ASSESSMENT OF ROD, CONE, AND INTRINSICALLY PHOTOSENS ITIVE RETINAL GANGLION CELL (iprgc) CONTRIBUTION TO THE CANINE CHROMATIC PUPILLARY RESPONSE

Transcription

1 ASSESSMENT OF ROD, CONE, AND INTRINSICALLY PHOTOSENS ITIVE RETINAL GANGLION CELL (iprgc) CONTRIBUTION TO THE CANINE CHROMATIC PUPILLARY RESPONSE By Connie Yun Yeh A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Comparative Medicine and Integrative Biology Master of Science 2016

2 ABSTRACT ASSESSMENT OF ROD, CONE, AND INTRINSICALLY PHOTOSENSITIVE RETINAL GANGLION CELLS (IPRGC) CONTRIBUTION TO THE CANINE CHROMATIC PUPILLARY RESPONSE By Connie Yun Yeh Functional testing is a crucial part of assessing retinal and optic nerve diseases in human patients and animals both in a clinical and research setting. This not only includes the diagnosis of disease, but also monitoring of disease progression and response to newly developed therapies. The electroretinogram (ERG) is the benchmark method for specific testing of retinal function and is routinely used in both clinical and research settings. Altering the testing condition (background light and intensity of light stimulus) allows recording of rod and cone photoreceptor-specific responses. For complete assessment of the visual system, the evaluation of the central visual pathways within the brain is also important. Chromatic pupillometry was recently developed and measures the rate and amplitude of pupil constriction in response to light stimuli of different color, mostly blue and red. Since pupillometry involves the evaluation of retina, optic nerve, and components of the central visual pathways, it complements ERG and visual behavioral testing for complete functional assessment of visually impaired human and animal patients. Chromatic pupillometry is likely going to be considered an additional standard technique in veterinary and comparative ophthalmology; therefore, validation in dogs is a high priority. The goal of this thesis is to develop chromatic pupillometry for specific assessment of canine retina and optic nerve function.

3 Copyright by CONNIE YUN YEH 2016

4 This thesis is dedicated to my husband, Brad, who has been a constant source of support and encouragement during the challenges of graduate school and residency. No matter where my career has taken me, he has always supported me 100%. Thank you for enduring living in Lansing, MI, home of the Spartans, for 4 years despite being an avid Wolverine fan! This work is also dedicated to my parents, Yung Hua and Pi Yun, who have always loved me unconditionally and whose good examples have taught me to work hard for the things that I aspire to achieve. Thank you for being my biggest supporters. I love all of you very much and I wouldn t be where I am without all your support. iv

5 ACKNOWLEDGMENTS My deepest appreciation goes to my mentor, Dr. Andras Komaromy, for his continuous guidance and support. I can never thank him enough for all the great opportunities he provided for me. Without him, I would not be where I am in my career currently, finishing a Masters degree and an ophthalmology residency! I have learned so much from him over the past several years and am truly grateful for having the chance to work with him. I thank him also for his never ending positive outlook on everything in life! I would also like to thank my committee members, Dr. Simon Petersen-Jones, Dr. Eric Eggenberger, and Dr. Arthur Weber for their support, expertise, and guidance through the program. Thank you to everyone in the Komaromy lab for all your help with this project, especially Christine Harman and Kristin Koehl for their technical and moral support, and Gabriel Stewart for his help compiling the anesthesia parameters into Excel. A special thanks to Dr. Gustavo Aguirre and Dr. William Beltran for the use of the animals at the Retinal Disease Study Facility (RDSF) at the University of Pennsylvania. A large portion of this project could not have been completed without their support. Thank you to Dr. Karina E. Guziewicz for assistance in primer design, Melinda Frame (Michigan State University) for assistance with confocal microscopy, Cheryl Craft (University of Southern California) for the hcar antibody, and the staff at the RDSF for their amazing technical support. v

6 This work was supported by the NIH (EY , , , T32-RR007063), the Foundation Fighting Blindness, and the Michigan State University College of Veterinary Medicine Endowed Research Funds. vi

7 TABLE OF CONTENTS LIST OF TABLES... ix LIST OF FIGURES... x KEY OF ABBREVIATIONS... xi CHAPTER 1 - INTRODUCTION Discovery of Melanopsin and a New Photoreceptor Morphology and Diversity of iprgcs Brain Targets of iprgcs iprgc Intraretinal Signaling iprgc Phototransduction Response of Melanopsin to Light iprgcs and the PLR Using Pupillary Light Reflex to Measure the iprgc Response Inherited Retinal Disease Canine Models Purpose of the Current Study CHAPTER 2 MATERIALS AND METHODS Animals Anesthesia Chromatic Pupillometry Data Analysis Electroretinography Sequencing and Cloning qrt-pcr Analysis of Retinal Gene Expression Immunohistochemistry (IHC) CHAPTER 3 - RESULTS Normal Chromatic Pupillometry Recorded From Wt Dogs Primary Rod Disease Leads to Lost Pupil Responses to Low-intensity Blue Stimulus Primary Cone Disease Leads to Reduced Pupil Responses to Bright Red Stimulus Response to Bright Blue Light is Maintained with Advanced Outer Retinal Disease vii

8 3.5 iprgc Function Loss Variation in Constriction Amplitude: Reproducibility and Light Intensity Series Presence of iprgc and Melanopsin Expression in Disease Retinas CHAPTER 4 - DISCUSSION AND CONCLUSIONS CHAPTER 5 - FUTURE DIRECTIONS REFERENCES viii

9 LIST OF TABLES Table 1. Comparison of the maximal sensitivity for canine and human photopigments 16 Table 2. Canine inherited retinal diseases with identified gene mutation and ERG changes Table 3. Summary of dogs used in the study Table 4. Summary of dogs used for the molecular work Table 5. Antibodies used in the study ix

10 LIST OF FIGURES Figure 1. Hematoxylin and eosin staining of wild type canine retina... 4 Figure 2. Phototranduction in vertebrate (ciliary) vs. invertebrate (rhabdomeric) photoreceptors Figure 3. Comparison of the spectral sensitivity curves for canine and human photoreceptors Figure 4. Chromatic pupillometry recording with the RETIport system with a Ganzfeld dome Figure 5. Machine measured pupil size vs. actual pupil size in 4 dogs Figure 6. Pupillogram showing the PLR parameters of interest Figure 7. Average (A) and individual (B) PLRs from the wt dogs Figure 8. Chromatic pupillometry and ERG results for dogs with primary rod disease 49 Figure 9. Chromatic pupillometry and ERG for dogs with primary cone disease Figure 10. Chromatic pupillometry and ERG results for dogs with advanced outer retinal disease Figure 11. PLRs from the dog affected by a mutation in RD3 and concurrent severe optic nerve head coloboma associated with the NEHJ1 mutation Figure 12. PLRs from the wt and 2 CNGB3-mutant dogs that had the same chromatic pupillometry testing protocol performed within 4-5 months after the first recording Figure 13. Light intensity series performed on wt and CNGB3-mutant dogs Figure 14. Alignment of the amino acid sequence of canine melanopsin with other species (mouse, chicken, lizard and human) Figure 15. qrt-pcr results for canine models of inherited retinal disease Figure 16. Melanopsin immunostaining of representative retinal sections from wt and PDE6B-mutant dogs Figure 17. Immunohistochemistry results of affected dogs compared to wt dog x

