclera (sc), cornea (c), choroid (ch), lens (l), iris (i), ciliary body (cb), and retina (r). (B) the degenerate eye with an epidermal cover and undifferentiated sclera/cornea – based on the hagfish eye. (C) The aquatic eye with ellipsoid shape and thick sclera and choroid – based on the grey whale eye. (D) A tubular eye for vision in low light environments with an accessory retina – based on the deep-sea fish Scopelarchus analis. (E) The eagle eye, with globose shape and concave indentations at periphery of cornea. (F) The snake eye, with spectacle covering and absent ciliary body – based on the eye of colubrid snakes.

1.2 - RETINA

The retina is a thin, highly ordered, neural tissue layer containing multiple stratified neural cell types lining the inner surface of the eye. The standard retina contains alternating layers of nuclei and synapses, consisting of three nuclear layers with cell bodies and two synaptic plexiform layers (Walls, 1942) (Figure 1.2). Beginning at the most interior layer, the first layer is the ganglion cell layer which contains primarily ganglion cells and displaced amacrine cells (Gregg et al., 2013; Kolb, 2007). The axons of the ganglion cells coalesce into the optic nerve and relay the visual signal to the brain (Walls, 1942). Below the ganglion cell layer is the inner nuclear layer containing the cell bodies from several different cell types: horizontal cells, bipolar cells, amacrine cells and Müller cells (Walls, 1942). Between the inner nuclear and the ganglion cell layer is the inner plexiform layer, containing the synaptic junction of the ganglion and amacrine cells with bipolar cells (Gregg et al., 2013; Walls,

1942). Proceeding from the inner nuclear layer, the next cellular level is the outer nuclear layer containing the photoreceptor cell nuclei (Gregg et al., 2013; Walls, 1942). Between the inner nuclear and the outer nuclear layer is the outer plexiform layer consisting of the synaptic junctions between bipolar or horizontal cells and the photoreceptor cells (Dowling,

2009). The photoreceptor cells ultimately absorb light and initiate the signal transduction cascade through the layers of the retina to the optic nerve, to the brain (Gregg et al., 2013). In vertebrates, there are two types of photoreceptors, rods and cones. Rods mediate dim-light vision in low scotopic light levels while cones mediate bright light and colour vision in high photopic light levels (Chen and Sampath, 2013). Most vertebrate retina contain both photoreceptor types and therefore possess duplex retina, though vertebrates with simplex retinae (one photoreceptor type) do exist, but rare (Walls, 1942).

The ganglion cell layer contains both ganglion cells and displaced amacrine cells.

Ganglion cells receive input from bipolar and amacrine cells and send the output from the retina as the axons of the ganglion cells form the optic nerve to carry the visual signal to the brain (Gregg et al., 2013). Among the separate cell types photoreceptors, bipolar cells and some ganglion cells transmit the visual signal through the layers of the retina, while the amacrine and horizontal cells transmit signal laterally with extensive interconnections in the plexiform layers (Gregg et al., 2013). Visual information is processed in both plexiform layers. The inner plexiform layer processes temporal changes in visual stimuli such as movement detection, while the outer plexiform layer processes spatial analyses with separate

ON and OFF pathways. Müller glial cells are interspersed as support cells for the retinal neurons (Gregg et al., 2013).

With multiple cells types and different synaptic connection combinations, specific functional pathways exist in the retina, such as rod-specific pathways. Currently three different rod-specific pathways have been characterized which are thought to process visual signals with different sensitivities (Gregg et al., 2013). The primary pathway utilizes rod- specific bipolar cells, AII amacrine cells in low scotopic light levels, while the secondary and

8 tertiary pathways utilize cone pathways and are thought to relay rod photoreceptor signaling under higher mesopic light levels (Gregg et al., 2013).

Figure 1.2 – Light image and simplified diagram of retinal layers. From Hartong et al. (2006)

1.3 - PHOTORECEPTORS

The vertebrate ciliary-derived photoreceptor cells (Fain et al., 2010) have four distinct segments: the outer segment, the inner segment, the cell body, and the synaptic terminals

(Chen and Sampath, 2013). The outer segment consists of membrane disks dense with light sensitive visual pigment proteins. The inner segment contains the cellular machinery of the cell: the Golgi, the endoplasmic reticulum and the mitochondria, while the nucleus resides in the cell body (Chen and Sampath, 2013; Kolb, 2007). The synaptic terminals transmit electrical signals to the horizontal and bipolar cells through the release of the neurotransmitter glutamate (Chen, 2005; Chen and Sampath, 2013). There are two types of photoreceptor cells in the retina, the cone and the rod (Figure 1.3). Most animals have duplex retinas, which are retinas with both cell types (Ebrey and Koutalos, 2001). Simplex retinas, or retinas with only one cell type, are possible although rare (Walls, 1942). The rod photoreceptor cells are usually responsible for dim light vision typically expressing the dim- light visual pigment rhodopsin (RH1), while the cone cells are responsible for colour vision in brighter light expressing photopic cone opsins: long wavelength-sensitive (LWS), middle wavelength-sensitive (RH2), short wavelength-sensitive 1 (SWS1) and short wavelength- sensitive 2 (SWS2) (Chen and Sampath, 2013). To facilitate the difference in function, there are several important physiological and morphological distinctions between rod and cone cells.

Figure 1.3 – General rod and cone photoreceptor anatomy. From Chen and Sampath (2013)

1.3.1 - ROD AND CONE PHOTORECEPTORS

The most distinct difference between rods and cones can be seen in the topology of the outer segment. Cones are characterized by smaller tapering outer segments with membrane disks open to the external environment packed with visual pigment cone opsins, whereas in rods the membranous discs are sealed off from the external environment by the plasma membrane of the rod (Chen and Sampath, 2013; Lamb, 2013) (Figure 1.3). In both photoreceptor cell types, nascent outer segments discs are formed near the ciliary neck close to the inner segment, with older discs gradually moving to the end of the rod (Chen and

Sampath, 2013). Another difference in the outer segment is the presence of deep longitudinal

11 indentations called incisions found in rod outer segments (Chen and Sampath, 2013). In the inner segment, cones may contain a paraboloid of glycogen or pigmented oil droplets that may act as spectral filter for incoming light (Chen and Sampath, 2013). Specializations in the inner segment of cone photoreceptors allow for the refraction and tunneling of incident light into the tapering outer segment (Harosi and Novales Flamarique, 2012). Rod-cone differences are also apparent at the synaptic terminals of the photoreceptors. Both photoreceptors make contact with bipolar and horizontal cells in the outer plexiform layer

(Gregg et al., 2013). Cone synaptic terminals form pedicles which make two types of contacts: ribbon synapse (formed by downstream dendrites of ON bipolar and horizontal cells in multiple invaginations of the cone pedicle) and flat contacts (along the base of the pedicle with OFF bipolar cells) (Chen and Sampath, 2013; Gregg et al., 2013). Rod synaptic terminals form smaller sphericules with ribbon synapses with single invaginations with connections to rod bipolar and horizontal cells (Chen and Sampath, 2013; Gregg et al., 2013).

Photoreceptor morphology is specialized for scotopic and photopic vision. Cone physiology maximizes wavelength specificity and kinetics while rod physiology maximizes sensitivity (Lamb, 2013). The long cylindrical rod outer segments maximize photosensitivity by increasing the number of visual pigment molecules available for light absorption (Lamb,

2013). The tapered outer segment funneling and spectral filtering in the inner segment of cones compensate for light loss caused by self-screening, by producing a higher signal-to- noise ratio and allowing for metabolic savings due to decreased structural components

(Harosi and Novales Flamarique, 2012).

1.3.2 - PHOTORECEPTOR TRANSMUTATION

Retinas with both photoreceptor types are known as duplex retina and are quite common, though simplex retina with one cell type have been found in multiple vertebrate classes

(Walls, 1942). However, simplex retina are quite common however among squamate reptiles

(Walls, 1942). Indeed, there are multiple closely related species with apparent duplex and simplex retina. The comparative ophthalmologist Gordon Walls proposed the theory of photoreceptor transmutation to explain the evolutionary transitions between simplex and duplex retinas in squamate reptiles (Walls, 1942). The theory of photoreceptor transmutation proposes the existence of an evolutionary process through which a rod photoreceptor can be converted into a cone or vice versa, rather than the repeated loss and gain of photoreceptor types between simplex and duplex species (Walls, 1942). Walls observed multiple retinal organizations in snakes and geckos from all-cone in diurnal species to all-rod in nocturnal species as well as intermediate-type simplex retina which lead him to believe that photoreceptor morphologies could evolutionarily transform from one type to another (Walls,

1942). Cone-to-rod transmutation has been well characterized in geckos while rod-to-cone transmutation is only beginning to be described in diurnal colubrid snakes (Schott et al.,

2016) (Figure 1.4).

The best described example of photoreceptor transmutation has been characterized in geckos, a diverse group of squamate lizards. The majority of geckos are nocturnal and seem to possess all-rod simplex retina. Walls (1942) proposed, based on retinal physiology, that the all-rod retinas evolved from all-cone retinas of ancestral diurnal lizards. He further predicted that geckos which had reverted to a diurnal lifestyle would also have transmuted the all-rod retina back to all-cone (Walls, 1942). Several studies over the past decades have

13 shown his predictions to be correct. The gecko rod-like photoreceptors were found to have exposed outer segment membrane discs characteristic of cones (Röll, 2001a; Röll, 2001b;

Tansley, 1964), while molecular studies show that the supposed all-rod geckos lack the rod visual pigment rhodopsin (RH1) and express only cone visual pigments (LWS, RH2, and

SWS1) in the outer segments of the rod-like photoreceptors (Kojima et al., 1992).

Investigation of the phototransduction cascade in these ‘rod’ photoreceptors showed the presence of the cone isoforms (Zhang et al., 2006) though the rod-like photoreceptors have been shown to function more similar to rods than cones. These studies together support the hypothesis that nocturnal geckos underwent cone-to-rod transmutation and that the “rods” in the retina are in fact transmuted rod-like cones, likely allowing for the redevelopment of scotopic vision due to the ancestral loss of rod photoreceptors.

Transmutation in snakes seems to be more complex. Snakes inhabit multiple different environments with varying lifestyles and are thus the most diverse group of squamates.

Excluding fossorial blind snakes, visual snakes can be broadly divided into henophidian snakes (pythons and boas) which are mainly nocturnal and caenophidian snakes (colubrid snakes, elapids and vipers) which have diverse lifestyles and ecologies. The visual system of snakes is equally diverse, several species depend on vision for predation (Drummond, 1985) or have well-developed visual systems with binocular vision (Baker et al., 2007) while other species rely more on chemoreception, tactile cues, and/or vision (Herzog and Burghardt,

1974). Snake visual systems are unique and differ from squamates, possibly due to the snake ancestor undergoing a prolonged fossorial phase where visual systems likely degraded

(Walls, 1942). After re-establishing a terrestrial lifestyle, the eyes and visual system redeveloped from its reduced state, acquiring unique features compensating for losses

14 previously incurred (Walls, 1942). As such, snake eyes have lost ciliary muscles and therefore depend on the lens being moved back and forth to focus, in contrast to reptilian eyes (Figure 1.1) (Caprette et al., 2004). The photoreceptors are enlarged in the canonical duplex retina of nocturnal henophidian snakes (Walls, 1942). As diurnal lifestyles emerged, the retinae of colubrid snakes appear to have lost rods and developed all-cone retinas with double cone photoreceptor types (Walls, 1942). Walls hypothesized that the nocturnal colubrid snakes possessing all-rod or intermediate retina were derived traits while the all- cone retina of diurnal colubrids was the ancestral colubrid retinal arrangement (Figure 1.4).

The apparent all-cone retina of diurnal colubrids was hypothesized by Walls to be the result of transmutation. Retinal characterization of the diurnal colubrid Thamnophis sirtalis

(garter snake) and others revealed an apparent physiologically all-cone retina with double cones and large single cones expressing a long-wavelength sensitive pigment and two small single cone types, one with an ultraviolet (UV) pigment and another with a middle- wavelength at ~480nm which was hypothesized as either RH2 or RH1 expressed in a transmuted rod photoreceptor (Sillman et al., 1997). A study from our lab on the diurnal colubrid Thamnophis proximus demonstrated that a small cone subtype had rod ultrastructure with fully enclosed membrane discs in the outer segment (Schott et al., 2016). Genetic sequencing revealed genes for LWS, SWS1 and RH1 which produced a functional blue- shifted rhodopsin in vitro (Schott et al., 2016). Immunohistochemistry showed the presence of rhodopsin in the outer segments with downstream rod G protein in the same subset of photoreceptors (Schott et al., 2016). Together we concluded that transmutation had produced cone-like rods in the retina of T. proximus though presence of transmutation in other

15 colubrids as well as the functional significance of transmutation in diurnal colubrid snakes has yet to be established.

Figure 1.4 – Photoreceptor transmutation as hypothesized by Walls (1942). Adapted from Walls (1942)

1.4 - PHOTOTRANSDUCTION

Light is translated into a biological signal in a process called visual phototransduction

(summarized in (Chen, 2005; Lamb, 2013; Pugh and Lamb, 1993)) (Figure 1.5). In the dark, the photoreceptors remain depolarized and cyclic nucleotide gated (CNG) cation channels maintain a dark current of approximately -35 mV and results in the continuous release of the neurotransmitter glutamate in the dark (Gregg et al., 2013). Light is absorbed by the visual pigments contained in the outer segment discs of the photoreceptor cells (Lamb, 2013).

Visual pigments are comprised of an apoprotein opsin and a covalently-bound, light- sensitive, vitamin A-derived 11-cis chromophore. In vertebrates this can be the 11-cis retinal

(A1) chromophore or the 11-cis 3,4-dehydroretinal (A2) chromophore (Wald, 1939a). The light activated cis-trans isomerization initiates conformational changes in the opsin protein to the active state which can bind the G-protein transducin (Choe et al., 2011). Transducin is comprised of three subunits a, b, and g. Upon binding the active opsin, the a subunit exchanges bound GDP for GTP which causes the a subunit to separate from the b and g subunits. The disassociated a subunit activates the phosphodiesterase (PDE) by binding the inhibitory g -subunit of PDE, thus activating the a and b subunits of PDE to hydrolyze cGMP. As the cGMP levels decrease, the CNG channels close, resulting in hyperpolarization of the photoreceptor cell, and slowing of glutamate release into the synaptic cleft. This initiates a signal cascade in the retina that eventually results in transmission to the brain that light has been perceived. In order to reset to the dark state, the photoreceptor cell undergoes a deactivation pathway (see (Chen and Sampath, 2013) for detailed summaries) (Figure 1.5).

In the dark, G protein-coupled receptor kinase (GRK) is bound to recoverin. Upon light activation, a decrease in cytosolic Ca2+ results in the dissociation of recoverin from GRK.

GRK will then proceed to phosphorylate active state opsin which decreases the ability of the opsin to bind transducin. The phosphorylated opsin is then bound by arrestin which completely inhibits transducin binding. Upon binding of the G protein signaling (RGS9) complex, the intrinsic GTPase activity in transducin is enhanced resulting in hydrolysis of the transducin-bound GTP. Transducin then disassociates from the inhibitory PDE g-subunit which rebinds the a and b PDE subunits, arresting cGMP hydrolysis. Hyperpolarization of the light-activated photoreceptor results in the activation of guanylate cyclase activating proteins (GCAP). The GCAPs activate guanylate cyclase which results in increased intracellular cGMP concentrations. This change in cytosolic cGMP concentrations reopen the

CNG channels to restore the dark, depolarized state of the photoreceptors. With the cation channels open, the higher intracellular Ca2+ concentrations deactivate GCAPs, and cause recoverin to rebind GRK (Figure 1.5) (summarized in Stryer, 1986). For the photoreceptor to be ready to signal again, the opsin needs to be to be dephosphorylated, possibly by phosphatase 2A (PP2A) (Kolesnikov et al., 2017). The opsin dephosphorylation pathway is still being investigated though it has been shown that it occurs faster in cones than rods

(Yamaoka et al., 2015). In addition to dephosphorylation, a new 11-cis chromophore needs to enter and bind to the opsin, regenerating the dark state opsin (Hubbard et al., 1958)

While the phototransduction cascade is generally identical in both rods and cones, several components have separate rod and cone isoforms (summarized in (Chen and Sampath, 2013;

Kawamura and Tachibanaki, 2008; Korenbrot, 2012)). Changes in activity, location and expression contribute to the differences in rod and cone sensitivities and kinetics. Rods and cones have separate transducin proteins, PDE subunits, CNG subunits, arrestins, GCAPs, and

GRKs (Lamb, 2013). In photopic light conditions, rod cycle transducin and recoverin out,

18 and arrestin into the outer segment as a protective measure to prevent over activation of rods which can be damaging to the retina (Lobanova et al., 2010). Cycling of transducin is light- activated but does not require ATP (Rosenzweig et al., 2007). It is hypothesized that cycling is mediated by differential affinity for outer segment membranes. Cones do not cycle transducin unless light levels reach damaging levels, though cones do actively cycle arrestin into the outer segment under photopic light conditions (Kennedy et al., 2004). With regards to variable expression levels, the G protein deactivation complex RGS9 is expressed at higher levels in cone photoreceptors, and is thought to accelerate photoresponse recovery

(Cowan et al., 1998; Zhang et al., 2003). Activity differences can be observed in the GRK proteins as the cone opsin kinase (GRK7) shows higher activity levels in comparison to the rhodopsin kinase (GRK1) which is thought to be a function of the higher sensitivity of rods as slower phosphorylation allows for individual activated rhodopsins to activate more transducin proteins leading to greater signal amplification in rods (Tachibanaki et al., 2005;

Vogalis et al., 2011; Wada et al., 2006). Rod-cone isoforms which directly interact with the opsin pigment have been shown to have identical efficacies when coupled to the opposite opsin type, therefore most of the rod-cone kinetic differences are mediated by members of the phototransduction cascade (Kefalov et al., 2003).

Figure 1.5 – Phototransduction cascade in the photoreceptor cell. From Chen and Sampath (2013)

1.5 - OPSINS

To initiate a biological response to light, it has to be absorbed by visual pigments in the retina. These visual pigments, consisting of an opsin protein moiety bound to a light-sensitive chromophore which absorbs light maximally at a specific wavelength (λmax), are located in rod and cone outer segments (Chen and Sampath, 2013). Opsins are members of the G protein-coupled receptor family and possess seven transmembrane a-helical domains

(Bowmaker, 2008). There are five classes of vertebrate opsins: rhodopsin (RH1 - sensitive to dim light at between about 460-530nm), short-wavelength sensitive 1 (SWS1 - sensitive to violet-ultraviolet light between about 355-440nm), short-wavelength sensitive 2 (SWS2 - sensitive to blue-violet light between about 410-490nm), middle-wavelength sensitive (RH2 - sensitive to green from about 480-535nm), and long-wavelength sensitive (LWS - sensitive to red-green light from about 490-570nm) (Bowmaker, 2008) (figure 1.6). RH1 evolved from a cone opsins, likely RH2 (Okano et al., 1992). The complex evolutionary history of cone opsin duplication followed by specializations to diverse spectral environments is well characterized in Bowmaker and Hunt (2006) and Hunt (2009).

The vitamin A-derived chromophore is either 11-cis retinal (A1) or the more rare 11-cis

3,4-dehydroretinal (A2) (Wald, 1939a) in vertebrates. When the opsin apoprotein is covalently linked with A2 chromophore, it has historically been called porphyropsin (Wald,

1939b), though in this thesis I will be referring to A2 chromophore bound to rhodopsin apoprotein as A2 rhodopsin and A1 chromophore bound to rhodopsin apoprotein as either rhodopsin or A1 rhodopsin. Although light is absorbed directly by the chromophore, the surrounding protein environment is what determines the λmax of the opsin (Bowmaker,

2008; Bowmaker and Hunt, 2006). The λmax can be shifted by changing the chromophore;

21 the A1 pigment will have a bluer λmax when compared to the equivalent A2 pigment (Wald et al., 1953). By adjusting λmax, the vertebrate can adapt to its surrounding spectral environment. After light activation, phosphorylation and arrestin binding, the covalent bond linking the all-trans form of the chromophore to the Schiff base is hydrolyzed and diffuses out of the opsin. New 11-cis chromophore then enters the opsin and regenerates the protein which is dephosphorylated and ready to respond to light again. The work in this thesis is mainly focused on rhodopsin, thus the structural and functional properties of rhodopsin will be expanded upon in the following sections.

Figure 1.6 – Schematic of vertebrate complement of photoreceptors and visual pigments. SWS1, SWS2, RH2, and LWS cone opsins typically are expressed in cone photoreceptors while RH1 is typically expressed in rod photoreceptors. Adapted from figure made by Ryan K. Schott.

1.6 - RHODOPSIN

The visual pigment rhodopsin found in rods is responsible for dim scotopic light vision and has thus evolved incredible sensitivity to light. Multiple studies have shown the ability of rhodopsin to detect single photons of light (Baylor et al., 1979; Doan et al., 2006;

Imai et al., 2007) with ultrafast kinetics on the femtosecond timescale (Johnson et al., 2015).

To maximize visual sensitivity and limit visual noise, rhodopsin has evolved remarkable dark state stability to minimize thermal activation and the false detection of light (Gozem et al.,

2012; Guo et al., 2014; Liu et al., 2011a; Rieke and Baylor, 2000), in contrast to cone opsins characterized by higher rates of spontaneous activation (Kefalov et al., 2003). Additionally, rods demonstrate larger signal amplification due to the longer lived active rhodopsin which activates hundreds of transducin protein complexes per second (Pugh and Lamb, 1993).

Figure 1.7 – Snake plot of human rhodopsin. Figure created by Alex Van Nynatten

1.6.1 - RHODOPSIN STRUCTURE

Rhodopsin is a typical G protein-coupled receptor with seven transmembrane helices

(TM), a cytoplasmic helix 8 (H8), a disulfide bond between TM3 and extracellular loop 2

(EL2) (C110-C187), and several post-translational modifications (N-terminal glycosylation and C-terminal palmitoylation) (Karnik et al., 1988; Kaushal et al., 1994; Ovchinnikov et al.,

1988; Palczewski et al., 2000) (Figure 1.7). Unique to opsins is the N-terminal cap and EL2 forming an antiparallel four stranded beta-sheet which extends from the intradiscal environment into the TM helix bundle to comprise one of the walls of the chromophore binding pocket (Janz, 2003; Okada et al., 2002). Due to the specialized function of rhodopsin, there are several structural features in the dark and light state of rhodopsin involved in stabilizing and constraining the conformation states which remain conserved along with several critical water molecules involved in hydrogen bonding networks within the protein structure.

The tight chromophore binding pocket is formed by the EL2 and the TM helix bundle. It is stabilized by an extensive hydrogen bonding network surrounding the 11-cis chromophore covalently bonded via Schiff base linkage to lysine at site 296 in rhodopsin (Bownds, 1967).

The chromophore binding pocket has been highly studied as the electronic environment it forms can directly tune the spectral sensitivity of rhodopsin (Yokoyama et al., 2008) and is thus of special interest to visual ecologists (Bowmaker, 2008; Bowmaker and Hunt, 2006).

The conserved hydrogen bonding network in the chromophore binding pocket has been shown to stabilize both the dark state and active state conformations (Figure 1.8). This network is formed between the residues Y192, E181, S186, E113, several water molecules and the Schiff base linking the chromophore to K296 (Janz and Farrens, 2004). Disruptions

24 of the network blue-shift the λmax, and affect the stability of dark and active state rhodopsin

(Janz and Farrens, 2004; Liu et al., 2011b). The hydrogen bonding network extends through water molecules to a conserved motif, the NPxxY motif, which connects both the chromophore binding pocket to the cytoplasmic face of the protein.

Two highly conserved domains outside the chromophore binding pocket are the NPxxY motif and the E(D)RY motif which stabilize the active and inactive states of rhodopsin and in other GPCR proteins (Ernst et al., 2013; Hofmann et al., 2009) (Figure 1.8). The NPxxY

(fully identified as NPxxY9(x)5,6F) is located at the end of TM7 towards the cytoplasmic face of the protein with the F residue on H8 (site 302, 303, 306, and 313 on rhodopsin) and interacts and structurally constrains adjacent transmembrane helices (Fritze et al., 2003;

Hofmann et al., 2009; Okada et al., 2002). The NPxxY motif forms part of the TM1-TM2-

TM7 hydrogen bonding network through connections with site 299, 298, and water molecules to bond with N55 and D83. This network is thought to play a role in the stabilization of dark state rhodopsin and is subsequently rearranged during light activation and the formation of the active state (MII) (Fritze et al., 2003; Hofmann et al., 2009). The

E(D)RY motif located at the cytoplasmic end of TM3 (sites 134-136) and creates an “ionic lock” via hydrogen bonding with TM3 and TM6 to stabilize the dark state of rhodopsin

(Hofmann et al., 2009; Okada et al., 2002).

Another additional hydrogen bonding network involved in the stabilization of the dark state rhodopsin is the TM3-TM5 network comprised of E122 H-bonding with W126 and

H211 (Hofmann et al., 2009) (Figure 1.8). This particular network is of interest due to the close proximity of E122 to the beta-ionone ring of the chromophore such that retinal analogs with differing ring structures can disrupt E122 H-bonding (Vogel et al., 2005). It is thought

25 that this site also plays a role in the active state of rhodopsin (MII) with a restructuring of the

H-bonding with W126 and the rearranged E122-H211 interaction involved in the maintenance of the active state (Ahuja et al., 2009; Choe et al., 2011; Lin and Sakmar, 1996;

Patel et al., 2005).

Figure 1.8 – Rhodopsin structural motifs highlighted on the 3D structure (PDB: 1U19). NPxxY-F (red), E(D)RY (blue), E122-H211-W126 (green), E181 hydrogen bonding network (purple), retinal chromophore (black)

1.6.2 - RHODOPSIN ACTIVATION

After light absorption, rhodopsin shifts through multiple conformations in a complex, branched double square schematic mechanism (summarized by (Szundi et al., 2017)).

Briefly, after photoisomerization of the chromophore a red-shifted intermediate conformation bathorhodopsin forms within picoseconds (Peters et al., 1977; Schoenlein et al., 1991).

Bathorhodopsin enters into an equilibrium with a blue-shifted intermediate (BSI) likely due to the conformational change in chromophore beginning within 50ns (Hug et al., 1990; Lewis et al., 1990). The BSI is followed by an equilibrium between two lumirhodopsin intermediates (Lumi-I and Lumi-II) occurring within nanoseconds/microseconds (Szundi et

26 al., 2003). From here, the mechanism branches as Lumi-II can form a 480nm absorbing metarhodopsin I (MetaIa480) and a 380 nm metarhodopsin I with deprotonated Schiff bass

(Meta I380) (Szundi et al., 2016). These Meta I intermediates subsequently convert into a second metarhodopsin I (Meta Ib480) and two MetaII intermediates (MetaIIa/MetaIIb)

(Szundi et al., 2016). The MetaI/MetaII equilibrium can either decay directly to opsin and released all-trans retinal or via another intermediate metarhodopsin III (Meta III) which is similar to dark state (Heck et al., 2003; Kolesnikov et al., 2003; Lewis et al., 1997).

After light absorption by the 11-cis bond of the chromophore, the bond isomerizes resulting in the all-trans conformation of the chromophore which has an elongated structure in comparison to the 11-cis conformation (Ritter et al., 2004). During the isomerization, the hydrogen bonding network surrounding the Schiff base is rearranged and the Schiff base counter ion shifts from E113 to E181 in MetaI (Lüdeke et al., 2005). As MetaI shifts into

MetaIIa, a proton is displaced from the Schiff base to the counter ion E113 (Jäger et al.,

1997). The transmembrane helices TM5 and TM6 rotate and shift (Ye et al., 2010), disrupting and breaking the E122-H211-W126 interaction (Beck et al., 1998; Hofmann et al.,

2009). This rotation and shifting of transmembrane helices creates the G protein binding site leading to MetaIIb (Knierim et al., 2007). The ionic locks stabilizing the dark state of rhodopsin mediated by the NPxxY and E(D)RY motifs are broken and reform in a new conformation in MetaIIb to assist in the stabilization of the active state of rhodopsin

(Hofmann et al., 2009). The MetaI and MetaII states exist in an equilibrium which can be shifted by varying pH and/or temperature, with MetaII, which absorbs maximally at 380nm, favouring lower pH and higher temperature (Parkes et al., 1999). In the MetaIIb state the cytoplasmic face of rhodopsin is bound by the G protein transducin which initiates GDP/GTP

27 exchange in transducin (Choe et al., 2011). Structural studies of the MetaII state of rhodopsin bound to the C-terminal peptide of the transducin α-subunit shows that R135 and Q312 of rhodopsin interact with the G protein (Choe et al., 2011). Additional studies have suggested that a single transducin molecule can bind the rhodopsin dimer, possibly modulating the sensitivity of the photoresponse (Govardovskii et al., 2009), with evidence suggesting that a single transducin protein will complex with two rhodopsin molecules (Jastrzebska et al.,

2004; Jastrzebska et al., 2013; Jastrzebska et al., 2010).

A recent study investigated the behaviour of all-trans retinal and rhodopsin after activation (Schafer et al., 2016). After light activation, the all-trans retinal chromophore separates from the activated apoprotein via hydrolysis of the Schiff base linkage (Janz and

Farrens, 2004). The exit of the all-trans retinal from the chromophore binding pocket can be measured as the fluorescence of a tryptophan residue is unquenched as the chromophore diffuses out of the binding pocket (Janz and Farrens, 2004; Morrow and Chang, 2015).

Schafer (2016) showed that after exit of the chromophore, the apoprotein temporarily persists in an empty active state conformation before collapsing into the dark state conformation.

Additionally, the study demonstrated that released all-trans retinal can be rebound by any empty active state apoprotein and the two exist in an equilibrium state governed by the affinity of the empty apoprotein for the all-trans retinal (Schafer et al., 2016).

In addition to photoactivation, rhodopsin can also be activated via thermal energy.

Thermal activation leads to the false perception of light, called dark noise, and places a limit on visual sensitivity (Aho et al., 1988). In Aho (1988), the visual sensitivity of the toad Bufo bufo was assayed under different light levels and temperatures. The study showed that the threshold of light required for vision in the toad lowers as the temperature lowers, suggesting

28 that the rate of thermal activation determines the threshold for accurate detection of light, as hypothesized in earlier studies (Aho et al., 1988; Baylor et al., 1984). However, rhodopsin is very thermally stable, with one rhodopsin molecule thermally activating once every 420 years (Baylor et al., 1984; Lórenz-Fonfría et al., 2010). However, with billions of rhodopsin molecules per rod cell, a rod photoreceptor can thermally activate approximately once every two and a half minutes (Baylor et al., 1984). The precise mechanisms of thermal activation have been debated with studies showing that thermal activation predominantly occurs via spontaneous isomerization however, it can occur due to spontaneous Schiff base hydrolysis as well (del Valle et al., 2003; Janz, 2003; Liu et al., 2009), though the exact energetics, transition states, and mechanics are still debated. The Barlow correlation suggests a relationship between the λmax of the pigment and the thermal stability (Barlow, 1957), as red photons contain lower energy therefore the barrier for thermal activation would be lower in redshifted pigments, and recent studies have successfully and computationally demonstrated the Barlow correlation (Gozem et al., 2012; Luo et al., 2011). However, other studies have demonstrated the role of spontaneous structural fluctuation and a stabilizing hydrogen bonding network in thermal activation independent of λmax (Liu et al., 2013; Lórenz-Fonfría et al., 2010).

1.6.3 - RHODOPSIN VARIATION

Studies of rhodopsin natural variation allow researchers to isolate the relationship between the protein/chromophore and the specialized function. Indeed the dual nature of rhodopsin, a specialized function which requires conservation while also modifying properties to better suit visual environments, has enabled the use of multiple non-model rhodopsin in the characterization and isolation of rhodopsin and opsins (Aho et al., 1988;

Ala-Laurila et al., 2007; Bickelmann et al., 2012; Imai et al., 2005; Starace and Knox, 1997;

Sugawara et al., 2010). The most common functional variation of rhodopsin studied is the

λmax of rhodopsin. Shifts in the λmax of rhodopsin to match spectral environments have been best characterized in fish rhodopsins as the spectral environment of water can shift

(Miyagi et al., 2012; Spady et al., 2005; Van Nynatten et al., 2015) depending on depth or marine/freshwater turbidity. Minimal changes in the amino acid composition of rhodopsin can have an effect on the λmax (Davies et al., 2012; Hunt et al., 2009; Yokoyama et al.,

2008). The sculpins of Lake Baikal are a set of closely related fishes demonstrating bluer rhodopsins as the species of fish have deeper habitat depths where blue light dominates

(figure 1.9). For example, Procottus jettelesi which has a habitat range of 1-120m, has a redshifted rhodopsin λmax of 505nm (Hunt et al., 1996), while Abyssocottus korotneffi which has a habitat range of 400-1500m has a blue-shifted rhodopsin with a λmax of 484 nm (Hunt et al., 1996). Though recent computational modelling suggests that this λmax blue-shifting to match the bluer light with increased depth may also be an adaptation for thermal stability as the environment temperature drops (Luk et al., 2016). Blue-shifting rhodopsin to match blue aquatic environments can also be observed in marine mammals (Dungan et al., 2016; Fasick and Robsinson, 1998). In the above examples, spectral tuning to match visual environments has been accomplished via amino acid variations in the protein sequence of rhodopsin, however the chromophore also plays a role in the tuning of spectral sensitivities.

Figure 1.9 – Spectral sensitivity shifting with environment visual spectrum. Figure demonstrates the gradual blue-shift of opsin complement of Lake Baikal sculpins with the spectral shift to blue light of the environment with increasing depth. From Bowmaker and Hunt (2006).

1.7 - VERTEBRATE CHROMOPHORE

Among vertebrates, there are two chromophores. The most common is 11-cis retinal (A1) which is used by all classes of vertebrates. However, there is also a second, rarer, chromophore, 11-cis 3,4 dehydroretinal (A2), which was first named by George Wald (1936) used by some reptiles, amphibians and fishes. A2 pigment was first isolated from the retina of frogs and labelled “visual purple” (Wald, 1937), A2 chromophore (called retinene2) and

A2 pigments were then named and characterized as “porphyropsin” by Wald in 1939, with the spectral differences between A1 and A2 pigments allowing for the isolation and comparison of the two chromophore types.

Figure 1.10 - Molecular structures of the two vertebrate chromophores (A) 11-cis retinal, also called the A1 chromophore. (B) the 11-cis 3-dehydroretinal, also called the A2 chromophore

1.7.1 - THE A2 CHROMOPHORE

Varying chromophore usage is another method of tuning visual pigment sensitivity to the external surroundings (Temple et al., 2005). The A2 retinal differs from the A1 with the presence of an additional double bond in the beta-ionone ring, elongating the electron chain found along the length of the chromophore (Gillam, 1938) (figure 1.10). The A2 chromophore, in comparison to the A1 chromophore, causes the λmax of the opsin to red shift, while also increasing the spectral bandwidth of absorption (Harosi, 1994). There are multiple different use strategies for the A2 chromophore in nature. There are species that

32 exclusively use the A2 chromophore such as the pike cichlid (Weadick et al., 2012) or the

American chameleon (Kawamura & Yokoyama 1998). Some species utilize both chromophores together in varying ratios (Cowing et al. 2002), while others switch between the two chromophores depending on differing factors. Chromophore switching has been observed in fish during seasonal shifts or migration (Beatty 1966; Muntz & Mouat 1984).

While other organisms utilize different chromophores during different developmental or metamorphic stages (Cohen et al. 1990). The significance of the chromophore switch is assumed to be solely spectral tuning, as the A2 chromophore would red-shift the sensitivity of an opsin protein without having to modify the protein sequence. The A2 visual pigments have mainly been studied in vivo with microspectrophotometry (MSP) measurements of

λmax (Ala-Laurila et al. 2007), dietary replacement studies to induce chromophore switches

(Shantz & Embree 1946; Suzuki & Miyata 1988), or electroretinogram (ERG) studies measuring the signaling of A2 pigments (Ala-Laurila et al. 2003). The historical in vitro studies with A2 involved regenerating extracted bovine or chicken opsins with minimal amounts of A2 chromophore (Wald et al. 1953) demonstrating the ability of A1-exclusive rhodopsins to form a functional pigment with A2 chromophore. More recently, A2 has been synthesized in vitro and A2 pigments have started to be characterized with the correct chromophore (Terai et al. 2017; Miyagi et al. 2012). However most of the in vitro characterization of A2 pigments has been done using the A1 chromophore, under the assumption that the two chromophores are functionally almost equivalent (Kawamura &

Yokoyama 1998; Hauser, Ilves, Schott, Chang, et al. 2017). Therefore, many studies investigating A1-A2 pigment pairs employ a mathematical relationship to predict the λmax

33 shift between chromophores (Carleton et al. 2008; Bowmaker et al. 2005; Saarinen et al.

2012).

There have been several attempts at defining the mathematical relationship between A1 and A2 λmax in opsins (Dartnall & Lythgoe 1965; Whitmore & Bowmaker 1989; Harosi

1994; Parry & Bowmaker 2000). But most attempts thus far depend on opsins extracted directly from the eyes of animals that use both chromophore (Dartnall & Lythgoe 1965;

Whitmore & Bowmaker 1989; Harosi 1994). There are complications with this method, as these opsin extracts typically have a mixture of both chromophores present making it difficult to create pure and separate samples of one chromophore in one opsin class. These studies have revealed that the magnitude of red-shift caused by the switch to A2 is dependent on the A1 λmax. The longer the A1 λmax, the larger the amount of red-shift (Whitmore &

Bowmaker 1989). While the earliest attempts at modelling the λmax shift were linear

(Dartnall 1968), recent modelling attempts have accommodated the differential shift by creating exponential or logarithmic equations to predict the A2 λmax for the entire range from UV to long wavelength visual pigments (Harosi 1994; Parry & Bowmaker 2000). The accuracy of these models across the range of sensitivities is limited by a small dataset of non- rhodopsin A2 pigment measurements, and as such the predictive equations are used to approximate a range for the shift caused by chromophore switching.

Other than spectral λmax measurements, only minimal characterization of the non- spectral properties of A2 pigments has been conducted. Some non-spectral properties of the chromophore shift have been observed in vivo. A2 rhodopsin has been shown to have higher rates of spontaneous activation in the dark compared to the equivalent A1 rhodopsin (Ala-

Laurila et al. 2003), suggesting a lower visual sensitivity with A2 rhodopsins (Aho et al.

1988). This increase in spontaneous activation is thought to be a function of the red-shift, as a redder wavelength of light corresponds to a lower energy photon requiring a lowered barrier of activation, allowing for thermal energy to activate the A2 rhodopsin in the dark, leading to the false perception of light (Barlow 1957). The increase of thermal noise in A2 pigments demonstrates the importance of the role of the chromophore in non-spectral properties of rhodopsin. Little else has been characterized in A2 rhodopsin, therefore it is unknown if other aspects of A2 rhodopsin are functionally equivalent to A1 rhodopsins.

The chromophore interacts extensively with the apoprotein. Pure chromophore in ethanol absorb at 380nm (A1) and 393nm (A2) (Hubbard et al. 1971), however, depending on the protein environment around the chromophore the sensitivity can vary from the UV to the near infrared (Bowmaker & Hunt 2006). The elongated electronic chain of the A2 chromophore, along with a more rigid beta-ionone ring could potentially interact differently with rhodopsins sequence. It is currently unknown if spectral tuning sites or important functional sites characterized in A1 pigments interact equivalently with the A2 chromophore.

1.7.2 - RETINOID CYCLE

In the eye, the visual cycle allows for the recycling of light isomerized all-trans retinal back to 11-cis retinal. There are two such cycles in the eye, the retinal pigment epithelium

(RPE) mediated visual cycle which provides recycled and new 11-cis retinal to both photoreceptor types, and the cone-specific retinal visual cycle which is mediated by Müller glial cells (summarized in (Kiser and Palczewski, 2016)). The retinal cycle in Müller cell for cone photoreceptors is less well studied, but it is understood that the additional retinal cycle

35 is required to provide the large amount of 11-cis retinal utilized by cone photoreceptors in bright light (Tang et al., 2013; Wang et al., 2014).

Newly released all-trans retinal is first reduced to all-trans retinol then bound by interphotoreceptor retinoid binding protein (IRBP) and transported to the RPE. In the RPE, the all-trans retinol is transferred to another chaperone protein before it is esterified into all- trans retinyl ester by lecithin:retinol acyltransferase (LRAT). After esterification, the retinoid isomerohydrolase (RPE65) hydrolyzes and isomerizes the all-trans retinyl ester into 11-cis retinol. After transfer to another chaperone protein, it is oxidized back into 11-cis retinal which is now ready for transport back to the photoreceptor by IRBP. Though it has not been explicitly studied, it is assumed that the retinal cycle is identical the A2 chromophore.

Though Vitamin A2 (3,4-didehydroretinol), the precursor for the A2 chromophore, has been isolated in other organs (Gillam, 1938; Jensen et al., 1943), for visual purposes, Vitamin A2 is produced by the RPE from Vitamin A1 (retinol). Vitamin A2 synthesis is mediated by the cytochrome P450 family member Cyp27c1 (Enright et al., 2015), which has been shown to be an evolutionarily ancient mechanism found to mediate the A1-to-A2 chromophore shift in the sea lamprey which diverged from jawed vertebrates during the Cambrian period (~500

Mya) (Morshedian et al., 2017). The exact substrate and reaction involved in the synthesis of vitamin A2 by CYP27C1 has yet to be determined, with every intermediate of the retinal cycle available in the RPE, there are a number of potential substrates. Additionally, it is currently hypothesized that the differential levels of A2 chromophore in retina with distinct zones of differential spectral sensitivity mediated by A1:A2 chromophore ratios is established by differential expression of the CYP27C1 gene (Enright et al., 2015). Though no

A2 visual pigments have ever been discovered in mammalian and avian species, orthologs of

36 the CYP27C1 gene are present in both avian and mammalian genomes, including human, with the presence of 3,4-didehydroretinoids (derived from vitamin A2) detected in elevated

levels in human hyperkeratotic skin lesions and in chicken embryos (Enright et al., 2015).

Figure 1.11 - The retinal pigment epithelium (RPE) visual cycle. All-trans retinal (RAL) is bound by interphotoreceptor retinoid binding protein (IRBP) upon release from the opsin in the photoreceptor. RAL is reduced to all-trans retinol (ROL) by NADPH-dependent retinol dehydrogenase (RDH). The IRBP-ROL complex is transported to the RPE where ROL is transferred to cellular retinol binding protein (CRBP) and esterified by lecithin:retinol acyltransferase (LRAT) to all-trans retinyl ester (RE) followed by hydrolysis and isomerization by retinoid isomerohydrolase (RPE65) into 11-cis retinol (ROL). 11-cis ROL is then bound by cellular retinaldehyde binding protein (CRALBP) and oxidized by 11- cis retinol dehydrogenase (RDH) into 11-cis retinal (RAL). 11-cis RAL is transferred back to IRBP and transported back to the photoreceptor cell. Taken from Tang and Crouch (2013)

1.8 - RETINAL DISEASE

The retina is highly ordered and highly specialized. Disruptions to the retina due to mutation or external force can induce retinal degradation and eventual blindness

(summarized in (Gregg et al., 2013)). As a common example, increased intraocular pressure due to glaucoma induces apoptosis in the ganglion cells leading to degradation of the ganglion cell layer and decreased visual acuity (Gregg et al., 2013). Due to the scope of this thesis, I will be focusing on one particular retinal disease, retinitis pigmentosa. Retinitis pigmentosa (RP) encompasses several heritable, highly heterogeneous diseases that involve the degeneration of photoreceptor cells, with patients exhibiting symptoms such as night blindness followed by decreasing visual fields, and ultimately progressive visual impairment that can result in legal or complete blindness (Hartong et al., 2006). Retinitis pigmentosa affects approximately 1:3000 to 1:5000 people worldwide (Athanasiou et al., 2018). On the basis of inheritance, non-syndromic RP (i.e. symptoms are restricted to the eyes without other systemic impairments) can be divided into autosomal dominant (30-40% of cases), autosomal recessive (50-60%), and X-linked (5-15%) forms (Athanasiou et al., 2018).

1.8.1 - RETINITIS PIGMENTOSA

The clinical phenotype of RP varies considerably (described in detail in (Athanasiou et al., 2018; Hartong et al., 2006)). Age of onset can vary, with patients showing symptoms in early childhood while others remain asymptomatic well into adulthood (Hartong et al., 2006).

The classic progression of the disease begins with night blindness and difficulties with adaptation to the dark during adolescence, progressing to loss of mid-peripheral retinal sensitivity in early adulthood. The patient loses peripheral vision completely in adulthood,

38 developing tunnel vision, eventually losing central vision and developing complete blindness by approximately 60 years of age (Gregg et al., 2013). The order of photoreceptor degradation is variable among patients with some patients losing rod photoreceptors initially, others beginning with the loss of cone photoreceptors, while other patients can have both photoreceptor cell types degrade together. Additionally, some patients see a degeneration of colour vision as the retina degrades over time, while others do not.

Retinal degradation can be measured and assessed by ophthalmologists in multiple ways.

Images of the fundus can reveal the characteristic infiltration of the RPE in response to photoreceptor cell death, with pigment deposits in the retina which gave RP its name.

Measurements of visual acuity of the central visual field can reveal possible methods of inheritance as, depending on the basis of inheritance, patients may see an improvement of central vision with corrective lenses, though some may not. Goldmann visual field measurements reveal the visual sensitivity of the periphery of the retina, while colour vision can be assessed with Ishihara plates. The adaptability of the eye to darkness can be measured to assess the magnitude of night blindness. ERGs can assess electrical response of the retina to flashes of light, with signaling decreasing as the retina degrades. A non-invasive method of directly assessing retinal morphology is optical coherence tomography, which can measure the thickness of the retina and determine the status of the photoreceptor layer. Certain patients with RP have high concentrations of lipofuscin in the RPE, which autofluoresce and can be measured with fundus images as areas of high autofluorescence have been shown to produce the lowest ERG signaling.

Figure 1.12 – RP mutations in rhodopsin (A) Schematic of potential pathogenic mechanism of rhodopsin retinitis pigmentosa mutations (B) Snake plot of rhodopsin with retinitis pigmentosa mutation sites highlighted and classified. From Athanasiou et al. 2018

1.8.2 - DISEASE MUTATIONS IN RHODOPSIN

Mutations in at least 45 different genes have been linked to RP, where most cases are monogenic (Hartong et al., 2006). Mutations in visual genes encoding members of the

40 phototransduction cascade and the retinoid cycle along with genes encoding for proteins involved in the cytoskeleton, phagocytosis, or RNA splicing have all been associated with

RP (Hartong et al., 2006). Most of these genes account for only a small proportion of RP cases, with the exception of the rhodopsin gene. Rhodopsin is the best characterized RP gene to date and accounts for 20-30% of autosomal dominant and 1% of autosomal recessive RP cases (Hartong et al., 2006). Given the highly specialized nature of rhodopsin as an ultrasensitive photodetector, it is not surprising that a variety of mutations can lead to visual defects and degenerative visual disease in humans. To date, over 150 rhodopsin mutations have been associated with degenerative retinal disease (RetNet, www.sph.uth.tmc.edu/RetNet/) (figure 1.10 and table 1.1). Patients with RP mutations in rhodopsin develop a classic form of RP with rod-cone dystrophy where rod photoreceptor cell death leads to cone photoreceptor degradation and, ultimately, retinal degradation

(Athanasiou et al., 2018). While the majority of pathogenic mutations in rhodopsin lead to adRP, there are mutations leading to autosomal recessive RP and a few rare mutations leading to another diseased state, congenital stationary night blindness (CSNB).