11 KEY TO ABBREVIATIONS ACHM Achromatopsia crd2 Cone-rod dystrophy 2 CEA DAG erd ERG IHC IP 3 IPL iprgc IV LCA LGN OPN PLC PIP 2 PIPR PLC PLR PRA PRCD qrt-pcr Collie eye anomaly Diacyglycerol Early retinal degeneration Electroretinogram Immunohistochemistry Inositol triphosphate Inner plexiform layer Intrinsically photosensitive retinal ganglion cell Intravenous Leber congenital amaurosis Lateral geniculate nucleus Olivary pretectal nucleus Phospholipase C Phosphatidylinositol 4,5-bisphosphate Post-illumination pupil response Phospholipase C Pupillary light reflex Progressive retinal atrophy Progressive rod cone degeneration Real-time quantitative reverse transcription-pcr xi

12 rcd1 Rod-cone dysplasia 1 rcd2 Rod-cone dysplasia 2 rcd3 Rod-cone dysplasia 3 RGC SARDS SCN wt Retinal ganglion cell Sudden acquired retinal degeneration syndrome Suprachiasmatic nucleus Wildtype XLPRA2 X-linked progressive retinal atrophy 2 xii

13 CHAPTER 1 - INTRODUCTION There has been compelling evidence emerging over the past decade about a novel class of photoreceptors in the mammalian retina: a subset of retinal ganglion cells (RGCs) that express the photopigment melanopsin. 1-7 These intrinsically photosensitive RGCs (iprgcs) differ in form and function from the classic rod and cone photoreceptors. They are important for a variety of non-image forming light responses due to their ability to measure ambient light levels (irradiance), including regulation of pupil size (pupillary light reflex, PLR), pineal melatonin production, sleep propensity, and the alignment of circadian clocks to the light-dark cycle (circadian photoentrainment). Recent studies suggest that iprgcs could also have image forming functions. Since the discovery of iprgcs in the early 2000 s, there has been a surge of studies on these cells, and it is a rapidly developing field. This study utilizes these new finding to better understand and assess the function of the classic photoreceptors and the iprgcs. 1.1 Discovery of Melanopsin and a New Photoreceptor Vision scientists have long thought of rods and cones as the retina s only photoreceptors but there have been clues that other photoreceptors exist, including the preserved circadian photoentrainment and pupillary light reflex in mice lacking rods and cones. 1-7 In 1927, Keeler identified a strain of mice that lacked rods, and he observed that they maintained a pupillary light response. 1 He speculated that in mammals the iris may function independent of vision perhaps by direct stimulation of the internal nuclear or ganglionic cells by light. 1, 8 Since Keeler s study in the 1920 s, other reports 1

14 have also supported the idea of a specialized, non-visual photoreceptor within the inner retina (nerve fiber layer, ganglion cell layer, and inner plexiform layer Figure 1). 2 Many vision scientists believed that the persistent light response were due to the residual cones in retinal degenerate mice. 9 Furthermore, since light suppression of melatonin production by the pineal gland was still present despite the loss of rod photoreceptors, it was thought that the cone photoreceptors contribute to photic melatonin suppression and circadian photoentrainment. 10 This theory was disclaimed when it was shown that a transgenic mouse model that lacked both photoreceptor types still had normal circadian photoentrainment and photic melatonin suppression. 4 Similar observations noted above were found in human patients with advanced outer retinal (loss of rod and cone photoreceptor Figure 1) disease and furthered the case for a 11, 12 novel inner retinal photoreceptor. Besides the preservation of circadian photoentrainment in mice and humans with retinal degeneration, the spectral sensitivity of photoentrainment in wt Mesocricetus auratus (hamster) was located near 500 nm and similar to the absorption spectrum for rhodopsin; however, there were certain characteristics, such as low light sensitivity, 13, 14 considered unusual for rods and cones and suggesting a different photoreceptor. These findings/clues motivated the discovery of a third retinal photoreceptor. It came in 1998 from studies on amphibian dermal melanophores by Provencio and the finding of the new photopigment melanopsin. 15 Since then, melanopsin has been found in the retina of primates, including human, rodents, rabbit, and cat In situ hybridization 2

15 histochemistry showed that melanopsin is expressed in a subset of RGCs. 16 Provencio et al. theorized that the unique inner retinal localization of melanopsin suggests that it is not involved in image formation but rather may mediate nonvisual photoreceptive tasks, such as the regulation of circadian rhythms and the acute suppression of pineal melatonin. 16 This theory was supported by the observation that melanopsin-expressing RGCs project to the suprachiasmatic nucleus (SCN), which controls circadian rhythm , 24 In 2002, two publications proved the existence of iprgcs. Berson et al. labeled RGCs that were found to innervate the SCN by injecting fluorescent microspheres into the rat hypothalamus. They exposed the isolated retinas to light and recorded the responses from the labeled RGCs. They responded to light by depolarizing and increasing their firing rate while the unlabeled RGCs lacked detectable responses. The labeled RGCs also responded to light when input from the rods and cones was blocked by cobalt chloride. 24 Berson et al. concluded that these RGCs are intrinsically photosensitive. Subsequently, Hattar et al. found that these iprgcs contain melanopsin and that their axons project to the SCN and other brain centers responsible to non-image forming functions. 17 In support of these findings, specific ablation of iprgcs eliminated non-image forming responses. 25 3

16 Figure 1. Hematoxylin and eosin staining of wild type canine retina. Layers of the retina: retinal pigment epithelium (RPE), photoreceptors (PR), outer nuclear layer (ONL), outer plexiform layer (OPL), inner nuclear layer (INL), inner plexiform layer (IPL), ganglion cell layer (GCL), nerve fiber layer (NFL). Calibration bar: 20 µm. 4

17 1.2 Morphology and Diversity of iprgcs Berson and Hattar initially identified iprgcs as a single type of RGC; however, since then multiple types of iprgc have been identified. There is morphological and functional heterogeneity with at least five types of iprgcs, M1-M5, with M1-M3 being the most studied , 18, M1 class cells were the first to be identified with somata of ~13-17 µm in diameter. 24, 29, 31 These M1 cells are mostly located in the RGC layer, with some displaced to the inner nuclear layer. M1 cells have a higher melanopsin immunoreactivity and intrinsic photosensitivity compared to the other iprgcs. 27, 29, 33 With a mean diameter of , 31 µm, the somata of M2 cells are slightly larger than those of M1. These cells express lower levels of melanopsin compared to M1 cells and therefore, are less intrinsically photosensitive. M2 cells project minimally to the SCN and strongly to the olivary pretectal nucleus (OPN). 27 M3 cells are similar to M2 cells in terms of dendritic 31, 32 field size, soma size, and complexity of their dendritic branching. Two additional types of iprgc with unique morphologies, M4 and M5, were revealed in a study by Ecker et al. 30 M4 cells have the largest soma of any iprgc subtypes and large, highly branched dendritic fields of about μm. M5 cells have compact, highly branched dendritic fields of about μm. 1.3 Brain Targets of iprgcs The use of mouse lines with labeled iprgcs allowed the identification of numerous targeted, non-image forming brain areas, such as SCN, OPN, intergeniculate leaflet 5

18 (IGL), ventral division of the lateral geniculate nucleus (vlgn), ventral 17, subparaventricular zone, and ventrolateral preoptic nucleus. Other brain regions such as the medial amygdale, lateral habenula, and periaqueductal gray are also targeted by iprgcs but their function remains unclear. 33 Distinct iprgc subtypes seem to project to specific brain regions. About 80% of iprgcs projecting to the SCN are M1 cells. 27 The OPN receives input from both M1 and M2 cells in similar p roportions. 27 M1 cells can be subdivided into Brn3b-positive M1 cells that mainly project to the OPN and Brn3b-negative M1 cells that project mainly to the SCN. 36 Genetic ablation of Brn3b-positive M1 cells in mice leaves just the Brn3b-negative M1 cells, leading to an impairment in PLR but functional circadian photoentrainment. 36 This suggests that different subtypes of iprgcs may be responsible for specific behaviors. Based on the projections to the non-image forming regions of the brain, it was previously thought that iprgcs were not involved in vision. However, recent studies have shown that iprgc projections may be more extensive than previously thought, and include image forming visual functions, such as the dorsal and ventral aspects of the LGN, the core of the OPN, the posterior pretectal nucleus, and the superior colliculus. 30, 37 Most of the M2, M4, and M5 cells project to the dorsal LGN. 30 The functional significance of iprgcs for image-forming vision still needs to be explored. Recent data indicate that in mice and humans, iprgcs may help with reception of spatial brightness. 38 6

19 1.4 iprgc Intraretinal Signaling Although iprgcs are intrinsically photosensitive, they are also subject to intraretinal synaptic influences and serve as principle channels for delivering rod- and cone-driven 25, information to non-image forming areas of the brain. One piece of evidence is the presence of their extensive dendritic arbors that stratify in the inner plexiform layer (IPL), the synaptic layer in which bipolar and amacrine cells convey rod and cone signals to RGCs. 17, 18, 24 Belenky et al. were able to identified synaptic contacts from both amacrine and ON bipolar cells on melanopsin-positive processes in the IPL of mouse retina. 39 Another piece of evidence that iprgcs are also subject to intraretinal synaptic influences is that melanopsin knock-out mice lack the intrinsic light response however, they still were able to photoentrain and retained PLR On the other hand, in mice lacking melanopsin, rods, and cones, all non-visual responses to light were 47, 48 abolished. iprgc subtypes receive different inputs from the outer retina based on the location of 32, 43, 49, 50 their dendrites in either the ON or OFF layers of the IPL. It still remains to be determined how the iprgcs integrate their intraretinal inputs to generate the signals that mediate non-image forming vision. Additional work is needed to establish the exact pattern of innervation of each iprgc type which may provide insight to the physiological functions performed by each type of iprgcs. 7

20 1.5 iprgc Phototransduction Two melanopsin gene families exist: Opn4x, originally discovered in Xenopus laevis by Provencio, is expressed by all vertebrates except mammals, and Opn4m is found in all vertebrates including mammals. 51 The sequence of both melanopsin families is more similar to invertebrate rhabdomeric photoreceptor opsins vs. vertebrate ciliary photoreceptor (rods and cones) opsins. 15 This suggests that melanopsin and invertebrate phototransduction may be similar. Although the precise cellular mechanisms of phototransduction in iprgcs have not been fully characterized, the current model of melanopsin phototransduction (Figure 2) is that melanopsin activation leads to activation of a G q/11 -type G-protein. There is then activation of phospholipase C beta 4 (PLC 4). Usually, PLC activation leads to the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP 2 ) which forms inositol triphosphate (IP 3 ) and diacyglycerol (DAG). Despite PLC s involvement in melanopsin phototransduction, the downstream effectors of PLC activity are unclear. Further studies need to be completed to determine if DAG and IP 3 are the exact second messengers. Then there is opening of TRP6/7-type channels, voltage gated calcium channel activation which leads to action potential generation. The chromophore for melanopsin is 11-cis retinal, similar to rod and cones, and after light stimulation, it is isomerized to all-trans retinal. The mechanism of regeneration of 11-cis retinal in iprgc has been a subject of much debate. It is thought that melanopsin may be similar to invertebrate photopigments in that it has two thermostable 8

21 states (bistable photopigment). This means that they have an ability to regenerate 11- cis-retinal without the need for a complex cycle of enzymatic reactions. Figure 2. Phototranduction in vertebrate (ciliary) vs. invertebrate (rhabdomeric) photoreceptors. In vertebrate photoreceptors, the activation of the opsin triggers G- protein transducin (G t ) leading to the activation of phosphodiesterase (PDE) that hydrolyzes cgmp. The decrease in cgmp closes the cyclic-nucleotide-gated (CNG) channels, decreasing the cation current and hyperpolarization of the photoreceptor cell. In rhabdomeric photoreceptors, the activation of the opsin triggers G q -protein leading to the activation of phospholipase C (PLC) that hydrolyzes phosphatidylinositol 4,5- bisphosphate (PIP 2 ) which forms inositol triphosphate (IP 3 ) and diacyglycerol (DAG). There is activation of TRP channels, an increase in ionic current and depolarization of the photoreceptor cell. 1.6 Response of Melanopsin to Light iprgcs response to light set them apart from rod and cone photoreceptors. The most notable difference is that iprgc depolarize upon light activation and photocurrents exhibit a TRP channel-like pharmacology, as mentioned in the previous section (Figure 24, 52 1). In addition, iprgcs are less sensitive to light and have slow response kinetics. The low sensitivity is thought to be due to the low membrane density of melanopsin in iprgcs, which results in a lower probability of photon absorption. 52 This is related to the fact that iprgcs lack structural specialization, such as the outer segments of rods 9