CSNB patients are characterized by a complete lack of scotopic vision and the absence of rod function detectible by ERG, while cone functions appear to be normal, though in older patients retinal degradation with pigmentation on the retinal periphery have been observed

(Sieving, 1995), suggesting some phenotypic overlap with RP in older patients. Currently, there are only 5 known mutations in rhodopsin that lead to CSNB, and its currently thought that the majority of these mutations lead to constitutively active rhodopsin and subsequent rod cell death (Gross et al., 2003). Four mutations in rhodopsin have been linked to arRP

(Athanasiou et al., 2018). Two of these mutations lead to premature stop codons and in arRP

41 patients the rod photoreceptors degrade as the development of the outer segment of the photoreceptor is dependent on the presence of rhodopsin (Chen and Sampath, 2013). The method of pathogenicity of the other two arRP mutations is understudied. Mice models suggest that E150K may not be a true arRP mutations, with retinal degradation occurring in the heterozygote, though very mild (Zhang et al., 2012). The effect of M253I on rhodopsin structure and function have yet to be studied biochemically.

Table 1.1 - Based on Athanasiou et al. (2018), all known and predicted retinitis pigmentosa mutations in rhodopsin classified

Classification Description Mutations

Outer segment trafficking I L328P, T342M, Q344R/P/ter, V345L/M, defect A346P, P347A/R/Q/L/ST, ter349/Q/E

N15S, T17M, v20G, P23A/H/L, Q28H, G51R/V, P53R, T58R/M, V87D, G89D, G106R/W, C110F/R/S/Y, E113K, Folding defect, ER II L125R, W161R, A164E/V, C167R/W, retention and instability P171Q/L/S, Y178N/D/C, E181K, G182S/V, C185R, C187G/Y, G188R/E, D190N/G/Y, H211R/P, C222R, P267R/L, S270R, K296N/E/M

Endocytosis and vesicular III R135G/L/P/W traffic defect

Post-translational IV modification defect and T4K, T17M, M39R, N55K, 690V instability

Transducin activation V M44T, V137M defect

VI Constitutively active G90D, T94I, A292E, A295V

VII Dimerization defect F45L, V209M, F220C

P12R, R21C, Q28H, L40R, L46R, L47R, F52Y, F56Y, L57R, Y60TER, Q64TER, R69H, N78I, L79P, V87L, L88P, T92I, No observed defect/not T97I, V104F, G109R, G114D/V, E122G, Unclassified W126L/ter, S127F, L131P, Y136TER, characterized C140S, T160T, M163T, A169P, P170H/R, S176F, P180A/S, Q184P, S186P/W, Y191C, T193M, M207R/K, V210F, I214N, P215L/T, M216R/L/K,

R252P, T289P, S29YR, A298D, K311E, N315ter, E341K, S343C

Autosomal dominant RP currently has over 150 mutations in rhodopsin associated with the disease phenotype (Hartong et al., 2006) (Figure 1.12 and Table 1.1). These autosomal dominant mutations are thought to function in a dominant-negative or gain-of- function manner, rather than haploinsufficiency, as heterozygous null arRP patients and heterozygous rhodopsin knock-out mice do not show retinal degradation (Athanasiou et al.,

2018). There have been multiple attempts to group RP mutations in rhodopsin via disease phenotype (Cideciyan et al., 1998; Krebs et al., 2010) or via biochemical/cellular properties

(Kaushal and Khorana, 1994; Krebs et al., 2010; Mendes et al., 2005; Rakoczy et al., 2011;

Sung et al., 1991). Difficulties with clinical grouping (early onset and severe vs later onset and mild) were the presence of interfamily variability of phenotype with the same RP mutation as different alleles of other retinal proteins (such as RPE65) and environmental light exposure are modifiers of disease phenotype in patients (Iannaccone et al., 2006). The most recent attempt at grouping RP mutations cautions that the classifications are not mutually exclusive due to the multiple mechanisms of RP rhodopsin pathogenicity

(Athanasiou et al., 2018) (Table 1.1). The two largest classes are Class I mutations, which result in proteins that resemble wildtype in function but interfere with cellular trafficking to the rod outer segments due to defects in the C-terminal trafficking motif recognized by rod photoreceptors. Meanwhile Class II mutations result in misfolded proteins that are trapped in the ER or in aggresomes, and are thought to eventually result in photoreceptor cell death

(Athanasiou et al., 2018; Hartong et al., 2006). Subsequent classes of mutations do not necessarily affect folding, but interfere with additional aspects of rhodopsin structure and

43 function, such as posttranslational modification, disrupted endocytosis and vesicular trafficking, altered transducin activation, constitutive activity, dimerization defects and unclassified/uncharacterized (Athanasiou et al., 2018; Hartong et al., 2006).

For this thesis, I will be focusing on the Class II RP mutations (for detailed summaries of the classes see Athanasiou et al. (2018)). Class II RP mutations cause rhodopsin to misfold in some manner and be retained and accumulated in the ER causing ER stress and possible aggregation, leading to photoreceptor cell death. Of the Class II RP mutations, P23H is the best studied and is the most common RP mutation in North America, likely due to the founder effect (Farrar et al., 1990). Accumulation of P23H in the ER increases ER stress, which initiates the unfolded protein response (UPR) (Athanasiou et al., 2012; Kaushal and

Khorana, 1994; Kosmaoglou et al., 2009; Sung et al., 1991), and is mediated by the three membrane receptors IRE1 (inositol requiring enzyme 1), PERK (double-stranded RNA- activated protein kinase PKR-like ER kinase) and ATF6 (activating transcription factor 6)

(summarized in (Lin et al., 2007)). The UPR can also initiate apoptosis if the ER stress is too severe or prolonged. The three branches of the UPR vary in longevity. IRE1 signal is quickly attenuated despite the continued presence of ER stress, the ATF6 pathway attenuates after a delay, while the PERK pathway persists with prolonged ER stress (Lin et al., 2007). The

IRE1 pathway is one of the best studied, as it is the most conserved pathway across eukaryotes, including yeast. IRE1 is a transmembrane kinase/endoribonuclease that when activated initiates non-conventional splicing mRNA encoding the transcription factor Xbp-1, which upregulates the transcription of ER chaperones. ATF6 is a transcription factor that, under normal circumstances, is membrane bound in the ER where it interacts with BiP

(binding immunoglobulin protein). When misfolded protein is encountered, BiP separates

44 from ATF6 and ATF6 is then sent to the Golgi where it is cleaved from the membrane-bound portion and travels to the nucleus to upregulate ER-resident proteins involved in protein folding, including BiP (Shen et al., 2002). PERK is an ER-resident transmembrane kinase.

When it is activated by ER stress, PERK oligomerizes and phosphorylates translation initiating factors to reduce mRNA translation in the ER. This also causes the upregulation of other transcription factors such as CHOP, which controls genes encoding apoptosis (Chiang et al., 2012). Therefore, the PERK pathway can contribute signals to cell death pathways.

Misfolded rhodopsin is targeted for proteosomal degradation by the ubiquitin proteasome system (UPS), though if degradation fails or is overwhelmed, rhodopsin will form aggregates in vitro, although not in animal models. This suggests that misfolded rhodopsin aggregation is either rare or leads to cell death before detection. The dominant negative phenotype of adRP mutations can be demonstrated in vitro where cells coexpressing both wildtype and

P23H rhodopsin results in ER retention and intracellular aggregation of both rhodopsin proteins (Mendes and Cheetham, 2008), though even this property of adRP mutations has shown variation in vitro (Gragg et al., 2016).

1.8.3 - RESCUE AND TREATMENT METHODS OF RETINITIS PIGMENTOSA

In addition to exhibiting variation in disease severity and progression, patients with RP can also exhibit variable responses to treatment, as treatment efficacy is often dependent on the nature of the rhodopsin mutation. For example, delivery of neurotrophic factors is effective for properly folded RHO mutants that are inefficiently transported to rod outer segments due to C-terminus mutations (Sanftner et al. 2001). However, this treatment may be ineffective for misfolded mutants. Pharmacological rescue is based on the precept that small

45 cell permeable molecules (pharmacological chaperones) can be introduced during protein synthesis to stabilize the conformation of otherwise misfolded or unstable proteins and thereby promoting their proper transport to their site of action (Bernier et al. 2004).

Retinoids (i.e., vitamin A derivatives such as 11-cis retinal and 9-cis retinal) may serve as effective molecular chaperones when rhodopsin is misfolded. As the chromophore acts as an inverse agonist that stabilizes the dark state structure, its administration can decelerate retinal degeneration (Noorwez et al. 2004; Saliba et al. 2002; Krebs et al. 2010). This strategy has been shown to be successful for a number of different GPCR conformational mutants, enhancing their expression at the plasma membrane, and restoring their abilities to bind ligand and activate cellular responses (Janovick et al. 2009; Generoso et al. 2015). This approach has also shown much promise for RP mutants of rhodopsin. In vitro studies using the native chromophore 11-cis retinal resulted in a dramatic rescue not only of the well- studied P23H mutant (Noorwez et al. 2004), but also RP mutants at other sites, with some variation in degree of effect (Krebs et al. 2010; Opefi et al. 2013). However, certain therapeutic strategies (metformin, sodium 4-phenylbutyrate) can also indirectly cause more degradation, as successfully transporting defective protein to the outer segment of photoreceptors can lead to the accelerated degradation of the photoreceptor (Athanasiou et al.

2017), emphasizing the need to rescue function along with folding. Compounds inhibiting the

UPR and/or apoptosis have also shown to prevent retinal degeneration and vision loss in some animal models of RP, which specifically activate the branch of the UPR being suppressed by the compound (Fernandez-Sanches 2017).

Currently the only prescribed treatment for RP patients is supplementation with Vitamin

A palmitate with docosahexaenoic acid (DHA) and lutein (Athanasiou et al., 2018). The

46 efficacy of this treatment is currently being debated as studies showing the protective effects of supplementation (Berson et al., 2012; Berson et al., 2010) all have the major flaw of not being performed on genotyped individuals. With the causal gene and mutation unknown in the patients included in the study, an effective supplementation may be ineffective, or worse detrimental, to individual patients with specific mutations in certain genes. Additionally, the high doses of vitamin A can cause liver damage and severe birth defects if taken during pregnancy. In vitro studies mentioned above have shown the variable effect of rescue with

11-cis retinal depending on RP mutation (for example (Opefi et al., 2013)). In patients, a second consideration is the effect of genetic modifiers on rescue methods. For example, treatment of a patient with a Rpe65 polymorphism which provides higher 11-cis retinal in addition to a rescuable RP mutation with excess vitamin A, could lead to toxic levels of 11- cis retinal and lead to photoreceptor death. Additional concerns with the treatment of high levels of vitamin A is that the large amounts of 11-cis retinal in the eye could, upon light bleaching, produce large amounts of the toxic all-trans retinal, again leading to additional damage to the photoreceptor (Chen et al., 2012). Therefore, RP is a genetic disease that would greatly benefit from personalized medicine which would take into consideration causal gene, disease mutation, and other genetic modifiers when developing a therapeutic strategy. Multiple therapies at the transcription/translation level have been attempted in animal models of RP, with RNA interference and gene therapy both suppressing the expression of the mutant rhodopsin and overexpressing wildtype rhodopsin (summarized in

(Athanasiou et al., 2018)). With CRISPR/Cas9 technology becoming the forefront of gene editing technologies, recent studies have successfully decreased mutant rhodopsin expression in the retina of animal models of RP with CRISPR, though the low electroporation efficiency

47 of the retina make complete rescues difficult (summarized in (Athanasiou et al., 2018)).

Additional difficulties with CRISPR/Cas9 targeting of adRP mutations is the need for allele specificity, as the inadvertent removal of the wildtype rhodopsin gene would greatly increase the severity of adRP phenotypes.

1.8.4 - RETINITIS PIGMENTOSA MUTATIONS SITES IN RHODOPSIN

RP mutations in rhodopsin can be found along the entire length of the protein. The N- terminus of rhodopsin is 35 amino acids long and contains 7 known RP sites (Athanasiou et al., 2018; Okada et al., 2002). RP patients with mutations in the N-terminus usually demonstrate a relatively mild disease with retinal degradation developing and progressing later in life (Cideciyan et al., 1998), though this is not universal for all N-terminal RP mutations. Mutations in the N-terminal are also associated with “sector RP”, where areas of the retina exposed to light degrade faster. Sector RP is thought to be caused by two possible pathways, either the RP mutation in rhodopsin does not affect misfolding but causes dysfunction upon light activation or the light exposure to the retina depletes 11-cis retinal which, as an inverse agonist, could otherwise be stabilizing the nascent mutant rhodopsin structure (Athanasiou et al., 2018). Folding of certain N-terminal RP mutations in rhodopsin can be rescued either by pharmacological rescue or by structural rescue with stabilizing mutations (Opefi et al., 2013), though function and stability of these rescued rhodopsins still show dysfunction. RP mutations can be found on every transmembrane helix, some mutations may be disrupting helix packing, or dimerization interfaces, or break critical functional motifs such as the E(D)RY motif or the E122-W126-H211 H-bonding or the chromophore binding site and counter ion. Mutations on the C-terminal of rhodopsin mainly

48 disrupt trafficking of rhodopsin to the outer segment of rod photoreceptors (for more detail see (Athanasiou et al., 2018; Hartong et al., 2006)).

RP mutations are found on 4 of the 6 intrahelical loops of rhodopsin, with the majority located on the extracellular loop 2 of rhodopsin. The extracellular loop 2 of rhodopsin is only

27 amino acids long but is involved in several interactions critical to the proper structure and function of rhodopsin. This can be seen by the presence of known RP sites found along the entire length of this loop. Starting at the extracellular end of transmembrane helix 4, the EL2 runs parallel to the membrane surface. RP mutations are known at site 174 (Fujiki et al.

1995), 176 (Li et al. 2010), and 178 (Sung et al. 1991) leading towards the beta3 strand, of these three mutants, only Y178C has been characterized in vitro (Kaushal and Khorana 1994) showing a failure of the mutant protein to bind to 11-cis retinal. The beta3 strand of the anti- parallel strand is made up of F177 to E182 with RP sites found at 179 (I179F Grondahl et al.

2007) and 181. Site 181 is also a critical site in rhodopsin structure function as it is the Schiff base counter ion in Meta I (Yan et al. 2003) and is part of the critical hydrogen bond network

(Janz and Farrens 2004). Leading out from the beta3 strand are the two RP sites of interest in this study, P180 and G182. The EL2 now forms a tight U-turn away from helix 7 and forms the beta4 strand underneath beta3, which makes up part of the chromophore binding pocket.

Along this stretch of amino acids are several critical to function sites: RP site Q184 (Dryja et al. 2000), C185 (Sohocki et al. 2001) which is involved in the pathogenesis of P23H RP

(McKibbin et al. 2007), and S186 (marking the start of beta4 (Dryja et al. 1991)), which is involved in the hydrogen bond network and the thermal stability of the protein (Liu et al.

2013). The cysteine at site 187 forms a disulfide bond with C110 (Karnik and Khorana 1990) and is also an RP site (Hwa et al. 1999). Additional RP sites in EL2 are G188 (Dryja et al.

1991; Iannaccone et al. 2006), D190 (Dryja et al. 1991) which is part of a conserved ion pair with R177, critical to dark state stability (Janz and Farrens 2003) and is found mutated in patients with night blindness (Cidecyan et al., 1998), and lastly, T193 (Cideciyan et al. 1998).

1.9 - THESIS OBJECTIVES

The primary goal of this thesis is to investigate and characterize natural variation in rhodopsin in the context of evolutionary adaptation, function, and pathogenic phenotype, using tissue culture, microscopic and spectroscopic techniques in combination with mutagenesis and in vitro expression assessing rhodopsin localization and functionality. This thesis highlights the use of wildtype and pathogenic rhodopsin to uncover novel aspects of rhodopsin structure and function.

The specific objectives of this thesis are:

1. Investigate the presence of rhodopsin in the all-cone retina of the diurnal colubrid

Pituophis melanoleucus and confirm the presence of rod-to-cone transmutation

2. Characterize the opsin complement of P. melanoleucus with in vitro expression and spectroscopy to investigate the possible functional significance of rod-to-cone transmutation in diurnal colubrid snakes

3. Measure and characterize the λmax red-shift caused by the switch from A1 to A2 chromophore in the rhodopsin of multiple vertebrate species using in vitro expression and absorbance spectroscopy

4. Characterize differences in the role of the chromophore in the active state decay of

A1 and A2 rhodopsins using fluorescence spectroscopy

5. Determine the activation energy for the light activated Schiff base hydrolysis in A1 and A2 bovine rhodopsin.

6. Characterize novel pathogenic mutations in rhodopsin using immunocytochemistry to assess intracellular trafficking and spectroscopy to quantify function of mutated rhodopsin

7. Determine the magnitude of response of retinitis pigmentosa (RP) mutations to pharmacological rescue by 11-cis retinal and structural rescue by stabilizing mutations

8. Utilize RP mutations P180L and G182V in the extracellular loop 2 to highlight the role of the beta3 sheet in rhodopsin folding and chromophore binding pocket structure

9. To assess dysfunction of the RP mutations A164V, A164E, and V81del in vitro with microscopy and spectroscopy and demonstrate the relationship between the in vitro phenotype and the clinical phenotype.

1.10 - THESIS OVERVIEW

Over the last several decades, the intricacies of rhodopsin structure and function have been extensively investigated and have revealed an exquisitely specialized light sensor.

Rhodopsin has shown incredible sensitivity, with the ability to detect mere photons of light with the covalently bound 11-cis retinal isomerizing and initiating conformational changes at ultrafast rates. However, much of this research has assumed rhodopsin as a monolith with universal function among all vertebrates. With the diversity of visual environments and ecological niches and the intense specializations of rhodopsin as a protein, the question of a possible gradient of rhodopsin functionality emerges. The extensive clinical work identifying numerous of mutations in rhodopsin causing dysfunction shows that rhodopsin has limited plasticity in protein sequence, though recent work in the characterizing of non-model organism rhodopsins (Bickelmann et al., 2012; Castiglione et al., 2017; Dungan et al., 2016;

Morrow and Chang, 2015) suggests that rhodopsin can accommodate a range in function that enables further insights into rhodopsin. This thesis contributes to expanding the known diversity of rhodopsin functionality by characterizing natural rhodopsin variation in an evolutionary, function and disease context.

In Chapter II, I characterize rhodopsin from a diurnal colubrid snake Pituophis melanoleucus which presents with a superficially all-cone retina. I use microscopy techniques to first demonstrate the presence of rhodopsin and rod machinery in the all-cone retina and conclude that the retinae of P. melanoleucus had undergone rod-to-cone transmutation. As the functional significance of transmutation in colubrid snakes is still unknown, I expressed and characterized the entire opsin complement (LWS, SWS1, and

RH1) and showed that all opsin genes produced functional visual pigments when expressed

53 in vitro. Consistent with other studies of colubrid snakes, I found that P. melanoleucus rhodopsin is extremely blue-shifted for a terrestrial vertebrate. Using spectroscopic assays, I showed that P. melanoleucus rhodopsin displayed typical cone opsin characteristics and suggest that transmutation may be an adaptation for diurnal, brighter-light vision, which could result in increased spectral sensitivity and chromatic discrimination with the potential for colour vision. The results of this chapter were published in Journal of Experimental

Biology: Bhattacharyya N, Darren B, Schott RK, Tropepe V, and Chang BSW (2017) Cone- like rhodopsin expressed in the all-cone retina of the colubrid pine snake as a potential adaptation to diurnality. The Journal of Experimental Biology, 220(13), pp.2418–2425.

With the previous chapter characterizing variation in rhodopsin protein, Chapter III aims to characterize variation in the covalently bound chromophore of rhodopsin. I expressed and purified 11 vertebrate rhodopsins in vitro and regenerated each with the common 11-cis retinal (A1) chromophore and the less abundant 11-cis 3,4 dehydroretinal (A2) chromophore.

Using UV-visible absorbance spectroscopy, I measured the λmax for each rhodopsin pair and found a relatively consistent λmax red-shift of 17-20 nm upon regeneration with A2 chromophore. With this data, I modelled a mathematical relationship to predict the red-shift of A2 chromophore though the outliers revealed the role of the protein in the determination of the magnitude of red-shift. Additionally, I used fluorescent spectroscopy to functionally characterize the decay of the active state in A1 and A2 rhodopsin and found that light- activated A2 rhodopsin decayed faster than the equivalent A1 rhodopsin, suggesting a shorter lived active state in A2 rhodopsins. Arrhenius plots revealed that the activation energy mediating the light-activated hydrolysis of the Schiff-base linkage is similar in both A1 and

A2 pigments, suggesting that the accelerated retinal release may be due to a lower affinity of

54 the all-trans retinal in A2 rhodopsin. This study represents the first comprehensive characterization of vertebrate rhodopsin proteins with the A2 chromophore and highlights the additional non-spectral effects of chromophore choice.

In this chapter, I investigated disease-causing mutations in rhodopsin to reveal mechanisms of rhodopsin dysfunction, as well as discovering new areas of rhodopsin critical to structure-function. In Chapter IV, I characterized two novel retinitis pigmentosa mutations found on the extracellular loop 2 of rhodopsin (P180L and G182V), and the response to pharmacological and compensatory mutations. Using mutagenesis and heterologous expression, I assessed the intracellular trafficking with fluorescent confocal microscopy. Both mutations demonstrated a severely deleterious phenotype, consistent with the clinical phenotype, resulting in minimal folded and functional protein sequestered in the

ER. Spectroscopic assays of purified mutant rhodopsin showed little to no response to pharmacological rescue with 11-cis retinal, however P180L is rescued when expressed with stabilizing mutations (N2C/D282C), while G182V requires both the stability mutant background and pharmacological rescue in order to produce properly trafficked and functional rhodopsin. The variation in results suggested that the mutant residues disrupted the protein structure in differing manners. These results illustrate the variability of RP mutant phenotypes and their response to rescue and emphasize the need for individual characterization of potential RP mutations in rhodopsin. This chapter highlights the use of pathogenic disruptive mutations to isolate areas of rhodopsin structure critical to function.

Finally, I investigated not only RP mutations in rhodopsin, but to also the relationship with disease phenotype. In Chapter V, I compared the in vitro effects of RP mutations on rhodopsin and the clinical phenotype to determine if in vitro characterizations of RP

55 mutations could be of use to clinicians in the determination of therapeutic strategies. I used microscopy and spectroscopy to characterize three novel and rare RP mutations in rhodopsin

(A164E, A164V, and V81del). Using confocal fluorescent microscopy, I assessed protein folding and trafficking intracellularly and then assayed purified mutant protein spectroscopically to determine expression levels, functionality, and response to functional rescue with 11-cis retinal. I then compared the in vitro results to clinical data evaluating retinal degradation in the corresponding RP patients. Our results showed that all three mutations exhibited differing levels of severity and response to pharmacological rescue (least severe to most severe: A164V < A164E < V81del) which was recapitulated in the patient data assessing degradation of the retina. This chapter demonstrates the validity of applying the results of in vitro techniques characterizing the variation of RP rhodopsin mutations to clinical interpretations of patient mutations and could help clinicians to create an effective therapeutic strategy, as not all RP mutations would respond to certain rescue techniques.

This thesis contributes to increasing the understanding of functional variation within rhodopsin. By characterizing rhodopsin variation found within differing species, chromophores and disease mutations, it has expanded the properties and functions attributed to rhodopsin. I hope that the results of this thesis will encourage further in-depth investigation into non-model, non-mammalian, and non-wildtype protein function in biology in general, as evolutionary, ecological and pathogenic variation can provide future researchers with a greater understanding about protein structure and function.

1.11 – REFERENCES Aho, A. C., Donner, K., Hydén, C., Larsen, L. O. and Reuter, T. (1988). Low retinal

noise in animals with low body temperature allows high visual sensitivity. Nature 334,

348–350.

Ahuja, S., Crocker, E., Eilers, M., Hornak, V., Hirshfeld, A., Ziliox, M., Syrett, N.,

Reeves, P. J., Khorana, H. G., Sheves, M., et al. (2009). Location of the retinal

chromophore in the activated state of rhodopsin. J Biol Chem 284, 10190–10201.

Ala-Laurila, P., Donner, K., Crouch, R. K. and Cornwall, M. C. (2007). Chromophore

switch from 11-cis-dehydroretinal (A2) to 11-cis-retinal (A1) decreases dark noise in

salamander red rods. J. Physiol. (Lond.) 585, 57–74.

Athanasiou, D., Aguilà, M., Bellingham, J., Li, W., McCulley, C., Reeves, P. J. and

Cheetham, M. E. (2018). The molecular and cellular basis of rhodopsin retinitis

pigmentosa reveals potential strategies for therapy. Prog Retin Eye Res 62, 1–23.

Athanasiou, D., Kosmaoglou, M., Kanuga, N., Novoselov, S. S., Paton, A. W., Paton, J.

C., Chapple, J. P. and Cheetham, M. E. (2012). BiP prevents rod opsin aggregation.

Mol Biol Cell 23, 3522–3531.

Baker, R. A., Gawne, T. J., Loop, M. S. and Pullman, S. (2007). Visual acuity of the

midland banded water snake estimated from evoked telencephalic potentials. J. Comp.

Physiol. A Neuroethol. Sens. Neural. Behav. Physiol. 193, 865–870.

Barlow, H. B. (1957). Noise and the Visual Threshold. Nature 180, 1405–1405.

Baylor, D. A., Lamb, T. D. and Yau, K. W. (1979). Responses of retinal rods to single

photons. J. Physiol. (Lond.) 288, 613–634.

Baylor, D. A., Nunn, B. J. and Schnapf, J. L. (1984). The photocurrent, noise and spectral

sensitivity of rods of the monkey Macaca fascicularis. J. Physiol. (Lond.) 357, 575–607.

Beck, M., Siebert, F. and Sakmar, T. P. (1998). Evidence for the specific interaction of a

lipid molecule with rhodopsin which is altered in the transition to the active state

metarhodopsin II. FEBS Lett. 436, 304–308.

Berson, E. L., Rosner, B., Sandberg, M. A., Weigel-DiFranco, C. and Willett, W. C.

(2012). ω-3 Intake and Visual Acuity in Patients With Retinitis Pigmentosa Receiving

Vitamin A. Arch. Ophthalmol. 130, 707–711.

Berson, E. L., Rosner, B., Sandberg, M. A., Weigel-DiFranco, C., Brockhurst, R. J.,

Hayes, K. C., Johnson, E. J., Anderson, E. J., Johnson, C. A., Gaudio, A. R., et al.

(2010). Clinical trial of lutein in patients with retinitis pigmentosa receiving vitamin A.

Arch. Ophthalmol. 128, 403–411.

Bickelmann, C., Morrow, J. M., Müller, J. and Chang, B. S. (2012). Functional

characterization of the rod visual pigment of the echidna (Tachyglossus aculeatus), a

basal mammal. Vis. Neurosci. 29, 1–7.

Bowmaker, J. K. (2008). Evolution of vertebrate visual pigments. Vision Research 48,

2022–2041.

Bowmaker, J. K. and Hunt, D. M. (2006). Evolution of vertebrate visual pigments. Curr.

Biol. 16, R484–9.

Bownds, D. (1967). Site of attachment of retinal in rhodopsin. Nature 216, 1178–1181.

Caprette, C. L. (2005). Conquering the Cold Shudder: The Origin and Evolution of Snake

Eyes. 1–122.

Caprette, C. L., Lee, M. S. Y., Shine, R., Mokany, A. and Downhower, J. F. (2004). The

origin of snakes (Serpentes) as seen through eye anatomy. Biological Journal of the

Linnean Society 81, 469–482.

Castiglione, G. M., Hauser, F. E., Liao, B. S., Lujan, N. K., Van Nynatten, A., Morrow,

J. M., Schott, R. K., Bhattacharyya, N., Dungan, S. Z. and Chang, B. S. (2017).

Evolution of nonspectral rhodopsin function at high altitudes. Proc Natl Acad Sci USA

114, 7385–7390.

Chen, C. K. (2005). The vertebrate phototransduction cascade: amplification and termination

mechanisms. Rev Physiol Biochem Pharmacol 155, 101–121.

Chen, J. and Sampath, A. P. (2013). Chapter 14 - Structure and Function of Rod and Cone

Photoreceptors. Fifth Edition. Elsevier Inc.

Chen, Y., Okano, K., Maeda, T., Chauhan, V., Golczak, M., Maeda, A. and Palczewski,

K. (2012). Mechanism of all-trans-retinal toxicity with implications for stargardt disease

and age-related macular degeneration. Journal of Biological Chemistry 287, 5059–5069.

Chiang, W. C., Hiramatsu, N., Messah, C., Kroeger, H. and Lin, J. H. (2012). Selective

Activation of ATF6 and PERK Endoplasmic Reticulum Stress Signaling Pathways

Prevent Mutant Rhodopsin Accumulation. Investigative Ophthalmology & Visual

Science 53, 7159–7166.

Choe, H.-W., Kim, Y. J., Park, J. H., Morizumi, T., Pai, E. F., Krauss, N., Hofmann, K.

P., Scheerer, P. and Ernst, O. P. (2011). Crystal structure of metarhodopsin II. Nature

471, 651–655.

Cideciyan, A. V., Hood, D. C., Huang, Y., Banin, E., Li, Z. Y., Stone, E. M., Milam, A.

H. and Jacobson, S. G. (1998). Disease sequence from mutant rhodopsin allele to rod

and cone photoreceptor degeneration in man. Proc. Natl. Acad. Sci. U.S.A. 95, 7103–

7108.

Cowan, C. W., Fariss, R. N., Sokal, I., Palczewski, K. and Wensel, T. G. (1998). High

expression levels in cones of RGS9, the predominant GTPase accelerating protein of

rods. Proc. Natl. Acad. Sci. U.S.A. 95, 5351–5356.

Davies, W. I. L., Collin, S. P. and Hunt, D. M. (2012). Molecular ecology and adaptation

of visual photopigments in craniates. Molecular Ecology 21, 3121–3158. del Valle, L. J., Ramon, E., Cañavate, X., Dias, P. and Garriga, P. (2003). Zinc-induced

decrease of the thermal stability and regeneration of rhodopsin. J Biol Chem 278, 4719–

4724.

DelMonte, D. W. and Kim, T. (2011). Anatomy and physiology of the cornea. Journal of

Cartaract & Refractive Surgery 37, 588–598.

Doan, T., Mendez, A., Detwiler, P. B., Chen, J. and Rieke, F. (2006). Multiple

phosphorylation sites confer reproducibility of the rod's single-photon responses. Science

313, 530–533.

Dowling, J. E. (2009). Retina: An overview. In Encyclopedia of Neuroscience (ed. LR, S.,

pp. 159–169. Oxford: Academic Press.

Drummond, H. (1985). The role of vision in the predatory behaviour of natricine snakes.

Animal Behaviour 33, 206–215.

Dungan, S. Z., Kosyakov, A. and Chang, B. S. (2016). Spectral Tuning of Killer Whale

(Orcinus orca) Rhodopsin: Evidence for Positive Selection and Functional Adaptation in

a Cetacean Visual Pigment. Mol. Biol. Evol. 33, 323–336.

Ebrey, T. and Koutalos, Y. (2001). Vertebrate photoreceptors. Prog Retin Eye Res 20, 49–

94.

Enright, J. M., Toomey, M. B., Sato, S.-Y., Temple, S. E., Allen, J. R., Fujiwara, R.,

Kramlinger, V. M., Nagy, L. D., Johnson, K. M., Xiao, Y., et al. (2015). Cyp27c1

Red-Shifts the Spectral Sensitivity of Photoreceptors by Converting Vitamin A1 into A2.

Curr. Biol. 25, 3048–3057.

Ernst, O. P., Lodowski, D. T., Elstner, M., Hegemann, P., Brown, L. S. and Kandori, H.

(2013). Microbial and Animal Rhodopsins: Structures, Functions, and Molecular

Mechanisms. Chem. Rev. 114, 126–163.

Fain, G. L., Hardie, R. and Laughlin, S. B. (2010). Phototransduction and the Evolution

Review of Photoreceptors. Curr. Biol. 20, R114–R124.

Farrar, G. J., Kenna, P., Redmond, R., McWilliam, P., Bradley, D. G., Humphries, M.

M., Sharp, E. M., Inglehearn, C. F., Bashir, R. and Jay, M. (1990). Autosomal

dominant retinitis pigmentosa: absence of the rhodopsin proline-histidine substitution

(codon 23) in pedigrees from Europe. Am. J. Hum. Genet. 47, 941–945.

Fasick, J. I. and Robsinson, P. R. (1998). Mechanism of spectral tuning in the dolphin

visual pigments. Biochemistry 37, 433–438.

Fernald, R. D. (1988). Aquatic Adaptations in Fish Eyes. In Sensory Biology of Aquatic

Animals, pp. 435–466. New York, NY: Springer New York.

Fritze, O., Filipek, S., Kuksa, V., Palczewski, K., Hofmann, K. P. and Ernst, O. P.

(2003). Role of the conserved NPxxY(x)5,6F motif in the rhodopsin ground state and

during activation. Proc. Natl. Acad. Sci. U.S.A. 100, 2290–2295.

Gillam, A. E. (1938). The vitamin A1 and A2 contents of mammalian and other animal

livers. Biochemical Journal 32, 1496–1500.

Govardovskii, V. I., Korenyak, D. A., Shukolyukov, S. A. and Zueva, L. V. (2009).

Lateral diffusion of rhodopsin in photoreceptor membrane: a reappraisal. Mol. Vis. 15,

1717–1729.

Gozem, S., Schapiro, I., Ferré, N. and Olivucci, M. (2012). The molecular mechanism of

thermal noise in rod photoreceptors. Science 337, 1225–1228.

Gragg, M., Kim, T. G., Howell, S. and Park, P. S. H. (2016). Wild-type opsin does not

aggregate with a misfolded opsin mutant. BBA - Biomembranes 1858, 1850–1859.

Gregg, R. G., McCall, M. A. and Massey, S. C. (2013). Chapter 15 - Function and

Anatomy of the Mammalian Retina. Fifth Edition. Elsevier Inc.

Gross, A. K., Rao, V. R. and Oprian, D. D. (2003). Characterization of Rhodopsin

Congenital Night Blindness Mutant T94I. Biochemistry 42, 2009–2015.

Guo, Y., Sekharan, S., Liu, J., Batista, V. S., Tully, J. C. and Yan, E. C. Y. (2014).

Unusual kinetics of thermal decay of dim-light photoreceptors in vertebrate vision. Proc

Natl Acad Sci USA 111, 10438–10443.

Hall, M. I. and Ross, C. F. (2006). Eye shape and activity pattern in birds. Journal of

Zoology 271, 437–444.

Harosi, F. I. (1994). An analysis of two spectral properties of vertebrate visual pigments.

Vision Research 34, 1359–1367.

Harosi, F. I. and Novales Flamarique, I. (2012). Functional significance of the taper of

vertebrate cone photoreceptors. J. Gen. Physiol. 139, 159–187.

Hartong, D. T., Berson, E. L. and Dryja, T. P. (2006). Retinitis pigmentosa. Lancet 368,

1795–1809.

Heck, M., Schadel, S. A., Maretzki, D., Bartl, F. J., Ritter, E., Palczewski, K. and

Hofmann, K. P. (2003). Signaling States of Rhodopsin: Formation of the Storage Form,

Metarhodopsin III, From Active Metarhodopsin II. Journal of Biological Chemistry 278,

3162–3169.

Herzog, H. A., Jr and Burghardt, G. M. (1974). Prey Movement and Predatory Behavior

of Juvenile Western Yellow-Bellied Racers, Coluber constrictor mormon. Herpetologica.

Hetling, J. R., Baig-Silva, M. S., Comer, C. M., Pardue, M. T., Samaan, D. Y., Qtaishat,

N. M., Pepperberg, D. R. and Park, T. J. (2005). Features of visual function in the

naked mole-rat Heterocephalus glaber. J. Comp. Physiol. A Neuroethol. Sens. Neural.

Behav. Physiol. 191, 317–330.

Hofmann, K. P., Scheerer, P., Hildebrand, P. W., Choe, H.-W., Park, J. H., Heck, M.

and Ernst, O. P. (2009). A G protein-coupled receptor at work: the rhodopsin model.

Trends Biochem Sci 34, 540–552.

Hubbard, R. & Kropf, A. (1958). The Action of Light on Rhodopsin. Proceedings of the

National Academy of Sciences of the United States of America, 44(2), pp.130–139.

Hug, S. J., Lewis, J. W., Einterz, C. M., Thorgeirsson, T. E. and Kliger, D. S. (1990).

Nanosecond photolysis of rhodopsin: evidence for a new blue-shifted intermediate.

Biochemistry 29, 1475–1485.

Hunt, D. M., Carvalho, L. S., Cowing, J. A. and Davies, W. L. (2009). Evolution and

spectral tuning of visual pigments in birds and mammals. Philosophical Transactions of

the Royal Society B: Biological Sciences 364, 2941–2955.

Hunt, D. M., Fitzgibbon, J., Slobodyanyuk, S. J. and Bowmaker, J. K. (1996). Spectral

tuning and molecular evolution of rod visual pigments in the species flock of cottoid fish

in Lake Baikal. Vision Research 36, 1217–1224.

Iannaccone, A., Man, D., Waseem, N., Jennings, B. J., Ganapathiraju, M., Gallaher, K.,

Reese, E., Bhattacharya, S. S. and Klein-Seetharaman, J. (2006). Retinitis

pigmentosa associated with rhodopsin mutations: Correlation between phenotypic

variability and molecular effects. Vision Research 46, 4556–4567.

Imai, H., Kefalov, V., Sakurai, K., Chisaka, O., Ueda, Y., Onishi, A., Morizumi, T., Fu,

Y., Ichikawa, K., Nakatani, K., et al. (2007). Molecular Properties of Rhodopsin and

Rod Function. J Biol Chem 282, 6677–6684.

Imai, H., Kuwayama, S., Onishi, A., Morizumi, T., Chisaka, O. and Shichida, Y. (2005).

Molecular properties of rod and cone visual pigments from purified chicken cone

pigments to mouse rhodopsin in situ. Photochem. Photobiol. Sci. 4, 667–8.

Janz, J. M. (2003). Stability of Dark State Rhodopsin Is Mediated by a Conserved Ion Pair

in Intradiscal Loop E-2. Journal of Biological Chemistry 278, 16982–16991.

Janz, J. M. and Farrens, D. L. (2004). Role of the retinal hydrogen bond network in

rhodopsin Schiff base stability and hydrolysis. J Biol Chem 279, 55886–55894.

Jastrzebska, B., Maeda, T., Zhu, L., Fotiadis, D., Filipek, S., Engel, A., Stenkamp, R. E.

and Palczewski, K. (2004). Functional Characterization of Rhodopsin Monomers and

Dimers in Detergents. J Biol Chem 279, 54663–54675.

Jastrzebska, B., Orban, T., Golczak, M., Engel, A. and Palczewski, K. (2013).

Asymmetry of the rhodopsin dimer in complex with transducin. The FASEB Journal 27,

1572–1584.

Jastrzebska, B., Tsybovsky, Y. and Palczewski, K. (2010). Complexes between

photoactivated rhodopsin and transducin: progress and questions. Biochem J 428, 1–10.

Jäger, S., Lewis, J. W., Zvyaga, T. A., Szundi, I., Sakmar, T. P. and Kliger, D. S. (1997).

Time-Resolved Spectroscopy of the Early Photolysis Intermediates of Rhodopsin Schiff

Base Counterion Mutants. Biochemistry 36, 1999–2009.

Jensen, J. L., Shantz, E. M. and Embree, N. D. (1943). The Biological Activity of Vitamin

A2. JBC.

Johnson, P. J. M., Halpin, A., Morizumi, T., Prokhorenko, V. I., Ernst, O. P. and

Miller, R. J. D. (2015). Local vibrational coherences drive the primary photochemistry

of vision. Nature Chemistry 7, 980–986.

Karnik, S. S., Sakmar, T. P., Chen, H. B. and Khorana, H. G. (1988). Cysteine residues

110 and 187 are essential for the formation of correct structure in bovine rhodopsin.

Proc. Natl. Acad. Sci. U.S.A. 85, 8459–8463.

Kaushal, S. and Khorana, H. G. (1994). Structure and function in rhodopsin. 7. Point

mutations associated with autosomal dominant retinitis pigmentosa. Biochemistry 33,

6121–6128.

Kaushal, S., Ridge, K. D. and Khorana, H. G. (1994). Structure and function in rhodopsin:

the role of asparagine-linked glycosylation. Proc. Natl. Acad. Sci. U.S.A. 91, 4024–4028.

Kawamura, S. and Tachibanaki, S. (2008). Rod and cone photoreceptors: Molecular basis

of the difference in their physiology. Comparative Biochemistry and Physiology Part A:

Molecular & Integrative Physiology 150, 369–377.

Kefalov, V., Fu, Y., Marsh-Armstrong, N. and Yau, K.-W. (2003). Role of visual pigment

properties in rod and cone phototransduction. Nature 425, 526–531.

Kennedy, M. J., Dunn, F. A. and Hurley, J. B. (2004). Visual pigment phosphorylation but

not transducin translocation can contribute to light adaptation in zebrafish cones. Neuron

41, 915–928.

Kiser, P. D. and Palczewski, K. (2016). Retinoids and Retinal Diseases. Annu Rev Vis Sci 2,

197–234.

Knierim, B., Hofmann, K. P., Ernst, O. P. and Hubbell, W. L. (2007). Sequence of late

molecular events in the activation of rhodopsin. Proc Natl Acad Sci USA 104, 20290–

20295.

Kojima, D., Okano, T., Fukada, Y., Shichida, Y., Yoshizawa, T. and Ebrey, T. G.

(1992). Cone visual pigments are present in gecko rod cells. Proc. Natl. Acad. Sci. U.S.A.

89, 6841–6845.

Kolb, H. (2007). Webvision: The Organization of the Retina and Visual System. (eds.

Fernandez, E. and Nelson, R. Salt Lake City (UT): University of Utah Health Sciences

Center.

Kolesnikov, A. V., Golobokova, E. Y. and Govardovskii, V. I. (2003). The identity of

metarhodopsin III. Vis. Neurosci. 20, 249–265.

Kolesnikov, A. V., Palczewski, K., Orban, T. and Kefalov, V. J. (2017).

Dephosphorylation of visual pigments by PP2A is required for timely dark adaptation of

rods and cones. Invest. Ophthalmol. Vis. Sci. 58, 3575–3575.

Korenbrot, J. I. (2012). Speed, sensitivity, and stability of the light response in rod and cone

photoreceptors: Facts and models. Prog Retin Eye Res 31, 442–466.

Kosmaoglou, M., Kanuga, N., Aguilà, M., Garriga, P. and Cheetham, M. E. (2009). A

dual role for EDEM1 in the processing of rod opsin. J Cell Sci 122, 4465–4472.

Krebs, M. P., Holden, D. C., Joshi, P., Clark, C. L., III, Lee, A. H. and Kaushal, S.

(2010). Molecular Mechanisms of Rhodopsin Retinitis Pigmentosa and the Efficacy of

Pharmacological Rescue. Journal of Molecular Biology 395, 1063–1078.

Lamb, T. D. (2013). Progress in Retinal and Eye Research. Prog Retin Eye Res 36, 52–119.

Lamb, T. D. and Pugh, E. N., Jr. (2004). Dark adaptation and the retinoid cycle of vision.

Prog Retin Eye Res 23, 307–380.

Land, M. F. (2005). The optical structures of animal eyes. Curr. Biol. 15, R319–23.

Lewis, J. W., Hug, S. J., Wallace-Williams, S. E. and Kliger, D. S. (1990). Direct

evidence for an equilibrium between early photolysis intermediates of rhodopsin. J. Am.

Chem. Soc. 112, 6711–6712.

Lewis, J. W., van Kuijk, F. J., Carruthers, J. A. and Kliger, D. S. (1997). Metarhodopsin

III formation and decay kinetics: comparison of bovine and human rhodopsin. Vision

Research 37, 1–8.

Lin, J. H., Li, H., Yasumura, D., Cohen, H. R., Zhang, C., Panning, B., Shokat, K. M.,

Lavail, M. M. and Walter, P. (2007). IRE1 signaling affects cell fate during the

unfolded protein response. Science 318, 944–949.

Lin, S. W. and Sakmar, T. P. (1996). Specific tryptophan UV-absorbance changes are

probes of the transition of rhodopsin to its active state. Biochemistry 35, 11149–11159.

Liu, J., Liu, M. Y., Fu, L., Zhu, G. A. and Yan, E. C. Y. (2011a). Chemical kinetic

analysis of thermal decay of rhodopsin reveals unusual energetics of thermal

isomerization and hydrolysis of Schiff base. Journal of Biological Chemistry 286,

38408–38416.

Liu, J., Liu, M. Y., Nguyen, J. B., Bhagat, A., Mooney, V. and Yan, E. C. Y. (2009).

Thermal Decay of Rhodopsin: Role of Hydrogen Bonds in Thermal Isomerization of 11-

cis Retinal in the Binding Site and Hydrolysis of Protonated Schiff Base. J. Am. Chem.

Soc. 131, 8750–8751.

Liu, J., Liu, M. Y., Nguyen, J. B., Bhagat, A., Mooney, V. and Yan, E. C. Y. (2011b).

Thermal properties of rhodopsin: insight into the molecular mechanism of dim-light

vision. Journal of Biological Chemistry 286, 27622–27629.

Liu, M. Y., Liu, J., Mehrotra, D., Liu, Y., Guo, Y., Baldera-Aguayo, P. A., Mooney, V.

L., Nour, A. M. and Yan, E. C. Y. (2013). Thermal Stability of Rhodopsin and

Progression of Retinitis Pigmentosa: A Comparison of S186W and D190N Rhodopsin

Mutants. Journal of Biological Chemistry.

Lobanova, E. S., Herrmann, R., Finkelstein, S., Reidel, B., Skiba, N. P., Deng, W. T., Jo,

R., Weiss, E. R., Hauswirth, W. W. and Arshavsky, V. Y. (2010). Mechanistic Basis

for the Failure of Cone Transducin to Translocate: Why Cones Are Never Blinded by

Light. Journal of Neuroscience 30, 6815–6824.

Lórenz-Fonfría, V. A., Furutani, Y., Ota, T., Ido, K. and Kandori, H. (2010). Protein

fluctuations as the possible origin of the thermal activation of rod photoreceptors in the

dark. J. Am. Chem. Soc. 132, 5693–5703.

Luk, H. L., Bhattacharyya, N., Montisci, F., Morrow, J. M., Melaccio, F., Wada, A.,

Sheves, M., Fanelli, F., Chang, B. S. and Olivucci, M. (2016). Modulation of thermal

noise and spectral sensitivity in Lake Baikal cottoid fish rhodopsins. Sci Rep 6, 38425.

Luo, D.-G., Yue, W. W. S., Ala-Laurila, P. and Yau, K.-W. (2011). Activation of visual

pigments by light and heat. Science 332, 1307–1312.

Lüdeke, S., Beck, M., Yan, E. C. Y., Sakmar, T. P., Siebert, F. and Vogel, R. (2005). The

Role of Glu181 in the Photoactivation of Rhodopsin. Journal of Molecular Biology 353,

345–356.

Mass, A. M. and Supin, A. Y. (2007). Adaptive features of aquatic mammals' eye. Anat Rec

290, 701–715.

Mendes, H. F. and Cheetham, M. E. (2008). Pharmacological manipulation of gain-of-

function and dominant-negative mechanisms in rhodopsin retinitis pigmentosa. Hum.

Mol. Genet. 17, 3043–3054.

Mendes, H. F., van der Spuy, J., Chapple, J. P. and Cheetham, M. E. (2005).

Mechanisms of cell death in rhodopsin retinitis pigmentosa: implications for therapy.

Trends in Molecular Medicine 11, 177–185.

Miyagi, R., Terai, Y., Aibara, M., Sugawara, T., Imai, H., Tachida, H., Mzighani, S. I.,

Okitsu, T., Wada, A. and Okada, N. (2012). Correlation between nuptial colors and

visual sensitivities tuned by opsins leads to species richness in sympatric Lake Victoria

cichlid fishes. Mol. Biol. Evol. 29, 3281–3296.

Morrow, J. M. and Chang, B. S. (2015). Comparative Mutagenesis Studies of Retinal

Release in Light-Activated Zebrafish Rhodopsin Using Fluorescence Spectroscopy.

Biochemistry 54, 4507–4518.

Morshedian, A., Toomey, M. B., Pollock, G. E., Frederiksen, R., Enright, J. M.,

McCormick, S. D., Cornwall, M. C., Fain, G. L. and Corbo, J. C. (2017). Cambrian

origin of the CYP27C1-mediated vitamin A 1-to-A 2switch, a key mechanism of

vertebrate sensory plasticity. R. Soc. open sci. 4, 170362–9.

Okada, T., Fujiyoshi, Y., Silow, M., Navarro, J., Landau, E. M. and Shichida, Y. (2002).

Functional role of internal water molecules in rhodopsin revealed by X-ray

crystallography. Proc. Natl. Acad. Sci. U.S.A. 99, 5982–5987.

Okano, T., Kojima, D., Fukada, Y., Shichida, Y. and Yoshizawa, T. (1992). Primary

structures of chicken cone visual pigments: vertebrate rhodopsins have evolved out of

cone visual pigments. Proc. Natl. Acad. Sci. U.S.A. 89, 5932–5936.

Opefi, C. A., South, K., Reynolds, C. A., Smith, S. O. and Reeves, P. J. (2013). Retinitis

Pigmentosa Mutants Provide Insight into the Role of the N-terminal Cap in Rhodopsin

Folding, Structure, and Function. J Biol Chem 288, 33912–33926.

Ovchinnikov, Y. A., Abdulaev, N. G. and Bogachuk, A. S. (1988). Two adjacent cysteine

residues in the C-terminal cytoplasmic fragment of bovine rhodopsin are palmitylated.

FEBS Lett. 230, 1–5.

Palczewski, K., Kumasaka, T., Hori, T., Behnke, C. A., Motoshima, H., Fox, B. A., Le

Trong, I., Teller, D. C., Okada, T., Stenkamp, R. E., et al. (2000). Crystal structure of

rhodopsin: A G protein-coupled receptor. Science 289, 739–745.

Parkes, J. H., Gibson, S. K. and Liebman, P. A. (1999). Temperature and pH Dependence

of the Metarhodopsin I−Metarhodopsin II Equilibrium and the Binding of Metarhodopsin

II to G Protein in Rod Disk Membranes †. Biochemistry 38, 6862–6878.

Patel, A. B., Crocker, E., Reeves, P. J., Getmanova, E. V., Eilers, M., Khorana, H. G.

and Smith, S. O. (2005). Changes in Interhelical Hydrogen Bonding upon Rhodopsin

Activation. Journal of Molecular Biology 347, 803–812.

Peters, K., Applebury, M. L. and Rentzepis, P. M. (1977). Primary photochemical event in

vision: proton translocation. Proc. Natl. Acad. Sci. U.S.A. 74, 3119–3123.

Pugh, E. N. and Lamb, T. D. (1993). Amplification and kinetics of the activation steps in

phototransduction. Biochimica et Biophysica Acta (BBA) - Bioenergetics 1141, 111–149.

Rakoczy, E. P., Kiel, C., McKeone, R., Stricher, F. and Serrano, L. (2011). Analysis of

Disease-Linked Rhodopsin Mutations Based on Structure, Function, and Protein Stability

Calculations. Journal of Molecular Biology 405, 584–606.

Rieke, F. and Baylor, D. A. (2000). Origin and Functional Impact of Dark Noise in Retinal

Cones. Neuron 26, 181–186.

Ritter, E., Zimmermann, K., Heck, M., Hofmann, K. P. and Bartl, F. J. (2004).

Transition of Rhodopsin into the Active Metarhodopsin II State Opens a New Light-

induced Pathway Linked to Schiff Base Isomerization. J Biol Chem 279, 48102–48111.

Rosenzweig, D. H., Nair, K. S., Wei, J., Wang, Q., Garwin, G., Saari, J. C., Chen, C. K.,

Smrcka, A. V., Swaroop, A., Lem, J., et al. (2007). Subunit Dissociation and Diffusion

Determine the Subcellular Localization of Rod and Cone Transducins. Journal of

Neuroscience 27, 5484–5494.

Röll, B. (2001a). Gecko vision - Visual cells, evolution, and ecological constraints. Vision

Research 41, 2043–2056.

Röll, B. (2001b). Gecko vision - retinal organization, foveae and implications for binocular

vision. Vision Research 41, 2043–2056.

Schafer, C. T., Fay, J. F., Janz, J. M. and Farrens, D. L. (2016). Decay of an active

GPCR: Conformational dynamics govern agonist rebinding and persistence of an active,

yet empty, receptor state. Proc. Natl. Acad. Sci. U.S.A. 113, 11961–11966.

Schoenlein, R. W., Peteanu, L. A., Mathies, R. A. and Shank, C. V. (1991). The first step

in vision: femtosecond isomerization of rhodopsin. Science 254, 412–415.

Schott, R. K., Müller, J., Yang, C. G. Y., Bhattacharyya, N., Chan, N., Xu, M., Morrow,

J. M., Ghenu, A.-H., Loew, E. R., Tropepe, V., et al. (2016). Evolutionary

transformation of rod photoreceptors in the all-cone retina of a diurnal garter snake. Proc

Natl Acad Sci USA 113, 356–361.

Shen, J., Chen, X., Hendershot, L. and Prywes, R. (2002). ER stress regulation of ATF6

localization by dissociation of BiP/GRP78 binding and unmasking of Golgi localization

signals. Dev. Cell 3, 99–111.

Sieving, P. A. (1995). Dark-light: model for nightblindness from the human rhodopsin Gly-

90--> Asp mutation. Proc Natl Acad Sci USA.

Sillman, A. J., Govardovskii, V. I., Röhlich, P., Southard, J. A. and Loew, E. R. (1997).

The photoreceptors and visual pigments of the garter snake (Thamnophis sirtalis): a

microspectrophotometric, scanning electron microscopic and immunocytochemical

study. J Comp Physiol A 181, 89–101.

Spady, T. C., Seehausen, O., Loew, E. R., Jordan, R. C., Kocher, T. D. and Carleton, K.

L. (2005). Adaptive Molecular Evolution in the Opsin Genes of Rapidly Speciating

Cichlid Species. Mol. Biol. Evol. 22, 1412–1422.

Starace, D. M. and Knox, B. E. (1997). Activation of transducin by a Xenopus short

wavelength visual pigment. J Biol Chem 272, 1095–1100.

Stryer, L. (1986). Cyclic GMP cascade of vision. Annual review of neuroscience, 9(1),

pp.87–119.

Sugawara, T., Imai, H., Nikaido, M., Imamoto, Y. and Okada, N. (2010). Vertebrate

Rhodopsin Adaptation to Dim Light via Rapid Meta-II Intermediate Formation. Mol.

Biol. Evol. 27, 506–519.

Sung, C. H., Davenport, C. M., Hennessey, J. C., Maumenee, I. H., Jacobson, S. G.,

Heckenlively, J. R., Nowakowski, R., Fishman, G., Gouras, P. and Nathans, J.

(1991). Rhodopsin mutations in autosomal dominant retinitis pigmentosa. Proc. Natl.

Acad. Sci. U.S.A. 88, 6481–6485.

Szundi, I., Funatogawa, C. and Kliger, D. S. (2016). Complexity of Bovine Rhodopsin

Activation Revealed at Low Temperature and Alkaline pH. Biochemistry 55, 5095–5105.

Szundi, I., Funatogawa, C., Guo, Y., Yan, E. C. Y. and Kliger, D. S. (2017). Protein

Sequence and Membrane Lipid Roles in the Activation Kinetics of Bovine and Human

Rhodopsins. Biophysj 113, 1934–1944.

Szundi, I., Lewis, J. W. and Kliger, D. S. (2003). Two Intermediates Appear on the

Lumirhodopsin Time Scale after Rhodopsin Photoexcitation †. Biochemistry 42, 5091–

5098.

Tachibanaki, S., Arinobu, D., Shimauchi-Matsukawa, Y., Tsushima, S. and Kawamura,

S. (2005). Highly effective phosphorylation by G protein-coupled receptor kinase 7 of

light-activated visual pigment in cones. Proc. Natl. Acad. Sci. U.S.A. 102, 9329–9334.

Tang, P. H., Kono, M., Koutalos, Y., Ablonczy, Z. and Crouch, R. K. (2013). New

insights into retinoid metabolism and cycling within the retina. Prog Retin Eye Res 32,

48–63.

Tansley, K. (1964). The gecko retina. Vision Research 4, 33–IN14.

Temple, S. E., Plate, E. M., Ramsden, S., Haimberger, T. J., Roth, W. M. and

Hawryshyn, C. W. (2005). Seasonal cycle in vitamin A1/A2-based visual pigment

composition during the life history of coho salmon (Oncorhynchus kisutch). J. Comp.

Physiol. A Neuroethol. Sens. Neural. Behav. Physiol. 192, 301–313.

Van Nynatten, A., Bloom, D., Chang, B. S. and Lovejoy, N. R. (2015). Out of the blue:

adaptive visual pigment evolution accompanies Amazon invasion. Biol. Lett. 11,

20150349–5.

Vogalis, F., Shiraki, T., Kojima, D., Wada, Y., Nishiwaki, Y., Jarvinen, J. L. P.,

Sugiyama, J., Kawakami, K., Masai, I., Kawamura, S., et al. (2011). Ectopic

expression of cone-specific G-protein-coupled receptor kinase GRK7 in zebrafish rods

leads to lower photosensitivity and altered responses. J. Physiol. (Lond.) 589, 2321–

2348.

Vogel, R., Siebert, F., Lüdeke, S., Hirshfeld, A. and Sheves, M. (2005). Agonists and

partial agonists of rhodopsin: retinals with ring modifications. Biochemistry 44, 11684–

11699.

Wada, Y., Sugiyama, J., Okano, T. and Fukada, Y. (2006). GRK1 and GRK7: Unique

cellular distribution and widely different activities of opsin phosphorylation in the

zebrafish rods and cones. Journal of Neurochemistry 98, 824–837.

Wald, G. (1937). Visual purple system in fresh-water fishes. Nature.

Wald, G. (1939a). On the distribution of vitamins A 1 and A 2. J. Gen. Physiol.

Wald, G. (1939b). THE PORPHYROPSIN VISUAL SYSTEM. J. Gen. Physiol. 22, 775–

794.

Wald, G., Brown, P. K. and Smith, P. H. (1953). Cyanopsin, a new pigment of cone vision.

Science 118, 505–508.

Walls, G. L. (1942). The vertebrate eye and its adaptive radiation [by] Gordon Lynn Walls.

Bloomfield Hills, Mich.,: Cranbrook Institute of Science.

Wang, J. S., Nymark, S., Frederiksen, R., Estevez, M. E., Shen, S. Q., Corbo, J. C.,

Cornwall, M. C. and Kefalov, V. J. (2014). Chromophore Supply Rate-Limits

Mammalian Photoreceptor Dark Adaptation. Journal of Neuroscience 34, 11212–11221.

Weadick, C. J., Loew, E. R., Rodd, F. H. and Chang, B. S. (2012). Visual pigment

molecular evolution in the Trinidadian pike cichlid (Crenicichla frenata): a less colorful

world for neotropical cichlids? Mol. Biol. Evol. 29, 3045–3060.

Yamaoka, H., Tachibanaki, S. and Kawamura, S. (2015). Dephosphorylation during

Bleach and Regeneration of Visual Pigment in Carp Rod and Cone Membranes. J Biol

Chem 290, 24381–24390.

Ye, S., Zaitseva, E., Caltabiano, G., Schertler, G. F. X., Sakmar, T. P., Deupi, X. and

Vogel, R. (2010). Tracking G-protein-coupled receptor activation using genetically

encoded infrared probes. Nature 464, 1386–1389.

Yokoyama, S., Tada, T., Zhang, H. and Britt, L. (2008). Elucidation of phenotypic

adaptations: Molecular analyses of dim-light vision proteins in vertebrates. Proc. Natl.

Acad. Sci. U.S.A. 105, 13480–13485.

Zhang, N., Kolesnikov, A. V., Jastrzebska, B., Mustafi, D., Sawada, O., Maeda, T.,

Genoud, C., Engel, A., Kefalov, V. J. and Palczewski, K. (2012). Autosomal recessive

retinitis pigmentosa E150K opsin mice exhibit photoreceptor disorganization. J Clin

Invest 123, 121–137.

Zhang, X., Wensel, T. G. and Kraft, T. W. (2003). GTPase regulators and photoresponses

in cones of the eastern chipmunk. J. Neurosci. 23, 1287–1297.

Zhang, X., Wensel, T. G. and Yuan, C. (2006). Tokay Gecko Photoreceptors Achieve Rod-

Like Physiology with Cone-Like Proteins†. Photochem. Photobiol. 82, 1452.

1.12 – COPYRIGHT Figure 1.2 - Reprinted from The Lancet, 368(9549), Hartong, DT., Berson, EL., Dryja, TP.,

Retinitis Pigmentosa, 1795-1809, Copyright (2006), with permission from Elsevier.

Figure 1.3, 1.5 - Reprinted from Retina, Fifth Edition, Chen, J., Sampath, AP., Structure and

Function of Rod and Cone Photoreceptors, 342-359, Copyright (2013), with permission from

Elsevier

Figure 1.9 - Reprinted from Current Biology, 16(13), Bowmaker, JK., Hunt DM., Evolution of Vertebrate Visual Pigments, R484-R489, Copyright (2006), with permission from Elsevier

Figure 1.11 - Reprinted from Progress in Retinal and Eye Research, 32, Tang, PH., Kono,

M., Koutalos, Y., Ablonczy, Z., Crouch, Rosalie, K., New Insights into Retinoid Metabolism and Cycling Within the Retina, 48-63, Copyright (2012), with permission from Elsevier

Figure 1.12 - Reprinted from Progress in Retinal and Eye Research, 62, Athanasiou, D.,

Aguila, M., Bellingham, J., Li, W., McCulley, C., Reeves, PJ., Cheetham, ME., The

Molecular and Cellular Basis of Rhodopsin Retinitis Pigmentosa Reveals Potential Strategies for Therapy, 1-23, Copyright (2018), with permission from Elsevier.

CHAPTER II: CONE-LIKE RHODOPSIN EXPRESSED IN THE ALL CONE RETINA OF THE COLUBRID PINE SNAKE AS A POTENTIAL ADAPTATION TO DIURNALITY

This chapter was published as: Bhattacharyya N, Darren B, Schott RK, Tropepe V, and Chang BSW (2017) Cone-like rhodopsin expressed in the all-cone retina of the colubrid pine snake as a potential adaptation to diurnality. The Journal of Experimental Biology, 220(13), pp.2418– 2425. doi: 10.1242/jeb.156430 url: http://jeb.biologists.org/content/early/2017/04/29/jeb.156430

Author contributions: NB, BD and BSWC designed the study. BD sequenced opsin genes and inserted them into expression vectors. NB performed all other experiments, with assistance from RKS for the phylogenetic analysis, and VT for theimmunofluorescence experiments. NB, RKS and BSWC analyzed and interpreted data. NB, RKS and BSWC wrote the manuscript with input from VT.

2.1 – ABSTRACT

Colubridae is the largest and most diverse family of snakes, with visual systems that reflect this diversity, encompassing a variety of retinal photoreceptor organizations. The transmutation theory proposed by Walls postulates that photoreceptors could evolutionarily transition between cell types in squamates, but few studies have tested this theory. Recently, evidence for transmutation and rod-like machinery in an all cone retina has been identified in a diurnal garter snake (Thamnophis), and it appears that the rhodopsin gene at least may be widespread among colubrid snakes. However, functional evidence supporting transmutation beyond the existence of the rhodopsin gene remains rare. We examined the all cone retina of another colubrid, Pituophis melanoleucus, thought to be more secretive/burrowing than

Thamnophis. We found that P. melanoleucus expresses two cone opsins (SWS1, LWS) and rhodopsin (RH1) within the eye. Immunohistochemistry localized rhodopsin to the outer

81 segment of photoreceptors in the all-cone retina of the snake and all opsin genes produced functional visual pigments when expressed in vitro. Consistent with other studies, we found that P. melanoleucus rhodopsin is extremely blue-shifted. Surprisingly, P. melanoleucus rhodopsin reacted with hydroxylamine, a typical cone opsin characteristic. These results support the idea that the rhodopsin-containing photoreceptors of P. melanoleucus are the products of evolutionary transmutation from rod ancestors and suggests that this phenomenon may be widespread in colubrid snakes. We hypothesize that transmutation may be an adaptation for diurnal, brighter-light vision, which could result in increased spectral sensitivity and chromatic discrimination with the potential for colour vision.

2.2 - INTRODUCTION

Reptiles are known for their impressive array of visual adaptations and retinal organizations, which reflect distinct ecologies and evolutionary histories (Underwood, 1970; Walls, 1942).

The family Colubridae is the most speciose family of snakes and encompasses a diverse range of lifestyles and ecologies. Colubrid snakes have recently emerged as a compelling group in which to study visual system evolution and adaptation (Schott et al., 2016; Simões et al., 2015; Simões et al., 2016).

In the vertebrate retina, photoreceptor cells can be divided into two types based on their overall structure and function: cones, which are active in bright light and contain cone visual pigments (SWS1, SWS2, RH2, LWS) in a tapered outer segment, and rods, which function in dim light and contain rhodopsin (RH1) in a longer, more cylindrical outer segment (Bowmaker, 2008; Lamb, 2013; Walls, 1942). Reptilian retinas are unique in having multiple retinal configurations (Underwood, 1970) among closely related species including all-rod (Kojima et al., 1992), rod and cone (Sillman et al., 2001), and all-cone (Sillman et al.,

1997). In 1942, physiologist Gordon Walls outlined his theory of transmutation to explain the evolutionary transformation of photoreceptors from one type to another (Walls, 1942). This phenomenon has since been investigated in nocturnal geckos, where cone opsins are expressed in an all-rod retina in order to compensate for the evolutionary loss of RH1 in a hypothesized diurnal, all-cone ancestor (Kojima et al., 1992; Taniguchi et al., 1999). While the nocturnal henophidian snakes, such as boas and pythons, are known to have duplex retinas expressing RH1, LWS and SWS1 in canonical photoreceptors (Davies et al., 2009), the more derived diurnal colubrid snakes have been primarily shown to possess simplex retinas comprising of all cone photoreceptors, with the fate of the rod photoreceptor unknown

(Caprette, 2005; Underwood, 1970; Walls, 1942). Early studies of the colubrid visual system found a green-sensitive visual pigment in addition to a red and a blue pigment (Sillman et al.,

1997) in the simplex retina, but were unable to distinguish between a spectrally shifted rhodopsin in a transmuted rod or a potentially resurrected RH2 cone opsin (Cortesi et al.,

2015). More recently, a study from our group identified a functional blue-shifted RH1 pigment in the all-cone retina of the ribbon snake (Thamnophis proximus), and proposed that this resulted from a rod to cone evolutionary transmutation in colubrid snakes that may have allowed for enhanced spectral discrimination and even trichromatic colour vision (Schott et al., 2016). A recent study that sequenced the opsins of several other colubrid snake species discovered the widespread presence of full-length rhodopsin genes in species with supposed simplex retinas that were previously presumed to have lost rod/rhodopsins (Simões et al.,

2016). However, detailed characterizations of colubrid snake opsins and photoreceptors in the context of the theory of evolutionary transmutation still remain rare.

To further test the hypothesis of widespread transmutation in colubrid snakes, and its potential functional consequences, we examined the visual system of the Northern Pine

Snake (Pituophis melanoleucus), a diurnal colubrid snake distantly related to T. proximus.

Pituophis melanoleucus inhabits the eastern half of the United States and Canada (Stull,

1940) and spends relatively short intervals on the surface during the day to forage for prey such as small mammals and birds, and to create new burrows (Diller and Wallace, 1996;

Himes, 2001). While P. melanoleucus has been found to possess an all-cone retina (Caprette,

2005), similar to previous diurnal colubrid snakes studied (Schott et al., 2016; Sillman et al.,

1997), unlike other strongly diurnal colubrids such as the garter snake, P. melanoleucus is

84 more secretive and is thought to spend a considerable amount of time burrowing (Gerald et al., 2006).

In this study, we investigate whether there is evidence of photoreceptor transmutation from rods into cones in the all-cone retina of P. melanoleucus via functional characterization, cellular localization, and molecular evolutionary analyses of its visual pigment (opsin) genes.

We isolated three opsin genes from P. melanoleucus: SWS1, LWS and RH1.

Immunohistochemistry of the retina localized rhodopsin (RH1) protein and rod transducin to the inner and outer segments of a small subset of photoreceptors, suggesting that P. melanoleucus exemplifies another rod-to-cone transmutation in diurnal colubrids. All three opsins were successfully expressed in vitro and displayed properties characteristic of fully functional visual pigments. Additionally, spectroscopic assays revealed that P. melanoleucus rhodopsin is sensitive to hydroxylamine, which is more typical of cone opsins and is suggestive of more cone-like functional properties. This study provides further evidence for a fascinating evolutionary transformation in the retinas of colubrid snakes, with implications for reptiles in general.

2.3 - MATERIALS AND METHODS

Animals

A Northern pine snake (Pituophis melanoleucus melanoleucus, adult) specimen and mice

(Mus musculus, adult, CD1) were obtained from a licensed source as commissioned by the

University Animal Care Committee (UACC). The specimen was sacrificed using an approved euthanasia protocol. The eyes were enucleated and preserved in RNAlater or 4% paraformaldehyde.

Total RNA extraction and cDNA synthesis

The dissected whole eye was homogenized with TRIzol, and total RNA was isolated using a phenol/chloroform extraction and ethanol precipitation. The first strand of complementary

DNA (cDNA) was synthesized using SuperScript III Reverse Transcriptase (Invitrogen,

Waltham, MA, USA) from RNA samples primed with a 3’ oligo-dT and a 5’ SMART primer, following the protocol outlined by the SMART cDNA Library Construction Kit (BD

Biosciences, Franklin Lakes, NJ, USA). The second strand complement was synthesized by long-distance PCR following the same protocol.

Visual pigment genes were isolated using a degenerate PCR strategy. Degenerate primers based on an alignment of reptilian visual pigment sequences were used in attempts to amplify partial sequences of the LWS, SWS1 and RH1 opsin genes with a heminested strategy. GenomeWalker (Clontech, Mountain View, CA, USA) was additionally used to obtain full-length sequences (Supplementary Table S2.1). Extracted PCR products were ligated into the pJET1.2 blunt plasmid vector.

Phylogenetic analysis

A representative set of vertebrate visual opsin sequences was obtained from Genbank. These sequences were combined with the three opsin genes sequenced from the pine snake and aligned using MUSCLE (Edgar, 2004). The poorly aligned 5' and 3' ends of the sequence were manually trimmed. Species list and accession numbers for all sequences used in the study are provided in Supplementary Table S2.2. In order to confirm the identities of the opsin genes from the pine snake, a gene tree was estimated using the resulting alignment in

MrBayes 3 (Ronquist and Huelsenbeck, 2003) using reversible jump MCMC with a gamma rate parameter (nst=mixed, rates=gamma), which explores the parameter space for the nucleotide model and the phylogenetic tree simultaneously. The analyses were run for five million generations with a 25% burn-in. Convergence was confirmed by checking that the standard deviations of split frequencies approached zero and that there was no obvious trend in the log likelihood plot.

Protein expression

Full-length opsin sequences (RH1, SWS1, and LWS) were amplified from pJET1.2 vector using primers that added the BamHI and EcoRI restriction sites to its 5' and 3' ends, respectively, and inserted into the p1D4-hrGFP II expression vector (Morrow and Chang,

2010). Expression vectors containing P. melanoleucus cone opsin and rhodopsin genes were transiently transfected into cultured HEK293T cells (ATCC CRL-11268) using

Lipofectamine 2000 (Invitrogen, Waltham, MA, USA; 8 µg of DNA per 10-cm plate) and harvested after 48 h. Visual pigments were regenerated with 11-cis retinal, generously

87 provided by Dr. Rosalie Crouch (Medical University of South Carolina), solubilized in 1% dodecylmaltoside, and purified with the 1D4 monoclonal antibody (University of British

Columbia #95-062, Lot #1017; Molday and MacKenzie, 1983) as previously described

(Morrow and Chang, 2015; Morrow and Chang, 2010; Morrow et al., 2011). RH1 and SWS1 pigments were purified in sodium phosphate buffers and LWS was purified in HEPES buffers containing glycerol (as described in van Hazel et al. (2013)). The ultraviolet-visible absorption spectra of purified visual pigments were recorded using a Cary 4000 double beam spectrophotometer (Agilent, Santa Clara, CA, USA). Dark-light difference spectra were calculated by subtracting light-bleached absorbance spectra from respective dark spectra.

Pigments were photoexcited with light from a fiber optic lamp (Dolan-Jenner, Boxborough,

MA, USA) for 60 s at 25°C. Absorbance spectra for acid bleach and hydroxylamine assays were measured following incubation in hydrochloric acid (100mM) and hydroxylamine

(NH2OH, 50mM), respectively. To estimate lmax, the dark absorbance spectra were baseline corrected and fit to Govardovskii templates for A1 visual pigments (Govardovskii et al.,

2000).

Immunohistochemistry

Fixation of pine snake eyes was conducted as previously described (Schott et al., 2016).

Briefly, after enucleating P. melanoleucus eyes in the light, they were rinsed in PBS (0.8%

NaCl, 0.02% KCl, 0.144% NaHPO4, and 0.024% KH2PO4, pH 7.4), fixed overnight at 4ºC in

4% paraformaldehyde, infiltrated with increasing concentrations of sucrose (5%, 13%, 18%,

22%, 30%) in PBS, and embedded in a 2:1 solution of 30% sucrose and O.C.T compound

(Tissue-Tek, Burlington, NC, USA) at -20º. The eyes were cryosectioned transversely at -

25ºC in 20 µm sections using a Leica CM3050 (Wetzlar, Germany) cryostat, placed onto positively charged microscope slides, and stored at -80ºC until use.

Slides were first rehydrated in PBS and then air-dried to ensure adhesion. Sections were rinsed three times in PBS with 0.1% Tween-20 (PBT) and then incubated in 4% paraformaldehyde PBS for 20 minutes. After rinsing in PBT and PDT (PBT with 0.1%

DMSO), the slides were incubated in a humidity chamber with blocking solution (1% BSA in

PDT with 2% normal goat serum) for one hour, incubated with primary antibody diluted in blocking solution overnight at 4º in a humidity chamber. Antibodies used were the K20 antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA, sc-389, lot#:C1909, dilution:

1:500) and RET-P1 anti-rhodopsin antibody (Sigma-Aldrich, St. Louis, MO, USA, O-4886, lot#: 19H4839, dilution: 1:200).

After extensive rinsing and soaking in PDT (3 times for 15 minutes), secondary antibody was added to the samples and incubated at 37ºC for one hour in a humidity chamber. Secondary antibodies used for the fluorescent staining were the AlexaFluor-488 anti-rabbit antibody (Life Technologies, Waltham, MA, USA, A11034, lot#: 1298480, dilution: 1:1000) and the Cy-3 anti-mouse antibody (Jackson ImmunoResearch, West Grove,

PA, USA, 115-165-003, dilution: 1:800). After rinsing with PBS, followed by PDT, sections were stained with 10 µg/mL Hoechst for 10 minutes at room temperature. The sections were then rinsed in PBS and PDT and mounted with ProLong® Gold antifade reagent (Life technologies, Waltham, MA, USA) and cover-slipped. Sections were visualized via a Leica

SP-8 confocal laser microscope (Wetzlar, Germany).

2.4 - RESULTS

Full-length RH1, SWS1 and LWS opsin sequences found in Pituophis melanoleucus cDNA

To determine the identities of the visual pigments in P. melanoleucus, eye cDNA and gDNA was screened for opsin genes. Three full-length opsins were amplified, sequenced, and analyzed phylogenetically with a set of representative vertebrate visual opsins (Table S2.1) using Bayesian inference (MrBayes 3.0) (Ronquist and Huelsenbeck, 2003). This analysis confirmed the identity of the three opsin genes as RH1, LWS, and SWS1 (Fig. S2.1-S2.3).

All three opsin gene sequences contained the critical amino acid residues required for proper structure and function of a prototypical opsin including K296, the site of the Schiff base linkage with 11-cis retinal (Palczewski et al., 2000; Sakmar et al., 2002), and E113, the counter-ion to the Schiff base in the dark state (Sakmar et al., 1989), as well as C110 and

C187, which form a critical disulfide bond in the protein (Karnik and Khorana, 1990). Both cone opsin genes also have the conserved P189 residue which is critical for faster cone opsin pigment regeneration (Kuwayama et al., 2002).

Interestingly, P. melanoleucus RH1 has serine at site 185 instead of the highly conserved cysteine, similar to several other snakes (Schott et al., 2016; Simões et al., 2016).

Mutations at site 185 have been shown to reduce both visual pigment stability (McKibbin et al., 2007) and transducin activation in vitro (Karnik et al., 1988). Also, the P. melanoleucus

RH1 has N83 and S292, which are often found in rhodopsins with blue-shifted lmax values, and can also affect all-trans retinal release kinetics following photoactivation (Bickelmann et al., 2012; van Hazel et al., 2016).

Based on known spectral tuning sites in LWS, P. melanoleucus has A285, compared to T285 in Thamnophis snakes. T285A is known to blue-shift the LWS pigment by 16-20 nm

(Asenjo et al., 1994; Yokoyama, 2000). This suggests that the P. melanoleucus LWS may be considerably blue-shifted relative to the LWS pigment in Thamnophis snakes. Within P. melanoleucus SWS1, the phenylalanine at site 86 suggests that the pigment will be absorbing in the UV, as is typical of reptilian SWS1 pigments (Hauser et al., 2014). Pituophis melanoleucus SWS1, as well as other colubrids SWS1 (Simões et al., 2016), have hydrophobic residues at two spectral tuning sites, A90 and V93. These sites usually have polar or charged amino acid side chains (Carvalho et al., 2011; Hauser et al., 2014). The functional significance of these hydrophobic residues has yet to be characterized, and therefore caution should be taken in applying spectral tuning predictions on squamates SWS1 pigments.

Rhodopsin and Rod Transducin Are Expressed in the outer segment of photoreceptors

Because P. melanoleucus has an anatomically all-cone retina, we used immunohistochemistry to determine if both rhodopsin and the rod G protein transducin are expressed in cone photoreceptors. We performed fluorescent immunohistochemistry on the transverse cryosections of the retina of P. melanoleucus with the rhodopsin antibody (RET-

P1) and a rod-specific transducin antibody (K20). Both antibodies have been previously shown to be selective across a range of vertebrates (Fekete and Barnstable, 1983; Hicks and

Barnstable, 1987; Osborne et al., 1999; Schott et al., 2016). We also used these antibodies on a CD1 mouse retina, following similar preparation, as a positive control.

Our results showed rhodopsin localized to the outer segments of select photoreceptors of the P. melanoleucus retina (red, Fig 2.1D), whereas the rod transducin localized to the inner segment (green, Fig 2.1E). The small amount of colocalization between rhodopsin and transducin in the inner segment (yellow, Fig 2.1F) is expected as the animal was not dark- adapted prior to sacrifice, as rod transducin translocates to the inner segment when exposed to bright light (Calvert et al., 2006; Elias et al., 2004). This pattern is consistent with rhodopsin and transducin staining in the T. proximus retina (Schott et al., 2016) and the previously unexplained results of rhodopsin detected in the retina of T. sirtalis (Sillman et al.,

1997).

As expected, CD1 mouse retina had strong rhodopsin fluorescence (red, Fig 2.1A) in the outer segment and strong rod transducin staining (green, Fig 2.1B) in the inner segment, consistent with the rod dominant mouse retina. The lack of colocalization is consistent with a light-adapted retina (Calvert et al., 2006; Elias et al., 2004) (Fig 2.1C).

P. melanoleucus opsins are all functional in vitro with a highly blue-shifted rhodopsin

Complete coding sequences of the P. melanoleucus RH1, LWS, and SWS1 opsins were cloned into the p1D4-hrGFP II expression vector, transfected into HEK293T cells and the expressed protein was then purified with the 1D4 monoclonal antibody (Morrow and

Chang, 2015; Morrow et al., 2011). Bovine wildtype rhodopsin was used as a control (Fig

2.2A). Pine snake rhodopsin has a lmax of 481nm (Fig 2.2B), which is similar to the measured lmax of rhodopsins from T. proximus, T. sirtalis, and Arizona elegans snakes

(Schott et al., 2016; Sillman et al., 1997; Simões et al., 2016). The drastic blue shift is expected given the presence of the blue-shifting N83 and S292 amino acid identities

(Bickelmann et al., 2012; Dungan et al., 2016; van Hazel et al., 2016). P. melanoleucus rhodopsin expressed similar to that of T. proximus, with a large ratio between total purified protein (absorbance at 280nm) and active protein (absorbance at lmax) that indicates that only a small proportion of the translated opsin protein is able to bind chromophore and become functionally active. One possible explanation for this effect is the S185 residue in P. melanoleucus rhodopsin, as mutations at this site have been shown to affect the retinal binding efficiency of rhodopsin pigments expressed in vitro (McKibbin et al., 2007).

Expression of pine snake SWS1 showed a more typical 280nm to lmax ratio (Fig

2.2C). We found that P. melanoleucus SWS1 pigment absorbs in the UV range with a lmax of

370nm, similar to the SWS1 lmax of Lampropeltis getula, Rhinocheilus lecontei, and

Hypsiglena torquata (Simões et al., 2016) all of which have the most red-shifted UV SWS measured among colubrid snakes.

Similar to the SWS1 expression, LWS also expressed quite well (Fig 2.2D). Fit to A1 templates gave a lmax of 534nm, which is blue-shifted relative to Thamnophis (Schott et al.,

2016; Sillman et al., 1997), but identical with LWS MSP measurements of H. torquata

(Simões et al., 2016) and very close to those of L. getula, A. elegans, and R. lecontei (Simões et al., 2016).

P. melanoleucus rhodopsin demonstrates cone opsin-like functional characteristics

In order to confirm the covalent attachment of the chromophore in P. melanoleucus

SWS1 pigments, the purified opsin was acid bleached (Fig 2.2C). We found a shift of the lmax from 370nm to 440nm, which indicates the presence of 11-cis retinal covalently bound

93 by a protonated Schiff base to a denatured opsin protein (Kito et al., 1968), suggesting that the UV sensitivity of the pigment may be established by only the presence of F86.

P. melanoleucus LWS (Fig 2.3A) and RH1 (Fig 2.3B) were tested for hydroxylamine reactivity, which assesses the accessibility of the chromophore-binding pocket to attack by small molecules. If hydroxylamine can enter the binding pocket, it will hydrolyze the Schiff base linkage, resulting in an absorbance decrease of the dark peak and the increase of a retinal oxime peak at 363nm. Rhodopsins are thought to be largely non-reactive in the presence of hydroxylamine (Dartnall, 1968) (Fig 2.3C) due to their highly structured and tightly packed chromophore binding pockets relative to cone opsins, which often react when incubated in hydroxylamine (van Hazel et al., 2013). P. melanoleucus LWS reacted to hydroxylamine, as expected, with a t1/2 of ~3.9 min (Fig 2.3A), a time within the range of cone opsins (Das et al., 2004; Ma et al., 2001). As the lmax of P. melanoleucus SWS1 is

370nm, it was not tested as we would not be able to distinguish the retinal oxime peak from the lmax peak. Interestingly, P. melanoleucus rhodopsin also reacted to hydroxylamine with a t1/2 of ~14 min (Fig 2.3B), unlike the bovine rhodopsin control that did not react (Fig 2.3C).

This implies that the chromophore binding pocket of P. melanoleucus rhodopsin has a more open configuration relative to other rhodopsin proteins, a property more typical of cone opsins.

Figure 2.1: Immunohistochemical staining of control (mouse, A-C, Scale bar = 40 µm) and pine snake (D-F, Scale bar = 20 µm) transverse retinal cryosections with rhodopsin (RET-P1) and rod-specific-transducin (K20) antibodies. Rhodopsin is found in a subset of cone-like photoreceptors localized to the outer segment (D). Rod-specific transducin is also found in the same photoreceptor, primarily to the inner segment (E). Double staining indicates that both rhodopsin and rod-specific transducin are found within the same cell (F). Nuclei are shown in blue, rhodopsin in red, and rod-specific transducin in green. Scale bar = 20 µm.

Figure 2.2: UV-visible dark absorption spectra of pine snake opsins. (A) Bovine wildtype rhodopsin was used as a control for expressions. Dark spectra for pine snake (B) rhodopsin (C) SWS1 and (D) LWS. Inset in (A), (B), and (D) is the dark-light spectra. Inset in (C) is the dark-acid bleach spectrum. lmax estimations are shown for each pigment.

Figure 2.3: Hydroxylamine reactivity of pine snake (A) LWS and (B) RH1 pigments and (C) bovine rhodopsin. Absorption values of the dark lmax peak decrease over time (open circles), while absorption of the retinal oxime at 360nm increase over time (solid circles). The half-lives of the reactive opsins were determined via curve fitting exponential rise and decay equations to data.

2.5 - DISCUSSION

Recently, there has been mounting evidence supporting the theory of transmutation in photoreceptor evolution, proposed by Walls in 1942, which outlines the evolutionary transformation of one photoreceptor type into another in reptilian retinas. Evidence of cone to rod transmutation in nocturnal geckos has been extensively demonstrated using both cellular and molecular techniques (Crescitelli, 1956; Dodt and Walther, 1958; Kojima et al., 1992;

McDevitt et al., 1993; Röll, 2001; Sakami et al., 2014; Tansley, 1959; Tansley, 1961;

Tansley, 1964; Zhang et al., 2006), while evidence of rod-to-cone transmutation in colubrid snakes remains somewhat sparse (Schott et al., 2016). In order to demonstrate rod-to-cone transmutation in the retina there needs to be evidence of a functional rod machinery in a photoreceptor with some rod-like features in a retina that appears, superficially, to consist of only cones. Certainly, the presence of RH1 genes and MSP data suggests transmutation has occurred in several colubrid species (Hart et al., 2012; Sillman et al., 1997; Simões et al.,

2015; Simões et al., 2016), but further investigation is required in order to firmly state transmutation is present in the retinas of these colubrid snakes as there are multiple alternate explanations possible (RH1 in the genome but not expressed, rhodopsin expressed but not functional, a cone cell co-opting rhodopsin, etc.). There is only one colubrid snake species for which cellular and molecular evidence for transmutation has been reported, Thamnophis proximus (Schott et al., 2016).

The present study provides strong evidence that supports the hypothesis that photoreceptor transmutation has occurred in the retina of P. melanoleucus. As P. melanoleucus is not closely related to snakes in the genus Thamnophis, this suggests that transmutation may be widespread in colubrid snakes. However, the functional significance of

98 transmutation in colubrid snakes still has not been established. In geckos, the advantage of cone-to-rod transmutation is more straightforward as these nocturnal animals are most likely compensating for the loss of RH1 in their diurnal ancestor. We propose that transmutation in colubrids may have occurred as an adaptation to diurnality that provided P. melanoleucus with a cone-like rod photoreceptor that operates at brighter light levels, perhaps as a compensation for the loss of the RH2 cone opsins. Our finding of a highly blue-shifted rhodopsin with more cone-like functional properties, as indicated by hydroxylamine reactivity, supports this hypothesis.

Pituophis melanoleucus rhodopsin shows hydroxylamine reactivity, a canonical cone opsin property (Wald et al., 1955). With a reaction half-life of ~14min, the P. melanoleucus rhodopsin reacts much quicker and closer to cone opsin speeds (Das et al., 2004; Ma et al.,

2001) than previous rhodopsins that have reacted when incubated in hydroxylamine, like the echidna (Bickelmann et al., 2012) and the anole (Kawamura and Yokoyama, 1998) which react over hours. The RH1 sequence contains both E122 and I189, which are known to mediate the slower decay and regeneration kinetics typical of rhodopsin (Imai et al., 1997;

Kuwayama et al., 2002). Conversely, the presence of serine rather than cysteine at site 185, in rhodopsin has been shown to activate fewer G proteins (Karnik et al., 1988) and mutation at site 185 has been shown to reduce the thermal stability of the protein (McKibbin et al.,

2007), both characteristics being more typical of cone opsins. Cones have been optimized for fast regeneration, with cone opsin meta-intermediate states being short lived compared to rhodopsin (Imai et al., 2005), and a cone-specific Müller cell retinoid cycle (Das et al., 1992) providing a dedicated pool of 11-cis retinal. These faster kinetic properties are hypothesized to be facilitated in cone opsins via the relative “openness” of the chromophore binding

99 pocket, which allows water molecules, and therefore other small molecules like hydroxylamine, to access the chromophore where they can participate in Schiff base hydrolysis (Chen et al., 2012; Piechnick et al., 2012; Wald et al., 1955). Rhodopsins, on the other hand, are optimized for sensitivity and signal amplification; therefore, E122/I189 and a tighter overall structure contribute to a slower active state decay allowing for the activation of multiple G proteins (Chen et al., 2012; Starace and Knox, 1997), increased thermal stability relative to cone opsins (Barlow, 1964), and a resistance to hydroxylamine (Dartnall,

1968). Pituophis melanoleucus rhodopsin shows adaptations to decrease the number of G- proteins activated, and hydroxylamine reactivity which suggests that an open chromophore binding pocket would enable water access to facilitate active state decay, Schiff base linkage hydrolysis, and retinal regeneration (Chen et al., 2012) rates similar to cone opsins.

Spectroscopic assays measuring G protein activation and retinal release rates have never been performed on colubrid rhodopsins, but would be an interesting direction for future research characterizing this cone opsin-like rhodopsin. Additionally, there are multiple instances of cone- and rod-specific members in the phototransduction cascade which vary in terms of expression levels (Cowan et al., 1998; Zhang et al., 2003) and functional efficiency

(Tachibanaki et al., 2005; Vogalis et al., 2011; Wada et al., 2006) consistent with rod (light sensitivity) and cone (rapid signaling) specializations. The effects of transmutation on the phototransduction cascade is uncharacterized and presents as another possible path for modifications to photoreceptor function.

Retinal immunohistochemistry localized P. melanoleucus rhodopsin protein in the outer segment of an anatomically cone-like photoreceptor, as well as the presence of rod transducin in the inner segment. Rod and cone transducin are thought to originate via

100 duplication from on ancestral gene (Larhammar et al., 2009) and both have been shown to function with all opsins (Sakurai et al., 2007), therefor the presence and preservation of rod transducin in the photoreceptor supports the theory that this is indeed a transmuted rod and not a cone photoreceptor co-opting rhodopsin expression. Because the retinas were not dark adapted prior to sacrifice, we can presume that under normal photopic light conditions, P. melanoleucus rod transducin is cycled out of the outer segment of the cone-like rod, a distinct rod property (Chen et al., 2007; Rosenzweig et al., 2007). In the light, rods cycle transducin and recoverin out of the outer segment, and arrestin into it (Calvert et al., 2006). This allows the rod to effectively shut down phototransduction under bleaching conditions to prevent damage to the photoreceptor. Cones generally do not cycle transducin out of the outer segment of the photoreceptor under normal light conditions (Chen et al., 2007). This suggests that the rhodopsin-expressing photoreceptors in the retina of P. melanoleucus would not be able to generate a photoresponse in normal daylight, and thus if this cone-like rod is participating in colour vision with the canonical cones in the retina, it would likely only be under mesopic light conditions where both photoreceptor cell types can be active.

Our microscopy results of the P. melanoleucus retina additionally revealed a cone- like rod, which still looks distinct in comparison to the other cones. The cone-like rod outer segment and inner segment had similar diameters with a relatively long outer segment, while the surrounding cones had distinctly large ellipsoids in the inner segment, and proportionally smaller outer segments. Rod photoreceptor morphology is also generally specialized to maximize sensitivity with long cylindrical outer segments (Lamb, 2013). Cone morphological specializations, however, are thought to enable selective colour vision, a faster phototransduction and visual pigment regeneration, while also minimizing metabolic load by

101 miniaturizing the overall structures with large ellipsoids that tunnel light onto smaller tapered outer segments (Harosi and Novales Flamarique, 2012). Previous EM studies on the retina of

T. proximus showed that the membrane discs unique to rods in the outer segment are still present in the transmuted photoreceptor (Schott et al., 2016). Interestingly, a reduction of

RH1 expression levels has been shown to reduce the size of the outer segment of rods, in addition to lowering the photosensitivity and altering the kinetics of the cell to be more cone- like (Makino et al., 2012; Rakshit and Park, 2015; Wen et al., 2009). Currently, the relative expression levels of RH1 in the retinas of colubrid snakes have not been measured. There are additional specializations in the synaptic structures that reflect the different priorities in rod and cone function (Lamb, 2013), but the synaptic structure of the cone-like rod also remains uninvestigated.