22 and cones, to maximize the probability of photon capture. The membrane density of melanopsin has been estimated to be about 8 thousand times lower than that of photopigments in the outer segments of rods and cones (~3 µm -2 compared to ~25,000 µm -2 ). 52 Although iprgcs have low sensitivity, each captured photon generates a large and prolonged response and the slow response kinetics may provide for long temporal integration and reporting of ambient light in the environment. 52 With increasing light intensities, iprgcs increase their firing rate which then reduces to a steady plateau. When stimulated by a prolonged light flash, the iprgc voltage response rises to a 53, 54 transient peak and then gradually decays to a lower steady state level. This finding is thought to be due to light adaptation. There is a shortening in the time to peak of responses to brighter flashes and a reduction of the response amplitude to repeated flashes. 53 The overall loss in sensitivity due to light adaptation is far less than the loss in sensitivity found in the rods and cones. 55 Dacey et al. evaluated the contributions to the primate iprgc response by obtaining response vs. intensity date under different light conditions and found that there is distinct rod, cone and melanopsin-initiated responses. 21 Under dark adapted conditions, rods primarily detected light and this light response depolarized iprgcs, triggering an action potential. As the light increases in intensity, but still within the working range of rods, there is an increase in iprgc firing. In light adapted conditions, the rods are bleached and the cones and melanopsin initiate light responses. Rods as well as the red (long wavelength sensitive or L) and green (medium wavelength 10

23 sensitive or M) cones trigger an ON response and the blue (short wavelength sensitive or S) cones trigger an OFF response iprgcs and the PLR Various non-image forming photic responses including circadian photoentrainment, PLR, and pineal melatonin suppression are mediated by the iprgcs. 4, 7, 10, 45, 48 Several investigations have sought to determine the contribution of iprgcs to the PLR. As previously discussed, the iprgcs project to the OPN of the dorsal midbrain, forming the afferent limb of the PLR. 27 Functional studies in mice have provided evidence for iprgc modulation of the PLR. In rodless and coneless (rd/rd cl ) mice, PLR is retained with rapid pupillary constriction following bright white light stimulation. 7 Although PLR was present, the latency to maximal constriction was increased and the irradiance needed to produce the constriction was higher than in mutant vs. wt mice. In mice with the melanopsin gene ablated PLR was similar to that of wt mice at low irradiances, but at high irradiances it was incomplete. 44 Mice lacking melanopsin as well as rods and 47, 48 cones, lack pupillary responses. The results from these studies show that rods, cones, and iprgcs all contribute to the afferent pupillomotor signal, and the contribution of each photoreceptor is dependent on the light condition. Studies in humans and non-human primates have also support a role for iprgcs in the PLR. 56 Following pharmacological blockade of rod and cone photoreceptor signals, PLR persisted and the spectral sensitivity of the response was close to that of melanopsin (~480 nm). In macaque eyes stimulated with equiluminant red and blue 11

24 light, fast onset of pupil constriction was noted for both stimuli but the maximum constriction amplitude was greater with the blue light under constant illumination. Also, a sustained, post-stimulus pupil constriction after high intensity light offset was noted and thought to be melanopsin driven. McDougal et al. evaluated the individual contributions of rods, cones, and iprgcs to the human PLR. 55 They found that there is a winner takes all effect on the PLR. If a stimulus is below the threshold for activating iprgcs, then the rod and cones determine the spectral sensitivity of the PLR: The spectral responses of the PLR are consistent with rods with little contribution from cones, since they rapidly adapt with steady-state light stimuli. In contrast, if the stimulus is above the threshold for iprgc activation, the spectral sensitivity is determined by melanopsin, and the iprgcs are solely responsible for driving the PLR. Furthermore, Gooley et al. showed that rods and cones are capable of driving the initial rapid pupil constriction following light stimulus onset and during exposure to continuous low-irradiance light. 57 Similar to what was found by McDougal, the contribution of cones to pupillary constriction decreases over time during exposure to continuous light. In blind individuals without rod and cone function, PLR is preserved at high irradiance 480 nm light but weak or absent at lower irradiance light. These findings suggest that iprgcs dominate PLR at high irradiances while rods and cones mediate PLR at continuous low irradiances. In addition, in the absence of rod and cone function, the pupillary response was slow and sustained and was unable to tract intermittent light 12

25 stimuli suggesting that the classic photoreceptors are required for fast modulations in light intensity. 1.8 Using Pupillary Light Reflex to Measure the iprgc Response Due to the recent advances in understanding the iprgcs and the previously mentioned studies on the contribution of rods, cones, and iprgcs to the PLR, there have been renewed interests in PLR and its value as a noninvasive test for neuroretinal function. PLR assessment is routinely performed as part of the ophthalmic examination of human patients and animals. While this testing is done qualitatively in a clinical setting by describing the rate and extent of pupil constriction, pupillometry uses digital imaging for precise quantitative measurement of the pupil (Chapter 2). Pupillometry with white light has been applied in as assessment of nociception during general anesthesia (for review, see Larson et al., 2015). 58 Pupillometry has been applied in intensive care medicine and in patients that have survived cardiopulmonary resuscitation, the return of PLR is a valuable prognostic indicator for return of neurologic function White light pupillometry has also been used to assess retinal function and outcome of retinal gene therapy Leber congenital amaurosis (LCA) due to a four base-pair deletion in the RPE65 gene has been described in the Briard dog and served as a large animal model for human RPE65-LCA. 68 Gene therapy using adeno-associated virus (AAV) vectors resulted in improved retinal function in affected Briard dogs as assessed by ERG and obstacle course vision testing. 66 In addition, there was improved transmission of signals from the retina to higher visual pathways, as the results from white light pupillometry 13

26 showed that the treated dogs had improved pupil constriction. 66 Chromatic pupillometry uses different light stimulus parameters based on wellestablished knowledge about the spectral sensitivity difference of photopigment in each photoreceptor, to activate each receptor system separately. For example, humans are trichromats having red (L), green (M), and blue (S) cone pigments with a maximal 69, 70 sensitivity of 560 nm, 530 nm, and 430 nm, respectively. The maximal sensitivity for human rhodopsin is ~495 nm The spectral sensitivity of human melanopsin is 16, 17, 21, 40, 47, nm, similar to other species. Chromatic pupillometry has been explored and optimized in a number of human studies Kardon et al. developed a protocol to assess the contributions of the rods, cones, and iprgcs to the PLR in both normal subjects and patients with neuroretinal visual loss. 75 They tested red (640 nm) and blue light (467 nm) at three different light intensities (1, 10, 100 cd/m 2 ). It was thought that the low intensity red and blue light stimuli were most likely driven by the cones and rods, respectively. The medium intensity red and blue light stimuli were most likely similar to the low intensity light stimuli but had a greater cone contribution. High intensity red and blue light stimuli were most likely driven by cones and a combination of cones and iprgcs, respectively. A patient with retinitis pigmentosa showed a reduced pupil response to low intensity blue light. There was a reduced pupil response to red light stimulation in a patient with achromatopsia. A patient with ganglion cell dysfunction due to anterior ischemic optic neuropathy had loss of pupil response to both red and blue light. A second study by 14