Results from this study suggest that transmutation is modifying the function of a subset of photoreceptors in the retina of P. melanoleucus. These modifications may serve to lower the sensitivity and signal amplification of the photoreceptor, supporting the hypothesis of a more cone-like function. However, the type of signal these transmuted rods send to the brain is still unknown. Rods and cones are known to have distinct ERG responses, but T. sirtalis is the only colubrid snake with ERG measurements performed at a variety of light levels (Jacobs et al., 1992). However, this study did not record any scotopic (rod) response, nor did it record any photopic response from the SWS1-type photoreceptors, which suggests that the results of the study may be incomplete or that the scotopic pathways in the colubrid eye have degraded. Indeed, in high scotopic and mesopic light levels, mammalian rod photoreceptors can and do use cone pathways (Daw et al., 1990; Gregg et al., 2013). And while the presence of amacrine cells has been demonstrated in the duplex retina of turtles

102

(Baylor and Fettiplace, 1977; Walls, 1942) and in the simplex retina of sea snakes (Hibbard and Lavergne, 1972), the presence of rod bipolar cells and AII amacrine cells, both of which are required in the rod-specific photoresponse pathway (Lamb, 2013), has never been established in the colubrid retina.

The evolution of cone-like functionality in a rod photoreceptor may be an attempt to compensate for the loss of the RH2 cone opsin and the lack of spectral overlap between the

LWS and SWS1 pigment, such that it could participate in colour vision. In addition to the molecular modifications to P. melanoleucus rhodopsin and the physiological modifications to the rod cell, the extreme blue-shift of the RH1 lmax, which is quite rare for terrestrial rhodopsins, may itself be an adaptation for colour vision, as a lmax of ~480 nm is in the range of typical RH2 pigments (Lamb, 2013). Pituophis melanoleucus, in comparison to the

Thamnophis genera (Schott et al., 2016; Sillman et al., 1997), has a narrower overall range of spectral sensitivities. There could be two possible reasons for this narrowing. It could be that this narrowing of the spectral ranges is to facilitate spectral overlap as an adaptation in P. melanoleucus. Or the narrowing of the spectral range may simply be due to phylogenetic history, as P. melanoleucus LWS and SWS1 absorb at similar wavelengths to its closest relatives (Simões et al., 2016), which in turn could be an adaptation, but not one due to the specific visual environment of P. melanoleucus. Trichromatic vision would be greatly advantageous for a diurnal species (Ankel-Simons and Rasmussen, 2008; Heesy and Ross,

2001), and perhaps sacrificing scotopic vision in order to achieve better mesopic and photopic vision is possible, since other snake sensory systems adaptations, such as chemoreception, could be sufficient in dim light environments (Drummond, 1985). However,

103 currently there is a lack of behavioral studies investigating trichromatic colour discrimination in colubrid snakes under mesopic light conditions.

We hypothesize that rod-to-cone transmutation may be allowing colubrid snakes to have a third cone-like photoreceptor, allowing for spectral sensitivity between SWS1 and

LWS, possibly also trichromatic colour perception in mesopic light conditions. The loss of

RH1 in nocturnal geckos and the resulting transmutation of cone into rod demonstrate that the visual system of squamates is capable of adapting to compensate for previous functionality loss in different photoreceptor types. In colubrid snakes, and possibly squamates in general, the rod/cone photoreceptor binary is not as distinct as it is in other vertebrates and caution should be taken in classifying rod or cone photoreceptors based on limited characterization.

In summary, we find that P. melanoleucus, like T. proximus, has an all-cone retina derived through evolutionary transmutation of the rod photoreceptors. Furthermore, P. melanoleucus rhodopsin is the first vertebrate rhodopsin to show hydroxylamine reactivity similar to cone opsins. This study is also the first to demonstrate the functional effects of transmutation in the retina of colubrid snakes. We suggest that transmutation in colubrid snakes is an adaptation to diurnality and is compensating for the loss of RH2 by establishing spectral sensitivity in a range where the existing SWS1 and LWS are not sensitive, and possibly establishing trichromatic colour vision. Perhaps transmutation in colubrid snakes may have contributed to the widespread success of the snake family across such a vast range of ecologies and lifestyle. Ultimately, future work investigating the functional effects of transmutation, from behavioral to molecular, will reveal the significance of rod-to-cone transmutation in colubrid snakes.

104

105

2.6 - REFERENCES

Ankel-Simons, F. and Rasmussen, D. T. (2008). Diurnality, nocturnality, and the evolution

of primate visual systems. Am. J. Phys. Anthropol. 137, 100–117.

Asenjo, A. B., Rim, J. and Oprian, D. D. (1994). Molecular determinants of human

red/green color discrimination. Neuron 12, 1131–1138.

Barlow, H. B. (1964). Dark-adaptation: a new hypothesis. Vision Research 4, 47–58.

Baylor, D. A. and Fettiplace, R. (1977). Transmission from photoreceptors to ganglion cells

in turtle retina. J. Physiol. (Lond.) 271, 391–424.

Bickelmann, C., Morrow, J. M., Müller, J. and Chang, B. S. W. (2012). Functional

characterization of the rod visual pigment of the echidna (Tachyglossus aculeatus), a

basal mammal. Vis. Neurosci. 29, 1–7.

Bowmaker, J. K. (2008). Evolution of vertebrate visual pigments. Vision Research 48,

2022–2041.

Calvert, P. D., Strissel, K. J., Schiesser, W. E., Pugh, E. N., Jr and Arshavsky, V. Y.

(2006). Light-driven translocation of signaling proteins in vertebrate photoreceptors.

Trends in Cell Biology 16, 560–568.

Caprette, C. L. (2005). Conquering the Cold Shudder: The Origin and Evolution of Snake

Eyes. 1–122.

Carvalho, L. S., Davies, W. L., Robinson, P. R. and Hunt, D. M. (2011). Spectral tuning

and evolution of primate short-wavelength-sensitive visual pigments. Proc. Biol. Sci.

106

279, rspb20110782–393.

Chen, J., Wu, M., Sezate, S. A. and McGinnis, J. F. (2007). Light Threshold–Controlled

Cone α-Transducin Translocation. Invest. Ophthalmol. Vis. Sci. 48, 3350–6.

Chen, M.-H., Kuemmel, C., Birge, R. R. and Knox, B. E. (2012). Rapid Release of Retinal

from a Cone Visual Pigment following Photoactivation. Biochemistry 51, 4117–4125.

Cortesi, F., Musilová, Z., Stieb, S. M., Hart, N. S., Siebeck, U. E., Malmstrøm, M.,

Tørresen, O. K., Jentoft, S., Cheney, K. L., Marshall, N. J., et al. (2015). Ancestral

duplications and highly dynamic opsin gene evolution in percomorph fishes. Proc. Natl.

Acad. Sci. U.S.A. 112, 1493–1498.

Cowan, C. W., Fariss, R. N., Sokal, I., Palczewski, K. and Wensel, T. G. (1998). High

expression levels in cones of RGS9, the predominant GTPase accelerating protein of

rods. Proc. Natl. Acad. Sci. U.S.A. 95, 5351–5356.

Crescitelli, F. (1956). The nature of the gecko visual pigment. J. Gen. Physiol. 40, 217–231.

Dartnall, H. (1968). The photosensitivities of visual pigments in the presence of

hydroxylamine. Vision Research 8, 339–358.

Das, J., Crouch, R. K., Ma, J.-X., Oprian, D. D. and Kono, M. (2004). Role of the 9-

Methyl Group of Retinal in Cone Visual Pigments †. Biochemistry 43, 5532–5538.

Das, S. R., Bhardwaj, N., Kjeldbye, H. and Gouras, P. (1992). Muller cells of chicken

retina synthesize 11-cis-retinol. Biochem J 285 ( Pt 3), 907–913.

107

Davies, W. L., Cowing, J. A., Bowmaker, J. K., Carvalho, L. S., Gower, D. J. and Hunt,

D. M. (2009). Shedding Light on Serpent Sight: The Visual Pigments of Henophidian

Snakes. Journal of Neuroscience 29, 7519–7525.

Daw, N. W., Jensen, R. J. and Brunken, W. J. (1990). Rod pathways in mammalian

retinae. Trends in Neurosciences 13, 110–115.

Diller, L. V. and Wallace, R. L. (1996). Comparative ecology of two snake species

(Crotalus viridis and Pituophis melanoleucus) in Southwestern Idaho. Herpetologica 52,

343–360.

Dodt, E. and Walther, J. B. (1958). [Spectral sensitivity and the threshold of gecko eyes;

electroretinographical studies on Hemidactylus turcicus & Tarentola mauritanica.].

Pflugers Archiv.

Drummond, H. (1985). The role of vision in the predatory behaviour of natricine snakes.

Animal Behaviour 33, 206–215.

Dungan, S. Z., Kosyakov, A. and Chang, B. S. W. (2016). Spectral Tuning of Killer Whale

(Orcinus orca) Rhodopsin: Evidence for Positive Selection and Functional Adaptation in

a Cetacean Visual Pigment. Mol. Biol. Evol. 33, 323–336.

Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high

throughput. Nucleic Acids Res 32, 1792–1797.

Elias, R. V., Sezate, S. S., Cao, W. and McGinnis, J. F. (2004). Temporal kinetics of the

light/dark translocation and compartmentation of arrestin and alpha-transducin in mouse

108

photoreceptor cells. Mol. Vis. 10, 672–681.

Fekete, D. M. and Barnstable, C. J. (1983). The subcellular localization of rat

photoreceptor-specific antigens. J. Neurocytol. 12, 785–803.

Gerald, G. W., Bailey, M. A. and Holmes, J. N. (2006). Movements and activity range

sizes of Northern Pinesnakes (Pituophis melanoleucus melanoleucus) in Middle

Tennessee. Journal of herpetology 40, 503–510.

Govardovskii, V. I., Fyhrquist, N., Reuter, T., Kuzmin, D. G. and Donner, K. (2000). In

search of the visual pigment template. Vis. Neurosci. 17, 509–528.

Gregg, R. G., McCall, M. A. and Massey, S. C. (2013). Chapter 15 - Function and

Anatomy of the Mammalian Retina. Fifth Edition. Elsevier Inc.

Harosi, F. I. and Novales Flamarique, I. (2012). Functional significance of the taper of

vertebrate cone photoreceptors. J. Gen. Physiol. 139, 159–187.

Hart, N. S., Coimbra, J. P., Collin, S. P. and Westhoff, G. (2012). Photoreceptor types,

visual pigments, and topographic specializations in the retinas of hydrophiid sea snakes.

J. Comp. Neurol. 520, 1246–1261.

Hauser, F. E., van Hazel, I. and Chang, B. S. W. (2014). Spectral tuning in vertebrate short

wavelength-sensitive 1 (SWS1) visual pigments: Can wavelength sensitivity be inferred

from sequence data? J. Exp. Zool. B Mol. Dev. Evol. 322, 529–539.

Heesy, C. P. and Ross, C. F. (2001). Evolution of activity patterns and chromatic vision in

primates: morphometrics, genetics and cladistics. Journal of Human Evolution 40, 111–

109

149.

Hibbard, E. and Lavergne, J. (1972). Morphology of the retina of the sea-snake, Pelamis

platurus. J. Anat. 112, 125–136.

Hicks, D. and Barnstable, C. J. (1987). Different rhodopsin monoclonal antibodies reveal

different binding patterns on developing and adult rat retina. Journal of Histochemistry

& Cytochemistry 35, 1317–1328.

Himes, J. G. (2001). Burrowing ecology of the rare and elusive Louisiana pine snake,

Pituophis ruthveni (Serpentes : Colubridae). Amphibia-Reptilia 22, 91–101.

Imai, H., Kojima, D., Oura, T., Tachibanaki, S., Terakita, A. and Shichida, Y. (1997).

Single amino acid residue as a functional determinant of rod and cone visual pigments.

Proc. Natl. Acad. Sci. U.S.A. 94, 2322–2326.

Imai, H., Kuwayama, S., Onishi, A., Morizumi, T., Chisaka, O. and Shichida, Y. (2005).

Molecular properties of rod and cone visual pigments from purified chicken cone

pigments to mouse rhodopsin in situ. Photochem. Photobiol. Sci. 4, 667–8.

Jacobs, G. H., Fenwick, J. A., Crognale, M. A. and Deegan, J. F., II (1992). The all-cone

retina of the garter snake: spectral mechanisms and photopigment. J. Comp. Physiol. A

Neuroethol. Sens. Neural. Behav. Physiol. 170, 701–707.

Karnik, S. S. and Khorana, H. G. (1990). Assembly of functional rhodopsin requires a

disulfide bond between cysteine residues 110 and 187. J Biol Chem 265, 17520–17524.

Karnik, S. S., Sakmar, T. P., Chen, H. B. and Khorana, H. G. (1988). Cysteine residues

110

110 and 187 are essential for the formation of correct structure in bovine rhodopsin.

Proc. Natl. Acad. Sci. U.S.A. 85, 8459–8463.

Kawamura, S. and Yokoyama, S. (1998). Functional characterization of visual and

nonvisual pigments of American chameleon (Anolis carolinensis). Vision Research 38,

37–44.

Kito, Y., Suzuki, T., Azuma, M. and Sekoguti, Y. (1968). Absorption spectrum of

rhodopsin denatured with acid. Nature 218, 955–957.

Kojima, D., Okano, T., Fukada, Y., Shichida, Y., Yoshizawa, T. and Ebrey, T. G.

(1992). Cone visual pigments are present in gecko rod cells. Proc. Natl. Acad. Sci. U.S.A.

89, 6841–6845.

Kuwayama, S., Imai, H., Hirano, T., Terakita, A. and Shichida, Y. (2002). Conserved

Proline Residue at Position 189 in Cone Visual Pigments as a Determinant of Molecular

Properties Different from Rhodopsins†. American Chemical Society.

Lamb, T. D. (2013). Progress in Retinal and Eye Research. Prog Retin Eye Res 36, 52–119.

Larhammar, D., Nordström, K. and Larsson, T. A. (2009). Evolution of vertebrate rod

and cone phototransduction genes. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 364, 2867–

2880.

Ma, J. X., Kono, M., Xu, L., Das, J., Ryan, J. C., Hazard, E. S., Oprian, D. D. and

Crouch, R. K. (2001). Salamander UV cone pigment: sequence, expression, and spectral

properties. Vis. Neurosci. 18, 393–399.

111

Makino, C. L., Wen, X.-H., Michaud, N. A., Covington, H. I., DiBenedetto, E., Hamm,

H. E., Lem, J. and Caruso, G. (2012). Rhodopsin Expression Level Affects Rod Outer

Segment Morphology and Photoresponse Kinetics. PLoS ONE 7, e37832–7.

McDevitt, D. S., Brahma, S. K., Jeanny, J.-C. and Hicks, D. (1993). Presence and foveal

enrichment of rod opsin in the “all cone” retina of the American chameleon. The

Anatomical Record 237, 299–307.

McKibbin, C., Toye, A. M., Reeves, P. J., Khorana, H. G., Edwards, P. C., Villa, C. and

Booth, P. J. (2007). Opsin stability and folding: The role of Cys185 and abnormal

disulfide bond formation in the intradiscal domain. Journal of Molecular Biology 374,

1309–1318.

Morrow, J. M. and Chang, B. S. W. (2015). Comparative Mutagenesis Studies of Retinal

Release in Light-Activated Zebrafish Rhodopsin Using Fluorescence Spectroscopy.

Biochemistry 54, 4507–4518.

Morrow, J. M. and Chang, B. S. W. (2010). The p1D4-hrGFP II expression vector: a tool

for expressing and purifying visual pigments and other G protein-coupled receptors.

Plasmid 64, 162–169.

Morrow, J. M., Lazic, S. and Chang, B. S. W. (2011). A novel rhodopsin-like gene

expressed in zebrafish retina. Vis. Neurosci. 28, 325–335.

Osborne, N. N., Safa, R. and Nash, M. S. (1999). Photoreceptors are preferentially affected

in the rat retina following permanent occlusion of the carotid arteries. Vision Research

39, 3995–4002.

112

Palczewski, K., Kumasaka, T., Hori, T., Behnke, C. A., Motoshima, H., Fox, B. A., Le

Trong, I., Teller, D. C., Okada, T., Stenkamp, R. E., et al. (2000). Crystal structure of

rhodopsin: A G protein-coupled receptor. Science 289, 739–745.

Piechnick, R., Ritter, E., Hildebrand, P. W., Ernst, O. P., Scheerer, P., Hofmann, K. P.

and Heck, M. (2012). Effect of channel mutations on the uptake and release of the

retinal ligand in opsin. Proc. Natl. Acad. Sci. U.S.A. 109, 5247–5252.

Rakshit, T. and Park, P. S. H. (2015). Impact of Reduced Rhodopsin Expression on the

Structure of Rod Outer Segment Disc Membranes. Biochemistry 54, 2885–2894.

Ronquist, F. and Huelsenbeck, J. P. (2003). MrBayes 3: Bayesian phylogenetic inference

under mixed models. Bioinformatics 19, 1572–1574.

Rosenzweig, D. H., Nair, K. S., Wei, J., Wang, Q., Garwin, G., Saari, J. C., Chen, C. K.,

Smrcka, A. V., Swaroop, A., Lem, J., et al. (2007). Subunit Dissociation and Diffusion

Determine the Subcellular Localization of Rod and Cone Transducins. Journal of

Neuroscience 27, 5484–5494.

Röll, B. (2001). Gecko vision - retinal organization, foveae and implications for binocular

vision. Vision Research 41, 2043–2056.

Sakami, S., Kolesnikov, A. V., Kefalov, V. J. and Palczewski, K. (2014). P23H opsin

knock-in mice reveal a novel step in retinal rod disc morphogenesis. Hum. Mol. Genet.

23, 1723–1741.

Sakmar, T. P., Franke, R. R. and Khorana, H. G. (1989). Glutamic acid-113 serves as the

113

retinylidene Schiff base counterion in bovine rhodopsin. Proc. Natl. Acad. Sci. U.S.A. 86,

8309–8313.

Sakmar, T. P., Menon, S. T., Marin, E. P. and Awad, E. S. (2002). Rhodopsin: insights

from recent structural studies. Annu Rev Biophys Biomol Struct 31, 443–484.

Sakurai, K., Onishi, A., Imai, H., Chisaka, O., Ueda, Y., Usukura, J., Nakatani, K. and

Shichida, Y. (2007). Physiological properties of rod photoreceptor cells in green-

sensitive cone pigment knock-in mice. J. Gen. Physiol. 130, 21–40.

Schott, R. K., Müller, J., Yang, C. G. Y., Bhattacharyya, N., Chan, N., Xu, M., Morrow,

J. M., Ghenu, A.-H., Loew, E. R., Tropepe, V., et al. (2016). Evolutionary

transformation of rod photoreceptors in the all-cone retina of a diurnal garter snake. Proc

Natl Acad Sci USA 113, 356–361.

Sillman, A. J., Govardovskii, V. I., Röhlich, P., Southard, J. A. and Loew, E. R. (1997).

The photoreceptors and visual pigments of the garter snake (Thamnophis sirtalis): a

microspectrophotometric, scanning electron microscopic and immunocytochemical

study. J Comp Physiol A 181, 89–101.

Sillman, A. J., Johnson, J. L. and Loew, E. R. (2001). Retinal photoreceptors and visual

pigments in Boa constrictor imperator. J. Exp. Zool. 290, 359–365.

Simões, B. F., Sampaio, F. L., Jared, C., Antoniazzi, M. M., Loew, E. R., Bowmaker, J.

K., Rodriguez, A., Hart, N. S., Hunt, D. M., Partridge, J. C., et al. (2015). Visual

system evolution and the nature of the ancestral snake. J. Evol. Biol. 28, 1309–1320.

114

Simões, B. F., Sampaio, F. L., Loew, E. R., Sanders, K. L., Fisher, R. N., Hart, N. S.,

Hunt, D. M., Partridge, J. C. and Gower, D. J. (2016). Multiple rod–cone and cone–

rod photoreceptor transmutations in snakes: evidence from visual opsin gene expression.

Proc. Biol. Sci. 283, 20152624–8.

Starace, D. M. and Knox, B. E. (1997). Activation of transducin by a Xenopus short

wavelength visual pigment. J Biol Chem 272, 1095–1100.

Stull, O. G. (1940). Variations and Relationship in the Snake of the Genus Pituophis, United

States.. National Museum. Washington, DC. Smithsonian Institution. Bulletin.

Tachibanaki, S., Arinobu, D., Shimauchi-Matsukawa, Y., Tsushima, S. and Kawamura,

(2005). Highly effective phosphorylation by G protein-coupled receptor kinase 7 of light-

activated visual pigment in cones. Proc. Natl. Acad. Sci. U.S.A. 102, 9329–9334.

Taniguchi, Y., Hisatomi, O., Yoshida, M. and Tokunaga, F. (1999). Evolution of visual

pigments in geckos. FEBS Lett. 445, 36–40.

Tansley, K. (1959). The retina of two nocturnal geckos Hemidactylus turcicus and Tarentola

mauritanica. Pflugers Arch Gesamte Physiol Menschen Tiere 268, 213–220.

Tansley, K. (1961). The retina of a diurnal gecko, Phelsuma madagascariensis longinsulae.

Pflugers Arch Gesamte Physiol Menschen Tiere 272, 262–269.

Tansley, K. (1964). The gecko retina. Vision Research 4, 33–IN14.

115

Underwood, G. (1970). Bioogy of the Reptilia. van Hazel, I., Dungan, S. Z., Hauser, F. E., Morrow, J. M., Endler, J. A. and Chang, B.

S. W. (2016). A comparative study of rhodopsin function in the great bowerbird

(Ptilonorhynchus nuchalis): Spectral tuning and light-activated kinetics. Protein Science

n/a–n/a. van Hazel, I., Sabouhanian, A., Day, L., Endler, J. A. and Chang, B. S. W. (2013).

Functional characterization of spectral tuning mechanisms in the great bowerbird short-

wavelength sensitive visual pigment (SWS1), and the origins of UV/violet vision in

passerines and parrots. BMC Evol. Biol. 13, 250.

Vogalis, F., Shiraki, T., Kojima, D., Wada, Y., Nishiwaki, Y., Jarvinen, J. L. P.,

Sugiyama, J., Kawakami, K., Masai, I., Kawamura, S., et al. (2011). Ectopic

expression of cone- specific G-protein-coupled receptor kinase GRK7 in zebrafish rods

leads to lower photosensitivity and altered responses. J. Physiol. (Lond.) 589, 2321–

2348.

Wada, Y., Sugiyama, J., Okano, T. and Fukada, Y. (2006). GRK1 and GRK7: Unique

cellular distribution and widely different activities of opsin phosphorylation in the

zebrafish rods and cones. Journal of Neurochemistry 98, 824–837.

Wald, G., Brown, P. K. and Smith, P. H. (1955). Iodopsin. J. Gen. Physiol. 38, 623–681.

Walls, G. L. (1942). The vertebrate eye and its adaptive radiation [by] Gordon Lynn Walls.

Bloomfield Hills, Mich.,: Cranbrook Institute of Science.

116

Wen, X.-H., Shen, L., Brush, R. S., Michaud, N., Al-Ubaidi, M. R., Gurevich, V. V.,

Hamm, H. E., Lem, J., DiBenedetto, E., Anderson, R. E., et al. (2009).

Overexpression of Rhodopsin Alters the Structure and Photoresponse of Rod

Photoreceptors. Biophys. J. 96, 939–950.

Yokoyama, S. (2000). Molecular evolution of vertebrate visual pigments. Prog Retin Eye

Res 19, 385–419.

Zhang, X., Wensel, T. G. and Kraft, T. W. (2003). GTPase regulators and photoresponses

in cones of the eastern chipmunk. J. Neurosci. 23, 1287–1297.

Zhang, X., Wensel, T. G. and Yuan, C. (2006). Tokay Gecko Photoreceptors Achieve Rod-

Like Physiology with Cone-Like Proteins†. Photochem. Photobiol. 82, 1452.

117

2.7 - SUPPLEMENTAL INFORMATION

Table S2.1. Degenerate Primers designed for sequencing opsin genes from genomic DNA

Opsin Primer Sequence (5’-3’) Source

RH1 SquamR1_1F AAGGAGTCTGARTCIACICARAARGC This paper

RH1 SquamR1_972R GCGGAACTGTCGATTCATRAAIACRTADAT This paper

LWS DIAPLMF1 AAGCGTATTYAYTTAYACCRACASCAACAA Davis et al., 2009

LWS DIAPLMR1 CATCCTBGACACYTCCYTCTCVGCCTTCTG’ Davis et al., 2009

LWS PM_LWS3’GW_877F TCTGGCAGCTTCCCTGCCTGCCTTCTT This paper

LWS PM_LWS3’GW_877FR GTAATACGACTCACTATAGGGC This paper

LWS PM_LWS3’GW_865F TGCCTTTCACCCTCTGGCAGCTTCCCT This paper

LWS PM_LWS3’GW_865FR GTAATACGACTCACTATAGGGC This paper

LWS PM_LWS3’GW_905F GCAAAAAGCGCCACCATTTACAACCCA This paper

LWS PM_LWS3’GW_905FR CGGGCTGGTGCGGAACTGTCGATTCAT This paper

LWS PM_LWS3’GW_936F TATACGTCTTCATGAATCGACAGTCCG This paper

LWS PM_LWS3’GW_936FR ACTATAGGGCACGCGTGGT This paper

LWS PM_LWS5’GW_212R GTAATACGACTCACTATAGGGC This paper

LWS PM_LWS5’GW_212RR TTGGCTGTGGCCACCAATACCAAACCA This paper

LWS PM_LWS5’GW_210R GTAATACGACTCACTATAGGGC This paper

LWS PM_LWS5’GW_210RR GGCTGTGGCCACCAATACCAAACCATT This paper

LWS PM_LWS5’GW_90R GTAATACGACTCACTATAGGGC This paper

LWS PM_LWS5’GW_90RR AGGGTCACGGGTATTGTTGCTGTTGGT This paper

LWS PM_LWS5’GW_89R GTAATACGACTCACTATAGGGC This paper

LWS PM_LWS5’GW_89RR GGGTCACGGGTATTGTTGCTGTTGGTG This paper

SWS1 SquamS1_84F TCCTCGCCTTCGAACGATATRTSGTSATCT This paper

SWS1 SquamS1_857R CATCATCCACTTTYTTSCCRAASAGCTGCA This paper

SWS1 PM_S13’GW_478F ATGTACATGGTGAACAACCCTCAGCAC This paper

118

SWS1 PM_S13’GW_478FR GTAATACGACTCACTATAGGGC This paper

SWS1 PM_S13’GW_477F CATGTACATGGTGAACAACCCTCAGCA This paper

SWS1 PM_S13’GW_477FR GTAATACGACTCACTATAGGGC This paper

SWS1 PM_S13’GW_519F CTTGGTCACCATCCCTGCCTTCTTC This paper

SWS1 PM_S13’GW_519FR GTAATACGACTCACTATAGGGC This paper

SWS1 PM_S13’GW_522F GGTCACCATCCCTGCCTTCTTCTCCAA This paper

SWS1 PM_S13’GW_522FR ACTATAGGGCACGCGTGGT This paper

SWS1 PM_S15’GW_128R GTAATACGACTCACTATAGGGC This paper

SWS1 PM_S15’GW_128RR ACTACCACAGCATGTTTGGAGTGGAAA This paper

SWS1 PM_S15’GW_120R GTAATACGACTCACTATAGGGC This paper

SWS1 PM_S15’GW_120RR AGCATGTTTGGAGTGGAAACGGAAGT This paper

SWS1 PM_S15’GW_99R ACTATAGGGCACGCGTGGT This paper

SWS1 PM_S15’GW_99RR GAAGTTCCCCAGCGGCTTGCAGATCAC This paper

SWS1 PM_S15’GW_95R ACTATAGGGCACGCGTGGT This paper

SWS1 PM_S15’GW_95RR TTCCCCAGCGGCTTGCAGATCACGATA This paper

119

Table S2.2. List of sequences used in the phylogenetic analyses of opsin genes

RH1 LWS SWS

Amerotyphlops brongersmianus KR336737 ------

Amphisbaena alba KR336729 KR336705 KR336720

Amphisbaena infraorbitale KR336730 KR336704 KR336719

Anilius scytale KR336736 ------

U08131 Anolis carolinensis NM_001291387 AF134194 XM_008103916

Arizona elegans KU324006 KU323986 KU323997

Atractus flammigerus KR336740 KR336712 KR336726

Bachia flavescens KR336731 KR336703 KR336715

Epictia collaris KR336735 ------

Feylinia KR336742 KR336714 KR336717

Gekko gecko ------M92036 AY024356

Gekko japonicus ------XM_015415465 XM_015427348

Hydrophis peronii KU324001 KU323990 KU323991

Hypsiglena jani KU324007 KU323988 KU323998

Iguana iguana ------AB626972

Lampropeltis californiae KU324004 KU323987 KU323992

Liotyphlops beui KR336734 ------

Melanoseps occidentalis KR336743 KR336713 KR336718

Natrix maura KU324002 KU323982 KU323993

Notechis scutatus KU324000 KU323989 KU323999

Ophiodes striatus KR336732 KR336708 KR336716

Phelsuma madagascariensis ------AF074043 AF074045

Phyllorhynchus decurtatus ------KU323985 KU323996

Pituophis melanoleucus xxxxxx xxxxxx xxxxxx

Polemon collaris KR336739 KR336710 KR336724

120

Protobothrops mucrosquamatus XM_015823472 XM_015812260 XM_015825841

Pseustes poecilonotus KR336741 KR336711 KR336725

Python bivittatus XM_007423262 XM_007420519 XM_007441636

Python regius FJ497236 FJ497238 FJ497237

Takydromus sexlineatus KR336727 KR336707 KR336722

Telescopus fallax KU324005 KU323984 KU323995

Thamnophis proximus KU306726 KU306727 KU306728

XM_014075668 Thamnophis sirtalis XM_014059138 XM_014068735 KU32399

Tropidophis feicki KR336738 KR336709 KR336723

Typhlophis squamosus KR336733 ------

Uta stansburiana DQ100323 DQ129869 DQ100325

Xenopeltis unicolor FJ497233 FJ497235 FJ497234

121

Melanoseps occidentalis 1 Feylinia Amerotyphlops brongersmianus 1Liotyphlops beui 1 Epictia collaris 1 Typhlophis squamosus Anilius scytale Arizona elegans 1 0.63 Pituophis melanoleucus 0.93 Pseustes poecilonotus

1 Telescopus fallax Atractus flammigerus 1 1 Lampropeltis californiae Hypsiglena jani 1 Hydrophis peronii 1 0.9 Notechis scutatus Polemon collaris 1 1 Natrix maura 1 0.7 Thamnophis proximus 1 0.97 Thamnophis sirtalis Protobothrops mucrosquamatus 1 Xenopeltis unicolor Python bivittatus 1 Python regius 0.83 Tropidophis feicki Amphisbaena alba 0.99 0.98 Amphisbaena infraorbitale Takydromus sexlineatus 1 Anolis carolinensis 1 0.5 Uta stansburiana Ophiodes striatus Bachia flavescens

0.04

Figure S2.1. Rhodopsin gene tree estimated using Bayesian inference illustrating the position of Pituophis melanoleucus RH1. Numbers at the nodes are posterior probability percentages. Generated by Ryan K. Schott.

122

Phelsuma madagascariensis 1 Gekko gecko 1 Gekko japonicus Amphisbaena alba 1Amphisbaena infraorbitale Anolis carolinensis 0.85 Iguana iguana 0.91 0.71 Uta stansburiana Takydromus sexlineatus Arizona elegans 1 1 Pituophis melanoleucus Lampropeltis californiae 1 Phyllorhynchus decurtatus 1 0.98 Pseustes poecilonotus Telescopus fallax Hydrophis peronii 1 0.74 0.99 Notechis scutatus Polemon collaris Natrix maura 1 Thamnophis proximus 1 1 1 Thamnophis sirtalis Atractus flammigerus 1 Hypsiglena jani

1 Protobothrops mucrosquamatus Python bivittatus 1 1 Python regius 0.52 0.97 Xenopeltis unicolor Tropidophis feicki Feylinia 1Melanoseps occidentalis Bachia flavescens Ophiodes striatus

0.04

Figure S2.2. SWS1 gene tree estimated using Bayesian inference illustrating the position of Pituophis melanoleucus SWS1. Numbers at the nodes are posterior probability percentages. Generated by Ryan K. Schott.

123

Phelsuma madagascariensis 1 Gekko gecko 1 Gekko japonicus Amphisbaena alba 1 0.68 Amphisbaena infraorbitale Takydromus sexlineatus 0.61 Anolis carolinensis (U08131) 1 0.6 Anolis carolinensis (XM008103916) Uta stansburiana Arizona elegans 1 Lampropeltis californiae 1 0.96 Pituophis melanoleucus Atractus flammigerus 1 0.99 Hypsiglena jani

0.86 Phyllorhynchus decurtatus 0.96 Telescopus fallax Pseustes poecilonotus 0.86 0.89 Natrix maura 1 Thamnophis proximus 1 1 0.99 Thamnophis sirtalis Polemon collaris Hydrophis peronii 0.98 1Notechis scutatus Feylinia 0.98 1 1 Protobothrops mucrosquamatus Python bivittatus 1 1 Python regius 0.98 Xenopeltis unicolor 0.99 Tropidophis feicki 0.91 Melanoseps occidentalis Ophiodes striatus Bachia flavescens

0.04

Figure S2.3. LWS gene tree estimated using Bayesian inference illustrating the position of Pituophis melanoleucus LWS. Numbers at the nodes are posterior probability percentages. Generated by Ryan K. Schott.

124

CHAPTER III: INVESTIGATING THE 11-CIS 3,4 DEHYDRORETINAL (A2) CHROMOPHORE IN DIVERSE VERTEBRATE RHODOPSINS REVEALS INSIGHTS INTO THE SPECTRAL AND NON-SPECTRAL ROLES OF THE CHROMOPHORE IN RHODOPSIN FUNCTION

Nihar Bhattacharya, Akimori Wada, Belinda S.W. Chang

Author contributions: NB and BSWC designed the study. AW synthesized A2 chromophore. NB performed experiments, data analysis and interpretation, with assistance from BSWC.

3.1 - ABSTRACT

Among vertebrates, visual pigments can utilize one of two light-sensitive, Vitamin A- based moieties, the well-studied 11-cis retinal (A1) chromophore or the lesser known 11-cis

3,4 dehydroretinal (A2) chromophore. A2 visual pigments can be found in some fishes, amphibians and reptiles, where the A2 chromophore is known to spectrally red-shift the maximal wavelength of absorption (lmax) relative to A1 visual pigments. However, in vitro studies of the spectral effects of A2 chromophore are rare and the effects of the A2 chromophore on the non-spectral functions of rhodopsin remain largely unknown. In this study, we heterologously expressed and purified rhodopsin apoprotein from 11 different vertebrates in vitro and regenerated each with both the A1 and the A2 chromophore, regardless of the native chromophore typically associated with each protein. Using UV- visible absorbance spectroscopy, we measured the lmax for each of the rhodopsin pigment pairs and found a lmax red-shift of approximately 17-20 nm upon regeneration with A2 chromophore, allowing us to model a mathematical relationship to estimate the magnitude of red-shift by the A2 chromophore. Additionally, we used fluorescence spectroscopy to functionally characterize the decay of the light-activated state in both A1 and A2 rhodopsins.

Our results showed that A2 rhodopsin decayed faster than the same opsin protein regenerated

125 with A1 chromophore, suggesting a shorter lived active state in rhodopsins when regenerated with the A2 chromophore. Arrhenius linear measurements revealed that the activation energy mediating the light-activated hydrolysis of the Schiff-base linking the all- trans chromophore to the light-activated apoprotein is similar in both A1 and A2 rhodopsins. This suggests that the accelerated retinal release may be due to a lowered affinity for all-trans A2 chromophore.

This study represents the first comprehensive characterization of vertebrate rhodopsin proteins with the A2 chromophore and highlights the critical nature of the chromophore in both spectral and non-spectral properties of rhodopsin.

126

3.2 - INTRODUCTION

The first step in the visual cascade occurs when opsin visual pigments in the photoreceptor of the eye absorb incoming light, enabling vertebrates to perceive light (Lamb,

2013). Vertebrate visual pigments consist of two components: the opsin apoprotein consisting of seven transmembrane helices, and a Vitamin A-derived chromophore covalently bound in a chromophore binding pocket within the apoprotein (Bownds, 1967;

Wald, 1968). Upon absorption of light of a maximal wavelength (λmax), the 11-cis chromophore isomerizes to initiate conformational shifts in the protein structure resulting in the activated state (Meta II) of the opsin (Ernst and Bartl, 2002). The light-activated Meta II state is the biologically active form that activates the downstream signaling transduction cascade within the outer segments of the photoreceptor cells (Ernst and Bartl, 2002) . Thus the chromophore is known to interact extensively with the opsin protein structure, and is thought to influence key functional properties of visual pigments (Bowmaker and Hunt,

2006; Liu et al., 2011; Schafer et al., 2016), such as the wavelength sensitivity of the opsin.

Among vertebrates, visual pigments can utilize two possible chromophores, of which the most common is 11-cis retinal (A1). A second, less common chromophore, 11-cis 3,4 dehydroretinal (A2), is utilized by some reptiles, amphibians and fishes (Ala-Laurila et al.,

2007; Kawamura and Yokoyama, 1998; Wald, 1939), and is known to shift spectral sensitivity towards longer wavelengths relative to A1 pigments and also increase the spectral bandwidth of absorption (Harosi, 1994). In natural systems, the A2 chromophore is incorporated into the visual system in different ways. There are species that exclusively use the A2 chromophore in their visual system (Kawamura and Yokoyama, 1998; Weadick et al.,

2012), while some other species can utilize both A1- and A2-containing opsins

127 simultaneously in varying ratios or even in different parts of the eye (summarized in Temple et al. 2011). Others may switch between the two chromophores depending on differing environments or migration or developmental stages (summarized in Temple et al. 2005).

The molecular structure of the A2 retinal chromophore differs from the A1 with the presence of an additional double bond in the beta-ionone ring, thus elongating the electron chain found along the length of the chromophore (Gillam et al., 1938) (Figure 3.1A, B). In an ethanol solution, the free 11-cis A1 and A2 chromophores spectrally absorb at 380 nm (A1) and 393 nm (A2) (Hubbard et al., 1971). However, when covalently bound to an opsin protein, the absorption wavelength can vary from the UV to the near infrared, depending on the electronic environment established by the protein sequence surrounding the chromophore

(Bowmaker and Hunt, 2006). The elongated electronic chain and the rigid beta-ionone ring of the A2 chromophore could potentially be interacting differently with the opsin protein sequence and the structurally stabilizing hydrogen bond networks which come into close proximity to the beta-ionone ring (Hofmann et al., 2009; Vogel et al., 2005). Thus both the identity of the chromophore and the specific protein sequence determine the spectral sensitivity of opsin, allowing for the specific tuning of the visual pigment λmax to potentially adapt to changing visual ecologies (Temple et al., 2005). Until recently, the biological significance of the chromophore switch had been assumed to be largely for spectral tuning, as the A2 chromophore is known to red-shift the absorption sensitivity of an opsin protein without having to modify the protein sequence (Shantz and Embree, 1946). Thus A2 visual pigment spectral sensitivity has mainly been studied in vivo using microspectrophotometry

(MSP) measurements of λmax (e.g. (Cowing et al., 2002; Harosi, 1994; Parry and

Bowmaker, 2000) and (Ala-Laurila et al., 2007)), dietary replacement studies to induce

128 chromophore switches (Shantz and Embree, 1946; Suzuki and Miyata, 1988), or ERG studies measuring the signalling of photoreceptors with A2 pigments (Ala-Laurila et al., 2003).

Although 11-cis A2 chromophore can be synthesized for use in in vitro experiments, in vitro

A2 studies remain quite rare (Miyagi et al., 2012; Terai et al., 2017), with most conducted using the A1 chromophore, under the assumption that the two chromophores are functionally equivalent (Hauser et al., 2017b; Kawamura and Yokoyama, 1998). Therefore when studying the visual systems of species utilizing the A2 chromophore, studies have employed a mathematical relationship to predict the λmax shift between chromophores (Bowmaker et al.,

2005; Carleton et al., 2008; Saarinen et al., 2012).

Since the discovery of the A2 chromophore in the vertebrate visual system, there have been several attempts at defining a mathematical relationship between the A1 and A2 λmax in visual pigments (Dartnall and Lythgoe, 1965; Harosi, 1994; Parry and Bowmaker, 2000;

Whitmore and Bowmaker, 1989), extrapolated from datasets of opsin λmax measurements from retinal extracts of animals utilizing both chromophores (Dartnall and Lythgoe, 1965;

Harosi, 1994; Whitmore and Bowmaker, 1989). This presents difficulties in that the opsin extracts typically have a mixture of both chromophores present, complicating the isolation of samples of opsins with only a single chromophore. Additionally, typical opsin/chromophore isolations from retinal extractions usually yield in only A1/A2 rhodopsin λmax measurements leading to a dearth of purified A2 cone opsin λmax data. Nevertheless, these studies have indicated that the magnitude of red-shift caused by the switch to A2 is dependent on the A1 λmax, as the longer the A1 λmax, the larger the magnitude of the red- shift (Whitmore and Bowmaker, 1989). While the earliest attempts at modelling the λmax shift were linear (Dartnall, 1968), recent modelling attempts have accommodated the

129 differential shift by using exponential or logarithmic equations to predict the A2 λmax for the entire range from UV- to long wavelength-sensitive visual pigments (Harosi, 1994; Parry and

Bowmaker, 2000). The accuracy of these newer mathematical models is limited by a smaller number of cone opsin A2 pigment measurements in the dataset. Thus, in the literature, predictive equations are used to approximate a range of potential λmax values when using another chromophore in order to accommodate the potential inaccuracy of predictive equations (e.g. (Carleton et al., 2005; Saarinen et al., 2012)).

There is also some evidence that non-spectral properties of visual pigments may also be associated with shifts in the chromophore. Among the opsins of the visual system, rhodopsin

(RHO) is the dim-light visual pigment found in the rod photoreceptors, and is characterized by incredible sensitivity (Baylor et al., 1979), thermal stability (Baylor et al., 1984), and ultrafast kinetics (Johnson et al., 2015; Schoenlein et al., 1991). A2 rhodopsin has been shown to have higher rates of spontaneous thermal activation in the dark compared to the equivalent A1 rhodopsin (Ala-Laurila et al., 2003), thus increasing the false detection of light, which suggests a lower sensitivity of visual systems relying on A2 rhodopsins for dim- light vision (Aho et al., 1988). This increase in spontaneous activation is thought to be a function of the λmax red-shift, as a redder wavelength of light corresponds to a lower energy photon requiring a lowered barrier of light activation of the cis-to-trans isomerization of the

A2 chromophore, increasing the probability of thermal energy activating the rhodopsin in the dark (Barlow, 1957; Gozem et al., 2012). However, thermal stability is only one non-spectral property of rhodopsin, other non-spectral functions, such as Meta II stability, have not yet been characterized in A2 rhodopsin, making our understanding of the role of the chromophore in rhodopsin incomplete.

130

In our study, using heterologous expression systems, we purified and characterized the

A1 and A2 rhodopsin from 11 different vertebrate species using absorbance and fluorescence spectroscopy. Using the data from the purified pigments we derived a mathematical relationship between the λmax values of rhodopsin A1-A2 pigment pairs. We find that, even with purified pigment, the relationship between the rhodopsin A1 and A2 λmax varied, suggesting a role of the protein in modulating the A1-A2 λmax shift. We also characterized the rate of retinal release from the light-activated rhodopsin A1-A2 pigment pairs and found that retinal released significantly faster in A2 rhodopsins, consistently across multiple species. To determine if this was due to the difference in activation energy for the hydrolysis of the Schiff base linkage, an Arrhenius plot of A2 bovine rhodopsin was constructed, demonstrating that the Ea was similar in both A1 and A2 pigments. This suggests that perhaps the structural and electronic differences between the two chromophores are differentially affecting the selectivity of active rhodopsin for the all-trans form of the chromophore, which may be significant enough to accelerate the retinal release of all-trans

A2 chromophore. Our study is the first comprehensive study of in vitro purified A2 rhodopsins from a diverse range of vertebrates and the first to functionally characterize the light activated form of A2 rhodopsin.

131

3.3 - MATERIALS AND METHODS In vitro expression and purification

Rhodopsin constructs in p1D4 (Morrow and Chang, 2010) were prepared as previously described (Bickelmann et al., 2012; Dungan et al., 2016; Hauser et al., 2017b; Luk et al., 2016; Morrow and Chang, 2015; Morrow et al., 2017). Briefly, full-length rhodopsin sequences were inserted into the p1D4-hrGFP II expression vector. HEK293T cells were transiently transfected with Lipofectamine 2000, media on the cells was changed 24 hours post-transfection and harvested after 48 hours and rinsed in PBS. A1 and A2 chromophores were reconstituted at a stock 5mM concentration in ethanol. Harvested cells were regenerated at a concentration of 5 µM retinal for 1.5 hours. A1 chromophore was generously provided by Dr. Rosalie Crouch (Medical University of South Carolina). 11-cis A2 chromophore was synthesized as previously described (Wada et al., 2008), and purity of chromophore was determined via NMR (Supplemental Figure S3.1). Regenerated cells were then solubilized in

1% N-dodecyl-D-maltoside and immunoaffinity purified with the 1D4 monoclonal antibody, as previously described. Purified rhodopsin samples were eluted with excess 1D4 peptide.

The UV-visible absorption spectra of the dark-state purified visual pigments were measured at 20ºC using a Cary4000 double-beam spectrophotometer (Varian). Pigments were light- bleached with white light generated by a fiber optic lamp (Dolan-Jenner) for 60s in order to measure light-activated spectra of the pigments. Difference spectra of purified protein was calculated by subtracting the light-activated spectra from the corresponding dark state spectra. All λmax values were calculated by fitting the absorbance spectra to A1 or A2 templates (Govardovskii et al., 2000). Purity of 11-cis A2 chromophore was verified via fit of A2 pigments to the A2 template (Govardovskii et al., 2000).

132

Fluorescence spectroscopy

The rate of retinal release from light-activated pigment was measured using a Cary

Eclipse fluorescence spectrophotometer (Varian) in accordance with a protocol modified from (Farrens and Khorana, 1995). Fluorescence measurements were then taken every 30 seconds with an excitation wavelength of 295nm and an emission wavelength of 330nm.

After 5 minutes of data collection, the purified rhodopsin samples were bleached for 30 seconds using a fiber optic lamp (Dolan-Jenner). The resulting data was fitted to an

-bx exponential rise to a maximum template (y= y0 + a (1 -e )) with half-life values calculated using the rate constant b (t1/2 = ln(2) / b). All curve fitting results in R-squared values greater than 0.9. Data for Arrhenius plots was collected at 16ºC, 20ºC, and 25ºC. The data was plotted with the reciprocal of the temperature on the x axis, and the natural logarithm of the retinal release on the y-axis. A linear regression line was fitted to the data and the slope of the line was used to calculate the activation energy (Ea) based on the Arrhenius equation (k =

A e-Ea/(RT)). The standard error of the estimated slope in the regression of 1/T vs. ln(k) was used to determine the error in activation energies for each chromophore. To determine if the activation energies were different between the two chromophores, a multiple-regression analysis was conducted using temperature, chromophore type, and an interaction term between temperature and chromophore type as independent variables. Chromophore type was inputted as a binary, 0 or 1, variable. If Ea values were different between A1 and A2, their estimated slopes would be significantly different as indicated by the interaction term.

However, there was no significant difference in activation energies for A1 and A2 (p<0.25)

Homology modelling and atomic modelling

133

Structural modelling of zebrafish and bowerbird rhodopsin was based on the dark-state bovine rhodopsin structure (PDB: 1U19) (Okada et al., 2004), using the homology modeling software Modeller (Eswar et al., 2002; Šali and Blundell, 1993). Minimization of the objective function generated several separate models, and the run with the lowest (most negative) discrete optimized protein energy (DOPE) score was chosen (Shen and Sali, 2006).

Distances were measured using Chimera (Pettersen et al., 2004).

Retinal (CID: 638015) and 3,4-dehydroretinal (CID: 5280866) molecular structures were acquired from PubChem. Electrostatic surface maps were created using Jmol: an open-source

Java viewer for chemical structures in 3D (http://www.jmol.org/).

134

3.4 - RESULTS

A2 chromophore can regenerate with rhodopsin apoprotein from multiple vertebrate species into functional rhodopsins with a red-shifted λmax

Rhodopsin apoprotein from 11 diverse vertebrate species were heterologously expressed in vitro and regenerated with either A1 or A2 chromophore, all forming functional protein irrespective of chromophore used (Table 3.1). Figure 3.1 shows the dark and light

UV-vis absorbance spectra of bovine rhodopsin regenerated with A1 (Figure 3.1C) and A2

(Figure 3.1D) chromophore. As expected, the A2 chromophore resulted in a red-shift of the

λmax of the dark state of bovine rhodopsin relative to the native A1 bovine pigment from

498.7 nm to 518.5 nm, a red-shift of 19.8 nm. Additionally, reconstitution with the A2 chromophore also red-shifted the light activated absorbance peak in A2 bovine rhodopsin by approximately the same amount.

To investigate the relationship between rhodopsin and the A1-A2 chromophores, we took a comparative approach, and expressed three mammalian rhodopsins -- human, echidna, and orca-- and reconstituted with both chromophores. All A1 λmax values agreed with previously reported measurements: human (494.1 nm; Figure 3.2B (Morrow et al., 2017)), echidna

(497.9 nm; Figure 3.2F, (Bickelmann et al., 2012)) and orca (486.2 nm; Figure 3.2C (Dungan et al., 2016)), while the corresponding A2 rhodopsin, which has never been measured before, showed varying degrees of red-shift: 19 nm in human (513.1 nm; Figure 3.2B), 18 nm in echidna (516nm; Figure 3.2F), and 17.9 nm in orca (504.1 nm; Figure 3.2C). To increase our sampling of native A1 rod pigments beyond mammals, we investigated red-shifts in rhodopsin proteins from two avian groups: Gallus gallus (chicken) and Ptilonorhychus

135 nuchalis (bowerbird). A1 λmax absorbance values of the chicken (503.4 nm; Figure 3.2D) and bowerbird (500.5 nm; Figure 3.2E) matched previously published figures (van Hazel et al., 2016), while the A2 pigments were red-shifted by 21.1 nm in chicken (524.5 nm; Figure

3.2D) and 19.8nm in bowerbird (520.3nm; Figure 3.2E). These results suggest that native rhodopsin pigment from vertebrates that exclusively use the A1 chromophore can still be consistently regenerated with A2 chromophore to produce functional red-shifted rhodopsins.

Since fish rhodopsins have evolved to use A1 or A2 chromophores depending on life history and visual ecology, we investigated the effects of both chromophores on five different fish rhodopsins. We expressed four exclusively A1 freshwater species (Danio rerio,

Paracottus jettelesi, Cottomocomephorus inermis, and Abyssocottus korotneffi) and a single freshwater A2 species (Crenicichla frenata). λmax values of all native A1 fish rhodopsins were consistent with previous findings (zebrafish (500.4 nm; Figure 3.2G(Morrow and

Chang, 2015), P. jettelesi (501.1 nm; Figure 3.2I; (Hunt et al., 1997; Luk et al., 2016)), C. inermis (496.6 nm; Figure 3.2H; (Hunt et al., 1997; Luk et al., 2016)), and A. korotneffi

(481.7 nm; Figure 3.2K; (Hope et al., 1997; Luk et al., 2016))). In contrast, native A2- containing C. frenata rhodopsin (517 nm; Figure 3.2J; (Weadick et al., 2012))) was red- shifted by 20.4 nm relative to previous λmax measurements with the A1 chromophore

(496.6 nm; Figure 3.2J; (Hauser et al., 2017a)). With the A2 chromophore, the zebrafish

λmax was shifted by 25.9 nm (526.3 nm; Figure 3.2G), P. jettelesi by 19.1 nm (520.2 nm;

Figure 3.2I; (Luk et al., 2016)), C. inermis by 20.4 nm (515.7 nm; Figure 3.2H; (Luk et al.,

2016)), and A. korotneffi by 17.2 nm (498.9 nm; Figure 3.2K; (Luk et al., 2016)). Taken together, these results suggest that A1/A2 chromophores can be interchanged across a wide range of vertebrate rhodopsins, despite only certain lineages utilizing A2 chromophores

136 natively. Plotting the λmax pairs with the A1 λmax on the x-axis and the A2 λmax on the y- axis reveals a linear relationship described by the equation λmaxA2 = 1.1634 * λmaxA1 -

61.591 with an R-squared value of 0.947 (Figure 3.3A).

Non-spectral functional effects of A2 rhodopsins vary across native A1 rhodopsins

We hypothesized that the A2 chromophore could potentially modulate other non-spectral properties of rhodopsin, such as the stability of the active Meta II state. The light-activated retinal release rates of A1 and A2 bovine rhodopsin at 20ºC showed that the A2 rhodopsin t1/2

(6.3 min +/- 0.43) was almost twice as fast as the A1 bovine rhodopsin t1/2 (12.37 min +/-

2.01). This represents the first time this assay has been done with an A2 pigment. These results suggest that the conformational selectivity of active state rhodopsin for all-trans retinal (ATR) may be decreased by the presence of the beta-ionone/double bond modifications present in the A2 chromophore. To determine if the trend of faster retinal release from rhodopsin was consistent across different species, the light-activated retinal release of zebrafish (an exclusively A1 rhodopsin) and C. frenata (an exclusively A2 rhodopsin) was measured with both chromophores at 20ºC (Figure 3.5). In both fishes, the

A2 rhodopsin released retinal faster than the A1 pigment. Zebrafish rhodopsin regenerated with A1 chromophore had a t1/2 of 7.98 +/- 0.9 min for retinal release (which is consistent with previously published data; (Morrow and Chang, 2015)), whereas the A2 zebrafish had a reaction half-life of 4.52 +/- 1.0 min. In the cichlid, the A2 rhodopsin had a half-life of 20.09

+/- 0.7 min, while the A1 cichlid rhodopsin took more than double the time with a t1/2 of

47.15 +/- 0.7 min, consistent with previously published data (Hauser et al., 2017a).

137

To ascertain if the difference in t1/2 observed was due to differences in the Schiff base hydrolysis activation energy, retinal release rates of bovine rhodopsin were measured at three different temperature points and used to construct an Arrhenius plot (Figure 3.4). In all runs, the A2 produced a noisier signal likely due to difference in pigment thermal stability and differences in extinction coefficients between A1 and A2 rhodopsins, as equivalent molar amounts of pigment produce a smaller signal in A2 rhodopsins.. At all temperature points, the A2 bovine rhodopsin reacted faster than the A1 bovine rhodopsin (Table 3.2, Figure

3.4A-B). The calculated Ea value for A1 bovine rhodopsin matches previously published data (Morrow and Chang, 2015) of 18.60 +/- 0.22 kcal. The calculated value for A2 pigments was 22.57 +/- 2.47 kcal. (Table 3.2, Figure 3.4C). There was no statistically significant difference (p-value of 0.25) in Ea for light-activated Schiff base hydrolysis between A1 and

A2 bovine rhodopsin, therefore the activation energy mediating the light-activated hydrolysis of the Schiff-base linkage is similar in both A1 and A2 pigments. This strongly suggests that the accelerated retinal release of A2 from both native A1 and A2 pigments is due to an increased conformational selectivity of Meta II active state rhodopsin for A2 all-trans chromophore. These results therefore raise the possibility that A2 chromophore usage may have also evolved to drive faster Meta II decay kinetics, which may be another functional mechanism explaining the distribution of A2 chromophore usage in natural systems.

138

Figure 3.1 - Molecular diagram of the (A) 11-cis retinal (A1) chromophore and the (B) 11-cis 3,4-dehydroretinal (A2) chromophore. The additional double bond on the beta- ionone ring is highlighted in red. (C) UV-Visible absorbance spectrum of bovine rhodopsin regenerated with A1 chromophore in the dark (black) and then light-bleached (red), inset shows the dark-light difference spectrum. (D) UV-Visible absorbance spectrum of bovine rhodopsin regenerated with A2 chromophore in the dark (black) and then light-bleached (red), inset shows the dark-light difference spectrum. The A2 bovine rhodopsin shows the characteristic red-shift in both the dark state and the light activated A2 rhodopsin in comparison to the equivalent A1 pigment.

139

Figure 3.2 - Normalized UV-Visible absorbance spectra of vertebrate rhodopsin with A1 (black) and A2 (red) chromophores with measured wavelength of maximal absorptions (λmax) for A1 (black text) and A2 (red text) pigments. Four mammals were expressed: (A)

140

Bovine, (B) Human, (C) Orca, (F) Echidna. Two avian species were expressed: (D) Chicken and (E) Bowerbird. Five species of fish were expressed: (G) Zebrafish, (H) Cottomocomephorus inermis, (I) Paracottus jettelesi, (J) Crenicichla frenata, and (K) Abyssocottus korotneffi.

141

Figure 3.3 - (A) A1 and A2 λmax pairs from 11 vertebrate rhodopsins expressed and purified in vitro. Linear regression produces the equation λmax(A2) = 1.1634*λmax(A1)- 61.591 with a R-squared value of 0.947. (B and C) show the λmax pairs from the current study and linear relationship compared to three previously determined absorbance relationships, Dartnall and Lythgoe (1965), Harosi (1994) and Parry (2000). (C) Data points were determined by taking the inverse of the λmax wavelength in centimeters and dividing by 1000.

142

Figure 3.4 - Light-activated retinal release of A1 and A2 bovine rhodopsin. Increase in fluorescence due to release of all-trans A1/A2 retinal following photoactivation in (A) A1 and (B) A2 bovine rhodopsin at 16ºC, 20ºC, and 25ºC. (A) Half-life of light-activated A1 retinal releases are 18.96 min at 16ºC, 12.37 min at 20ºC, and 6.79 min at 25ºC. (B) The half- life of A2 bovine rhodopsin are considerably faster at 12.11 min at 16ºC, 6.30 min at 20ºC, and 3.71 min at 25ºC. (C) An Arrhenius plot of the natural logarithm of retinal release rates of bovine A1 and A2 rhodopsin were used to estimate activation energies (Ea) based on the negative reciprocal of the slope of a linear regression line that best fit the data.

143

Figure 3.5 - Light activated retinal release of A1/A2 cichlid and zebrafish, demonstrating the acceleration of retinal release in A2 pigments. Half-lives of the A1 zebrafish was 7.98 min with the A2 zebrafish half-life faster at 4.52 min. Cichlid A1 had a half-life of 47.15 min, with the A2 cichlid pigment having a faster half-life at 20.09 min.

144

Figure 3.6 - Homology modelling of zebrafish and bowerbird rhodopsin. (A) Overlaid 3D structures of zebrafish (blue) and bowerbird (pink). Amino acid differences in dark blue (zebrafish) and dark pink (bowerbird) with side chains displayed. 11-cis retinal in lime green. (B) E122 sidechain distances from beta-ionone ring in zebrafish and bowerbird. (C) E181 and E113 side chain differences from the 11-cis bond and/or the Schiff base.

145

Figure 3.7 - Chromophore charge surface. (A) the electrostatic surface map of the all-trans A1 chromophore showing the distribution of charge across the surface of the chromophore. (B) the electrostatic surface map of the all-trans A2 chromophore showing the distribution of charge across the surface of the chromophore. The elongated electron chain shows additional negative charge on the beta-ionone ring of A2 chromophore. (C) Shape of beta-ionone ring bulk in A1 chromophore and (D) the A2 chromophore

146

Table 3.1 - Measured λmax values of A1 and A2 pairs of 11 expressed and purified vertebrate rhodopsins and the magnitude of the redshift. Included in the study were four mammals (Bovine, human, orca and echidna), two birds (chicken and bowerbird) and five fish species (Zebrafish, Pike cichlid, and three Lake Baikal sculpin species)

Species lmax(A1) lmax(A2) A2-A1

Bovine 498.7 nm 518.5 nm 19.8 (Bos taurus)

Human 494.1 nm 513.1 nm 19 (Homo sapiens)

Orca 486.2 nm 504.1 nm 17.9 (Orcinus orca)

Echidna 497.9 nm 516.0 nm 18.1 (Tachyglossus aculeatus)

Chicken 503.4 nm 524.5 nm 21.1 (Gallus gallus)

Great Bowerbird 500.5 nm 519.9 nm 19.4 (Ptilonorhynchus nuchalis)

Zebrafish 500.4 nm 526.3 nm 25.9 (Danio rerio)

Cichlid 496.6 nm 517.0 nm 20.4 (Crenicichla frenata)

Paracottus jettelesi 501.1 nm 520.2 nm 19.1

Cottomocomephorus inermis 495.3 nm 515.7 nm 20.4

Abyssocottus korotneffi 481.7 nm 498.9 nm 17.2

147

Table 3.2 - Arrhenius plot data. Using three different temperature points, an Arrhenius plot of the light activated retinal release was constructed. The negative reciprocal of the slope of a linear regression line was used to calculate the activated energy of the light-activated Schiff base hydrolysis in A1 and A2 bovine rhodopsin.

Bovine A1 Bovine A2

t1/2 (min, 16º C) 18.96 ± 2.71 12.11 ± 0.29

t1/2 (min, 20º C) 12.37 ± 2.01 6.30 ± 0.43

t1/2 (min, 25º C) 6.79 ± 1.14 3.71 ± 0.69

ln(k) - 16º C -7.397 -6.955

(lower, upper) (0.0815, 0.0753) (0.0143, 0.0141)

ln(k) - 20º C -6.967 -6.299

(lower, upper) (0.1029, 0.0933) (0.0424, 0.0407)

ln(k) - 25º C -6.424 -5.760

(lower, upper) (0.0673, 0.0631) (0.1018, 0.0924)

Equation of the line y = -9.3098x + 24.814 y = -11.358x + 32.388 (R2) (0.999) (0.987)

Ea (kcal) 18.60 ± 2.3 22.57 ± 3

A value 24.814 ± 3.65 32.388 ± 5.22

148

3.5 - DISCUSSION

In this study, we heterologously expressed and purified A1 and A2 rhodopsins from a diverse range of vertebrates in order to assess the effects of the chromophore on both spectral and non-spectral properties of rhodopsin. Using absorbance spectroscopy to measure the

λmax and fluorescence spectroscopy to assay rhodopsin active state stability, our study was the first to characterize a diverse dataset of in vitro purified A1 and A2 rhodopsins from 11 vertebrates. Our results show that, among rhodopsin proteins, the A1-A2 λmax relationship appears effectively linear, yet interesting deviations from this relationship reveal a possible role of the protein in modulating the λmax shift between chromophores. We also experimentally measured the retinal release rates and active state decay of A2 rhodopsin to provide the first functional in vitro characterization of an A2 pigment. Using fluorescence spectroscopy, we show that in A2 rhodopsin the all-trans chromophore is released faster after light bleach than A1 rhodopsins in multiple species, suggesting a shorter lived active state.

Arrhenius measurements revealed a similar activation energy of the Schiff-base hydrolysis of

A1 versus A2 rhodopsin, suggesting differences in affinity between the two chromophores mediating the retinal release rates. Our results of expressed A2 rhodopsin in vitro highlighted the role of the rhodopsin apoprotein in the magnitude of the λmax shift between chromophores, while revealing the conservation of activation energies of key functional processes in A1 and A2 rhodopsin proteins.

Vertebrate rhodopsin can be expressed and regenerated in vitro with both A1 or A2, regardless of native chromophore

149

The significance of an alternate chromophore for vertebrate opsins, and the possible effect it could have on non-spectral rhodopsin function has been somewhat ignored in the literature, with most A2 studies focusing on the spectral differences between A1 and A2 pigments. Although it has been known for decades that bovine rhodopsin and chicken cone opsins could be regenerated with A2 chromophore (Wald et al., 1953), there has been little investigation into the effects of A2 chromophore on A1 pigments following that discovery.

Until recently, most the studies involving A2 visual pigments have either been conducted in vivo (Ala-Laurila et al., 2007; Enright et al., 2015; Kefalov et al., 2003; Reuter et al., 1971;

Suzuki and Miyata, 1988; Temple et al., 2005; Toyama et al., 2008) or in vitro substituting the A1 in place of the A2 chromophore (Darden et al., 2003; Hauser et al., 2017b; Kawamura and Yokoyama, 1998; Starace and Knox, 1997; Sugawara et al., 2005). More recently, there has been increasing interest in reconstituting opsins with A2 chromophore, but these have been limited largely to native A2 pigments (Miyagi et al., 2012; Terai et al., 2017). In comparison, native A1 rhodopsins have been extensively studied, and are functionally well characterized (Karnik et al., 1988; Kaushal and Khorana, 1994; Palczewski et al., 2000;

Sakmar et al., 1989; Wald, 1968; Zhukovsky and Oprian, 1989). Of the 11 species of rhodopsin expressed in this study, 10 were exclusively A1 rhodopsin pigments which demonstrated no observable dysfunction when regenerated with A2 chromophore and appear to be functional rhodopsins. Future investigations of G protein activation or structural studies of native A1 rhodopsins regenerated with A2 chromophore should be conducted to fully verify the functionality of these pigments, though mammalian dietary replacement studies have demonstrated functional visual systems with A2 chromophore (Shantz and Embree,

1946; Suzuki and Miyata, 1988). With the demonstrated ease of expression of A2

150 rhodopsins, future in vitro studies of A2 pigments could be expanded to encompass cone pigments and other spectral and non-spectral effects of chromophore choice on the visual system.

Protein environment likely governs A2 λmax red-shift across vertebrate rhodopsins

It is known that the electrostatic and physicochemical properties of the opsin apoprotein environment surrounding the covalently bound retinal chromophore can modulate the λmax of the opsin (Wang et al., 2014). Relative to A1 chromophore, A2 red-shifts visual pigment

λmax due to increased conjugation in the electron chain, thereby lowering the isomerization energy (Luk et al., 2016). There have been several attempts to mathematically describe the relationship between the A1 λmax and the A2 λmax of a pigment pair since the discovery of

A2 visual pigments and their effect on the λmax (Dartnall and Lythgoe, 1965; Harosi, 1994;

Parry and Bowmaker, 2000; Whitmore and Bowmaker, 1989). All, except Parry et al (2000), mainly relied on data collected from species where a change in chromophore occurred seasonally or developmentally and the majority consist of pigment extracts from rod photoreceptors. Dartnall and Lythgoe (1965) was the first to describe a linear relationship between 16 measured A1 and A2 λmax values from teleost visual pigments, likely rhodopsins, with chicken A2 cone opsin (Wald, 1953) to expand the range of the dataset. In comparison to our data, the Darnall and Lythgoe (1965) relationship has a higher slope overestimating the A2 λmax for rhodopsin with an A1 λmax above 490nm and underestimating for those with A1 λmax below 490nm (Figure 3.3B). Soon after, it was realized that the magnitude of λmax shift between the pigment pairs increases in longer wavelength sensitive opsins, and decreases in shorter wavelength pigments, which suggests an exponential relationship (Bridges, 1966). First proposed by Whitmore and Bowmaker

151

(1989) with a limited dataset, and then with a larger, dataset in Harosi (1994), the exponential mathematical relationship proposed accommodate the increasing λmax shift for long wavelength pigments. However, the exponential equation was modelled with a dataset containing only minimal measurements of A1/A2 pigment pairs from short wavelength and long wavelength sensitive opsins. Our data from purified pigment pairs clusters mainly below the Harosi exponential curve, accurately predicting the two pigments with the most and the least red-shift in our dataset but not for those in the mid-range. (Figure 3.3C). We show that the red-shift caused by the A2 chromophore demonstrates a consistent trend with the bluest rhodopsins having the smallest shift (~17 nm) while the reddest rhodopsin had the most (~21 nm). The only exception to this was zebrafish, which had the biggest red-shift of

~25nm, which will be discussed further below. In Parry et al (2000), the entire goldfish opsin complement was measured in vivo, where it exclusively used the A2 chromophore, and then bleached and reconstituted with A1 chromophore and measured again. The described logarithmic relationship from this data slightly underestimates the A2 λmax in our rhodopsin dataset (Figure 3.3B).

The λmax relationship derived from our data is linear, but as it was developed using only rhodopsins, we propose that this relationship is most applicable to rhodopsin proteins.

However, we believe that there are caveats for the use of our modelled equation (or any other

λmax predictive equation) to estimate λmax, and that there are exceptions to this rule. For example, our data shows that although zebrafish A1 rhodopsin and bowerbird A1 rhodopsin have similar λmax measurements, when we reconstitute each apoprotein with A2 chromophore, the λmax red-shifts by 19.8 nm in bowerbird and 25.9 nm in zebrafish. This suggests that the A2 chromophore is possibly interacting with the apoprotein in a different

152 manner in each of these proteins. Comparing the rhodopsin sequence of bowerbird and zebrafish, there is a 22.6% difference in amino acid sequence between the two species

(Figure 3.6A). Thus, there exists many possible sites that could mediate the red-shift when reconstituted with 11-cis A2 chromophore between these two species, and there may be variation in the nature of the A1/A2 relationship among different vertebrate groups.

With respect to the zebrafish and bowerbird rhodopsins, three regions of interest may underlie the difference in the A2 λmax. Site 83 has been well studied in A1 rhodopsins and is known to modulate both spectral and non-spectral properties (Breikers et al., 2001; van Hazel et al., 2016). Zebrafish rhodopsin has the more common D83, while bowerbird has N83 which has been shown to blue-shift the λmax by 2-5 nm and slow retinal release rates

(Dungan et al., 2016; Hauser et al., 2017b; van Hazel et al., 2016). Since this site has a known effect on the protein-chromophore interaction, this difference in sequence between zebrafish and bowerbird could be causing the larger red-shift in zebrafish A2 rhodopsin. Yet among the other vertebrate rhodopsins in this study, five have N83 (echidna, orca, A. korotneffi, C. inermis, and C. frenata) and all five of those species do not have the extended red-shift exhibited by A2 zebrafish rhodopsin. This suggests that other sites besides site 83 may underlie differences in A2-mediated λmax shift.

Another site, 122, is known to be critical in differentiating rhodopsins from cone opsins

(Imai et al., 1997) and has been shown to interact with the beta-ionone ring of the chromophore (Hofmann et al., 2009; Vogel et al., 2005). Although the rhodopsins of the bowerbird and zebrafish both have the conserved residue E122, the amino acids differ at sites

123 and 124. Recent studies of these nearby residues suggest that they may also influence rhodopsin function, possibly through site 122 (Castiglione and Chang, in review; (Morrow

153 and Chang, 2015)). Zebrafish has two uncommon residues, M123 and G124, at two of these sites. Homology modelling suggests that E122 is slightly shifted away from the beta-ionone ring in zebrafish compared to bowerbird (Figure 3.6B), and mutagenesis at these sites in zebrafish have been shown to affect both the λmax and the retinal release of A1 zebrafish rhodopsin (Morrow and Chang, 2015). Similarly, natural variation between bowerbird and zebrafish rhodopsin also exists at several sites near the important rhodopsin residue E181, which is involved in stabilizing the Schiff base upon light activation (Yan et al., 2003), as well as in the vicinity of E113, which is the conserved counter ion for the Schiff base

(Sakmar et al., 1989). Homology modelling suggests that variation at these sites in zebrafish in comparison to bowerbird may induce movement in both E113 and E181 (Figure 3.6C).

Subtle shifting in these residues, especially those near the beta-ionone ring, may be involved in the increased red-shift in A2 zebrafish. Other residues near the beta-ionone ring are of special interest, as the A2 chromophore electron chain extends into the ring (Figure 3.7A, B) and therefore subtle charge modulations near the ring could affect the electronic chain across the entire chromophore in A2 pigments, but not in A1. Additionally, the presence of another double bond in the beta-ionone ring introduces rigidness to the structure, changing the shape of this area (Figure 3.7C, D). Mutagenesis studies targeting these variable sites could help elucidate the role of E113, E122, and E181 in mediating the red-shift in A2 rhodopsins.

Activated rhodopsin apoprotein appears to have a higher affinity for the A1 all-trans chromophore regardless of native chromophore

As in vitro studies of A2 pigments have been quite rare, all functional characterization of

A2 rhodopsin has been done in vivo (Ala-Laurila et al., 2007; Kefalov et al., 2003; Ma et al.,

2001; Shantz and Embree, 1946). Conversely, the majority of in vitro studies expressing and

154 characterizing opsins from species that exclusively use A2 chromophore has been done using the A1 chromophore instead (Hauser et al., 2017b; Kawamura and Yokoyama, 1998). In addition to characterizing the effects of chromophore on the λmax, we also investigated non- spectral effects of the A2 chromophore on rhodopsin, such as the rate of retinal release from the light-activated A1/A2 rhodopsin.

After light activation, rhodopsin undergoes several rapid conformational changes before reaching the active state, Meta II (Choe et al., 2011; Kibelbek et al., 1991). In this state, the all-trans chromophore is still covalently linked and the protein is in the active-conformation capable of activating the G protein transducin, leading to signal transduction (Kibelbek et al.,

1991). Eventually, the Schiff base covalent linkage between rhodopsin and the all-trans chromophore hydrolyzes, releasing the chromophore from the apoprotein allowing it to diffuse out of the protein. Afterwards, a new 11-cis chromophore enters and binds the apoprotein, thereby regenerating the dark-state (Hildebrand et al., 2009; Morrow and Chang,

2015; Piechnick et al., 2012). With purified pigment, we can use fluorescence spectroscopy to measure the rate of release of the all-trans chromophore from the protein after light- activation, and thereby infer the stability of the Meta II state (Schafer et al., 2016). Our results suggest that A2 rhodopsins have a significantly faster retinal release than A1 pigments, and this is consistent across multiple species.

It is currently thought that two factors determine the rate of release of all-trans retinal from the light-activated rhodopsin: The activation energy barrier to the Schiff base hydrolysis

(Piechnick et al., 2012), and the affinity of the apoprotein for the all-trans form of the chromophore (Schafer and Farrens, 2015). In order to isolate the cause of the difference in half-lives in A1-A2 rhodopsins, we calculated the Ea of the Schiff base linkage hydrolysis in

155

A1 and A2 bovine rhodopsin and determined that they were not significantly different. This suggests that the elongated electron chain in the A2 chromophore as compared to the A1 chromophore does not affect hydrolysis (Figure 3.7A, B). Any disruption to the extensive hydrogen bonding network that surrounds the Schiff base linkage has been shown to affect the activation energy of the hydrolysis (Janz and Farrens, 2004). The similar Ea suggests that the A2 chromophore interacts with this hydrogen bonding network in a similar manner to the

A1 chromophore around the Schiff base. Interestingly, it has been hypothesized that the A value of the Arrhenius equation should be determined by the electronic environment around the Schiff-base linkage, which should be identical between A1 and A2 pigments as the apoprotein is the same in both pigments (Ala-Laurila et al., 2004). Our data is the first to show that this may be true as the A values of A1 and A2 bovine rhodopsin are within one standard deviation of each other.

We suggest that the most probable explanation for the faster release rate is due to the different shape and electronic structure of the A2 chromophore, allowing it to diffuse more rapidly out the protein due to reduced steric hindrance (Figure 3.7C, D). Mutagenesis studies have shown that steric hindrance by amino acid side chains around the opening have been shown to affect the retinal release rates (Morrow and Chang, 2015; Piechnick et al., 2012).

The faster t1/2 in A2 pigments could also be explained if the protein had a lesser affinity for the A2 all-trans form via increases to MII conformational selectivity for all-trans retinal, which could also be due do the differing steric and electronic structure of the chromophore.

Specifically, recent studies have shown that an equilibrium develops between the released all-trans chromophore and the rhodopsin protein still in an active conformation after Schiff base hydrolysis (Schafer and Farrens, 2015). Interestingly, even the native A2 pigments

156 released A2 faster, and since the Ea is similar for Schiff base hydrolysis, this suggests a higher selectivity for all-trans A1 retinal relative to the A2 is a general feature of vertebrate rhodopsins likely due to differences in the chromophore electronic configuration. If the protein has a lower affinity for A2 versus A1 all-trans, the decay of the active state would be completed faster and the fluorescent signal during the retinal release assay would plateau faster.

Different retinal release rates with A1 vs A2 chromophores could be of significance to animals that utilize both chromophores. Some vertebrate species are known to switch chromophores in response to changes in the visual environment (due to migration, day length, or development; (Temple et al., 2005)), modulating rhodopsin function to better tailor their visual system to the external environment. The faster half-life of A2 rhodopsins suggests a shorter lived Meta II active state. The short-lived active state could be interpreted as enabling faster rhodopsin regeneration in order to respond to additional signaling (Imai et al., 1997). Species shifting between the two chromophores could potentially be utilizing the

A2 chromophore in environments with the expectation of multiple light bleaches occurring frequently, requiring faster recovery from light bleaching. However, it should be noted that the functional relationship between light-activated A2 rhodopsin and the G protein transducin

(which stabilizes the active state (Schafer and Farrens, 2015)) or the deactivating members of the phototransduction cascade (rhodopsin kinase and arrestin) is currently uncharacterized. It assumed to be equivalent to A1 pigments, but it may be that the physiological ramifications of a shorter lived active state of A2 rhodopsin is mitigated in vivo due to modified interactions with other members of the phototransduction cascade. It may be that the physiological ramifications of a shorter lived active state of A2 rhodopsin is mitigated in vivo

157 due to modified interactions with other members of the phototransduction cascade. Recently, the affinity for rhodopsin protein for all-trans retinal has been hypothesized to mitigate the effects of phototoxicity caused by excess all-trans released by photobleaching (Castiglione and Chang, in preparation; (Hauser et al., 2017b) (Rózanowska and Sarna, 2005)). Therefore, if the accelerated active state decay measured in A2 pigments is due to decreased affinity for all-trans A2, shifting to use the A2 chromophore in environments where frequent light bleaching is possible could be detrimental due to effects of phototoxicity caused by the sudden release of all-trans A2 retinal. Future investigations of adaptive evolution in A2 chromophore usage should also consider the non-spectral effects of chromophore switching, such as spontaneous activation in the dark and now the shorter-lived Meta II state of A2 rhodopsin.

This chapter represents the first comprehensive study of purified A2 rhodopsins in vitro.

We show that the λmax shift caused by the chromophore switch is relatively predictable for rhodopsins proteins, but that the role of the protein sequence may also be important in the determination of the red-shift. Biochemically, we have shown that A1 and A2 rhodopsins can have similar activation energies mediating the Schiff-base linkage hydrolysis, however this suggests that the faster retinal release of A2 rhodopsins could be due to steric and electronic differences in the A2 chromophore. Our results from this study further clarify the effect of the chromophore on the spectral properties of rhodopsin but also on the non-spectral properties of rhodopsin. This study marks the beginning of in vitro studies utilizing multiple vertebrate chromophores to isolate the specific role of the chromophore in rhodopsin function.

158

3.6 - REFERENCES

Aho, A. C., Donner, K., Hydén, C., Larsen, L. O. and Reuter, T. (1988). Low retinal

noise in animals with low body temperature allows high visual sensitivity. Nature 334,

348–350.

Ala-Laurila, P., Albert, R.-J., Saarinen, P., Koskelainen, A. and Donner, K. (2003). The

thermal contribution to photoactivation in A2 visual pigments studied by temperature

effects on spectral properties. Vis. Neurosci. 20.

Ala-Laurila, P., Donner, K. and Koskelainen, A. (2004). Thermal activation and

photoactivation of visual pigments. Biophys. J. 86, 3653–3662.

Ala-Laurila, P., Donner, K., Crouch, R. K. and Cornwall, M. C. (2007). Chromophore

switch from 11-cis-dehydroretinal (A2) to 11-cis-retinal (A1) decreases dark noise in

salamander red rods. J. Physiol. (Lond.) 585, 57–74.

Barlow, H. B. (1957). Noise and the Visual Threshold. Nature 180, 1405–1405.

Baylor, D. A., Lamb, T. D. and Yau, K. W. (1979). Responses of retinal rods to single

photons. J. Physiol. (Lond.) 288, 613–634.

Baylor, D. A., Nunn, B. J. and Schnapf, J. L. (1984). The photocurrent, noise and spectral

sensitivity of rods of the monkey Macaca fascicularis. J. Physiol. (Lond.) 357, 575–607.

Bickelmann, C., Morrow, J. M., Müller, J. and Chang, B. S. (2012). Functional

characterization of the rod visual pigment of the echidna (Tachyglossus aculeatus), a

basal mammal. Vis. Neurosci. 29, 1–7.

159

Bowmaker, J. K. and Hunt, D. M. (2006). Evolution of vertebrate visual pigments. Curr.

Biol. 16, R484–9.

Bowmaker, J. K., Loew, E. R. and Ott, M. (2005). The cone photoreceptors and visual

pigments of chameleons. J. Comp. Physiol. A Neuroethol. Sens. Neural. Behav. Physiol.

191, 925–932.

Bownds, D. (1967). Site of attachment of retinal in rhodopsin. Nature 216, 1178–1181.

Breikers, G., Bovee-Geurts, P. H., DeCaluwé, G. L. and DeGrip, W. J. (2001). A

structural role for Asp83 in the photoactivation of rhodopsin. Biol. Chem. 382, 1263–

1270.

Bridges, C. D. (1966). The grouping of fish visual pigments about preferred positions in the

spectrum. Vision Research 5, 223–238.

Carleton, K. L., Parry, J. W. L., Bowmaker, J. K., Hunt, D. M. and Seehausen, O.

(2005). Colour vision and speciation in Lake Victoria cichlids of the genus Pundamilia.

Molecular Ecology 14, 4341–4353.

Carleton, K. L., Spady, T. C., Streelman, J. T., Kidd, M. R., McFarland, W. N. and

Loew, E. R. (2008). Visual sensitivities tuned by heterochronic shifts in opsin gene

expression. BMC Biol. 6, 22.

Choe, H.-W., Kim, Y. J., Park, J. H., Morizumi, T., Pai, E. F., Krauss, N., Hofmann, K.

P., Scheerer, P. and Ernst, O. P. (2011). Crystal structure of Metarhodopsin II. Nature

471, 651–655.

160

Cowing, J. A., Poopalasundaram, S., Wilkie, S. E., Bowmaker, J. K. and Hunt, D. M.

(2002). Spectral Tuning and Evolution of Short Wave-Sensitive Cone Pigments in

Cottoid Fish from Lake Baikal. Biochemistry 41, 6019–6025.

Darden, A. G., Wu, B. X., Znoiko, S. L., Hazard, E. S., Kono, M., Crouch, R. K. and

Ma, J.-X. (2003). A novel Xenopus SWS2, P434 visual pigment: structure, cellular

location, and spectral analyses. Mol. Vis. 9, 191–199.

Dartnall, H. (1968). The photosensitivities of visual pigments in the presence of

hydroxylamine. Vision Research 8, 339–358.

Dartnall, H. and Lythgoe, J. N. (1965). The spectral clustering of visual pigments. Vision

Research.

Dungan, S. Z., Kosyakov, A. and Chang, B. S. (2016). Spectral Tuning of Killer Whale

(Orcinus orca) Rhodopsin: Evidence for Positive Selection and Functional Adaptation in

a Cetacean Visual Pigment. Mol. Biol. Evol. 33, 323–336.

Enright, J. M., Toomey, M. B., Sato, S.-Y., Temple, S. E., Allen, J. R., Fujiwara, R.,

Kramlinger, V. M., Nagy, L. D., Johnson, K. M., Xiao, Y., et al. (2015). Cyp27c1

Red-Shifts the Spectral Sensitivity of Photoreceptors by Converting Vitamin A1 into A2.

Curr. Biol. 25, 3048–3057.

Ernst, O. P. and Bartl, F. J. (2002). Active states of rhodopsin. Chembiochem 3, 968–974.

161

Eswar, N., Webb, B., Marti-Renom, M. A., Madhusudhan, M. S., Eramian, D., Shen,

M.-Y., Pieper, U. and Sali, A. (2002). Comparative Protein Structure Modeling Using

Modeller. Hoboken, NJ, USA: John Wiley & Sons, Inc.

Farrens, D. L. and Khorana, H. G. (1995). Structure and Function in Rhodopsin. Journal

of Biological Chemistry.

Gillam, A. E., Heilbron, I. M., Jones, W. E. and Lederer, E. (1938). On the occurrence

and constitution of the 693mmu chromogen (vitamin A(2)?) of fish liver oils. Biochem J

32, 404.1–416.

Govardovskii, V. I., Fyhrquist, N., Reuter, T., Kuzmin, D. G. and Donner, K. (2000). In

search of the visual pigment template. Vis. Neurosci. 17, 509–528.

Gozem, S., Schapiro, I., Ferré, N. and Olivucci, M. (2012). The molecular mechanism of

thermal noise in rod photoreceptors. Science 337, 1225–1228.

Harosi, F. I. (1994). An analysis of two spectral properties of vertebrate visual pigments.

Vision Research 34, 1359–1367.

Hauser, F. E., Ilves, K. L., Schott, R. K., Castiglione, G. M., López-Fernández, H. and

Chang, B. S. (2017a). Accelerated Evolution and Functional Divergence of the Dim

Light Visual Pigment Accompanies Cichlid Colonization of Central America. Mol. Biol.

Evol. 34, 2650–2664.

162

Hauser, F. E., Ilves, K. L., Schott, R. K., Chang, B. S. and Castiglione, G. M. (2017b).

Accelerated evolution and functional divergence of the dim light visual pigment

accompanies cichlid colonization of Central America. Molecular Biology ….

Hildebrand, P. W., Scheerer, P., Park, J. H., Choe, H.-W., Piechnick, R., Ernst, O. P.,

Hofmann, K. P. and Heck, M. (2009). A ligand channel through the G protein coupled

receptor opsin. PLoS ONE 4, e4382.

Hofmann, K. P., Scheerer, P., Hildebrand, P. W., Choe, H.-W., Park, J. H., Heck, M.

and Ernst, O. P. (2009). A G protein-coupled receptor at work: the rhodopsin model.

Trends Biochem Sci 34, 540–552.

Hope, A. J., Partridge, J. C., Dulai, K. S. and Hunt, D. M. (1997). Mechanisms of

wavelength tuning in the rod opsins of deep-sea fishes. Proceedings of the Royal Society

B: Biological Sciences 264, 155–163.

Hubbard, R., Brown, P. K. and Bownds, D. (1971). Methodology of vitamin A and visual

pigments. Meth. Enzymol. 18, 615–653.

Hunt, D. M., Fitzgibbon, J., Slobodyanyuk, S. J., Bowmaker, J. K. and Dulai, K. S.

(1997). Molecular evolution of the cottoid fish endemic to Lake Baikal deduced from

nuclear DNA evidence. Molecular Phylogenetics and Evolution 8, 415–422.

Imai, H., Kojima, D., Oura, T., Tachibanaki, S., Terakita, A. and Shichida, Y. (1997).

Single amino acid residue as a functional determinant of rod and cone visual pigments.

Proc. Natl. Acad. Sci. U.S.A. 94, 2322–2326.

163

Janz, J. M. and Farrens, D. L. (2004). Role of the retinal hydrogen bond network in

rhodopsin Schiff base stability and hydrolysis. J Biol Chem 279, 55886–55894.

Johnson, P. J. M., Halpin, A., Morizumi, T., Prokhorenko, V. I., Ernst, O. P. and

Miller, R. J. D. (2015). Local vibrational coherences drive the primary photochemistry

of vision. Nature Chemistry 7, 980–986.

Karnik, S. S., Sakmar, T. P., Chen, H. B. and Khorana, H. G. (1988). Cysteine residues

110 and 187 are essential for the formation of correct structure in bovine rhodopsin.

Proc. Natl. Acad. Sci. U.S.A. 85, 8459–8463.

Kaushal, S. and Khorana, H. G. (1994). Structure and function in rhodopsin. 7. Point

mutations associated with autosomal dominant retinitis pigmentosa. Biochemistry 33,

6121–6128.

Kawamura, S. and Yokoyama, S. (1998). Functional characterization of visual and

nonvisual pigments of American chameleon (Anolis carolinensis). Vision Research 38,

37–44.

Kefalov, V., Fu, Y., Marsh-Armstrong, N. and Yau, K.-W. (2003). Role of visual pigment

properties in rod and cone phototransduction. Nature 425, 526–531.

Kibelbek, J., Mitchell, D. C., Beach, J. M. and Litman, B. J. (1991). Functional

equivalence of Metarhodopsin II and the Gt-activating form of photolyzed bovine

rhodopsin. Biochemistry 30, 6761–6768.

Lamb, T. D. (2013). Progress in Retinal and Eye Research. Prog Retin Eye Res 36, 52–119.

164

Liu, J., Liu, M. Y., Nguyen, J. B., Bhagat, A., Mooney, V. and Yan, E. C. Y. (2011).

Thermal properties of rhodopsin: insight into the molecular mechanism of dim-light

vision. Journal of Biological Chemistry 286, 27622–27629.

Luk, H. L., Bhattacharyya, N., Montisci, F., Morrow, J. M., Melaccio, F., Wada, A.,

Sheves, M., Fanelli, F., Chang, B. S. and Olivucci, M. (2016). Modulation of thermal

noise and spectral sensitivity in Lake Baikal cottoid fish rhodopsins. Sci Rep 6, 38425.

Ma, J. X., Kono, M., Xu, L., Das, J., Ryan, J. C., Hazard, E. S., Oprian, D. D. and

Crouch, R. K. (2001). Salamander UV cone pigment: sequence, expression, and spectral

properties. Vis. Neurosci. 18, 393–399.

Miyagi, R., Terai, Y., Aibara, M., Sugawara, T., Imai, H., Tachida, H., Mzighani, S. I.,

Okitsu, T., Wada, A. and Okada, N. (2012). Correlation between nuptial colors and

visual sensitivities tuned by opsins leads to species richness in sympatric Lake Victoria

cichlid fishes. Mol. Biol. Evol. 29, 3281–3296.

Morrow, J. M. and Chang, B. S. (2015). Comparative Mutagenesis Studies of Retinal

Release in Light-Activated Zebrafish Rhodopsin Using Fluorescence Spectroscopy.

Biochemistry 54, 4507–4518.

Morrow, J. M. and Chang, B. S. (2010). The p1D4-hrGFP II expression vector: a tool for

expressing and purifying visual pigments and other G protein-coupled receptors. Plasmid

64, 162–169.

165

Morrow, J. M., Castiglione, G. M., Dungan, S. Z., Tang, P. L., Bhattacharyya, N.,

Hauser, F. E. and Chang, B. S. (2017). An experimental comparison of human and

bovine rhodopsin provides insight into the molecular basis of retinal disease. FEBS Lett.

Okada, T., Sugihara, M., Bondar, A.-N., Elstner, M., Entel, P. and Buss, V. (2004). The

Retinal Conformation and its Environment in Rhodopsin in Light of a New 2.2Å Crystal

Structure. Journal of Molecular Biology 342, 571–583.

Palczewski, K., Kumasaka, T., Hori, T., Behnke, C. A., Motoshima, H., Fox, B. A., Le

Trong, I., Teller, D. C., Okada, T., Stenkamp, R. E., et al. (2000). Crystal structure of

rhodopsin: A G protein-coupled receptor. Science 289, 739–745.

Parry, J. W. and Bowmaker, J. K. (2000). Visual pigment reconstitution in intact goldfish

retina using synthetic retinaldehyde isomers. Vision Research 40, 2241–2247.

Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng,

E. C. and Ferrin, T. E. (2004). UCSF Chimera--a visualization system for exploratory

research and analysis. J Comput Chem 25, 1605–1612.

Piechnick, R., Ritter, E., Hildebrand, P. W., Ernst, O. P., Scheerer, P., Hofmann, K. P.

and Heck, M. (2012). Effect of channel mutations on the uptake and release of the

retinal ligand in opsin. Proc. Natl. Acad. Sci. U.S.A. 109, 5247–5252.

Reuter, T. E., White, R. H. and Wald, G. (1971). Rhodopsin and porphyropsin fields in the

adult bullfrog retina. J. Gen. Physiol. 58, 351–371.

166

Rózanowska, M. and Sarna, T. (2005). Light-induced damage to the retina: role of

rhodopsin chromophore revisited. Photochem. Photobiol. 81, 1305–1330.

Saarinen, P., Pahlberg, J., Herczeg, G., Viljanen, M., Karjalainen, M., Shikano, T.,

Merila, J. and Donner, K. (2012). Spectral tuning by selective chromophore uptake in

rods and cones of eight populations of nine-spined stickleback (Pungitius pungitius). J.

Exp. Biol. 215, 2760–2773.

Sakmar, T. P., Franke, R. R. and Khorana, H. G. (1989). Glutamic acid-113 serves as the

retinylidene Schiff base counterion in bovine rhodopsin. Proc. Natl. Acad. Sci. U.S.A. 86,

8309–8313.

Schafer, C. T. and Farrens, D. L. (2015). Conformational Selection and Equilibrium

Governs the Ability of Retinals to Bind Opsin. J Biol Chem 290, 4304–4318.

Schafer, C. T., Fay, J. F., Janz, J. M. and Farrens, D. L. (2016). Decay of an active

GPCR: Conformational dynamics govern agonist rebinding and persistence of an active,

yet empty, receptor state. Proc. Natl. Acad. Sci. U.S.A. 113, 11961–11966.

Schoenlein, R. W., Peteanu, L. A., Mathies, R. A. and Shank, C. V. (1991). The first step

in vision: femtosecond isomerization of rhodopsin. Science 254, 412–415.

Shantz, E. M. and Embree, N. D. (1946). The replacement of vitamin A1 by vitamin A2 in

the retina of the rat. J Biol Chem 163, 455–464.

Shen, M.-Y. and Sali, A. (2006). Statistical potential for assessment and prediction of

protein structures. Protein Sci. 15, 2507–2524.

167

Starace, D. M. and Knox, B. E. (1997). Activation of transducin by a Xenopus short

wavelength visual pigment. J Biol Chem 272, 1095–1100.

Sugawara, T., Terai, Y., Imai, H., Turner, G. F., Koblmüller, S., Sturmbauer, C.,

Shichida, Y. and Okada, N. (2005). Parallelism of amino acid changes at the RH1

affecting spectral sensitivity among deep-water cichlids from Lakes Tanganyika and

Malawi. Proc. Natl. Acad. Sci. U.S.A. 102, 5448–5453.