27 Kardon et al. applied the protocol to a group of patients with retinitis pigmentosa. 76 They found that the median rod-, cone-, and melanopsin- weighted pupil responses were significantly reduced compared to normal subjects. They compared the pupil response to the patient s degree of ERG abnormality and found that patients with nonrecordable ERG tended to have the greatest loss of pupil response. Although these studies showed that chromatic pupillometry was a useful noninvasive diagnostic tool for monitoring outer and inner retinal function, the protocol did not seem to selectively activate rod, cone, and iprgc responses i ndependently. For example, retinitis pigmentosa patients with abnormal but recordable scotopic ERGs did not have a rodweighted pupil response that was significantly different than the normal subjects, indicating that the 1 cd/m 2 blue light stimulus may not be selective enough to measure pure rod activity. Based on Kardon s proof of concept studies, Park et al. aimed to develop a chromatic pupillometry protocol that would optimally assess the rods, cones, and iprgcs Their testing on normal subjects and patients with retinitis pigmentosa and LCA, found that low intensity (0.001 and 0.01 cd/m 2 ) blue light stimulus (467 nm) in the dark, high intensity (~ 400 cd/m 2 ) red light stimulus (640 nm) on a blue background, and high intensity (~ 400 cd/m 2 ) blue light stimulus (467 nm) was able to preferentially test rods, cones, and iprgcs, respectively. Only the high intensity blue light stimulus in the dark was able to produce a sustained pupil response and this sustained response is most likely melanopsin driven. 15

28 Chromatic pupillometry has also been introduced over recent years for the assessment of the canine visual pathway The absorption spectra of canine photoreceptors are similar to humans (except for human M-cones) (Table 1 and Figure 3). Dogs are function dichromats, having combined red-green (L/M) and blue (S) cone pigments with 81, 82 a maximal sensitivity of 555 nm and nm, respectively. Canine rhodopsin has peak sensitivity of nm. 82 The spectral sensitivity of canine melanopsin has not been validated but it is assumed to be 480 nm, similar to other species, including 16, 17, 21, 40, 47, 56 humans. Table 1. Comparison of the maximal sensitivity for canine and human photopigments Canine Human Rhodopsin nm 495 nm L - opsin M - opsin 560 nm 530 nm L/M - opsin 555 nm S - opsin nm 430 nm Melanopsin 480 nm 480 nm 16

29 Figure 3. Comparison of the spectral sensitivity curves for canine and human photoreceptors. Courtesy of Dr. Randy Kardon, University of Iowa. Grozdanic et al. used a handheld light source (Melan-100; BioMed Vision Technologies, Ames, IA) to performed chromatic PLR analysis in healthy dogs and dogs with sudden acquired retinal degeneration syndrome (SARDS) using light stimulus of different intensity and wavelength. 78 Dogs affected with SARDS present with sudden onset of blindness and have a complete loss of photoreceptor activity based on the flatline ERG 83, 84 recording. The etiology of SARDS is still unknown but it is thought to be immune mediated. 85 The Melan-100 has a diode-based light source with blue light (480 nm) and red light (630 nm) and a function to change the light stimulus intensity. Grozdanic et al. tested both red and blue light intensities of 3.47, 4.3, 5.3, 5.7, and 6 log cd/m 2. In 17

30 normal dogs, low intensity red and blue light stimuli can elicit a PLR but in dogs with SARDS, only the high intensity blue light stimulus (5.3 log cd/m 2 ) was able to elicit a PLR. Based on the spectral sensitivity of the photopigment in canine rods and cones, it is likely that their red light protocol primarily activated the L/M cones and was unable to full test the rods. The blue light most likely activated the iprgcs. The finding suggests a loss of outer retinal mediated PLR and the preservation of the inner retinal (iprgc) mediated PLR. As a follow-up to their initial study, Grozdanic tried to determine the reliability of the chromatic PLR results from the Melan-100 to diagnose retina and optic nerve diseases in canine cataract patients. 79 From the results of the initial study performed in 2007, blue (480 nm) and red (630 nm) light stimuli at an intensity of 5.3 log cd/m 2 (200 kcd/m 2 ) were used for testing the PLR. There was a high sensitivity and specificity in detecting retinal degeneration and retinal detachment in canine cataract patients by chromatic PLR. Furthermore, chromatic PLR can be used to assess optic nerve disease, as optic neuritis patients were found to have absent chromatic PLR. Whiting et al. developed a custom-made recording system and a chemical restraint protocol that was able to reliably quantify the canine PLR. 80 PLR was recorded in normal dogs for red (627 nm) light stimuli at 13.5 and 15.0 log photons/cm 2 /s and blue (470 nm) light stimuli at 13.0 and 14.5 log photons/cm 2 /s. There were robust PLRs with both red and blue light stimuli at the two intensities tested. At the lower intensities of light (13.0 and 13.5 log photons/cm 2 /s), the constriction amplitude for the blue and red 18

31 light was similar. At the higher intensities of light (14.5 and 15.0 log photons/cm 2 /s), the pupil redilated quickly after offset of the red light stimulus but there was a sustained pupil constriction after offset of the blue stimulus. The sustained response to the bright blue light stimulus suggested that iprgc input to the PLR is conserved in the dog. Whiting et al. also evaluated injectable vs. combination of injectable and inhalant (isoflurane) chemical restraint for PLR recording. The combination of injectable and inhalant chemical restraint allowed for more consistent baseline pupil size measurements with minimal noise and the use of an eyelid speculum and conjunctival stay suture for eye positioning without too much stress on the dog. Although the baseline pupil size was significantly reduced in dogs that received a combination of injectable and inhalant chemical vs. injectable chemical restraint alone, the PLR was still reliably measured (as small as 5% constriction). Large PLR constriction amplitudes were observed even with the use of isoflurane. Based on the results of both human and canine studies, chromatic pupillometry appears to be a reliable diagnostic tool for retina and optic nerve disease, and these studies provide a basis for the work described in this thesis. 1.9 Inherited Retinal Disease Canine Models Similar to humans, dogs are affected by various forms of inherited retinal diseases. 86 There are many similarities between dogs and human eyes, including size, morphology, and density of photoreceptor distribution; these allow dogs to be good models for equivalent human diseases in order to study pathogenesis, and novel diagnostic 19