Suzuki, T. and Miyata, S. (1988). Substitution of porphyropsin for rhodopsin in mouse

retina. Experimental Eye Research 46, 161–172.

Šali, A. and Blundell, T. (1993). Comparative protein modelling by satisfaction of spatial

restraints. Elsevier 234, 779–815.

Temple, S. E., Plate, E. M., Ramsden, S., Haimberger, T. J., Roth, W. M. and

Hawryshyn, C. W. (2005). Seasonal cycle in vitamin A1/A2-based visual pigment

composition during the life history of coho salmon (Oncorhynchus kisutch). J. Comp.

Physiol. A Neuroethol. Sens. Neural. Behav. Physiol. 192, 301–313.

Terai, Y., Miyagi, R., Aibara, M., Mizoiri, S., Imai, H., Okitsu, T., Wada, A.,

Takahashi-Kariyazono, S., Sato, A., Tichy, H., et al. (2017). Visual adaptation in Lake

Victoria cichlid fishes: depth-related variation of color and scotopic opsins in species

from sand/mud bottoms. 1–12.

Toyama, M., Hironaka, M., Yamahama, Y., Horiguchi, H., Tsukada, O., Uto, N., Ueno,

Y., Tokunaga, F., Seno, K. and Hariyama, T. (2008). Presence of rhodopsin and

porphyropsin in the eyes of 164 fishes, representing marine, diadromous, coastal and

168

freshwater species--a qualitative and comparative study. Photochem. Photobiol. 84, 996–

1002. van Hazel, I., Dungan, S. Z., Hauser, F. E., Morrow, J. M., Endler, J. A. and Chang, B.

S. (2016). A comparative study of rhodopsin function in the great bowerbird

(Ptilonorhynchus nuchalis): Spectral tuning and light-activated kinetics. Protein Science

25, 1308–1318.

Vogel, R., Siebert, F., Lüdeke, S., Hirshfeld, A. and Sheves, M. (2005). Agonists and

partial agonists of rhodopsin: retinals with ring modifications. Biochemistry 44, 11684–

11699.

Wada, A., Wang, F. and Ito, M. (2008). A convenient and stereoselective synthesis of 11Z-

3, 4-didehydroretinal by Horner-Emmons reaction using diphenyl phosphonate.

Chemical and Pharmaceutical Bulletin 56, 112–114.

Wald, G. (1939). On the distribution of vitamins A 1 and A 2. J. Gen. Physiol.

Wald, G. (1968). Molecular basis of visual excitation. Science 162, 230–239.

Wald, G., Brown, P. K. and Smith, P. H. (1953). Cyanopsin, a new pigment of cone vision.

Science 118, 505–508.

Wang, W., Geiger, J. H. and Borhan, B. (2014). The photochemical determinants of color

vision: Revealing how opsins tune their chromophore's absorption wavelength. Bioessays

36, 65–74.

169

Weadick, C. J., Loew, E. R., Rodd, F. H. and Chang, B. S. (2012). Visual pigment

molecular evolution in the Trinidadian pike cichlid (Crenicichla frenata): a less colorful

world for neotropical cichlids? Mol. Biol. Evol. 29, 3045–3060.

Whitmore, A. V. and Bowmaker, J. K. (1989). Seasonal variation in cone sensitivity and

short-wave absorbing visual pigments in the rudd Scardinius erythrophthalmus. J Comp

Physiol A 166.

Yan, E. C. Y., Kazmi, M. A., Ganim, Z., Hou, J.-M., Pan, D., Chang, B. S., Sakmar, T.

P. and Mathies, R. A. (2003). Retinal counterion switch in the photoactivation of the G

protein-coupled receptor rhodopsin. Proc. Natl. Acad. Sci. U.S.A. 100, 9262–9267.

Zhukovsky, E. A. and Oprian, D. D. (1989). Effect of carboxylic acid side chains on the

absorption maximum of visual pigments. Science 246, 928–930.

170

3.7 – SUPPLEMENTAL

Figure S3.1 – NMR spectra of 11-cis A2 chromophore. Data provided and collected by Dr. Akimori Wada

171

CHAPTER IV: CHARACTERIZING THE CLINICAL AND IN VITRO PHENOTYPE OF TWO RETINITIS PIGMENTOSA MUTATIONS, P180L AND G182V, IN THE EXTRACELLULAR LOOP 2 OF RHODOPSIN

Nihar Bhattacharya, James Morrow, Portia Tang, Elise Heon, Belinda S-W Chang

Author contributions: NB and BSWC conceptualized study. EH provided clinical data. JM and PT created constructs. NB conducted all experiments. NB wrote manuscript with input from BSWC.

4.1 - ABSTRACT

Retinitis pigmentosa (RP) is an inherited retinal degenerative disease that is characterized by extensive phenotypic variation. RP mutations in the dim light visual pigment rhodopsin have been found throughout the entire protein and account for approximately 30% of autosomal dominant RP. The extracellular loop 2 (EL2) and the N-terminus of rhodopsin forms a four stranded anti-parallel beta sheet creating a protein cap at the extracellular face, extending into the chromophore binding pocket, and is essential for rhodopsin function.

Structural and pharmacological methods of rescuing RP phenotypes in vitro have been used to characterize the highly heterogenic effect of RP mutations on rhodopsin structure and function. Here, we present novel mutations found in two RP patients located in the critical

EL2 structure of rhodopsin, P180L and G182V. We characterize these pathogenic mutations in rhodopsin, including the in vitro response to pharmacological and structural rescue, and present clinical data for the patient with the P180L rhodopsin mutation. The P180L patient exhibited a rapid and severe degradation of the visual field and retina and showed a severe

RP phenotype. In vitro, both of the mutations in rhodopsin are severely deleterious, resulting in minimal folded and functional protein sequestered in the ER, recapitulating the severity observed in the clinical data from the P180L patient. The EL2 RP mutants show little to no

172 response to in vitro pharmacological rescue with 11-cis retinal, however P180L is rescued when expressed with N-terminal stabilizing mutations (N2C/D282C), while G182V requires both methods of rescue to produce properly trafficked and functional rhodopsin. Using two methods of rescue with EL2 RP mutations in rhodopsin, we suggest that the beta3 strand of

EL2 is critical for the proper folding and assembly of the EL2 and the N-terminal cap. Our clinical and in vitro results illustrate the variability of RP mutant phenotypes and their response to rescue and emphasize the need for molecular analysis in the characterization of patients with RP to better understand disease severity and response to treatment.

173

4.2 - INTRODUCTION

Retinitis pigmentosa (RP) is a heritable, highly heterogeneous disease that is characterized by the progressive degeneration of the retina. In patients, clinical RP phenotypes vary considerably (Berger et al., 2010); characteristics such as age of onset are highly variable, with patients exhibiting symptoms in early childhood while others remain asymptomatic well into adulthood (Hartong et al., 2006). The classic progression of the disease begins with night blindness during adolescence, progressing to mid-peripheral vision loss in early adulthood. Complete peripheral vision loss and tunnel vision usually occurs in later adulthood. Eventually, patients can develop complete blindness by approximately 60 years of age (Gregg et al., 2013). RP affects approximately 1:3000 to 1:5000 people worldwide and can be autosomal dominant (30-40% of cases), autosomal recessive (50-

60%), or X-linked (5-15%) (Athanasiou et al., 2018). Currently, the only prescribed treatment for RP patients is supplementation with Vitamin A palmitate with docosahexaenoic acid (DHA) and lutein (Athanasiou et al., 2018), although patients with RP can also exhibit variable responses to treatment (Berson et al., 2012; Berson et al., 2010). This and the spectrum of RP severity are thought to be a reflection of the effects of the particular mutations of RP-associated proteins (Berger et al., 2010). Mutations in at least 45 different genes have been linked to monogenic RP (Hartong et al., 2006). Most of these genes account for only a small proportion of RP cases, with the exception of the rhodopsin gene. Rhodopsin is the best characterized RP gene to date and accounts for 20-30% of autosomal dominant and 1% of autosomal recessive RP cases (Malanson and Lem, 2009).

Rhodopsin, found in rod photoreceptors in the retina, is a seven transmembrane helical G protein-coupled receptor that initiates the phototransduction cascade when dim

174 light isomerizes the covalently bound 11-cis Vitamin A-derived retinal chromophore

(Sakmar et al., 2002; Whitmore and Bowmaker, 1989). Rhodopsin is a specialized light sensor with high sensitivity (Aho et al., 1988), ultrafast kinetics (Johnson et al., 2015;

Schoenlein et al., 1991), and incredible stability (Baylor et al., 1980). These specialized properties are established via multiple conserved structural and functional elements throughout the protein (Bownds, 1967; Fritze et al., 2003; Kaushal et al., 1994; Mahalingam et al., 2008; Ovchinnikov et al., 1988). To date, over 150 rhodopsin mutations along the entire length of the protein have been associated with degenerative retinal disease (Figure

4.1A) (DeCaluwé and DeGrip, 1996; RetNet, www.sph.uth.tmc.edu/RetNet/). Patients with

RP mutations in rhodopsin typically develop a classic form of RP with rod-cone dystrophy where rod photoreceptor cell death leads to cone photoreceptor degradation and, ultimately, retinal degradation (Athanasiou et al., 2018).

RP mutations in rhodopsin are known to disrupt rhodopsin structure/function via multiple methods and are thus often classified based on their biochemical and cellular phenotypes when expressed in vitro (Mendes et al., 2005). RP mutations that result in misfolded rhodopsin are the most common, which result in rhodopsin being trapped in the ER or in aggresomes, and are thought to eventually result in photoreceptor cell death via apoptosis due to ER stress (Sung et al., 1993; Sung et al., 1991). Other classes of mutations do not necessarily affect folding, but interfere with additional aspects of rhodopsin structure and function (Mendes et al., 2005). This biochemical heterogeneity of RP rhodopsin phenotype is thought to underlie the variation seen in rhodopsin RP patients.

The function of RP-associated rhodopsin mutants can be rescued in vitro with the addition of small cell permeable molecules (pharmacological chaperones). Inverse agonist

175 molecules like retinoids (i.e., vitamin A derivatives such as 11-cis and 9-cis retinal chromophores), stabilize the conformation of otherwise misfolded or unstable rhodopsin protein when introduced during protein synthesis, thereby promoting proper folding and decelerating retinal degradation (Bernier et al., 2004; Krebs et al., 2010; Noorwez et al.,

2004; Saliba et al., 2002). This phenomenon underlies the rationale for the prescription of high doses of Vitamin A for RP patients regardless of patient genotype (Berson et al., 2012).

In addition to pharmacological rescue of RP mutations in rhodopsin, structural rescue of the

N-terminus has been shown to be effective at rescuing N-terminal RP mutations in vitro

(Opefi et al., 2013).

The N-terminus of rhodopsin forms a structured cap that folds over the seven- transmembrane protein bundle, interacting with all the helices, and extracellular loops

(Figure 4.1B). Multiple RP mutations can be found in the N-terminus (Mendes et al., 2005).

Some of these mutations have been successfully rescued via pharmacological rescue with 11- cis retinal (Noorwez et al., 2004; Krebs et al., 2010; Opefi et al., 2013), but also the pathogenic effects of multiple N-terminal RP mutations have additionally been ameliorated by the stabilization of the N-terminal cap structure by the addition of the mutations

N2C/D282C, which creates a disulfide bond between the N-terminal cap and the EL-3 (Opefi et al., 2013; Xie et al., 2003). The N-terminal cap contains the first two strands of a four strand anti-parallel beta sheet, while the extracellular loop 2, over which the N-terminal cap lies, contains the other two strands (Palczewski et al., 2000). The beta sheet begins at the extracellular face and continues to the chromophore binding pocket where the fourth beta strand, beta4, comprises one of the walls of the chromophore binding pocket (Figure 4.1B).

The role of the beta4 strand in the chromophore binding pocket has been the primary focus of

176 studies investigating the EL2 (Liu et al., 2013; Yan et al., 2003) but few studies have investigated the beta3 strand which interacts with the underside of the N-terminal cap (Doi et al., 1990).

The extracellular loop 2 of rhodopsin is 27 amino acids long (from site 173 to 200)

(Okada et al., 2002), and is involved in several interactions critical for proper structure and function rhodopsin. Several structural and functional elements appear in the EL2 (Figure

4.1B), including the beta3 and beta4 strands, and other critical sites (Karnik and Khorana,

1990; Palczewski et al., 2000; Yan et al., 2003). Additionally the EL2 makes contact, or comes into close proximity, with the chromophore, the N-terminus, the EL-1 and EL-3 and several of the transmembrane helices (Palczewski et al., 2000). 13 sites with RP mutations have been identified in humans (Li et al., 2010; Mendes et al., 2005).

We recently identified P180L and G182V as novel RP-associated mutations in rhodopsin.

P180 is located at the end of the beta3 strand of the EL-2 (Figure 4.1B)(Okada et al., 2004).

While P180L is a novel RP mutation, another RP mutation at this site has been previously characterized, P180A. P180A has a relatively mild pathogenic phenotype producing mutant rhodopsin which folds but is thermally unstable in vitro, and a mild patient phenotype

(Iannaccone et al., 2006). The effects of the P180L RP mutation in rhodopsin, however, have yet to be characterized. G182V, also on the beta3 strand of the EL2, was also relatively recently discovered as a mutation causing RP in humans (Fernandez-San Jose et al., 2014;

Yang et al., 2014). In the rhodopsin crystal structure, G182 appears in the hairpin turn linking the beta3 sheet to the beta4 sheet which folds the EL2 back away from helix 7 (Okada et al.,

2004). Another RP mutation at this site, G182S (Sheffield et al., 1991), has a severe in vitro phenotype with no functional protein produced (Sung et al., 1993), however the patient

177 phenotype is relatively mild with preserved vision well into old age (Fishman et al., 1992).

The heterogeneity between mutations at the same sites (P180 and G182) suggests that mutations in the EL2 RP may disrupt rhodopsin through different molecular mechanisms and may therefore require variable rescue strategies.

In this study, we investigate two RP mutations (P180L and G182V) that occur in a critical region of rhodopsin, the EL2 loop. We investigate these rhodopsin mutations in vitro using a heterologous expression system, microscopy and spectroscopy, and present clinical data for a patient with the P180L rhodopsin mutation, including assessment of the visual field and measurements of visual acuity. Additionally, we conducted both pharmacological and structural rescue experiments in vitro in order to gain insight into how these mutations may disrupt rhodopsin function, and to infer how they may be disrupting rhodopsin structure. Our findings allow us to clarify the role of the beta3 strand in the overall structure and function of rhodopsin, while also highlighting the variable nature of RP mutations in rhodopsin.

178

4.3 - MATERIALS AND METHODS pGFP construct and mutants

Previous studies have shown that GFP tagged rhodopsin is stable as the C-terminus of rhodopsin is quite flexible and can accommodate the GFP protein in vitro and in vivo (Moritz et al., 2001). We constructed the pGFP expression vector to allow us to easily insert rhodopsin full-length sequences into a plasmid that would result in a rhodopsin-GFP fusion protein. pGFP was constructed by cloning the humanized recombinant GFP II (hrGFP II) gene in pIRES-hrGFPII (Stratagene) and modifying the multiple cloning site (MCS). The hrGFP II sequence was amplified out of the pIRES-hrGFPII vector using primers that added the BamHI restriction site to the 5’ end and the XhoI restriction site to the 3’ end of the sequence. The MCS of the pIRES vector was then mutated to change the BamHI cut site to

BglII and the EcoRI to SalI. This pIRES was then digested to remove the hrGFPII reporter gene and ligated with a peptide containing a BamHI cut site and the previously amplified hrGFPII sequence. This resulted in an expression plasmid that would fuse GFP to the C-

Terminus of any inserted rhodopsin sequence missing a stop codon with a 2-amino acid linker (See figure 4.2).

Bovine rhodopsin, the model system for in vitro studies of rhodopsin structure function, was used as the template for all constructs in this study. Site-directed mutagenesis primers were designed to introduce the stabilizing amino acid substitutions N2C and D282C in the wildtype bovine rhodopsin sequence to create the bovine stability mutant construct using the

QuickChange II protocol. Additional mutagenesis primers were used to introduce the retinitis pigmentosa mutations P180L and G182V in both the wildtype and stability mutant

179 background. Constructs were sequenced to verify successful mutation and cloned into the p1D4 and pGFP expression vectors.

Immunocytochemistry

Human SK-N-SH neuroblastoma (ATCC HTB-11) cells, an in vitro model for aggregation-based diseases (Mendes and Cheetham, 2008; Westhoff et al., 2005), were grown and cultured in full media (DMEM (Life technologies), 10% FBS (Invitrogen), and

Penicillin-Streptomycin (Invitrogen)) at 37º in 5% CO2 and seeded into 24-well plates with coverslips (Sarstedt) while under 5 passages. Once cells reached approximately 75% confluence, they were transfected with 645ng of construct in pGFP using Lipofectamine

2000 (Invitrogen) protocols. Wildtype bovine rhodopsin was used as a positive control for proper translation and trafficking of proteins inside SK-N-SH cells.

After 24 hours, half the wells were incubated with WGA in HBSS for 10 minutes at 37º to label the cell membrane. All cells were then rinsed with PBS and fixed with 2% paraformaldehyde in PBS. To label cells with the endoplasmic reticulum marker antibody anti-calreticulin (1:400, Abcam), cells were washed and permeabilized in PBS containing 1% bovine serum albumin (Sigma) and 0.1% saponin (PBS-BS). Anti-calreticulin was diluted in

PBS-BS and incubated for 1hr at room temperature. After washing with PBS-BS, secondary antibody (Cy3-conjugated goat anti-rabbit IgGt, 1:200, Jackson Immunoresearch) was diluted in PBS-BS and added to the wells for 1 hour. Nuclei were stained with Hoechst

(1:1000 in PBS, Hoechst type 33258 Invitrogen) for 10 minutes. Cells were mounted with

ProLong Gold Antifade (Thermofisher), coverslipped and allowed to cure for 24 hours in the

180 dark prior to imaging on Leica TCSSP8 confocal microscope. ImageJ was used to construct

Z-stacks, maximum projection images and scale bars.

Expression

In order to produce purified rhodopsin for functional assays, heterologous expression and purification took place as previously described (Bhattacharyya et al., 2017). Briefly, a 10 cm plate of HEK293T cells were transfected with 8 ug of construct in p1D4 using Lipofectamine

2000 protocols. Cells were then either given 10 µM 11-cis retinal for 24 hrs and 20 µM for an additional 24hrs or given ethanol as a control. Cells were cultured in the dark at 37º in 5%

CO2. 48 hours post-transfection, cells were harvested and regenerated in 5 µM 11-cis retinal and solubilized in DM and affinity purified with 1D4 antibody coupled to sepharose beads.

Purified protein was eluted using WB2 buffers and 1D4 peptide. Wildtype bovine rhodopsin was used as control for both the purification protocol and for wildtype rhodopsin function.

Spectroscopy

UV-Vis spectra of purified rhodopsins was collected using a Cary 4000 dual-beam spectrophotometer (Agilent) at 20ºC. Samples were light activated with a 30 second light bleach with a fiber optic lamp (Dolan-Jenner) and difference specs were calculated. To calculate percentage rescued, the ratios of the active peak to the total protein peak (280nm) were compared between rescued and unrescued samples, both pharmacological and structural. Absorbance spectra for hydroxylamine assays were measured following the addition in 50mM NH2OH (Sigma Aldrich) until completion of reaction or 2 hours.

Patient phenotyping

181

This study was approved by the Human Research Ethics Board of the Hospital for Sick

Children (Toronto, ON) and met the tenets of the Declaration of Helsinki. The cases were selected from an internal database and the phenotype information was collected retrospectively. Other than basic demographic and genetic information, we collected information about Visual acuity (VA), color vision, Goldmann visual fields (GVF), electroretinography (ERG) and imaging. Imaging included fundus photography

(VisucamNM/FA - Carl Zeiss Meditec, Dublin, California, USA and Optos), optical coherence tomography (OCT, Cirrus from Carl Zeiss Meditec, Dublin, California, USA). Genetic testing was using gene panels-based sequencing by CLIA approved laboratories.

182

4.4 - RESULTS

Patient with P180L shows progressive rod-cone degeneration and a severely constricted field of vision

Case 1 (P180L) was a female of Irish/Dutch descent with difficulty adapting to a dim lit environment (nyctalopia) since the first decade of life. Though her central visual acuity and color vision remained within normal limits at age 22 years, the retinal function documented by ERG and the visual field showed progressive deterioration. The first ERG at age 9 years showed moderate rod cone dysfunction, which slowly worsened within 3 years. The horizontal field of vision at age 11 years was 25º in each eye (normally 120º-130º) and constricted to 8∘ in each eye by the age of 22 years (Figure 4.3, Supplemental table 4.1). Her retinal changes were typical of a progressive rod-cone degeneration. Case 2 (G182V) also had symptoms since the first decade. However, she was lost to follow-up and no clinical details could be obtained.

P180L and G182V cause rhodopsin to misfold and be retained inside the cell

Rhodopsin wildtype protein fused to GFP was used as a positive control to represent proper protein translation, folding, and transportation to the cell membrane in vitro. SK-N-

SH neuroblastoma cells were transfected with wildtype rhodopsin in the pGFP vector where the wildtype-GFP fusion protein was successfully transported to the cell membrane to colocalize with wheat germ agglutinin, a cell membrane marker (Figure 4.4A). We observed no retained WT-GFP protein within the cell, suggesting that the neuroblastoma cell could fold and process the fusion protein with little stress (Athanasiou et al., 2012; Liu et al., 2013).

183

P180L-GFP in the wildtype background almost completely colocalized with the anti- calreticulin staining, an ER marker (Figure 4.4B). In the cells, the tagged protein was generally dispersed throughout the ER, although the presence of a few bright puncta outside the ER was suggestive of aggregation. With pharmacological rescue using 11-cis retinal chromophore, some of the P180L fusion protein did appear to be on the cell membrane and the periphery of the cell, suggesting a small population of folded protein, with some of the expressed protein in the ER as well (Figure 4.4C). However, the majority of the protein appeared to be in puncta throughout the cell, which did not colocalize with either the ER marker or the cell membrane. With the GFP staining mainly perinuclear, this suggested that the rescued P180L-GFP protein may be contained within the Golgi, where there may be possible delayed post-translational modification.

The extracellular loop 2 retinitis pigmentosa mutant G182V-GFP was retained in the cell with the majority of the protein colocalizing with the ER marker (Figure 4.4D); however, there were several bright cytosolic puncta, suggesting aggregation and ER stress as they were not perinuclear and do not colocalize with the ER marker. When pharmacologically rescued, the G182V-GFP protein remained inside the cell (Figure 4.4E), as no protein appeared to be on the cell membrane. However, inside the cell, the distribution of the protein changed. The majority of the protein remained in the ER, although staining separate of the ER was seen perinuclearly and other intracellular puncta decreased.

Structural rescue, in addition to pharmacological rescue, allows for partial transport of mutant rhodopsins to the cell membrane

184

To determine if N-terminal stabilization can rescue RP mutant folding within the cell, the retinitis pigmentosa extracellular loop 2 mutants were additionally expressed in the stability mutant background (sP180L-GFP, and sG182V-GFP), which has an additional disulfide bond linking the N-terminus to extracellular loop 3. sP180L-GFP in the stability mutant background was transported to the cell membrane of SK-N-SH cells, where it colocalized with the cell membrane marker (Figure 4.5B), indicating that the stability mutant was somewhat successful in rescuing folding. The protein did not colocalize with the ER marker and sP180L-GFP formed puncta on the surface of the cell. The presence on the cell membrane remained consistent when the cells were transfected and incubated with 11-cis retinal, however sP180L-GFP appeared perinuclearly (Figure 4.5C). The fusion protein on the cell membrane and in the Golgi could be indicative of higher expression levels saturating the protein translation machinery of the cell or defects with post-translational modifications.

Without pharmacological rescue, sG182V-GFP was found on both the surface of the cell and in large quantities in the ER (Figure 4.5D). Some of the non-ER intracellular staining appeared to be perinuclear (Golgi), while other intracellular puncta appeared in the periphery of the cytosol. However, with pharmacological rescue, we observed that sG182V-GFP was present mainly on the cell membrane, indicating that both rescue methods together are required for a full rescue of G182V trafficking, and presumably folding, in SK-N-SH cells.

G182V expression and function cannot be rescued in the wildtype background

Wildtype and RP mutant constructs in p1D4 were heterologously expressed in

HEK293T cells and affinity purified. The UV-Vis spectra of these purified pigments had two

185

peaks, one at 280nm (A280) and one at ~500nm (Aλmax). The A280 peak is composed of all protein affinity purified with the 1D4 tag, regardless of whether it is properly folded. The

Aλmax peak represents dark-state rhodopsin regenerated with 11-cis retinal and is referred to as the wavelength of maximal absorption (λmax). A high A280:Aλmax ratio is desirable as it implies that the majority of the protein expressed by the cell can be regenerated with chromophore to produce functional protein. Exposure to light causes the λmax peak to shift to 380nm, the light-activated form of rhodopsin, which in rod photoreceptors would trigger the signal transduction cascade.

Wildtype rhodopsin was used as a positive control to represent non-pathogenic and functional rhodopsin which expresses with a high A280:Aλmax. The expression of wildtype bovine rhodopsin was not significantly increased by pharmacological rescue, with only a 4% increase in relative yield (Figure 4.6A, Table 4.1), suggesting that the expression of wildtype protein is already ideal in our heterologous system without pharmacological rescue. Without rescue, expression of P180L rhodopsin protein yielded a minimal amount of dark-state rhodopsin (black trace, Figure 4.6C, Table 4.1). The presence of protein at 280nm suggests that most of the protein expressed was misfolded or could not bind to 11-cis retinal and is therefore non-functional. Providing 11-cis retinal at the moment of transfection, the

A280:Aλmax increased by 63% (red trace, Figure 4.6C, Table 4.1). Interestingly, both peaks did not increase proportionally; the Aλmax peak increases by almost 160% while the A280 only increases by 61% (Table 4.1). This suggests that the 11-cis retinal allowed for a higher proportion of translated protein to regenerate, and also that the higher A280 peak could be due to increased overall expression values or lowered degradation of translated protein by the

186 cell. The presence of 11-cis retinal during translation may be stabilizing the EL2 to form a rudimentary chromophore binding pocket.

The G182V mutation produced no active protein when expressed in the wildtype background (black trace, Figure 4.6E, Table 4.1). Additionally, pharmacological rescue with

11-cis retinal cannot rescue folding or expression of G182V (red trace, Figure 4.5E, Table

4.1) suggesting G182V has a more severe phenotype in vitro. These expression results recapitulate the results from the immunocytochemistry.

The stabilizing mutation rescues folding and function in both extracellular loop 2 mutants

G182V and P180L were also expressed in the stability mutant background to determine if stabilization of the N-terminal cap structure could rescue rhodopsin function.

The wildtype stability mutant (referred to hereafter as sRho) expresses well in the heterologous expression system with high A280 and a high Aλmax (black trace, Figure 4.6B,

Table 4.1). Interestingly, sRho showed a less favourable A280:Aλmax ratio when compared to wildtype rhodopsin (0.36 vs 0.51) (Table 4.1), suggesting that sRho translation and folding is less ideal in vitro, possibly due to complications with proper disulfide bond formation.

Pharmacological rescue only increased the A280:Aλmax by 6% (red trace, Figure 4.5B, Table

4.1). The stability mutant λmax matches the λmax of wildtype rhodopsin (499nm), showing that the additional stability mutations did not affect the spectroscopic properties of wildtype rhodopsin.

P180L in the stability mutant background (sP180L) showed a ~16% increase in active rhodopsin when compared to the pharmacologically rescued P180L (black trace, Figure

4.5D, Table 4.1), but the amount of total protein was lower in the stability mutant

187 background. When 11-cis retinal was administered, the expression of sP180L increased (red trace, Figure 4.6D, Table 4.1). There was a higher yield of both total protein purified and active protein when both rescue methods were combined. The stability mutant backbone appears to rescue the G182V mutation (sG182V) only when 11-cis retinal is administered at the time of transfection. Without pharmacological rescue, the A280 of sG182V (black trace,

Figure 4.6F, Table 4.1) is higher when compared to the same mutation in the wildtype background. This suggests that the expressed protein is not being degraded by the cell in the stability mutant background. While there appears to be dark state functional protein, there is only a very small amount. When the transfected cells were incubated with 11-cis retinal, however, the λmax peak increased by 300% while the total protein decreased (red trace, Figure

4.6F, Table 4.1).

Stabilizing the N-terminus rescues chromophore binding pocket instability in RP extracellular loop 2 mutants

Hydroxylamine can react with the covalently bound 11-cis chromophore to form a retinal oxime product. When rhodopsin is incubated with hydroxylamine, the tight binding pocket typically prevents the small molecule from reacting with retinal chromophore. The ability to resist attack by hydroxylamine is a putative property of rhodopsins in general

(Figure 4.6A). Should the rhodopsin chromophore binding pocket structural integrity be compromised due to mutation, hydroxylamine would be able to enter and react with the covalently bound retinal. This reaction can be measured spectroscopically with the appearance of the retinal oxime peak at 360nm and the disappearance of the dark-state rhodopsin λmax peak. Thus, this assay allowed us to quantify the stability or the relative

‘open-ness’ or structural integrity of the chromophore binding pocket. Rhodopsin molecules

188 with the stabilizing mutation did not react upon incubation with hydroxylamine (Figure

4.7B).

When the P180L mutant was exposed to hydroxylamine in the dark, the λmax peak rapidly disappears, and a corresponding retinal oxime 360nm peak appeared (Figure 4.7C) as the hydroxylamine entered the binding pocket and reacted with the retinal. The t1/2 of the reaction of the retinal with the hydroxylamine was ~14 minutes. However, the stabilizing mutant appeared to allow the binding pocket of the P180L mutant to form more stably, as sP180L had a longer reaction half-life of 24.5 minutes (Figure 4.7D). G182V in the wildtype background could not be assayed as the mutation produced no functional protein. In the stability mutant background, the sG182V binding pocket showed surprising integrity with a slower t1/2 of 40 minutes (Figure 4.7E).

189

Figure 4.1 - Structure of wildtype rhodopsin (PDB: 1U19) (A) Whole structure of wildtype rhodopsin. Known sites of RP mutations highlighted in red. (B) Zoom in on the N- terminal and extracellular loop 2 anti-parallel sheet. Transmembrane helix 2 was removed to show inner chromophore binding pocket. N-terminal cap in green, extracellular loop 2 in tan, beta sheet strands in purple, site 180 and 182 shown in red, 11-cis retinal in black, water molecules represented as dark blue spheres.

190

Figure 4.2 - Physical map of the pGFP expression vector, derived from p1D4-hrGFP II. Important features include the CMV promoter, and the C-terminal hrGFPII tag. Rhodopsin genes, or other genes of interest, can be ligated into the multiple cloning site between the unique BamHI and EcoRI restriction sites.

191

Figure 4.3 - Ocular phenotype of Case 1 (carrying the P180L variant). A) High magnification view of the posterior pole of the retina centered on the fovea/macular area (arrow). The retina around the nerve and macular shows atrophy (a). The foveal/macular area shows preserved (p) retina. This correlates with B) the OCT image centered on the fovea (*). The residual preserved outer retina is defined by the bracket. For orientation, c: choroid. Insert shows area scanned. C) Wide field view of the retina of the right eye (Optos). The optic nerve (ON) shows some pallor, the vessels are attenuated and there is bone spiculing pigmentation typical of retinitis pigmentosa. Inferior are patient lashes. D) Goldmann visual field showing constriction of the field to the central 8-10 degrees (full blue line). Normal field would meet the gray dotted line. The width of the central field corresponds to the preserved outer retina on the OCT (B). Data and figure created by Elise Heon.

192

Figure 4.4 - Rhodopsin RP mutants are retained inside the cell and cannot be rescued with 11-cis retinal. Confocal immunocytochemistry of SK-N-SH neuroblastoma cells expressing rhodopsin RP mutations in a bovine wildtype background with a C-terminus GFP tag. (A) Wildtype bovine rhodopsin (green) is trafficked to the membrane where it colocalizes with the cell membrane marker (lower row, red) as seen in the merge (yellow) (B) rhodopsin P180L RP-GFP fusion mutant (green) is retained inside the cell and colocalizes with the ER marker (top row, red) and does not appear on the cell membrane, (C) Trafficking is not rescued with 11-cis retinal as it remains inside the cell. (D) rhodopsin G182V RP-GFP fusion mutant (green) colocalizes with the ER marker (top row, red) as it is retained inside the cell. (E) trafficking cannot be rescued with the addition of 11-cis retinal as the G182V-GFP protein is still retained inside the cell. White arrowheads indicate cytosolic puncta. Scale bar represents 30 µm.

193

Figure 4.5 - The stability mutant can rescue RP mutant trafficking to the cell membrane. Confocal immunocytochemistry of SK-N-SH neuroblastoma cells expressing rhodopsin RP mutations in the bovine stability mutant background with a C-terminus GFP tag. (A) Bovine stability mutant rhodopsin (green) is trafficked to the membrane where it colocalizes with the cell membrane marker (top row, red) as seen in the merge (yellow) (B) rhodopsin P180L RP- GFP mutant (green) in the stability mutant background is partially retained inside the cell, (C) trafficking to the cell membrane is fully rescued with 11-cis retinal. (D) rhodopsin G182V RP-GFP mutant (green) in the stability mutant background colocalizes with the ER marker (red) as it is partially retained inside the cell (E) with the addition of 11-cis retinal, the protein is trafficked to the cell membrane, though a large proportion is still seen inside the cell. Scale bars represent 30 µm.

194

Figure 4.6 - UV-visible dark absorption spectra of wildtype and RP rhodopsin mutants (A) Wildtype bovine rhodopsin was used as a control for expressions. Wildtype responds minimally to pharmacological rescue (B) Stability mutant rhodopsin was used to rescue RP mutants. The stability mutant responds minimally to rescue (C) P180L dark state protein is only purified when rescued with 11-cis retinal. (D) P180L in the stability mutant background (sP180L) shows a small amount of dark-state protein, but the expression is significantly rescued with the addition of 11-cis retinal. (E) G182V shows no properly folded protein even with pharmacological rescue. (F) Even in the stability mutant background (sG182V), only a very small amount of sG182V protein is purified without rescue. With rescue, expression increases. Inset has same axes as main graph.

195

Figure 4.7 - Comparison of hydroxylamine reactivity of RP mutants with wildtype rhodopsin. Hydroxylamine assays to determine binding pocket integrity. (A) Wildtype bovine rhodopsin does not react in the presence of hydroxylamine. (B) the stability mutant is also resistant to attack by hydroxylamine (C) P180L mutant rhodopsin protein in the wildtype background reacts the fastest with a half-life of 14.6 minutes. (D) P180L in the stability mutant background reacts slower than the same mutation in the wildtype background with a half-life of 24.5 minutes (E) G182V in the stability mutant background reacts relatively slowly at 40.1 minutes.

196

Figure 4.8: Schematic diagram illustrating a model for pharmacological rescue and stability mutant rescue in EL2 RP rhodopsin mutations. (A) Wildtype and Stability mutant protein folds correctly inside the cell and is processed through the Golgi apparatus and is exported to the cell membrane. (B) P180L mutation cannot fold correctly and is retained in the ER and degraded. Pharmacological repair may facilitate partial folding, but a disordered N-terminus retains the protein in the Golgi. Pharmacological rescue in combination with the stability mutation allows for folding and export of the protein to the cell membrane, though the chromophore binding pocket remains compromised. (C) G182V protein remains retained inside the ER in the wildtype background. The stability mutant allows for partial folding and Golgi retention. Both rescue methods together allow for sufficient folding to enable export to the cell membrane, though the chromophore binding pocket remains somewhat open.

197

Table 4.1 - Effects of pharmacological and structural rescue of EL2 RP mutants.

Percentage rescue

Almax A280 Almax: A280 Pharmacological Stability Mutant Both

(Almax, A280) (Almax, A280) (Almax, A280)

Wildtype 0.139 0.268 0.518 104.4 70.1 74.3 rWildtype 0.117 0.217 0.541 (84.6, 81.0) (36.0, 51.4) (51.7, 69.5)

P180L 0.005 0.101 0.047 163.1 281.0 470.6 rP180L 0.012 0.163 0.076 (263.9, 161.87) (308.0, 109.6) (787.6, 167.4)

G182V -- 0.122 ------406.1 rG182V -- 0.121 -- (--, 99.3) (--, 128.5) (285.7, 96.0) sRho 0.050 0.138 0.364 106.0 rsRho 0.072 0.186 0.385 (143.4, 135.4) sP180L 0.014 0.110 0.131 167.5 rsP180L 0.037 0.168 0.220 (255.7, 152.7) sG182V 0.007 0.156 0.042 406.1 rsG182V 0.020 0.117 0.172 (303.2, 74.6)

198

4.5 - DISCUSSION

In this study, we characterized the clinical phenotype of a patient that was found to possess a P180L mutation in rhodopsin using multiple ophthalmologic methods to assess degradation of the retina. Additionally, we characterized the rhodopsin mutations P180L and

G182V in vitro using microscopy and spectroscopy assays to evaluate protein translation, function, and structure. We found the clinical phenotype of the P180L patient to be severe, presenting with rod-cone dystrophy leading to rapid degradation of peripheral vision early in life. In vitro, we found that both P180L and G182V rhodopsin mutants were highly deleterious and retained inside the ER of the cell with signs of protein aggregation. The mutations also produced little to no purified functional protein in our heterologous expression system. While G182V showed no response to pharmacological rescue, P180L showed a minor increase in expression. In the stability mutant, the presence of an additional disulfide bond stabilizing the N-terminal cap structure allowed for the rescue of expression levels and trafficking of P180L. G182V expression and trafficking was only fully recovered when the disulfide bond-mediated rescue was combined with pharmacological rescue. Here, we

199 discuss our results in the context of rhodopsin structure and function with specific focus on the interaction between the N-terminal cap and the extracellular loop 2.

Stabilization of the N-terminus with an additional disulfide bond rescues EL2 RP rhodopsin mutant function

The rhodopsin stability mutant (N2C/D282C) has been extensively utilized in the study of rhodopsin structure-function as it greatly increases yield and stabilizes functionality, while preserving wildtype structure and functionality (Gross et al., 2003; Sinha et al., 2014;

Standfuss et al., 2011; Vishnivetskiy et al., 2013). The stability mutation has allowed for the functional characterization of mutations found in different parts of the protein (constitutively active (Vishnivetskiy et al., 2013) and dimerization defects (Ploier et al., 2016)), while also rescuing function in RP mutants (Opefi et al., 2013; Singhal et al., 2013). While the structure and function of the stability mutant is thought to recapitulate that of the wildtype, the additional disulfide bond has been shown to minimally affect G protein activation (Xie et al.,

2003). We utilized the stability mutant in our study to stabilize the potentially disrupted N- terminal cap over the two EL2 beta3 RP mutants, and to assess the ability of the stabilized N- terminal cap to rescue structure and function of these mutants.

The P180L mutation likely also prevents the N-terminal cap from forming over the

EL2 (Figure 4.8B), as demonstrated by our trafficking and hydroxylamine results. The instability of the rescued P180L mutant protein was shown by the rapidly reaction with hydroxylamine (Figure 4.7B). The disrupted N-terminal may be causing Golgi retention, which has been previously characterized in EL2 autosomal recessive RP mutations with delayed post-translational modifications, like E150K (Zhu et al., 2006). When P180L was

200 expressed in the stability mutant background (sP180L), the now stabilized N-terminal cap appears to allow the chromophore binding pocket to form to produce a functional rhodopsin, but still not to the wildtype level. This suggests the structural rescue of the N-terminal cap may minimally rescue the disrupted interaction between beta3 and beta2 due to the disordered EL2 folding, and that the N-terminal cap was likely strongly disrupted in P180L.

For G182V, only when the two methods of rescue are combined, does the protein appear to fold properly. We believe the two rescues together could be acting as a kind of "sandwich" rescue in G182V. It is likely that the EL2 cannot fold compactly enough to form a functional protein in the G182V mutant as glycine plays a large role in beta turns (Chou and Fasman,

2002). We hypothesize that with the chromophore rescue, the presence of 11-cis retinal in the binding pocket could stabilize the beta4 strand below beta3, while the stabilized N-terminal cap would potentially be helping to fold the rest of EL2 above beta3, with both rescues needed to produce a functional structure (Figure 4.7C). The stability rescue alone likely was ineffective for two possible reasons: the additional disulfide bond could not have formed as the transmembrane helix bundle was not ordered enough (due to a disordered EL2) to bring the two cysteine residues into proximity, or, the disulfide bond did form but the EL2 could be unstructured under the N-terminal cap to then accommodate the 11-cis retinal post- translation.

The two methods of rescue acting additively and/or separately between the two EL2 mutants was somewhat expected, as some N-terminal mutations have been shown to respond differently to different rescue methodologies (Opefi et al., 2013). Our results are consistent with previous studies suggesting that pharmacological rescue allows for better EL2 packing but a disordered N-terminal cap, with the magnitude of the N-terminal cap disorder

201 determining how well the pharmacological repair can affect expression levels. With the EL2

RP mutations studied here, we see similar trends, specifically with P180L where an ordered

N-terminal cap greatly increased expression and function.

Though the stability mutant is not a clinically viable treatment method for RP patients, the underlying mechanisms of rescuing rhodopsin functionality should be pursued to aid in the search for clinically relevant compounds. Compounds targeting and stabilizing the N- terminal cap structure could potentially develop into an effective treatment to rescue rhodopsin function. GPCR binding drugs have been extensively studied and developed

(Kypreos et al., 2014) and, as the N-terminus of rhodopsin is on the extracellular face, which is canonically the ligand binding face of other non-opsin GPCRs, there is interest in that region. Recent work in developing nanobodies to stabilize specific GPCR conformations

(Manglik et al., 2017) could be applied to research into RP drug design. Also, it should be emphasized that potential RP drug development should be focused on restoring function of the protein, as there are now studies showing that simply down-regulating ER stress in vivo allows for the trafficking of defective rhodopsins to the outer segment, where their dysfunction causes the photoreceptor to degrade faster (Athanasiou et al., 2017). Currently there are few studies that have looked at the effect of “stabilizing” rhodopsin states with different compounds, such as antibodies (Piscitelli et al., 2006), lipids (Dong et al., 2015), or small molecules (Mattle et al., 2018) to varying degrees of success. Our study helps highlight the use of individual RP mutant characterization, as these results show that while P180L patients would likely respond to high doses of Vitamin A, G182V patients would not.

However, if in the future, an N-terminal stabilizing drug is developed for RP mutations,

202

G182V patients would likely only benefit from N-terminal stabilization with presence of excess Vitamin A.

The mutant residue at P180 determines severity of RP phenotype

Another retinitis pigmentosa mutation at site 180 has been previously characterized,

P180A (Iannaccone et al., 2006). Interestingly, the P180A clinical phenotype and in vitro phenotype are both milder in comparison to P180L. P180A expresses at high levels and produces a folded functional protein (Iannaccone et al., 2006), in contrast to P180L. The

P180A mutant protein is thermally unstable, activating frequently in the dark (Iannaccone et al., 2006), while P180L fails to produce any functional protein. Typically, the presence of a proline in a structural motif is thought to introduce a kink to the structure, thus there could be an assumption that the presence of a proline at 180 would be critical to the beta sheet folding.

However, the variation of phenotype between P180A and P180L in RP patients suggests that the absence of a proline at site 180 is not the contributing factor for the mutant phenotype, but rather the magnitude of disruption caused by the specific mutant residue is the determining factor for severity. This study continues to elucidate the role of site 180, and we hypothesize that the proline at 180 could be participating mainly electrostatically in the structure of rhodopsin, as the large hydrophobic leucine sidechain could be more disruptive to the hydrophilic space between the beta2 and beta3 strands compared to the smaller alanine, thus producing a more severe phenotype.

Glycine is required at site 182 as RP mutations at site 182 produce a similar in vitro phenotype

203

In contrast to RP mutations at P180, in vitro characterization of another RP mutation found at site 182, G182S (Sheffield et al., 1991), showed a similar disease phenotype to

G182V characterized in this study, with no functional protein purified in vitro and the mutant protein sequestered in the ER (Sung et al., 1993). This suggests the disease phenotype seen with mutations at site 182 could be due to the absence of the glycine, rather than the disruptive presence of the mutant amino acid. This could be due to the glycine contributing to the flexibility of the beta turn linking the beta3 and beta4 strands of the protein (Chou et al;

2002), and removal of G182 could be preventing the proper compact folding of the EL2 between the helices of rhodopsin (Figure 4.8C).

However, with RP mutations at G182, there appears to be a mismatch between the in vitro and clinical phenotype. Both G182S and G182V patients exhibited night blindness in the first decade of life, while G182S patients demonstrated only significant losses of peripheral vision very late in life (Fishman et al., 1992), despite the observation that both mutations have severe in vitro phenotypes (Sung et al., 1993). This mismatch reveals a potential caveat of in vitro characterization, in that the heterologous expression and functional assessment is done in isolation. As both mutations are autosomal dominant, these

RP mutations would be coexpressed in the rod photoreceptor with a wildtype protein.

Dimerization with the wildtype protein has been shown to rescue some misfolded proteins in vitro (Zhang et al., 2016), and this may be a possible explanation of the discrepancy between the in vitro RP G182 mutation phenotypes and the clinical phenotype.

The in vitro phenotype severity of P180L recapitulates the severe P180L patient phenotype

204

Our clinical observations of the P180L patient visual system suggest a severe phenotype due to the rapid degradation of the patient’s visual field early in life. Our in vitro results also demonstrate a severe phenotype with only minimal rescue of functional protein with pharmacological rescue in the wildtype background. Additionally, the rapid hydroxylamine reactivity of the rescued P180L protein in vitro suggests an unstable and potentially dysfunctional mutant rhodopsin. These results combined with the lack of trafficking rescue in vitro suggests that patients with P180L treated with high doses of

Vitamin A may respond minimally to this particular therapeutic treatment. As the G182V patient was lost to follow up, with only nyctalopia known in the first decade of life, we cannot sufficiently judge the patient phenotype in these patients, though the G182V patient phenotype in the first decade is similar to the G182S phenotype. G182S is characterized by nyctalopia in the early teens followed by loss of the inferior visual field very late in life

(Fishman et al., 1992). This mild patient phenotype is in direct contrast to the severe in vitro phenotype of G182V/S (Sung et al., 1993), though this may be due to in vitro expression conditions as discussed previously.

In this study, we investigated two retinitis pigmentosa mutants found on the extracellular loop 2 of rhodopsin, and the ability to rescue function pharmacologically and with a stabilizing disulfide bond. Our results suggest that 1) the disruption to the hydrophilic pocket between the beta3 strand of the EL2 and the beta2 strand of the N-terminus could be disrupting the protein and 2) introducing inflexibility into the EL2 likely disrupts packing.

We propose that the interface between the N-terminus and EL-2 is critical for rhodopsin

205 structure-function, specifically the four stranded anti-parallel beta sheet formed by the two structures. Although the two studied RP mutations are separated by only one amino acid, both show differing levels of severity as well as different responses to rescue methodologies.

Our study highlights the importance of in vitro characterization of individual RP mutations as even different mutations at the same site can lead to differing (P180A, mild vs P180L severe)

(Iannaccone et al., 2006) or similar (G182V and G182S, both severe) (Sung et al., 1993) phenotypes. Clinical observations show that the P180L mutation produces a more severe phenotype in patients in comparison to P180A, recapitulating the in vitro characterization and highlighting the variable nature of RP. Characterization is especially important for therapeutic strategies to slow retinal degradation as certain compounds may, depending on the mutation identity, be ineffectual or even accelerate retinal degeneration. With high throughput and rapid sequencing techniques becoming common, personalized medicine has started to become a viable concept, therefore methodology should be developed to rapidly determine not only disease phenotype but also response to treatment. The characterization of disease mutations is also of significant interest to basic researchers as disease mutations could help isolate and characterize new critical areas of importance in the protein.

206

4.6 - REFERENCES

Aho, A. C., Donner, K., Hydén, C., Larsen, L. O. and Reuter, T. (1988). Low retinal

noise in animals with low body temperature allows high visual sensitivity. Nature 334,

348–350.

Athanasiou, D., Aguilà, M., Bellingham, J., Li, W., McCulley, C., Reeves, P. J. and

Cheetham, M. E. (2018). The molecular and cellular basis of rhodopsin retinitis

pigmentosa reveals potential strategies for therapy. Prog Retin Eye Res 62, 1–23.

Athanasiou, D., Aguilà, M., Opefi, C. A., South, K., Bellingham, J., Bevilacqua, D.,

Munro, P. M., Kanuga, N., Mackenzie, F. E., Dubis, A. M., et al. (2017). Rescue of

mutant rhodopsin traffic by metformin-induced AMPK activation accelerates

photoreceptor degeneration. Hum. Mol. Genet. ddw387–15.

Athanasiou, D., Kosmaoglou, M., Kanuga, N., Novoselov, S. S., Paton, A. W., Paton, J.

C., Chapple, J. P. and Cheetham, M. E. (2012). BiP prevents rod opsin aggregation.

Mol Biol Cell 23, 3522–3531.

Baylor, D. A., Matthews, G. and Yau, K. W. (1980). Two components of electrical dark

noise in toad retinal rod outer segments. J. Physiol. (Lond.) 309, 591–621.

Berger, W., Kloeckener-Gruissem, B. and Neidhardt, J. (2010). The molecular basis of

human retinal and vitreoretinal diseases. Prog Retin Eye Res 29, 335–375.

Bernier, V., Bichet, D. G. and Bouvier, M. (2004). Pharmacological chaperone action on

G-protein-coupled receptors. Curr Opin Pharmacol 4, 528–533.

207

Berson, E. L., Rosner, B., Sandberg, M. A., Weigel-DiFranco, C. and Willett, W. C.

(2012). ω-3 Intake and Visual Acuity in Patients With Retinitis Pigmentosa Receiving

Vitamin A. Arch. Ophthalmol. 130, 707–711.

Berson, E. L., Rosner, B., Sandberg, M. A., Weigel-DiFranco, C., Brockhurst, R. J.,

Hayes, K. C., Johnson, E. J., Anderson, E. J., Johnson, C. A., Gaudio, A. R., et al.

(2010). Clinical trial of lutein in patients with retinitis pigmentosa receiving vitamin A.

Arch. Ophthalmol. 128, 403–411.

Bhattacharyya, N., Darren, B., Schott, R. K., Tropepe, V. and Chang, B. S. (2017).

Cone-like rhodopsin expressed in the all-cone retina of the colubrid pine snake as a

potential adaptation to diurnality. J. Exp. Biol. 220, 2418–2425.

Bownds, D. (1967). Site of attachment of retinal in rhodopsin. Nature 216, 1178–1181.

Budzynski, E., Gross, A. K., McAlear, S. D., Peachey, N. S., Shukla, M., He, F.,

Edwards, M., Won, J., Hicks, W. L., Wensel, T. G., et al. (2010). Mutations of the

opsin gene (Y102H and I307N) lead to light-induced degeneration of photoreceptors and

constitutive activation of phototransduction in mice. J Biol Chem 285, 14521–14533.

Chou, P. Y. and Fasman, G. D. (2002). Prediction of protein conformation. Biochemistry

13, 222–245.

DeCaluwé, G. L. and DeGrip, W. J. (1996). Point mutations in bovine opsin can be

classified in four groups with respect to their effect on the biosynthetic pathway of opsin.

Biochem J 320 ( Pt 3), 807–815.

208

Doi, T., Molday, R. S. and Khorana, H. G. (1990). Role of the intradiscal domain in

rhodopsin assembly and function. Proc. Natl. Acad. Sci. U.S.A. 87, 4991–4995.

Dong, X., Ramon, E., Herrera-Hernández, M. G. and Garriga, P. (2015). Phospholipid

Bicelles Improve the Conformational Stability of Rhodopsin Mutants Associated with

Retinitis Pigmentosa. Biochemistry 54, 4795–4804.

Fernandez-San Jose, P., Blanco-Kelly, F., Corton, M., Trujillo-Tiebas, M.-J., Gimenez,

A., Avila-Fernandez, A., Garcia-Sandoval, B., Lopez-Molina, M.-I., Hernan, I.,

Carballo, M., et al. (2014). Prevalence of Rhodopsinmutations in autosomal dominant

Retinitis Pigmentosa in Spain: clinical and analytical review in 200 families. Acta

Ophthalmol 93, e38–e44.

Fishman, G. A., Stone, E. M., Sheffield, V. C., Gilbert, L. D. and Kimura, A. E. (1992).

Ocular findings associated with rhodopsin gene codon 17 and codon 182 transition

mutations in dominant retinitis pigmentosa. Arch. Ophthalmol. 110, 54–62.

Fritze, O., Filipek, S., Kuksa, V., Palczewski, K., Hofmann, K. P. and Ernst, O. P.

(2003). Role of the conserved NPxxY(x)5,6F motif in the rhodopsin ground state and

during activation. Proc. Natl. Acad. Sci. U.S.A. 100, 2290–2295.

Gregg, R. G., McCall, M. A. and Massey, S. C. (2013). Chapter 15 - Function and

Anatomy of the Mammalian Retina. Fifth Edition. Elsevier Inc.

Gross, A. K., Xie, G. and Oprian, D. D. (2003). Slow binding of retinal to rhodopsin

mutants G90D and T94D. Biochemistry 42, 2002–2008.

209

Hartong, D. T., Berson, E. L. and Dryja, T. P. (2006). Retinitis pigmentosa. Lancet 368,

1795–1809.

Iannaccone, A., Man, D., Waseem, N., Jennings, B. J., Ganapathiraju, M., Gallaher, K.,

Reese, E., Bhattacharya, S. S. and Klein-Seetharaman, J. (2006). Retinitis

pigmentosa associated with rhodopsin mutations: Correlation between phenotypic

variability and molecular effects. Vision Research 46, 4556–4567.

Johnson, P. J. M., Halpin, A., Morizumi, T., Prokhorenko, V. I., Ernst, O. P. and

Miller, R. J. D. (2015). Local vibrational coherences drive the primary photochemistry

of vision. Nature Chemistry 7, 980–986.

Karnik, S. S. and Khorana, H. G. (1990). Assembly of functional rhodopsin requires a

disulfide bond between cysteine residues 110 and 187. J Biol Chem 265, 17520–17524.

Kaushal, S., Ridge, K. D. and Khorana, H. G. (1994). Structure and function in rhodopsin:

the role of asparagine-linked glycosylation. Proc. Natl. Acad. Sci. U.S.A. 91, 4024–4028.

Krebs, M. P., Holden, D. C., Joshi, P., Clark, C. L., III, Lee, A. H. and Kaushal, S.

(2010). Molecular Mechanisms of Rhodopsin Retinitis Pigmentosa and the Efficacy of

Pharmacological Rescue. Journal of Molecular Biology 395, 1063–1078.

Kypreos, M., Banerjee, T. and Mukherjee, D. (2014). G protein-coupled receptors -

potential roles in clinical pharmacology. Cardiovasc Hematol Agents Med Chem 12, 29–

33.

210

Li, S., Xiao, X., Wang, P., Guo, X. and Zhang, Q. (2010). Mutation spectrum and

frequency of the RHO gene in 248 Chinese families with retinitis pigmentosa.

Biochemical and Biophysical Research Communications 401, 42–47.

Liu, M. Y., Liu, J., Mehrotra, D., Liu, Y., Guo, Y., Baldera-Aguayo, P. A., Mooney, V.

L., Nour, A. M. and Yan, E. C. Y. (2013). Thermal Stability of Rhodopsin and

Progression of Retinitis Pigmentosa: A Comparison of S186W and D190N Rhodopsin

Mutants. Journal of Biological Chemistry.

Mahalingam, M., Martínez-Mayorga, K., Brown, M. F. and Vogel, R. (2008). Two

protonation switches control rhodopsin activation in membranes. Proc Natl Acad Sci

USA 105, 17795–17800.

Malanson, K. M. and Lem, J. (2009). Rhodopsin-mediated retinitis pigmentosa. Prog Mol

Biol Transl Sci 88, 1–31.

Manglik, A., Kobilka, B. K. and Steyaert, J. (2017). Nanobodies to Study G Protein-

Coupled Receptor Structure and Function. Annu. Rev. Pharmacol. Toxicol. 57, 19–37.

Mattle, D., Kuhn, B., Aebi, J., Bedoucha, M., Kekilli, D., Grozinger, N., Alker, A.,

Rudolph, M. G., Schmid, G., Schertler, G. F. X., et al. (2018). Ligand channel in

pharmacologically stabilized rhodopsin. Proc. Natl. Acad. Sci. U.S.A. 56, 201718084–6.

Mendes, H. F. and Cheetham, M. E. (2008). Pharmacological manipulation of gain-of-

function and dominant-negative mechanisms in rhodopsin retinitis pigmentosa. Hum.

Mol. Genet. 17, 3043–3054.

211

Mendes, H. F., van der Spuy, J., Chapple, J. P. and Cheetham, M. E. (2005).

Mechanisms of cell death in rhodopsin retinitis pigmentosa: implications for therapy.

Trends in Molecular Medicine 11, 177–185.

Moritz, O. L., Tam, B. M., Papermaster, D. S. and Nakayama, T. (2001). A Functional

Rhodopsin-Green Fluorescent Protein Fusion Protein Localizes Correctly in Transgenic

Xenopus laevisRetinal Rods and Is Expressed in a Time-dependent Pattern. J Biol Chem

276, 28242–28251.

Noorwez, S. M., Malhotra, R., McDowell, J. H., Smith, K. A., Krebs, M. P. and

Kaushal, S. (2004). Retinoids Assist the Cellular Folding of the Autosomal Dominant

Retinitis Pigmentosa Opsin Mutant P23H. Journal of Biological Chemistry 279, 16278–

16284.

Okada, T., Fujiyoshi, Y., Silow, M., Navarro, J., Landau, E. M. and Shichida, Y. (2002).

Functional role of internal water molecules in rhodopsin revealed by X-ray

crystallography. Proc. Natl. Acad. Sci. U.S.A. 99, 5982–5987.

Okada, T., Sugihara, M., Bondar, A.-N., Elstner, M., Entel, P. and Buss, V. (2004). The

Retinal Conformation and its Environment in Rhodopsin in Light of a New 2.2Å Crystal

Structure. Journal of Molecular Biology 342, 571–583.

Opefi, C. A., South, K., Reynolds, C. A., Smith, S. O. and Reeves, P. J. (2013). Retinitis

Pigmentosa Mutants Provide Insight into the Role of the N-terminal Cap in Rhodopsin

Folding, Structure, and Function. J Biol Chem 288, 33912–33926.

212

Ovchinnikov, Y. A., Abdulaev, N. G. and Bogachuk, A. S. (1988). Two adjacent cysteine

residues in the C-terminal cytoplasmic fragment of bovine rhodopsin are palmitylated.

FEBS Lett. 230, 1–5.

Palczewski, K. (2010). Retinoids for treatment of retinal diseases. Trends in

Pharmacological Sciences 31, 284–295.

Palczewski, K., Kumasaka, T., Hori, T., Behnke, C. A., Motoshima, H., Fox, B. A., Le

Trong, I., Teller, D. C., Okada, T., Stenkamp, R. E., et al. (2000). Crystal structure of

rhodopsin: A G protein-coupled receptor. Science 289, 739–745.

Piscitelli, C. L., Angel, T. E., Bailey, B. W., Hargrave, P., Dratz, E. A. and Lawrence, C.

M. (2006). Equilibrium between metarhodopsin-I and metarhodopsin-II is dependent on

the conformation of the third cytoplasmic loop. J Biol Chem 281, 6813–6825.

Ploier, B., Caro, L. N., Morizumi, T., Pandey, K., Pearring, J. N., Goren, M. A.,

Finnemann, S. C., Graumann, J., Arshavsky, V. Y., Dittman, J. S., et al. (2016).

Dimerization deficiency of enigmatic retinitis pigmentosa-linked rhodopsin mutants.

Nature Communications 7, 1–11.

Sakmar, T. P., Menon, S. T., Marin, E. P. and Awad, E. S. (2002). Rhodopsin: insights

from recent structural studies. Annu Rev Biophys Biomol Struct 31, 443–484.

Saliba, R. S., Munro, P. M. G., Luthert, P. J. and Cheetham, M. E. (2002). The cellular

fate of mutant rhodopsin: quality control, degradation and aggresome formation. J Cell

Sci 115, 2907–2918.

213

Schoenlein, R. W., Peteanu, L. A., Mathies, R. A. and Shank, C. V. (1991). The first step

in vision: femtosecond isomerization of rhodopsin. Science 254, 412–415.

Sheffield, V. C., Fishman, G. A., Beck, J. S., Kimura, A. E. and Stone, E. M. (1991).

Identification of novel rhodopsin mutations associated with retinitis pigmentosa by GC-

clamped denaturing gradient gel electrophoresis. Am. J. Hum. Genet. 49, 699–706.

Singhal, A., Ostermaier, M. K., Vishnivetskiy, S. A., Panneels, V., Homan, K. T.,

Tesmer, J. J. G., Veprintsev, D., Deupi, X., Gurevich, V. V., Schertler, G. F. X., et

al. (2013). Insights into congenital stationary night blindness based on the structure of

G90D rhodopsin. Nature Publishing Group 14, 520–526.

Sinha, A., Jones Brunette, A. M., Fay, J. F., Schafer, C. T. and Farrens, D. L. (2014).

Rhodopsin TM6 Can Interact with Two Separate and Distinct Sites on Arrestin:

Evidence for Structural Plasticity and Multiple Docking Modes in Arrestin–Rhodopsin

Binding. Biochemistry 53, 3294–3307.

Standfuss, J., Edwards, P. C., D'Antona, A., Fransen, M., Xie, G., Oprian, D. D. and

Schertler, G. F. X. (2011). The structural basis of agonist-induced activation in

constitutively active rhodopsin. Nature 471, 656–660.

Standfuss, J., Xie, G., Edwards, P. C., Burghammer, M., Oprian, D. D. and Schertler,

G. F. X. (2007). Crystal structure of a thermally stable rhodopsin mutant. Journal of

Molecular Biology 372, 1179–1188.

214

Sung, C. H., Davenport, C. M. and Nathans, J. (1993). Rhodopsin mutations responsible

for autosomal dominant retinitis pigmentosa. Clustering of functional classes along the

polypeptide chain. J Biol Chem 268, 26645–26649.

Sung, C. H., Schneider, B. G., Agarwal, N., Papermaster, D. S. and Nathans, J. (1991).

Functional heterogeneity of mutant rhodopsins responsible for autosomal dominant

retinitis pigmentosa. Proc. Natl. Acad. Sci. U.S.A. 88, 8840–8844. van Hazel, I., Sabouhanian, A., Day, L., Endler, J. A. and Chang, B. S. (2013).

Functional characterization of spectral tuning mechanisms in the great bowerbird short-

wavelength sensitive visual pigment (SWS1), and the origins of UV/violet vision in

passerines and parrots. BMC Evol. Biol. 13, 250.

Vishnivetskiy, S. A., Ostermaier, M. K., Singhal, A., Panneels, V., Homan, K. T.,

Glukhova, A., Sligar, S. G., Tesmer, J. J. G., Schertler, G. F. X., Standfuss, J., et al.

(2013). Constitutively active rhodopsin mutants causing night blindness are effectively

phosphorylated by GRKs but differ in arrestin-1 binding. Cellular Signalling 25, 2155–

2162.

Westhoff, B., Chapple, J. P., van der Spuy, J., Höhfeld, J. and Cheetham, M. E. (2005).

HSJ1 Is a Neuronal Shuttling Factor for the Sorting of Chaperone Clients to the

Proteasome. Current Biology 15, 1058–1064.

Whitmore, A. V. and Bowmaker, J. K. (1989). Seasonal variation in cone sensitivity and

short-wave absorbing visual pigments in the rudd Scardinius erythrophthalmus. J Comp

Physiol A 166.

215

Xie, G., Gross, A. K. and Oprian, D. D. (2003). An opsin mutant with increased thermal

stability. Biochemistry 42, 1995–2001.

Yan, E. C. Y., Kazmi, M. A., Ganim, Z., Hou, J.-M., Pan, D., Chang, B. S., Sakmar, T.

P. and Mathies, R. A. (2003). Retinal counterion switch in the photoactivation of the G

protein-coupled receptor rhodopsin. Proc. Natl. Acad. Sci. U.S.A. 100, 9262–9267.

Yang, G., Xie, S., Feng, N., Yuan, Z., Zhang, M. and Zhao, J. (2014). Spectrum of

rhodopsin gene mutations in Chinese patients with retinitis pigmentosa. Mol. Vis. 20,

1132–1136.

Zhang, T., Cao, L.-H., Kumar, S., Enemchukwu, N. O., Zhang, N., Lambert, A., Zhao,

X., Jones, A., Wang, S., Dennis, E. M., et al. (2016). Dimerization of visual pigments

in vivo. Proc. Natl. Acad. Sci. U.S.A. 201609018–6.

Zhu, L., Imanishi, Y., Filipek, S., Alekseev, A., Jastrzebska, B., Sun, W., Saperstein, D.

A. and Palczewski, K. (2006). Autosomal recessive retinitis pigmentosa and E150K

mutation in the opsin gene. J Biol Chem 281, 22289–22298.

216

4.7 – SUPPLEMENTAL INFORMATION

Table S1 – Clinical Patient Summaries. Data collected by Elise Heon.

Case 1

p.P180L c.539C>T NM_000539.3

Age at first symptoms VA (age) GVF (age)

Nyctalopia age 10 yrs 20/20 (14) 20 degrees (14)

20/20 (17) 10 degrees (17)

20/25 (19) 10 degrees (19)

VA: visual acuity (Snellen). GVF: Goldmann visual field.

Case 2 p. G182V, c. 545G>T, NM_000539.3

Symptoms of nyctalopia in the first decade. Positive family history. No further clinical information available. Patient lost to follow-up.

217

CHAPTER V: COMPARING RETINITIS PIGMENTOSA RHODOPSIN MUTATIONS IN VITRO TO CLINICAL PHENOTYPES

Nihar Bhattacharyya, James Morrow, Elise Heon, Belinda S. W. Chang

Author Contributions: NB, JM, and BSWC conceptualized study. JM created constructs and vectors. NB and JM conducted expressions. NB performed all microscopy. EH genotyped and collected patient data. NB wrote chapter with input from BSWC.

Portions of this data will appear in a manuscript: Scott BM, Chen SK, Bhattacharyya N, Heon E, Peisajovich SG, Chang BSW “Coupling of Human Rhodopsin to a Yeast Signaling Pathway Enables High-Throughput Characterization of Mutations Associated with Retinal Disease”, submitted to Genetics

5.1 - ABSTRACT

The retinal degenerative disease retinitis pigmentosa (RP) is characterized by variation in disease phenotype. Similarly, RP-causing mutations in the dim-light sensor rhodopsin are also highly variable in the degree of severity and mechanism of disruption.

Thus far, few studies have sought to demonstrate if in vitro characterizations of RP mutation phenotype severity correlate to the severity of disease in RP patients. In this study, three RP mutations in rhodopsin were investigated (A164E, A164V, and V81del), of which two were newly discovered and recently reported (A164E and V81del). We first assessed protein folding and intracellular trafficking with a GFP-fusion expression vector followed by spectroscopic assays of purified mutant protein to determine functionality, and response to pharmacological rescue with 11-cis retinal chromophore. We then compared the in vitro results to clinical data evaluating the visual field and the degradation of the retina in patients with the corresponding RP mutations. Our in vitro results revealed that all three mutations exhibited differing levels of severity and response to pharmacological rescue. A164V produced a mild phenotype in vitro, demonstrating no observable intracellular trafficking defect in SK-N-SH cells, a model system for mutational studies of trafficking and misfolding

218 defects, and little effect on expression of functional protein. The other RP mutation at site

164, A164E, was found to be more severe, with protein retained in the endoplasmic reticulum and only minimal levels of functional protein expression. However, A164E did respond to rescue with 11-cis retinal chromophore showing increases in both trafficking and expression levels. The most severe mutation characterized was V81del, which was retained inside the cell, with no functional protein recovered even after pharmacological rescue with

11-cis retinal. Clinical studies of patients with these rhodopsin mutations showed a ranked order of severity that is comparable to the results of our in vitro studies (least severe to most severe, A164V < A164E < V81del). The A164V patient had the latest age of onset and retained the most intact visual field, while the V81del patient showed a more advanced progression, and possessed a severely degraded retina, and the A164E patient presenting an intermediate phenotype. This chapter demonstrates the utility of applying the results of in vitro techniques characterizing RP rhodopsin mutations to clinical interpretations of patient mutations, particularly the variability of the response to pharmacological rescue. Such studies could in the future assist clinicians with developing more accurate prognoses and therapeutic strategies for preserving or slowing the degradation of the retina in RP patients.

219

5.2 - INTRODUCTION

Mutations in the dim-light sensor rhodopsin are thought to account for 30% of autosomal dominant retinitis pigmentosa (RP) cases (Malanson and Lem, 2009), with mutations found throughout the structure of rhodopsin affecting the stability and function of the protein in a highly variable manner (Mendes et al., 2005) (Figure 5.1). Clinical RP phenotypes vary considerably with respect to severity, age of onset, and response to therapeutics (Berger et al., 2010). However, the precise relation between the in vitro phenotype RP rhodopsin and the clinical phenotype of RP patients with mutations in rhodopsin has not been extensively investigated. In vitro characterization of RP mutations in rhodopsin could determine if the biochemical phenotype of the specific mutation underlies the heterogeneity observed in the patient phenotype and assist in creating a customized treatment regimen specific to the causal

RP mutation.

Rhodopsin is a seven transmembrane G protein coupled receptor found in rod photoreceptor cells and is responsible for vision in dim light. Due to the nature of its function, rhodopsin is highly specialized (Fritze et al., 2003), incredibly sensitive (Sampath and Baylor, 2002) and possess ultrafast kinetics (Sakmar et al., 2002). Upon photon absorption, the covalently bound 11-cis retinal chromophore isomerizes to the all-trans form to initiate the conformational shift in the protein structure to the active state, ultimately leading to signal transduction inside the photoreceptor. The bound 11-cis retinal acts not only as a light sensor but also as an inverse agonist (Han et al., 1997), stabilizing the dark state structure of rhodopsin when bound. Researchers have exploited the stabilizing nature of the chromophore in vitro to stabilize mutant rhodopsin protein and rescue misfolding. With this method, several RP mutations have been shown to be pharmacologically rescued by 11-cis

220 retinal (Krebs et al., 2010). Indeed, this strategy has been used to slow retinal degradation in transgenic animals and patients (Berson et al., 2012; Berson et al., 2010; Li et al., 1998;

Palczewski, 2010), and underlies the therapeutic treatment of RP patients with high doses of

Vitamin A regardless of genotype. However, not all RP mutations respond to pharmacological rescue with retinoids in vitro ((Opefi et al., 2013) and Chapter IV), and with no clear predictor of rescue response, specific mutations require individual assays in order to determine if retinoids would be effective as therapeutic agents.

Several attempts have been made to classify the different biochemical phenotypes of RP mutations found in rhodopsin (Mendes et al., 2005). Class II mutations are the largest class and includes RP mutations that cause rhodopsin to misfold inside the cell, resulting in ER stress and/or aggregation, which eventually leads to photoreceptor cell death and retinal degradation (Sung et al., 1993; Sung et al., 1991). Studies of Class II RP mutations show a range of misfolding and retention inside the cell and an array of responses to rescue methods

((Opefi et al., 2013), Chapter IV). For example, RP mutations at site 23 have been very well studied, as P23H is one of the most common RP mutations in North America (Athanasiou et al., 2018). P23A and P23L are also RP mutations found at this site, but all three show different levels of severity and different responses to pharmacological and structure rescue methods (Opefi et al., 2013). This level of phenotype variability signifies that, given a novel

RP mutation, the severity or response to rescue likely cannot be accurately predicted without some manner of in vitro characterization.

Recently, two novel RP mutations were uncovered in patients with reduced visual function: A164E and V81del (Figure 5.1). While another RP mutation at site 164 (A164V) had been studied, the impact of these new RP mutations on rhodopsin function had yet to be

221 investigated. A164V had previously been shown to prevent packing of the transmembrane helices in rhodopsin due to steric hindrance, but was an otherwise mild RP mutant with increased thermal instability (Stojanovic et al., 2003). A164 is found on the fourth transmembrane domain in close proximity to the chromophore binding pocket and the small alanine side chain directs at and interacts with transmembrane helix 3 (Palczewski et al.,

2000; Venkatakrishnan et al., 2013). V81 is another hydrophobic transmembrane residue found in the middle of helix 2 and the hydrophobic side chain juts out into the lipid space, and the deletion of this residue could likely destabilize the alpha helical structure of helix 2

(Palczewski et al., 2000). Due to the hydrophobic nature of these amino acids and their location, we can hypothesize that these residues are involved in helix packing or other interhelical interactions.

In vitro studies of RP mutations have characterized critical components of rhodopsin structure and function (Janz and Farrens, 2004; McKibbin et al., 2007; Opefi et al., 2013)

(Chapter IV), but often this is not accompanied by investigations of corresponding patient phenotypes. Here, we have characterized three RP mutations in rhodopsin, V81del, A164E, and A164V to assess the intracellular translation and trafficking using fluorescence microscopy. Additionally, we have expressed and purified these mutant rhodopsin proteins to assess their function in response to light. We then evaluated the effects of pharmacological rescue with 11-cis retinal chromophore on both trafficking and expression. Our experimental results were compared to newly reported clinical data from RP patients with corresponding rhodopsin mutations, to investigate the relationship between the severity of our in vitro and clinical data.

222

5.3 – MATERIALS AND METHODS

Vector and sequence construction

Previous studies have shown that GFP tagged rhodopsin is very stable as the C-terminus of rhodopsin is quite flexible and can accommodate the GFP protein in vitro and in vivo

(Moritz et al. 2001). We constructed the pGFP expression vector to allow us to easily insert rhodopsin full-length sequences into a plasmid that would result in a rhodopsin-GFP fusion protein. pGFP was constructed by cloning and ligating the humanized recombinant GFP II

(hrGFP II) gene in pIRES-hrGFPII (Stratagene) and modifying the multiple cloning site

(MCS). The hrGFP II sequence was amplified out of the pIRES-hrGFPII vector using primers that added the BamHI restriction site to the 5’ end and the XhoI restriction site to the

3’ end of the sequence. The MCS of the pIRES vector was then mutated to change the

BamHI cut site to BglII and the EcoRI to SalI. Subsequently, this pIRES was digested to remove the hrGFPII reporter gene and ligated with a peptide containing a BamHI cut site and the previously amplified hrGFPII sequence. This resulted in an expression plasmid that would fuse GFP to the C-terminus of any inserted rhodopsin sequence missing a stop codon with a 2-amino acid linker.

Bovine rhodopsin, the most widely used model system for studies of rhodopsin structure and function (Janz and Farrens, 2004; Palczewski et al., 2000), was used as the template for all constructs in this study. Mutagenesis primers were used to introduce the retinitis pigmentosa mutations P23H, V81del, A164V, and A164E in the wildtype background using the QuickChange II protocol (Table S5.1). Constructs were sequenced to verify successful mutation and cloned into the p1D4 and pGFP expression vectors.

223

Immunocytochemistry

Human SK-N-SH neuroblastoma (ATCC HTB-11) cells, an in vitro model for aggregation-based diseases (Westhoff et al. 2005; Mendes & Cheetham 2008), were grown and cultured in full media (DMEM (Life technologies), 10% FBS (Invitrogen), and

Penicillin-Streptomycin (Invitrogen)) at 37º C in 5% CO2 and seeded into 24-well plates with coverslips (Sarstedt) while under 5 passages. Once cells reached approximately 75% confluence, they were transfected with 645ng of construct in pGFP using Lipofectamine

2000 (Invitrogen) protocols. Cells were then either given 10 uM 9-cis retinal (Sigma Aldrich) for 24 hrs and 20uM for an additional 24hrs or given ethanol as a control. Wildtype bovine rhodopsin was used as a positive control for proper translation and trafficking, while P23H was used as a control for the RP disease phenotype and ER retention/rescue.

After 24 hours, half the wells were incubated with WGA in HBSS for 10 minutes at 37º

C to label the cell membrane. All cells were then rinsed with PBS and fixed with 2% paraformaldehyde in PBS. To label cells with the endoplasmic reticulum marker antibody, anti-calreticulin (1:400, Abcam), cells were washed and permeabilized in PBS containing 1% bovine serum albumin (Sigma) and 0.1% saponin (PBS-BS). Anti-calreticulin was diluted in

PBS-BS and incubated for 1 hour at room temperature. After washing with PBS-BS, secondary antibody (Cy3-conjugated goat anti-rabbit IgGt, 1:200, Jackson Immunoresearch) was diluted in PBS-BS and added to the wells for 1 hour. Nuclei were stained with Hoechst

(1:1000 in PBS, Hoechst type 33258 Invitrogen) for 10 minutes. Cells were mounted with

ProLong Gold Antifade (Thermofisher), coverslipped and allowed to cure for 24 hours in the dark prior to imaging on Leica TCSSP8 confocal microscope. ImageJ was used to construct

Z-stacks, maximum projection images and scale bars.

224

Expression

In order to produce purified rhodopsin for functional assays, heterologous expression and purification took place as previously described (Bhattacharyya et al. 2017). Briefly, a 10 cm plate of HEK293T cells were transfected with 8 ug of construct in p1D4 using Lipofectamine

2000 protocols. Cells were then either given 10 uM 11-cis retinal for 24 hrs and 20uM for an additional 24hrs or given ethanol as a control. Cells were cultured in the dark at 37ºC in 5%

CO2. 48 hours post-transfection, cells were harvested and regenerated in 5uM 11-cis retinal then solubilized in DM and affinity purified with 1D4 antibody coupled to sepharose beads.

Purified protein was eluted using WB2 buffers and 1D4 peptide. Wildtype bovine rhodopsin was used as control for both the purification protocol and for wildtype rhodopsin function.

UV-Vis spectra of purified rhodopsins were collected using a Cary 4000 dual-beam spectrophotometer (Agilent) at 20ºC.

Retinitis pigmentosa patient disease phenotype

The patient cases were selected from an internal database and the phenotype information was collected retrospectively. Other than basic demographic and genetic information, we collected information about visual acuity (VA), color vision, Goldmann visual fields (GVF), electroretinography (ERG) and imaging. Imaging included fundus photography

(VisucamNM/FA - Carl Zeiss Meditec, Dublin, California, USA and Optos), and optical coherence tomography (OCT, Cirrus from Carl Zeiss Meditec, Dublin, California, USA).

Genetic testing was conducted using gene panels-based sequencing by CLIA approved laboratories. This study was approved by the Human Research Ethics Board of The Hospital for Sick Children and met the tenets of the Declaration of Helsinki.

225

5.4 - RESULTS

Retinitis pigmentosa mutants V81del and A164E are retained inside the cell, while

A164V is successfully trafficked to the cell membrane

We used an in vitro cellular system to assess protein aggregation and trafficking within the cell. The human neuroblastoma cell line, SK-N-SH, had been previously used to study other protein aggregation diseases (Westhoff et al., 2005), and is therefore ideal for qualitatively assessing protein aggregation and trafficking of RP mutants in vitro. Wildtype and RP mutants were inserted into the novel pGFP expression vector, which generated rhodopsin protein fused with a GFP protein on the C-terminus with a 2-amino acid linker.

P23H was used as a positive control, as a rhodopsin mutant that can be rescued with addition of retinal, but is otherwise retained in the ER (Mendes and Cheetham, 2008).

While wildtype rhodopsin-GFP was properly trafficked to the membrane of the cell

(Figure 5.2A) where it colocalized with the cell membrane marker, P23H-GFP was retained inside the cell and colocalized with the ER marker (Figure 5.2C) as previously reported

(Mendes and Cheetham, 2008). A164V-GFP, consistent with the relatively mild disease phenotype characterized previously in vitro (Stojanovic et al., 2003), displayed a phenotype similar to wildtype as it was properly trafficked to the cell membrane (Figure 5.2B). A164E-

GFP and V81del-GFP, however, were both retained inside the cell and partially colocalized with the ER-marker (Figure 5.2E, G). Both A164E-GFP and V81del-GFP showed intracellular localization that did not overlap with the ER marker, suggesting potential cytosolic localization where the fusion protein could be degraded by cellular machinery or aggregated.

226

RP mutants at site 164 express in vitro and respond to light, while V81del produces no functional protein

RP mutations in the wildtype rhodopsin background were constructed, expressed and purified in parallel with P23H and wildtype bovine rhodopsin. Purified rhodopsin samples were light bleached to assess protein function. The UV-Vis dark-light difference absorbance spectra of the wildtype rhodopsin (Figure 5.3A) shows the typical expression levels of wildtype rhodopsin with a dark-state λmax of 499nm, that fully activated after 30 seconds of light-bleach at 20º C with an active peak at 380nm. The P23H control expressed at very low levels, consistent with previous studies (Opefi et al., 2013)(Figure 5.3C, black trace), although the minimal amount of protein expressed was activated by light. In our system,

A164V expressed well, similar to (Stojanovic et al., 2003) at approximately 1/5 of wildtype levels (Figure 5.3B) and produced protein that responded to light. The novel RP mutants

A164E and V81del both expressed at lower levels than A164V. A164E expressed at levels similar to P23H (Figure 5.3D, black trace) and responded to light bleach. V81del showed no dark peak and no light peak, indicating the absence of functional rhodopsin protein with the ability to bind 11-cis retinal (Figure 5.3E, black trace).

RP mutant trafficking and function can be pharmacologically rescued in A164E but not

V81del

Past studies have shown that 11-cis retinal, and the retinal analog 9-cis retinal, can pharmacologically rescue expression, trafficking, and function of rhodopsin mutants in vitro

(Mendes and Cheetham, 2008; Opefi et al., 2013). The retinal chromophore of rhodopsin acts as an inverse agonist, stabilizing the dark state structure. Therefore, by treating tissue culture

227 cells with 11-cis or 9-cis retinal upon transfection, the chromophore can be incorporated into the nascent rhodopsin protein during translation and folding in the ER, thereby potentially stabilizing the overall structure of the mutant rhodopsin, and negating to a certain degree the destabilizing effects of the disease mutation (Han et al., 1997).

Here, P23H was used as a positive control for a rescuable RP phenotype as previous studies have established the pharmacological rescue of P23H with both 11-cis and 9-cis retinal (Mendes and Cheetham, 2008; Opefi et al., 2013). Levels of heterologously expressed

P23H were increased by approximately 4-fold with the addition of 11-cis retinal (Figure

5.3C, red trace). This pharmacological rescue was also observed using immunocytochemistry, as P23H-GFP was no longer sequestered inside the cell but successfully trafficked to the cell membrane (Figure 5.2D). A164E also demonstrated a rescuable phenotype, as expression levels increased ~1.5-fold (Figure 5.3D, red trace). The

GFP fusion protein was also successfully trafficked to the cell membrane upon treatment with 11-cis retinal, where it colocalized with the cell membrane marker (Figure 5.2E).

However, V81del proved to be resistant to rescue, as even with chromophore exposure the

GFP-fusion protein was retained inside the cell and colocalized with the ER marker (Figure

5.2H), with no presence of active mutant rhodopsin when heterologously expressed (Figure

5.3E, red trace). These results suggest a clear trend in increasing in vitro severity of RP mutants, with A164V being the least severe, and V81del being the most severe while A164E showing an intermediate level of severity.

The retina of all three patients show differing amounts of retinal degradation over time

228

The comparative severity in vitro correlated with the patient phenotype information

(Table S5.2). The patient with the V81del mutation first had symptoms of difficulty adapting to a dim lit environment (nyctalopia) around 10 years of age. This slowly progressed, and at

39 years she has moderate visual acuity loss (20/50), mildly abnormal color vision, and constriction of the visual field to the central 5 degrees. At age 26, electroretinography documented severe reduction of rod and cone function (Figure 5.4). The phenotype of the patient with the A164V variant was milder than that of the patient with the A164E variant.

Although the A164V patient had symptoms of nyctalopia since childhood, the progression of his disease was extremely slow. At age 64 years, his electroretinogram was recordable and only mildly abnormal. His central visual acuity at 67 years was 20/40 and despite a paracentral scotoma (area of decreased vision), he maintained a peripheral field (Figure 5.4).

In contrast, the patient with the A164E mutation has good central visual acuity and normal color vision at age 53. However, her paracentral scotoma was more severe and progressed to form an annular scotoma at the age of 53. Unlike the patient with the V81del mutation, she also preserved some good peripheral field of vision at the age of 45, and her ERG was only moderately abnormal (Figure 5.4).

Of the 3 patients, the V81del patient presented with highly reduced retinal function at the youngest age, while the A164V patient had the best retinal function at a more advanced age

(Figure 5.4). The patient with the A164E mutation again appeared intermediate to both.

These results show that the overall trend of clinical severity corresponded well with the in vitro characterization of their rhodopsin mutations.

229

Figure 5.1: 2D and 3D visualizations of dark-state rhodopsin structure with retinitis pigmentosa mutations highlighted. (A) A snake plot showing known RP sites in red. RP sites investigated in this study are highlighted in green. P23H is highlighted in blue. (B) Tertiary structure model of dark state rhodopsin (PDB: 1U19) with known RP sites highlighted in red. RP sites investigated in this study are highlighted in green. P23H is highlighted in blue.

230

Figure 5.2: Confocal fluorescent microscopy images of SK-N-SH cells expressing rhodopsin constructs (green) with endoplasmic reticulum and cell membrane markers (red). (A) Wildtype bovine rhodopsin-GFP fusion protein (green) colocalizes with the cell membrane (red, top row) and is not found in the ER (red, bottom row). (B) RP mutant

231

A164V rhodopsin-GFP fusion protein (green) colocalizes with the cell membrane (red, top row) and is not found in the ER (red, bottom row). (C) RP mutant P23H rhodopsin-GFP fusion protein (green) colocalizes with the ER (red, bottom row) and is not found on the cell surface (red, top row). (D) Pharmacological rescue with 11-cis retinal causes the RP mutant P23H rhodopsin-GFP fusion protein (green) to colocalize with the cell membrane (red, top row) and is found in the cell, but not in the ER (red, bottom row). (E) RP mutant A164E rhodopsin-GFP fusion protein (green) colocalizes with the ER (red, bottom row) and is not found on the cell surface (red, top row). (F) Pharmacological rescue with 11-cis retinal causes the RP mutant A164E rhodopsin-GFP fusion protein (green) to colocalize with the cell membrane (red, top row) and is found in the cell, but not in the ER (red, bottom row). (G) RP mutant V81del rhodopsin-GFP fusion protein (green) colocalizes with the ER (red, bottom row) and is not present on the cell surface (red, top row). (F) Pharmacological rescue with 11-cis retinal causes the RP mutant V81del rhodopsin-GFP fusion protein (green) to colocalize with the ER (red, bottom row) and is not found on the cell surface (red, top row).

232

Figure 5.3: UV-visible absorbance difference spectra of immunoaffinity purified rhodopsin constructs. Dark-state spectra were obtained, and then the samples were light bleached to obtain the light-state spectra. The light spectra were then subtracted from the dark-state spectra to obtain difference spectra. (A) Wildtype rhodopsin expresses at high levels and responds completely to light bleach. (B) A164V protein produces less protein, but still responds to light. (C) P23H produces minimal functional protein (black trace) but can be pharmacologically rescued (red trace). (D) A164E produces minimal functional protein (black trace) but does respond to pharmacological rescue (red trace). (E) V81del produces no functional protein with (red trace) or without (black trace) pharmacological rescue.

233

Figure 5.4: Clinical Assessment of Patients with Rhodopsin Mutations V81Δ, A164E, and A164V. (A) Goldmann visual fields of the right eye at two time points. Normal fields would reach the gray dotted line. The solid blue line outlines the actual field. The hatched areas are scotoma, i.e. areas of loss in sensitivity. Darker areas refer to denser scotoma. (B)Structural retinal phenotype of the right eye from cases carrying the A164E and V81Δ mutations. Optical coherence tomography (OCT) above showing the different retinal layers. Brackets show area of preserved outer retina; A164E > V81Δ. Unlike for A164E, the

234

OCT of V81Δ shows disturbed lamination of the retina with degenerative cysts, reflecting more advanced disease. The retinal photograph below centered on the posterior pole. Photograph on the right is taken with a wider field camera. ON: optic nerve. The dotted white line indicated the foveal area at the center of the macula. Double white arrow indicates vessel attenuation, while single arrow shows typical pigmentary deposits (few in these cases). The width of the central visual field corresponds to the area of preserved outer retina on the OCT. Data collected and figure created by Elise Heon.

235

Figure 5.5: Homology modelling showing the effects of RP mutations at site 164 on the orientation of E122 and H211. (A) Entire tertiary structure of rhodopsin (PDB: 1U19) (B) In wildtype rhodopsin, the E122-H211 lock appears between A164 and the chromophore. (C) In A164V, the steric bulk of the mutated residue (blue) causes the orientation of E122-H211 to change somewhat, but mainly the larger sidechain disrupts the packing between helix 3 and helix 4. (D) In A164E, the larger negatively charged sidechain (pink) causes larger shifts between E122-H211, however E164 adds additional steric hindrance to helix packing.

236

5.5 - DISCUSSION

In this study, we compared the in vitro characterization of novel mutations in rhodopsin with clinical disease phenotype in RP patients with the corresponding mutations, in order to assess the validity of using in vitro characterization of RP mutants to predict patient outcomes and response to treatment. We investigated three RP mutations in vitro, A164V and A164E which were identified previously (Stojanovic et al., 2003; Vaithinathan et al.,

1994), and V81del which was newly reported here. Our in vitro investigations found that the three mutations studied demonstrated a gradient of severity using three primary methodologies: functional assessment via spectroscopy of purified mutant pigment; fluorescent microscopy assessing mutant protein folding and trafficking in a SK-N-SH neuroblastoma cell line; and rescue of phenotype in both systems via exposure to 11-cis or 9- cis retinal. These methods revealed A164V as the mildest mutation and V81del as the most severe. Goldmann visual fields, electroretinography, and visual acuity tests assessing the degradation of vision in RP patients in conjunction with the age of the patients was used to assess disease severity in vivo. In comparing the in vitro data to the in vivo phenotype of the disease, we suggest that the relative severity of the three RP mutants is the same in both systems, thus indicating that in vitro characterization of RP mutants is a useful approach that can be used to assist with improving patient prognoses and possible clinical treatment options.

Retinitis pigmentosa mutations exhibit differing severities consistent with type of mutation

Two RP mutations characterized in this study occur at the same residue in rhodopsin.

This is not the first example of multiple RP mutation identities occurring at the same site in

237 the protein ((Janz and Farrens, 2003; Opefi et al., 2013) and Chapter IV). Multiple RP mutations at a single residue can occur when the specific amino acid identity is critical to the protein structure and function, thus any mutation at that loci may result in a disease phenotype (e.g. cysteines involved in disulfide bonding (McKibbin et al., 2007). Another possible cause for multiple RP mutations found at a site is when the general properties of the wildtype amino acid are integral at the specific location in the protein (e.g. electrostatic interactions or hydrophobic residues (Krebs et al., 2010)) and as such, any mutation to an amino acid of a different class could cause a disease phenotype. This is the case for two common examples of RP with multiple amino acid identities at one site: P23 and G51. P23H is a typical Class II RP mutant that causes the photoreceptor cell to undergo apoptosis due to accumulated ER stress and aggregation of the misfolded mutant rhodopsin. P23L and P23A are also known RP mutations which are less severe than P23H and pharmacological rescue is more effective on these mutants. With the removal of the rigid proline, the backbone of the

N-terminus becomes flexible, causing the protein to misfold (Opefi et al., 2013). Similarly, the RP mutants G51A, V, and L all have been shown to disrupt the packing between the transmembrane helices I and II of rhodopsin due to the replacement of the small glycine with the sterically larger alanine, valine, or leucine (Bosch et al., 2003).

Previously, A164V has been characterized in vitro. It produces a thermally unstable pigment that has a very short-lived active state compared to wildtype, which was thought to be due to steric interference in helical packing due to the larger valine sidechain (Stojanovic et al., 2003). Despite this, our results demonstrate that in vitro, A164V appeared to be translated, processed and trafficked to the cell membrane, and these aspects of protein function appeared unaffected. By disrupting the helical packing, A164V on helix 4 may be

238 disrupting an ionic salt bridge between Glu122 on helix 3 and His211 on helix 5, which is a critical interhelical interaction in maintaining conformations stability (Ahuja et al., 2009;

Choe et al., 2011). Homology modelling of A164V (Figure 5.5C) shows that the increased steric hindrance from the larger residue at site 164 could be causing a change in distance between the sidechains of E122 and H211. However, as valine is only slightly larger than alanine and hydrophobic, the in vitro phenotype of A164V was quite mild in comparison to the other RP mutation at site 164, A164E.

We characterized the disease phenotype of A164E in vitro and our results revealed a RP mutation more severe than A164V. In comparison to alanine and valine, glutamine is a much larger amino acid and introduces a negative charge to the region. Given the knowledge about

A164V, A164E likely disrupts E122-H211 to such an extent where only a small proportion of rhodopsin protein can fold and bind 11-cis retinal successfully. As E122-H211 falls between A164E and the retinal molecule, it is logical that this mutant responds well to pharmacological rescue. Crystal structures have shown that the presence of retinal in the protein does affect the orientation of E122 and H211 sidechains (Choe et al., 2011), therefore the stabilizing retinal binding during translation likely supports the formation of E122-H211 and possibly compensates for the added instability from A164E. Homology modelling of this

RP mutation suggests that the bond distance between E122 and H211 decreased even more compared to A164V (Figure 5.5D), possibly disrupting the direct interaction between the two residues and the larger interaction with the extended hydrogen bonding network (Hofmann et al., 2009; Li et al., 2004). However, the best fit homology model is with the negatively charged glutamine side chain at 164 directed away from the protein, out into the hydrophobic lipid layer which was not modelled. It is likely that the negatively charged glutamine

239 sidechain would rotate to point inward, away from the hydrophobic lipids which would potentially disrupt the helical packing and the E122-H211 interaction to a large degree.