32 techniques and therapies. The availability of an unparalleled group of well-characterized canine retinal disease models at both Michigan State University (MSU) and the University of Pennsylvania (UPENN) provided us with a unique opportunity to test the hypothesis that rod-, cone-, and iprgc-mediated pupil responses in canines can be specifically assessed by use of blue and red light stimuli. The dogs available at these two universities included dogs with CNGB1 progressive retinal atrophy (PRA), CNGB3 achromatopsia (ACHM), PDE6B rod-cone dysplasia 1 (rcd1), RD3 rod-cone dysplasia 2 (rcd2), PDE6A rod-cone dysplasia 3 (rcd3), RPE65 LCA, PRCD progressive rod cone degeneration (prcd), IQCB1 cone rod dystrophy 2 (crd2), RPGR X-linked progressive retinal atrophy 2 (XLPRA2), and STK38L early retinal degeneration (erd) (Table 2). 20

33 Table 2. Canine inherited retinal diseases with identified gene mutation and ERG changes 21

34 Table 2 (cont d). 22

35 1.10 Purpose of the Current Study Although chromatic pupillometry has been introduced over the recent years for the assessment of the canine visual pathway, the contribution of different cell populations to the response to specific light stimuli has not been determined in detail The chromatic pupillometry methods and results from a human study form the basis on which we have formulated the work described in this thesis. 77 The evaluation of the previously established human protocol was justified by the fact that the absorption spectra of canine photoreceptors are similar to humans (except for human M-cones) (Table 1 and Figure 3). As mentioned previously, the availability of an unparalleled group of well-characterized canine retinal disease models provided us with an unique opportunity to test the hypothesis that rod-, cone-, and iprgc-mediated pupil responses in canines can be specifically assessed by use of blue and red light stimuli. The study resulted in an additional effective functional assessment tool for the canine retina, and helped to verify previously made predictions about the cell-specificity of particular pupillometry protocols 76, 77 used in human patients. Our hypothesis was that pupil responses elicited by light stimuli of different color and intensity allow functional assessment of canine rod and cone photoreceptors as well as iprgcs. The specific aims were: 23

36 Specific Aim 1: Develop a chromatic pupillometry protocol for reproducible assessment of cone-mediated retinal function. Rationale: By testing dogs with isolated cone function loss (achromatopsia caused by mutations in the CNGB3 gene; CNGB3-ACHM 92 ) and comparing the results to normal controls, we have a unique tool to verify that our stimulation protocol is specific for canine cones. Currently, the most likely testing protocol for cones is the use of a bright red stimulus (640 nm, 400 cd/m 2 ) on a blue background (480 nm, 25 cd/m 2 76, 77 ). Specific Aim 2: Develop a chromatic pupillometry protocol for reproducible assessment of rod-mediated retinal function. Rationale: Dogs with progressive retinal atrophy (PRA) due to a mutation in the rod specific gene CNGB1 (CNGB1-PRA) 87 and young animals (<6 months of age) affected by rod-cone dysplasia (rcd type 1 PDE6B gene mutation 99 ; type 2 RD3 gene 103 ; and type 3 PDE6A gene 106 ) completely lack rod function while cone function is still intact. By comparing the pupil responses of these dogs to age-matched normal controls, we can define responses that are rod-specific. The most likely testing protocol consists of a dim blue stimulus (470 nm, 1 cd/m 2 76, 77 ) under dark adaptation. Note that in older (>1 year) rcd-affected dogs, both rod and cone function is lost, providing an additional tool to support Specific Aims 1 and 2. Because optic nerve function is not affected in these dogs, they should still have normal iprgc-mediated responses, likely elicited by use of a bright blue stimulus (470 nm, 400 cd/m 2 ) under dark adaptation (see Specific Aim 3). 24

37 Specific Aim 3: Develop a chromatic pupillometry protocol for reproducible assessment of iprgc-mediated retinal function. Rationale: A dog with optic nerve disease (severe optic nerve head coloboma) was used to verify that pupil constriction elicited by bright blue light originates from RGCs. We can also draw some conclusions about testing conditions specific for iprgcs based on the results from other dogs: All the dogs tested under Specific Aims 1 and 2 are expected to have comparable pupil responses following the presumed iprgc specific light stimulus. Should this not be the case, then the testing protocol will have to be modified accordingly. The most likely testing protocol consists of a bright blue stimulus (470 nm, 400 cd/m 2 76, 77 ) under dark adaptation. Specific Aim 4: Establish normal reference values for chromatic pupillometry in dogs. Rationale: The testing of normal dogs will not only provide the age-matched (<6 months and >1 year) controls for the mutant dogs tested under Specific Aims 1 and 2, but will also yield normal reference values for future studies. Specific Aim 5: Apply chromatic pupillometry for testing of dogs with inherited retinal diseases and variable severity of both rod and cone degeneration. Rationale: Once rod-, cone-, and iprgc-specific testing protocols have been established, we will apply them to a number of other well defined canine retinal disease models with known genetic defects and decreased rod- and cone-function, such as RPE65-LCA 68, 121, early retinal degeneration (erd; STK38L gene mutation) 122, cone-rod 25

38 dystrophy 2 (crd2; IQCB1 gene mutation) 118, and X-linked PRA type 2 (XLPRA2; RPGR gene mutation) 120. The protocol established could help evaluate melanopsin vs. rod and cone contribution to the PLR and could be used clinically to estimate the degree of RGC vs. the photoreceptor loss. This is particularly important to assess novel therapies such as gene therapy and retinal prosthesis. 26

39 CHAPTER 2 MATERIALS AND METHODS 2.1 Animals For the chromatic pupillometry testing, 37 dogs with different stages of primary loss of rod-, cone-, combined rod/cone-, and optic nerve function were tested and compared to 5 wt dogs (Table 3). This included dogs with CNGB1 progressive retinal atrophy (PRA), CNGB3 achromatopsia (ACHM), PDE6B rod cone dysplasia 1 (rcd1), RD3 rod cone dysplasia 2 (rcd2), PDE6A rod cone dysplasia 3 (rcd3), RPE65 Leber congenital amaurosis (LCA), PRCD progressive rod cone degeneration (prcd), IQCB1 cone rod dystrophy 2 (crd2), RPGR X-linked progressive retinal atrophy 2 (XLPRA2), STK38L early retinal degeneration (erd), and NEHJ1 Collie eye anomaly (CEA). 68, 87, 88, 92, 99, 103, 104, 106, , The dogs were grouped and tested based on the previously reported retinal and optic nerve disease phenotype. Dogs with primary loss of rod function included dogs affected by CNGB1-PRA, and young dogs (<1 year of age) affected by PDE6A-rcd3 and PDE6B-rcd1. 101, 102 Dogs with primary loss of cone function included dogs affected by CNGB3-ACHM. Dogs with primary loss of optic nerve function included a dog with severe optic nerve head coloboma due to NEHJ1- CEA. This dog was also affected by RD3-rcd2. Once rod-, cone-, and iprgc-specific testing protocols were established based on testing of the group of dogs, we tested dogs with variable severity of rod and cone degeneration. These included older dogs affected with PDE6B-rcd1, RD3-rcd2, and PDE6A-rcd3, and dogs affected with RPE65- LCA, PRCD-prcd, IQCB1-crd2, STK38L-erd, and RPGR-XLPRA2. A light intensity series was also performed on 2 additional wt dogs and 2 CNGB3 affected dogs. 27