Based on the in vitro phenotype of both A164V and A164E, we would propose both mutations as Class II mutations, causing pathogenesis in vivo due to misfolded protein.

Additionally, for A164V, the ability for the mutant protein to activate G protein has yet to be measured. Due to disruption to the E122-H211 ionic lock important in the conformational shift to the active state (Ahuja et al., 2009) and the lowered thermal stability of the mutant protein (Stojanovic et al., 2003), this mutant could also be causing pathogenicity by over activation of the rod photoreceptor (Tam et al., 2014).

The most severe RP mutation we investigated was V81del. As site 81 is located mid transmembrane helix 1, a deletion here would likely misalign subsequent helices within the lipid bilayer. Additionally, intramolecular/interhelical interactions would no longer be able to align, therefore it is somewhat unsurprising that a deletion would cause the protein to potentially misfold so severely. This is likely why deletion RP mutations in rhodopsin are quite rare (Keen et al., 1991; Krebs et al., 2010). V81del is also a possible Class II mutant, as we detected mutant protein in the ER and in the cytosol where it was likely being degraded or aggregated.

Severity of RP mutant phenotype follows the same gradient in in vitro assays as well as patient progression

Studies of retinitis pigmentosa mutations in rhodopsin are generally divided into two broad categories: clinical papers sequencing rhodopsin, including pedigrees and describing the disease phenotypes in patients (Berson et al., 2002; Bonilha et al., 2015; Dryja et al.,

240

1991; Dryja et al., 2000; Dryja et al., 1990; Pan et al., 2012); or experimental investigations, where known RP mutations in rhodopsin are expressed and characterized separate from clinical contexts (Dong et al., 2015; Liu et al., 2013; Ploier et al., 2016; Tam et al., 2014).

However, there are few studies that incorporate both types of data to compare the patient disease phenotype to the in vitro investigation (Iannaccone et al., 2006; Zeitz et al., 2008).

Thus, the literature demonstrates a lack of cross talk between basic scientists and clinicians so that the characterized properties of RP mutations in rhodopsin in vitro rarely impact strategies for clinical therapies. The current suggested therapeutic regimen for retinitis pigmentosa patients are high doses of vitamin A palmitate, omega-3 fatty acids and lutein

(Berson et al., 2010; Berson et al., 2012; Marmor, 1993), in an effort to increase levels of retinal and stabilize presumably rhodopsin in the eye. As clinical trials showing the effectiveness of this strategy have not been conducted on genotyped individuals, this regiment could be undertaken by RP patients with rhodopsin RP mutations that do not respond to pharmacological rescue, even though there is known risk of toxicity for patients taking high doses of vitamin A (Hathcock et al., 1990).

All three mutations showed variable response to pharmacological rescue in vitro and showed a range of severity in vitro, from mild (A164V) to intermediate (A164E) and severe

(V81del). In the disease progression of the RP patients, this same order of severity was preserved with the oldest patient with A164V having the most intact vision, while the youngest patient possessing V81del having the most limited vision, and the A164E patient showing an intermediate phenotype. Characterization is especially important for therapeutic strategies to slow retinal degradation as certain compounds may, depending on the mutation identity, be ineffectual or even accelerate retinal degeneration (Athanasiou et al., 2017).

241

Potential mismatches seen between in vitro and patient data could be attributed to variable

RNA transcript stability in the different cell types. Additionally, missense mutations in the genome could be resulting in missplicing in the patient (Rosenfeld et al., 1992), but as in vitro techniques utilize intron-less sequences in the expression vector, this property remains uncharacterized in vitro. Personalized medicine, which seeks to pair disease diagnoses with patients’ genotypes, is becoming a reality with the advancements in high throughput and rapid sequencing techniques. Therefore, methodology should be developed to rapidly determine not only disease phenotype but also response to treatment. Individual characterization of mutant phenotypes would be of great use to clinicians and basic scientists alike, as demonstrated by this study. For example, given the three RP mutations characterized here, one would not respond to treatment with Vitamin A (V81del) which is of clinical relevance, and A164E has expanded on the body of knowledge surrounding rhodopsin helix packing and the H211-E122 interaction, which is of relevance to rhodopsin researchers.

In this study, we characterized novel RP mutations in rhodopsin in vitro and compared them to the disease phenotype. We used a new expression vector and fluorescent microscopy to assess folding and trafficking in vitro and absorbance spectroscopy to assay mutant rhodopsin expression and response to light. We additionally observed the response to pharmacological rescue with retinal analogs with both the microscopy and spectroscopy methods. Our results highlight the significant heterogeneity of RP mutations in rhodopsin. As demonstrated by the two mutations at site 164 showing differing trafficking and expression levels and also the more severe mutations, A164E and V81del, responding differently to pharmacological rescue. When comparing our in vitro results to the clinical measurements of

242 retinal field degradation over time, we see the recapitulation of our in vitro results with the patient with the most degraded retina possessing the RP mutation that was most severe in vitro and vice versa. Our study serves to highlight, that due to the highly variable nature of

RP, the pairing of in vitro characterization with clinical treatment strategies would be valuable, especially in this emerging age of personalized medicine.

5.6 - REFERENCES

Ahuja, S., Hornak, V., Yan, E. C. Y., Syrett, N., Goncalves, J. A., Hirshfeld, A., Ziliox,

M., Sakmar, T. P., Sheves, M., Reeves, P. J., et al. (2009). Helix movement is coupled

to displacement of the second extracellular loop in rhodopsin activation. Nat. Struct. Mol.

Biol. 16, 168–175.

Athanasiou, D., Aguilà, M., Bellingham, J., Li, W., McCulley, C., Reeves, P. J. and

Cheetham, M. E. (2018). The molecular and cellular basis of rhodopsin retinitis

pigmentosa reveals potential strategies for therapy. Prog Retin Eye Res 62, 1–23.

Athanasiou, D., Aguilà, M., Opefi, C. A., South, K., Bellingham, J., Bevilacqua, D.,

Munro, P. M., Kanuga, N., Mackenzie, F. E., Dubis, A. M., et al. (2017). Rescue of

mutant rhodopsin traffic by metformin-induced AMPK activation accelerates

photoreceptor degeneration. Hum. Mol. Genet. ddw387–15.

Berger, W., Kloeckener-Gruissem, B. and Neidhardt, J. (2010). The molecular basis of

human retinal and vitreoretinal diseases. Prog Retin Eye Res 29, 335–375.

243

Berson, E. L., Rosner, B., Sandberg, M. A., Weigel-DiFranco, C. and Willett, W. C.

(2012). ω-3 Intake and Visual Acuity in Patients With Retinitis Pigmentosa Receiving

Vitamin A. Arch. Ophthalmol. 130, 707–711.

Berson, E. L., Rosner, B., Sandberg, M. A., Weigel-DiFranco, C., Brockhurst, R. J.,

Hayes, K. C., Johnson, E. J., Anderson, E. J., Johnson, C. A., Gaudio, A. R., et al.

(2010). Clinical trial of lutein in patients with retinitis pigmentosa receiving vitamin A.

Arch. Ophthalmol. 128, 403–411.

Berson, E. L., Rosner, B., Weigel-DiFranco, C., Dryja, T. P. and Sandberg, M. A.

(2002). Disease progression in patients with dominant retinitis pigmentosa and rhodopsin

mutations. Investigative Ophthalmology & Visual Science 43, 3027–3036.

Bonilha, V. L., Rayborn, M. E., Bell, B. A., Marino, M. J., Beight, C. D., Pauer, G. J.,

Traboulsi, E. I., Hollyfield, J. G. and Hagstrom, S. A. (2015). Retinal histopathology

in eyes from patients with autosomal dominant retinitis pigmentosa caused by rhodopsin

mutations. Graefes Arch Clin Exp Ophthalmol 1–9.

Bosch, L., Ramon, E., del Valle, L. J. and Garriga, P. (2003). Structural and functional

role of helices I and II in rhodopsin. A novel interplay evidenced by mutations at Gly-51

and Gly-89 in the transmembrane domain. J Biol Chem 278, 20203–20209.

Chiang, W. C., Messah, C. and Lin, J. H. (2012). IRE1 directs proteasomal and lysosomal

degradation of misfolded rhodopsin. Mol Biol Cell 23, 758–770.

244

Choe, H.-W., Kim, Y. J., Park, J. H., Morizumi, T., Pai, E. F., Krauss, N., Hofmann, K.

P., Scheerer, P. and Ernst, O. P. (2011). Crystal structure of metarhodopsin II. Nature

471, 651–655.

Dong, X., Ramon, E., Herrera-Hernández, M. G. and Garriga, P. (2015). Phospholipid

Bicelles Improve the Conformational Stability of Rhodopsin Mutants Associated with

Retinitis Pigmentosa. Biochemistry 54, 4795–4804.

Dryja, T. P., Hahn, L. B., Cowley, G. S., McGee, T. L. and Berson, E. L. (1991).

Mutation spectrum of the rhodopsin gene among patients with autosomal dominant

retinitis pigmentosa. Proc. Natl. Acad. Sci. U.S.A. 88, 9370–9374.

Dryja, T. P., McEvoy, J. A., McGee, T. L. and Berson, E. L. (2000). Novel rhodopsin

mutations Gly114Val and Gln184Pro in dominant retinitis pigmentosa. Investigative

Ophthalmology & Visual Science 41, 3124–3127.

Dryja, T. P., McGee, T. L., Reichel, E., Hahn, L. B., Cowley, G. S., Yandell, D. W.,

Sandberg, M. A. and Berson, E. L. (1990). A point mutation of the rhodopsin gene in

one form of retinitis pigmentosa. Nature 343, 364–366.

Fritze, O., Filipek, S., Kuksa, V., Palczewski, K., Hofmann, K. P. and Ernst, O. P.

(2003). Role of the conserved NPxxY(x)5,6F motif in the rhodopsin ground state and

during activation. Proc. Natl. Acad. Sci. U.S.A. 100, 2290–2295.

Han, M., Lou, J., Nakanishi, K., Sakmar, T. P. and Smith, S. O. (1997). Partial Agonist

Activity of 11-cis-Retinal in Rhodopsin Mutants. J Biol Chem 272, 23081–23085.

245

Hathcock, J. N., Hattan, D. G., Jenkins, M. Y., Mcdonald, J. T., Sundaresan, P. R. and

Wilkening, V. L. (1990). Evaluation of Vitamin-a Toxicity. Am. J. Clin. Nutr. 52, 183–

202.

Hofmann, K. P., Scheerer, P., Hildebrand, P. W., Choe, H.-W., Park, J. H., Heck, M.

and Ernst, O. P. (2009). A G protein-coupled receptor at work: the rhodopsin model.

Trends Biochem Sci 34, 540–552.

Iannaccone, A., Man, D., Waseem, N., Jennings, B. J., Ganapathiraju, M., Gallaher, K.,

Reese, E., Bhattacharya, S. S. and Klein-Seetharaman, J. (2006). Retinitis

pigmentosa associated with rhodopsin mutations: Correlation between phenotypic

variability and molecular effects. Vision Research 46, 4556–4567.

Janz, J. M. and Farrens, D. L. (2003). Assessing structural elements that influence Schiff

base stability: mutants E113Q and D190N destabilize rhodopsin through different

mechanisms. Vision Research 43, 2991–3002.

Janz, J. M. and Farrens, D. L. (2004). Role of the retinal hydrogen bond network in

rhodopsin Schiff base stability and hydrolysis. J Biol Chem 279, 55886–55894.

Keen, T. J., Inglehearn, C. F., Lester, D. H., Bashir, R., Jay, M., Bird, A. C., Jay, B. and

Bhattacharya, S. S. (1991). Autosomal dominant retinitis pigmentosa: four new

mutations in rhodopsin, one of them in the retinal attachment site. Genomics 11, 199–

205.

246

Krebs, M. P., Holden, D. C., Joshi, P., Clark, C. L., III, Lee, A. H. and Kaushal, S.

(2010). Molecular Mechanisms of Rhodopsin Retinitis Pigmentosa and the Efficacy of

Pharmacological Rescue. Journal of Molecular Biology 395, 1063–1078.

Li, J., Edwards, P. C., Burghammer, M., Villa, C. and Schertler, G. F. X. (2004).

Structure of bovine rhodopsin in a trigonal crystal form. Journal of Molecular Biology

343, 1409–1438.

Li, T., Sandberg, M. A., Pawlyk, B. S., Rosner, B., Hayes, K. C., Dryja, T. P. and

Berson, E. L. (1998). Effect of vitamin A supplementation on rhodopsin mutants

threonine-17 --> methionine and proline-347 --> serine in transgenic mice and in cell

cultures. Proc. Natl. Acad. Sci. U.S.A. 95, 11933–11938.

Lin, J. H., Li, H., Yasumura, D., Cohen, H. R., Zhang, C., Panning, B., Shokat, K. M.,

Lavail, M. M. and Walter, P. (2007). IRE1 signaling affects cell fate during the

unfolded protein response. Science 318, 944–949.

Liu, M. Y., Liu, J., Mehrotra, D., Liu, Y., Guo, Y., Baldera-Aguayo, P. A., Mooney, V.

L., Nour, A. M. and Yan, E. C. Y. (2013). Thermal Stability of Rhodopsin and

Progression of Retinitis Pigmentosa: A Comparison of S186W and D190N Rhodopsin

Mutants. Journal of Biological Chemistry.

Malanson, K. M. and Lem, J. (2009). Rhodopsin-mediated retinitis pigmentosa. Prog Mol

Biol Transl Sci 88, 1–31.

Marmor, M. F. (1993). A Randomized Trial of Vitamin A and Vitamin E Supplementation

for Retinitis Pigmentosa. Arch. Ophthalmol. 111, 1460–1461.

247

McKibbin, C., Toye, A. M., Reeves, P. J., Khorana, H. G., Edwards, P. C., Villa, C. and

Booth, P. J. (2007). Opsin stability and folding: The role of Cys185 and abnormal

disulfide bond formation in the intradiscal domain. Journal of Molecular Biology 374,

1309–1318.

Mendes, H. F. and Cheetham, M. E. (2008). Pharmacological manipulation of gain-of-

function and dominant-negative mechanisms in rhodopsin retinitis pigmentosa. Hum.

Mol. Genet. 17, 3043–3054.

Mendes, H. F., van der Spuy, J., Chapple, J. P. and Cheetham, M. E. (2005).

Mechanisms of cell death in rhodopsin retinitis pigmentosa: implications for therapy.

Trends in Molecular Medicine 11, 177–185.

Opefi, C. A., South, K., Reynolds, C. A., Smith, S. O. and Reeves, P. J. (2013). Retinitis

Pigmentosa Mutants Provide Insight into the Role of the N-terminal Cap in Rhodopsin

Folding, Structure, and Function. J Biol Chem 288, 33912–33926.

Palczewski, K. (2010). Retinoids for treatment of retinal diseases. Trends in

Pharmacological Sciences 31, 284–295.

Palczewski, K., Kumasaka, T., Hori, T., Behnke, C. A., Motoshima, H., Fox, B. A., Le

Trong, I., Teller, D. C., Okada, T., Stenkamp, R. E., et al. (2000). Crystal structure of

rhodopsin: A G protein-coupled receptor. Science 289, 739–745.

Pan, Z., Lu, T., Zhang, X., Dai, H., Yan, W., Bai, F. and Li, Y. (2012). Identification of

two mutations of the RHO gene in two Chinese families with retinitis pigmentosa:

correlation between genotype and phenotype. Mol. Vis. 18, 3013–3020.

248

Ploier, B., Caro, L. N., Morizumi, T., Pandey, K., Pearring, J. N., Goren, M. A.,

Finnemann, S. C., Graumann, J., Arshavsky, V. Y., Dittman, J. S., et al. (2016).

Dimerization deficiency of enigmatic retinitis pigmentosa-linked rhodopsin mutants.

Nature Communications 7, 1–11.

Rosenfeld, P.J., Cowley, G.S., McGee, T.L., Sandberg, M.A., Berson, E.L., Dryja, T.P.

(1992). A null mutation in the rhodopsin gene causes rod photoreceptor dysfunction and

autosomal recessive retinitis pigmentosa. Nature genetics, 1(3), pp.209–213.

Sakmar, T. P., Menon, S. T., Marin, E. P. and Awad, E. S. (2002). Rhodopsin: insights

from recent structural studies. Annu Rev Biophys Biomol Struct 31, 443–484.

Sampath, A. P. and Baylor, D. A. (2002). Molecular mechanism of spontaneous pigment

activation in retinal cones. Biophys. J. 83, 184–193.

Stojanovic, A., Hwang, I., Khorana, H. G. and Hwa, J. (2003). Retinitis Pigmentosa

Rhodopsin Mutations L125R and A164V Perturb Critical Interhelical Interactions: NEW

INSIGHTS THROUGH COMPENSATORY MUTATIONS AND CRYSTAL

STRUCTURE ANALYSIS. Journal of Biological Chemistry 278, 39020–39028.

Sung, C. H., Davenport, C. M. and Nathans, J. (1993). Rhodopsin mutations responsible

for autosomal dominant retinitis pigmentosa. Clustering of functional classes along the

polypeptide chain. J Biol Chem 268, 26645–26649.

Sung, C. H., Schneider, B. G., Agarwal, N., Papermaster, D. S. and Nathans, J. (1991).

Functional heterogeneity of mutant rhodopsins responsible for autosomal dominant

retinitis pigmentosa. Proc. Natl. Acad. Sci. U.S.A. 88, 8840–8844.

249

Tam, B. M., Noorwez, S. M., Kaushal, S., Kono, M. and Moritz, O. L. (2014).

Photoactivation-Induced Instability of Rhodopsin Mutants T4K and T17M in Rod Outer

Segments Underlies Retinal Degeneration in X. laevis Transgenic Models of Retinitis

Pigmentosa. Journal of Neuroscience 34, 13336–13348.

Vaithinathan, R., Berson, E. L. and Dryja, T. P. (1994). Further screening of the

rhodopsin gene in patients with autosomal dominant retinitis pigmentosa. Genomics 21,

461–463.

Venkatakrishnan, A. J., Deupi, X., Lebon, G., Tate, C. G., Schertler, G. F. and Babu,

M. M. (2013). Molecular signatures of G-protein-coupled receptors. Nature 494, 185–

194.

Westhoff, B., Chapple, J. P., van der Spuy, J., Höhfeld, J. and Cheetham, M. E. (2005).

HSJ1 Is a Neuronal Shuttling Factor for the Sorting of Chaperone Clients to the

Proteasome. Current Biology 15, 1058–1064.

Zeitz, C., Gross, A. K., Leifert, D., Kloeckener-Gruissem, B., McAlear, S. D., Lemke, J.,

Neidhardt, J. and Berger, W. (2008). Identification and functional characterization of a

novel rhodopsin mutation associated with autosomal dominant CSNB. Invest.

Ophthalmol. Vis. Sci. 49, 4105–4114.

250

5.7 – SUPPLEMENTAL INFORMATION

Table S5.1. Site-directed mutagenesis primers.

Mutation Primers (3' to 5')

P23H CGTGGTGCGCAGCCACTTCGAGGCCCCGC

GCGGGGCCTCGAAGTGGCTGCGCACCACG

V81del CCTGCTCAACCTGGCCGCCGACCTCTTCATGG

CCATGAAGAGGTCGGCGGCCAGGTTGAGCAGG

A164E CTTCACCTGGGTCATGGAGCTGGCCTGTGCCGCGC

GCGCGGCACAGGCCAGCTCCATGACCCAGGTGAAG

A164V CTTCACCTGGGTCATGGTTCTGGCCTGTGCCGCGC

GCGCGGCACAGGCCAGAACCATGACCCAGGTGAAG

251

Table S5.2 Phenotype summary. Data collected, and table created by Elise Heon.

Case 1 (V81Δ) Case 2 (A164V) Case 3 (A164E)

Onset symptoms 10 yo Teenage years 10 yo

VA (age) 20/60 (34) 20/25 (63) 20/20 (40)

20/80 (36) 20/40 (67) 20/25 (44)

20/50 (38) 20/30 (47)

20/30 (53)

Color vision Mild defect Normal Normal

GVF (age) 10 degrees (34) Paracentral scotoma Paracentral scotoma (60) (47)

5 degrees (39) denser paracentral annular Scotoma (53) scotoma (64)

ERG amplitude (age) rod: ND (26) Rod: LLNormal (63) rod: ND (45)

Rod/cone: ⇓99% Rod/cone: Mild Rod/cone: ⇓98% decrease

Cone: ⇓60% Cone: ⇓30%

Legend: Ref sequence NM_000539.3. VA: visual acuity, ND: response not detected, LL: lower limit.

252

CHAPTER VI: GENERAL DISCUSSION

6.1 – GENERAL SUMMARY

This thesis contributes to the broadening our understanding of rhodopsin functional variation through evolutionary adaptation, chromophore use and disease. By characterizing the visual system of non-model and non-mammalian organisms, I characterized an unusually blue-shifted terrestrial rhodopsin exhibiting cone opsin-like properties as a possible evolutionary adaptation for trichromatic colour vision in diurnal colubrid snakes (Chapter

II). By characterizing the less studied A2 chromophore used by fishes, amphibians and reptiles, I identified novel non-spectral properties of A2 rhodopsin and refined the relationship between the λmax and chromophore (Chapter III). By studying rhodopsin disease mutations and their response to rescue, I isolated the role of the beta3 strand in rhodopsin structure (Chapter IV) and characterized four new disease phenotypes in rhodopsin while demonstrating the correlation between in vitro and in vivo phenotype

(Chapter V). I believe that rhodopsins from understudied species or with uncommon chromophores or with unique disease mutations are an endless resource for studies of protein structure and function. For example, recent investigations into rhodopsins from catfish and orca (Castiglione et al., 2018; Dungan and Chang, 2017) have revealed previously uncharacterized regions of rhodopsin participating in intramolecular epistasis. While another study investigating novel methods of rescuing function in pathogenic rhodopsin mutants has isolated previously uncharacterized conformational flexibility in rhodopsin (Mattle et al.,

2018). This chapter will discuss the relevance of my results while proposing future directions of research, expanding on prior concepts.

253

6.2 – GENERAL DISCUSSION

In Chapter II, I showed that the diurnal colubrid snake Pituophis melanoleucus expressed an unusual cone opsin-like rhodopsin in the superficially all-cone retina. We concluded that it was likely that the photoreceptor expressing rhodopsin and rod transducin was a rod photoreceptor that had undergone transmutation thus developing a cone-like morphology.

We proposed that the functional purpose of transmutation in diurnal colubrid snakes could be compensating for past retinal degeneration occurring during an evolutionary bottleneck in a prolonged fossorial or aquatic period. During this period, ancestral snakes lost several cone opsins in the middle-wavelength region, precluding the ability of colour vision due to minimal spectral overlap between the cone opsins LWS and SWS1. As a possible adaptation to diurnality, we suggest that the cone-like qualities of both the photoreceptor and the opsin could enable further light activity and enable chromatic vision under mesopic light conditions where both cones and rods can be active. While we characterized the relative “openness” of the chromophore binding pocket of P. melanoleucus rhodopsin, further characterizations of non-spectral properties, such as the stability of the light activated state and thermal stability of the dark-state rhodopsin, could reveal more cone opsin-like properties. The effects of S185 on G-protein activation are currently uncharacterized thus the effects of the residue identity on modulating G-protein activation are currently theoretical. Also of interest is the low expression of P. melanoleucus rhodopsin in vitro which was unusual as rhodopsin proteins are typically more stable than cone opsins in vitro, yet the more “unstable” cone opsins of P. melanoleucus expressed at high levels in the same mammalian heterologous expression system. The unusual spectral curve of P. melanoleucus rhodopsin (very high A280, very low

Aλmax) is intriguing as, in our system, rhodopsin which do not successfully form dark-state

254

functional protein (low Aλmax) are typically degraded intracellularly (low A280) as can be seen with RP mutations in Chapters V and IV. Intracellular fluorescent microscopy of tagged P. melanoleucus could reveal if the snake rhodopsin is properly processed by the mammalian cells, and if it is not we could incubate the cells with 11-cis retinal to act as an inverse agonist to stabilize the nascent protein as it is being translated. The high A280-low Aλmax however suggests that regeneration of the P. melanoleucus apoprotein with chromophore may be different, which is also a future line of inquiry.

While we, and others (Simões et al., 2016), conclude that transmutation is widespread throughout colubrid snakes, the question remains if transmutation in colubrids was limited to the photoreceptor layer of the retina. In one study (Jacobs and Deegan, 1992), the scotopic

(rod) response in the diurnal colubrid snake, Thamnophis sirtalis, was not detected. This puts the status of rod-specific pathways in the retina and brain into question. The presence of rod bipolar cells and AII amacrine cells, both of which are specific to the rod photoresponse pathway, has never been established in the colubrid retina. However, it is possible that the transmuted rod photoreceptors are signaling through the secondary and tertiary rod pathways

(summarized in Gregg et al., 2013) which signal via connections to adjacent cones and cone bipolar cells. These pathways are thought to be less sensitive, functioning at mesopic light levels with cone photoreceptors. The possible use of these secondary and tertiary rod pathways due to the degradation of the scotopic photo response would lend support to our theory of trichromatic colour vision in mesopic light levels when both the cone-like rod and the canonical cone photoreceptors can function. A study from our lab showed that snakes had lost the phototransduction gene GRK1 which codes for the rod-specific kinase which phosphorylates activated rhodopsin to initiate deactivation (Schott et al., 2018). Rhodopsin

255 kinase has been thought to contribute to the higher sensitivity of rod photoreceptors, as

GRK1 phosphorylates rhodopsin slower when compared to cone opsin kinase (GRK7). This is thought to allow more rod transducin to be activated before shut off, amplifying signal transduction (Tachibanaki et al., 2005). With GRK1 absent in snakes, we assume that GRK7 is expressed and functional in rod photoreceptors, as the presence of GRK7 has been observed in zebrafish rods (Wada et al., 2006) though this has not been established in snakes.

Expression of GRK7 in zebrafish rod photoreceptors produced rods which were less sensitive to light but also produced smaller cone-like response to single photo stimulus (Vogalis et al.,

2011). If the snake cone opsin kinase was indeed expressed and functioning in the cone-like rod photoreceptor, this would be yet another replicated cone-like function of the rod photoreceptor. In the phototransduction cascade, there are several components that have rod/cone specific identities, such as the G-protein, arrestin, cGMP phosphodiesterase, and the cyclic nucleotide gated channel, and it is currently unknown if transmutation has had any effect on the functionality of these proteins.

We believe that the cone opsin-like rhodopsin may be enabling trichromatic colour vision in the diurnal colubrid snake. However, in diurnal colubrids, behavioural studies testing colour vision in mesopic light levels (where the both the cones and the cone-like rod would be active) have never been conducted. While it is rare to perform behavioural studies on colubrid snakes, there have been successful behavioural studies conducted to assay chemoreception in colubrids (Drummond, 1985). There are several potential uses of trichromatic colour vision in snakes, for example colour vision could be assisting in species or mate identification, as colubrid snakes can and do have coloured patterning that could be used for intraspecific signaling.

256

In Chapter III, we characterized the effect of two different vertebrate chromophores on rhodopsin sensitivity and function. Two significant conclusions of this study were that the

λmax red-shift when utilizing the A2 chromophore may be tuned by the protein sequence and that the A2 chromophore causes the active state to decay faster, suggesting another potential adaptation for those species switching chromophores. For the first conclusion, we showed that our A1-A2 λmax dataset formed a linear relationship that did not fit previous modelled relationships. We suggest that, due to the apparent role of the protein sequence in the magnitude of the shift, that perhaps individual mathematical relationships for the separate opsin classes would be advisable. As the λmax pairs from zebrafish and bowerbird demonstrate, the protein sequences may be modulating the magnitude of red-shift with the

A2 chromophore. Due to the different electronic structures of the two chromophores, the presence of A2-specific spectral tuning sites is a possibility.

Zebrafish are a diurnal freshwater fish species that is a part of the Cyprinidae family, the largest fish family, where the majority of the species inhabits freshwater environments. The majority of Cyprinids investigated utilize the A2 chromophore with or without A1 (Toyama et al., 2008), however there are cyprinids that exclusively use the A1 chromophore, including zebrafish (Wang et al., 2008). Though zebrafish utilizes the A1 chromophore under normal conditions, treatment with thyroid hormone has been shown to switch the retinal chromophore to A2 in zebrafish and other fishes (Enright et al., 2015). As the bowerbird and zebrafish A1 λmax are identical, but the zebrafish has a higher magnitude of red-shift with the A2, we can compare the amino acid sequence of bowerbird and zebrafish to identify potential A2-specific spectral tuning sites. Though with an only 80% sequence identity between the two species, it may be difficult to isolate the specific residues responsible.

257

However, another closely related cyprinid, the goldfish (Carassius auratus) exclusively uses

A2 chromophore and has had purified visual pigments characterized in vitro with both A1 and A2 chromophore (Parry and Bowmaker, 2000). In Parry et al (2000), the A2 λmax of goldfish rhodopsin was measured at 522.2 nm with the A1 blue-shifting the λmax only 19.5 nm to 502.7 nm, suggesting again that the 25 nm spectral shift is zebrafish specific. There are only 25 amino acid differences between goldfish and zebrafish rhodopsin, with certain notable differences. Site 112, adjacent to the counter ion at site 113, is a proline in goldfish and a leucine in zebrafish. Goldfish and zebrafish have identical residues at site 123 and 124, so the differences seen at those sites in bowerbird may be mediating the large difference in retinal release between bowerbird and zebrafish, instead of the A2 λmax (Morrow and

Chang, 2015; van Hazel et al., 2016). Also of interest is the sequence variability seen in the extracellular loop 2, a conserved region which comprises part of the chromophore binding pocket. A comparative study of zebrafish, goldfish and bowerbird utilizing site-directed mutagenesis could isolate the specific residues mediating the larger 25 nm spectral shift observed in zebrafish. For future investigations into A2 pigments, the effect of the chromophore changes on other vertebrate A2 species such as amphibians (Xenopus (Kefalov et al., 2003)) and reptiles (such as the American gecko (Kawamura and Yokoyama, 1998)) would be an intriguing line of inquiry. It would be fascinating to not only see if other chromophore effects on rhodopsin properties become apparent in these studies, but we could expand our search for A2 specific spectral tuning sites to include cone opsins which have a larger range of spectral sensitivities.

Investigating the role of the chromophore in non-spectral properties of rhodopsin can now be conducted due to the presence of two canonical chromophores. We showed that the

258

A2 chromophore had no effect on the activation energy of Schiff base linkage hydrolysis, but other studies have shown a lower thermal stability in A2 pigments, and the chromophore interacts with rhodopsin in many different ways. After isomerization, the beta-ionone ring pushes against the transmembrane helices to cause conformational changes, the effect of A2 on rhodopsin activation is unknown. How the all-trans A2 effects the meta equilibria is unknown. The effect of A2 on the extensive hydrogen bonding throughout the structure is unknown. How other known spectral tuning sites interact with the A2 chromophore is unknown. A1:A2 ratios in the retina have been shown to have geographic zones where the

A2 chromophore is favoured dorsally/ventrally in fish (Temple, 2011). These differing regions of chromophore content are possible adaptations for aerial or aquatic fields of view differences in spectral sensitivity or light levels or as a potential adaptation against phototoxicity from repeated light bleaches. But unless the non-spectral properties of A2 rhodopsin are fully characterized, the physiological significance of chromophore use will remain theorized. There have been numerous studies demonstrating adaptation of non- spectral properties rhodopsin, many from our lab (Castiglione et al., 2018; Castiglione et al.,

2017; Hauser et al., 2017; Weadick et al., 2012) therefore the use of A2 chromophore could have a significant effect on opsin functionality beyond red-shifting the λmax sensitivities of the pigment.

In Chapter IV and V, I showed the variable response of multiple RP mutations in rhodopsin to two forms of rescue: pharmacological rescue with retinal (inverse agonist to stabilize the dark state), or structural rescue with the introduction of an additional disulfide bond (stabilizing the N-terminal cap over the seven transmembrane helix bundle). I showed in Chapter V that the characterization of RP mutation rescues in vitro correlates to patient

259 phenotype severity. This is significant as, when studying rhodopsin in vitro, the protein is isolated and is functionally characterized away from other molecular components found in rod photoreceptors. A continuous caveat of in vitro studies is whether properties characterized remain valid in vivo environment, here I will discuss the potential impact of dimerization and lipid composition on in vitro characterization of rhodopsin, in particular RP mutant rhodopsin.

For some time now, it has been known that rhodopsin exists as dimers and oligomers in the outer segment of rod photoreceptors (Fotiadis et al., 2003). The homodimers, formed through dimerization interfaces found on transmembrane helices 1, 2, 4, and 5 (Liang, 2003), have been shown to asymmetrically interact with both G protein transducin (Filipek et al.,

2004) and arrestin (Zhuang et al., 2012). Dimerization of rhodopsin can stabilize the overall structure of the protein as one member of the homodimer could even be without chromophore, as the rhodopsin paired with the apoprotein would be enough to stabilize the structure (Zhang et al., 2016). Mutations in rhodopsin which disrupt dimerization have been shown to cause autosomal dominant retinitis pigmentosa (Ploier et al., 2016). In rhodopsin adRP patients, the wildtype and mutant rhodopsin are both expressed in the same rod photoreceptor, and the heterodimer of RP-wildtype rhodopsin has been shown to both rescue folding of the RP mutant protein (Zhang et al., 2016) or cause the wildtype to misfold

(Mendes and Cheetham, 2008). Therefore, with regards to the characterization of disease mutations in rhodopsin, the mutant phenotype may be different if co-expressed with wildtype rhodopsin. Studies have shown that in vitro, rhodopsin can form dimers in DM detergent micelles in concentrations of detergent below 3 mM (Jastrzebska et al., 2004) and that dimeric rhodopsin in vitro has a slower rate of retinal release and a higher affinity for

260 transducin. We solubilize our cells in 19 mM DM, yet we purify at concentrations of 1.9 mM, therefore we may or may not be characterizing the properties of monomer rhodopsin, however heterologously expressed rhodopsin is known to form dimers in the membrane of

HEK293T cells prior to harvest (Jastrzebska et al., 2004). Therefor future studies investigating the stabilizing effects of dimerization, or heterodimers, or mutations disrupting dimerization can utilize fluorescent tagging for intracellular assessment of dimerization, or utilize lowered concentrations of detergent, or perhaps a different detergent all together to characterize protein function in purified samples.

In our studies of rhodopsin, all in vitro characterization of purified rhodopsin occurred separate from the native membrane. A recent study demonstrated the functional rescue of an

RP mutation with lipids (Dong et al., 2015) and may be why fatty acids are recommended and are possibly effective as a therapeutic for retinitis pigmentosa patients (Berson et al.,

2012; Surette, 2008), thus it is known that lipid composition can and does affect rhodopsin functionality and/or stability. The importance of lipid composition on in vivo rhodopsin functionality was shown in a recent animal study where mice that were deficient in docosahexaenoic acid (DHA), which enriches photoreceptor membranes (Brown, 1994), had impairment in rhodopsin signal transduction (Senapati et al., 2018). Lipid bicelle nanodiscs have recently been in the in vitro characterization of rhodopsin that replicate the effects of in vivo membrane on rhodopsin properties (Szundi et al., 2017; Van Eps et al., 2017). With nanodiscs, future studies investigating rhodopsin from disparate species can tailor lipid composition to better replicate in vivo environments as lipid composition can change depending on multiple factors (Spector and Yorek, 1985; Weinstein et al., 1969).

261

Rhodopsin evolved from cone opsins and is considered functionally distinct from cone opsins. Cone opsins are optimized for fast regeneration, with short-lived active states

(Imai et al., 2005) compared to rhodopsins with more open and looser chromophore binding pockets (Piechnick et al., 2012) that enable faster regeneration (Lamb, 2013). This more open binding pocket allows for the entry of small molecules, and thus all cone opsins react when incubated with hydroxylamine with a half-life of mere minutes (Das et al., 2004; Ma et al.,

2001; Wald et al., 1953). Conversely, rhodopsins are characterized by slower meta II decay rates (Imai et al., 2005), a slower regeneration time due to closed binding pocket that is inaccessible to small molecules in the dark (Imai et al., 2005; Piechnick et al., 2012). Thus, rhodopsins do not react with hydroxylamine (Crescitelli, 1956; Wald et al., 1953), though previous to studies in this thesis there were two exceptions, the echidna (Bickelmann et al.,

2012) and the anole (Kawamura and Yokoyama, 1998) which reacted with hydroxylamine over several hours of incubation. In comparison to cone opsins, rhodopsins are characterized with very low rates of spontaneous activity (Barlow, 1956; Rieke and Baylor, 2000) and the ability to detect minute amounts of light (Barlow, 1956) of a typical wavelength of 500nm.

Among the many rhodopsin proteins studied in this thesis, we see that the above distinctions begin to blur. In comparison to the canonical rhodopsin, A2 rhodopsins and

A164V exhibit shorter lived active states as well as lowered thermal stability (Ala-Laurila et al., 2004; Stojanovic et al., 2003). When considering the EL2 rhodopsin mutants, all three assayed (sG182V, sP180L, and P180L) show more open chromophore binding pockets allowing for reactions with hydroxylamine. The most cone-like rhodopsin characterized was the rhodopsin protein from the diurnal colubrid Pituophis melanoleucus which was not only unusually blue-shifted, and reacted with hydroxylamine but the presence of S185 suggests

262 less G-protein activation, a short lived active state, and increased thermal instability (Karnik and Khorana, 1990; Karnik et al., 1988). P. melanoleucus rhodopsin is a wildtype rhodopsin yet it reacts to hydroxylamine faster and expresses worse in comparison several of the RP disease mutations characterized in this thesis. And while the lowered expression levels of P. melanoleucus rhodopsin may be due to incompatibilities between the mammalian heterologous expression system and the snake protein, the cone opsin genes from P. melanoleucus (which should be more unstable, thermally and structurally than rhodopsin by definition) express at high levels in the same system. Therefore, RP mutations in rhodopsin could be considered as functional variants of rhodopsin. While phylogenetic methods allow us to use sequence information to differentiate between rhodopsin protein and cone opsin

(Chapter 2, (Schott et al., 2016; Weadick et al., 2012)) and even find new rhodopsin genes

(Morrow et al., 2011), I believe that, given the degree of natural variation of rhodopsin and the range of visual environments, it will become very difficult to functionally differentiate between rhodopsin and cone opsins as more non-model rhodopsin become characterized.

Chapter II and Schott et al. (2016) also demonstrated this blurring of lines in photoreceptors as several structural properties typically separating a rod photoreceptor and a cone photoreceptor had become shared in species undergone transmutation.

263

6.3 - CONCLUSION

Too often in science we find ourselves constrained to model organisms or lines of inquiry in the context of human relevance. Though there are very valid rationalizations for this focus as model organisms provide a wealth of background knowledge to help contextualize results, most in vitro techniques have been designed and tailored to model organisms and studies investigating biological systems in the context of human disease is necessary in the treatment and cure for disease. However, by limiting ourselves to model organisms or mammalian systems, we deny ourselves access to the wealth of information contained within natural variation. Natural variation in extant species is essentially the result of a deep mutational scanning and varied optimization experiments that has been ongoing over millions of years, over an endless array of ecological niches. And by limiting ourselves to mammals/humans/model organisms, we can miss functional variants fine-tuned by evolution.

In this thesis alone, through the study of diurnal colubrid snakes, we have characterized novel functional adaptations of a cone opsin-like rhodopsin. Through the study of a rare and understudied vertebrate chromophore, we have shown novel non-spectral methods of adaptation in addition to potentially isolating an uncharacterized method of modulating

λmax. Even through the study of human disease, this thesis has for the first time highlighted the importance of the beta3 strand in rhodopsin function and demonstrated a connection between in vitro and in vivo phenotype. It is my hope that this thesis has shown that the advantages of the study of non-model organisms or dysfunctional proteins and how they can help increase our body of knowledge of protein structure and function, which, in the future, would aid in contextualizing studies of model organisms or disease phenotypes.

264

6.3 - REFERENCES

Ala-Laurila, P., Pahlberg, J., Koskelainen, A. and Donner, K. (2004). On the relation

between the photoactivation energy and the absorbance spectrum of visual pigments.

Vision Research 44, 2153–2158.

Barlow, H. B. (1956). Retinal noise and absolute threshold. J Opt Soc Am 46, 634–639.

Berson, E. L., Rosner, B., Sandberg, M. A., Weigel-DiFranco, C. and Willett, W. C.

(2012). ω-3 Intake and Visual Acuity in Patients With Retinitis Pigmentosa Receiving

Vitamin A. Arch. Ophthalmol. 130, 707–711.

Bickelmann, C., Morrow, J. M., Müller, J. and Chang, B. S. (2012). Functional

characterization of the rod visual pigment of the echidna (Tachyglossus aculeatus), a

basal mammal. Vis. Neurosci. 29, 1–7.

Brown, M. F. (1994). Modulation of rhodopsin function by properties of the membrane

bilayer. Chem. Phys. Lipids 73, 159–180.

Castiglione, G. M., Hauser, F. E., Liao, B. S., Lujan, N. K., Van Nynatten, A., Morrow,

J. M., Schott, R. K., Bhattacharyya, N., Dungan, S. Z. and Chang, B. S. (2017).

Evolution of nonspectral rhodopsin function at high altitudes. Proc Natl Acad Sci USA

114, 7385–7390.

Castiglione, G. M., Schott, R. K., Hauser, F. E. and Chang, B. S. (2018). Convergent

selection pressures drive the evolution of rhodopsin kinetics at high altitudes via

nonparallel mechanisms. Evolution 72, 170–186.

265

Crescitelli, F. (1956). The nature of the gecko visual pigment. J. Gen. Physiol. 40, 217–231.

Das, J., Crouch, R. K., Ma, J.-X., Oprian, D. D. and Kono, M. (2004). Role of the 9-

Methyl Group of Retinal in Cone Visual Pigments †. Biochemistry 43, 5532–5538.

Dong, X., Ramon, E., Herrera-Hernández, M. G. and Garriga, P. (2015). Phospholipid

Bicelles Improve the Conformational Stability of Rhodopsin Mutants Associated with

Retinitis Pigmentosa. Biochemistry 54, 4795–4804.

Drummond, H. (1985). The role of vision in the predatory behaviour of natricine snakes.

Animal Behaviour 33, 206–215.

Dungan, S. Z. and Chang, B. S. (2017). Epistatic interactions influence terrestrial–marine

functional shifts in cetacean rhodopsin. Proc. Biol. Sci. 284, 20162743–9.

Enright, J. M., Toomey, M. B., Sato, S.-Y., Temple, S. E., Allen, J. R., Fujiwara, R.,

Kramlinger, V. M., Nagy, L. D., Johnson, K. M., Xiao, Y., et al. (2015). Cyp27c1

Red-Shifts the Spectral Sensitivity of Photoreceptors by Converting Vitamin A1 into A2.

Curr. Biol. 25, 3048–3057.

Filipek, S., Krzysko, K. A., Fotiadis, D., Liang, Y., Saperstein, D. A., Engel, A. and

Palczewski, K. (2004). A concept for G protein activation by G protein-coupled receptor

dimers: the transducin/rhodopsin interface. Photochem. Photobiol. Sci. 3, 628–638.

Fotiadis, D., Liang, Y., Filipek, S., Saperstein, D. A., Engel, A. and Palczewski, K.

(2003). Atomic-force microscopy: Rhodopsin dimers in native disc membranes. Nature

421, 127–128.

266

Gregg, R. G., McCall, M. A. and Massey, S. C. (2013). Chapter 15 - Function and

Anatomy of the Mammalian Retina. Fifth Edition. Elsevier Inc.

Hauser, F. E., Ilves, K. L., Schott, R. K., Chang, B. S. and Castiglione, G. M. (2017).

Accelerated evolution and functional divergence of the dim light visual pigment

accompanies cichlid colonization of Central America. Molecular Biology ….

Imai, H., Kuwayama, S., Onishi, A., Morizumi, T., Chisaka, O. and Shichida, Y. (2005).

Molecular properties of rod and cone visual pigments from purified chicken cone

pigments to mouse rhodopsin in situ. Photochem. Photobiol. Sci. 4, 667–8.

Jacobs, G. H. and Deegan, J. F. (1992). Cone photopigments in nocturnal and diurnal

procyonids. J Comp Physiol A 171, 351–358.

Jastrzebska, B., Maeda, T., Zhu, L., Fotiadis, D., Filipek, S., Engel, A., Stenkamp, R. E.

and Palczewski, K. (2004). Functional Characterization of Rhodopsin Monomers and

Dimers in Detergents. J Biol Chem 279, 54663–54675.

Karnik, S. S. and Khorana, H. G. (1990). Assembly of functional rhodopsin requires a

disulfide bond between cysteine residues 110 and 187. J Biol Chem 265, 17520–17524.

Karnik, S. S., Sakmar, T. P., Chen, H. B. and Khorana, H. G. (1988). Cysteine residues

110 and 187 are essential for the formation of correct structure in bovine rhodopsin.

Proc. Natl. Acad. Sci. U.S.A. 85, 8459–8463.

267

Kawamura, S. and Yokoyama, S. (1998). Functional characterization of visual and

nonvisual pigments of American chameleon (Anolis carolinensis). Vision Research 38,

37–44.

Kefalov, V., Fu, Y., Marsh-Armstrong, N. and Yau, K.-W. (2003). Role of visual pigment

properties in rod and cone phototransduction. Nature 425, 526–531.

Lamb, T. D. (2013). Progress in Retinal and Eye Research. Prog Retin Eye Res 36, 52–119.

Liang, Y. (2003). Organization of the G Protein-coupled Receptors Rhodopsin and Opsin in

Native Membranes. J Biol Chem 278, 21655–21662.

Ma, J., Znoiko, S., Othersen, K. L., Ryan, J. C., Das, J., Isayama, T., Kono, M., Oprian,

D. D., Corson, D. W., Cornwall, M. C., et al. (2001). A visual pigment expressed in

both rod and cone photoreceptors. Neuron 32, 451–461.

Mattle, D., Kuhn, B., Aebi, J., Bedoucha, M., Kekilli, D., Grozinger, N., Alker, A.,

Rudolph, M. G., Schmid, G., Schertler, G. F. X., et al. (2018). Ligand channel in

pharmacologically stabilized rhodopsin. Proc. Natl. Acad. Sci. U.S.A. 56, 201718084–6.

Mendes, H. F. and Cheetham, M. E. (2008). Pharmacological manipulation of gain-of-

function and dominant-negative mechanisms in rhodopsin retinitis pigmentosa. Hum.

Mol. Genet. 17, 3043–3054.

Morrow, J. M. and Chang, B. S. (2015). Comparative Mutagenesis Studies of Retinal

Release in Light-Activated Zebrafish Rhodopsin Using Fluorescence Spectroscopy.

Biochemistry 54, 4507–4518.

268

Morrow, J. M., Lazic, S. and Chang, B. S. (2011). A novel rhodopsin-like gene expressed

in zebrafish retina. Vis. Neurosci. 28, 325–335.