40 A complete ophthalmic examination, including vision assessment via testing of the menace response, slit lamp biomicroscopy (SL15, Kowa Optimed, Inc., Torrance, CA), indirect ophthalmoscopy (All Pupil II, Keeler Instruments Inc., Broomall, PA), and retinal photography (RetCamII, Clarity Medical Systems, Pleasanton, CA), was performed on all dogs included in this part of the study. Twenty three additional dogs, both wt dogs and dogs with different stages of retinal diseases, were used for the molecular work of this study. Seventeen dogs were used for sequencing/cloning and real-time quantitative reverse transcription-pcr (qrt-pcr) and 6 dogs were used for immunohistochemistry (IHC) (Table 4). Two of the dogs used for IHC also had chromatic pupillometry performed. The eyes were enucleated immediately following euthanasia with an overdose of sodium pentobarbital. All studies were performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by Michigan State University and the University of Pennsylvania Institutional Animal Care and Use Committees. 28

41 Table 3. Summary of dogs used in the study Disease Mutated Gene Functional Change Dog Gender Age (years) Normal CNGB1-PRA CNGB1 No/reduced rod function CNGB3-ACHM CNGB3 No/reduced cone function PDE6B-rcd1 (young) PDE6B No/reduced rod function PDE6B-rcd1 (old) RD3-rcd2 (old) PDE6B RD3 No/reduced rod and cone function No/reduced rod and cone function N1 F 1 N2 F 1 N3 F 0.6 N4 M 1 N5 ǁ F 2.5 N6* F 0.8 N7* M 0.8 1CNGB1 M 0.3 2CNGB1 M 5.2 1CNGB3* M 4.8 2CNGB3 F 4.6 3CNGB3 ǁ M 1.6 4CNGB3 F 1.3 5CNGB3 ǁ M 1.4 6CNGB3* M 0.6 1PDE6B M 0.7 2PDE6B F 0.7 3PDE6B F 0.4 4PDE6B F 6 5PDE6B F 5.3 1RD3 M 1.3 3RD3 F 3.7 4RD3 F 3.4 PDE6A-rcd3 (young) PDE6A No/reduced rod function 1PDE6A F 0.7 PDE6A-rcd3 (old) PDE6A No/reduced rod and cone function 2PDE6A M 3.8 1RPE65 F 4.4 2RPE65 F 4.4 RPE65-LCA RPE65 No/reduced rod and cone function 3RPE65 F 2.1 4RPE65 F 0.8 5RPE65 F 0.8 6RPE65 F

42 Table 3 (cont d). Disease Mutated Gene Functional Change Dog Gender Age (years) PRCD-prcd IQCB1-crd2 STK38L-erd RPGR-XLPRA2 RD3-rcd2 (old) and NEHJ1-CEA PRCD IQCB1 STK38L RPGR RD3 and NEHJ1 No/reduced rod and cone function No/reduced rod and cone function No/reduced rod and cone function No/reduced rod and cone function No/reduced rod, cone, and iprgc function 1PRCD F 8.3 2PRCD M 5.9 1crd2 F 1.5 2crd2 F 1.5 3crd2 M 3.8 4crd2 M 3.8 1erd M 8.4 2erd M 3.5 3erd M 3.5 1XLPRA2 F 7.2 2XLPRA2 F 6.5 3XLPRA2 F 5.8 2RD3 M 3.3 * Light intensity series performed with multiple increases in stimulus intensity dogs Electroretinogram (ERG) completed during the same session as pupillometry Immunohistochemistry (IHC) available Animal also had severe optic nerve head coloboma ǁ Chromatic pupillometry reproducibility evaluated by repeat testing 4-5 months later 30

43 Table 4. Summary of dogs used for the molecular work Disease Dog Gender Age (years) Studies Normal CNGB3-ACHM PRCD-prcd 1984 F 0.8 Sequence/cloning 1985 F 0.9 Sequence/cloning/qRT-PCR V qrt-pcr M656 M 0.5 qrt-pcr GS170 M 0.8 qrt-pcr Mateo M 0.5 IHC GS86 M 0.6 qrt-pcr M550 M 0.08 qrt-pcr GS53 F 2.2 qrt-pcr M676 M 0.4 qrt-pcr M681 F 0.4 qrt-pcr M614 F 2.5 qrt-pcr GS171 M 0.8 qrt-pcr M501 F 7.3 IHC X168 F 4.8 qrt-pcr X225 M 3.5 qrt-pcr P774 F 3 qrt-pcr P1450* F 8.2 IHC RPGR-XLPRA2 Z234 M 0.06 qrt-pcr PDE6B-rcd F 0.06 qrt-pcr 1888 M 0.3 IHC 2016 F 1.5 IHC STK38L-erd E1044* M 8.4 IHC * Chromatic pupillometry performed 31

44 2.2 Anesthesia In an initial pilot study, 5 dogs underwent chromatic pupillometry recordings using injectable chemical restraint alone with intravenously (IV) administered dexmedetomidine (Dexdomitor, Zoetis, Florham Park, NJ) at a dose of 4 µg/kg. This sedation alone did not provide adequate restraint for reliable PLR recordings due to fluctuations in pupil size and movement of the eyes (data not shown). Therefore, for all pupillometry recording sessions, dogs were placed under isoflurane gas anesthesia and positioned in sternal recumbency on a custom made table with a head rest. The dogs were premedicated with acepromazine at a dose of 0.02 mg/kg IV (AceproJect; Henry Schein, Dublin, Ohio) and induced with propofol (Propoflo 28; Abbott; Abbott Park, Illinois) given IV to effect (starting dose 4 mg/kg). The dogs were then intubated and general anesthesia was maintained with isoflurane (2-3% vaporizer setting; Isothesia ; Henry Schein, Dublin, Ohio) in oxygen. Heart rate, respiratory rate, and body temperature were monitored throughout the procedure. A portable multiparameter veterinary monitor (PM-9000Vet, Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Nanshan, Shenzhen, P.R. China) was used to assess blood pressure, oxygen saturation, and end-tidal CO 2. The level anesthesia was evaluated by monitoring changes in respiration or heart rate. 2.3 Chromatic Pupillometry Both the left and right eyes were tested separately, and the right eye was tested first. A Barraquer eyelid speculum was inserted to ensure that the nictitating membrane and eyelids did not interfere with testing. Stay sutures (4-0 Perma-Hand Silk, Ethicon, 32

45 Inc., Somerville, NJ) were placed in the superior and inferonasal bulbar conjunctiva 2 mm posterior to the limbus. The stay sutures allowed globe manipulation and maintenance of the pupil in the central optical axis for pupillometry recording. The eye was lubricated (Optixcare Eye Lube; CLC Medica; Waterdown, Ontario, Canada) prior to the start of the recording. The untested eye was covered with a black plastic ocular shield (Oculo-Plastik, Inc., Montréal, Canada) which contained a hypermellose 2.5% ophthalmic demulcent (Goniosoft; OCuSOFT, Inc.; Rosenberg, Texas). Once the testing on the right eye was completed, the left eye was tested with the same process outlined above. Recordings were performed with the RETIport system with a Ganzfeld dome (Roland Consult; Germany) (Figure 4). This system allowed timed delivery of an LED light of a particular wavelength and intensity. An infrared-sensitive camera located within the Ganzfeld dome was used for real-time pupil recording. The system also contained an automatic pupil detector that measured the size of the pupil throughout the recording with a sample rate of 30 fps. The RETIport pupillometry system was developed for testing of human subjects, but we found that there was a linear correlation between the actual canine pupil size and the system s measurements; the machine overestimated the pupil size by ~33% (Figure 5). The actual canine pupil size was measured via a ruler that was placed temporal to the eye when the dog s head was in the Ganzfeld dome. Since this pupillometry system was new, we continued to improve it for use in animals in collaboration with the Roland Consult company. 33

46 Figure 4. Chromatic pupillometry recording with the RETIport system with a Ganzfeld dome. The dog s head is placed into the Ganzfeld dome and correctly positioned so that the tested eye can be visualized by the infrared camera. The system has a pupil detector that automatically detects the pupil (green circle) and measures the diameter of the pupil. A light stimulus is delivered and the pupil response is recorded on the computer. 34

47 Machine Pupil Size (mm) Figure 5. Machine measured pupil size vs. actual pupil size in 4 dogs. D1 and D2 are normal dogs that were not included in the main study. Solid black line shows x = y D1 D2 1CNGB3 6CNGB Actual Pupil Size (mm) 35

48 76, 77 The testing parameters were adapted from previously published human protocols. After 10 minutes of dark adaptation (scotopic conditions), the eye was stimulated with a 1 second dim (1 cd/m 2 ) and bright (400 cd/m 2 ) blue light stimulus (470 nm). Following 5 minutes of light adaptation to a blue background (480 nm, 25 cd/m 2 ) the eye was stimulated with a 1 second bright red (640 nm, 400 cd/m 2 ) light. The blue background light remained on throughout the testing (photopic condition). Two runs were performed for each of the light intensities and the results were averaged. While these intensities were photopically matched for the human eye, we found good agreement with canine spectral sensitivity (Figure 3). It is important to note that as we were evaluating the system, in dogs that still had cone function, a small light artifact (movement of the pupil detector) was observed for several milliseconds after the bright red light stimulus onset but the pupil response quickly overcame the artifact. In dogs that had an absence of cone function, the light artifact was seen throughout the time the bright red light stimulus was on. We found that it was important to observe the real time video recording for the bright red light testing with blue background due to the light artifact to determine if there is a true pupil response. Test-retest variability was assessed in 3 dogs (Table 3) by performing the same chromatic pupillometry testing protocol 4-5 months after the first recording. In order to determine if variation in constriction amplitude found between the normal dogs could be due to entrance pupil size, pupil responses were measured over a range 36

49 of light stimulus intensities under both dark and light adaptation. For this light intensity series, the right eye of 2 wt dogs and 2 CNGB3-mutant dogs were tested. After a 10 minute dark adaptation, alternating red and blue light stimuli were presented over an intensity range from 1 to 400 cd/m 2 in ~0.5-log steps. Light stimulus duration of 1 second was used. Two runs were performed for each of the light intensities. The same protocol outlined above was followed for the photopic condition except the addition of a blue background light to suppress rod function. Blue background light intensity of 6 cd/m 2 and 25 cd/m 2 were tested. Light adaptation of 5 minutes with the blue background light was given prior to testing and remained on throughout the testing. 2.4 Data Analysis The data was exported to Microsoft Excel for analysis. Absolute PLRs were recorded and averaged for the two runs performed for each of the light intensities. The baseline pupil size was calculated by obtaining the median pupil size during 1 second prior to each stimulus onset. The normalized pupil size was then calculated by dividing the averaged absolute pupil size by the baseline pupil size. When both the right and left eyes were tested, the normalized pupil size for both eyes was averaged. The maximum constriction amplitude was defined as the minimum pupil size after stimulus-onset. The latency was defined as the time between stimulus onset and the beginning of pupil constriction. 77 A method to determine the onset of the pupil light reflex was devised by Bergamin et al. and it involved filtering, as well as analysis of the first and second derivative. 130 Filtering of the data as well as analysis of the first and second derivative of the pupil movement as described in Bergamin et al. was attempted but was unsuccessful due to the low sample rate of 30 fps. 130 Therefore, latency was 37

50 determined by evaluating the pupil response graph for the time from stimulus onset to the beginning of pupil constriction. PLR parameters that can be calculated are shown in Figure 6. Anesthesia records of the tested dogs were evaluated to determine if PLRs were affected by the level of anesthesia as measured by heart rate, respiratory rate, isoflurane level, end tidal CO 2 level, and blood pressure. Results were presented as mean ± SD. F test and unpaired Student s t-test or Mann- Whitney U test were performed to measure any statistically significant difference in latency and constriction amplitude between wt dogs and affected dogs in the 3 different disease groups for all testing protocols. The relationship between baseline pupil size and constriction amplitude was evaluated by Pearson correlation. Statistical computations were performed in Microsoft Excel, GraphPad ( and VassarStats ( Differences were considered statistically significant if the P-value <

51 Figure 6. Pupillogram showing the PLR parameters of interest. The parameters can be calculated as indicated. 39

52 2.5 Electroretinography Standard scotopic and phototopic full-field electroretinograms (ERG) were recorded in select dogs (Table 3) under isoflurane anesthesia during the same chromatic pupillometry recording session using the Roland Consult system with a Ganzfeld dome, Jet contact lens electrodes (Fabrinal Eye Care; Switzerland), and commercially available platinum subdermal needle electrodes (Grass Technologies; Warwick, RI). In these selected dogs, chromatic pupillometry was only performed in one eye. The ERG was recorded first in the left eye followed by chromatic pupillome try in the right eye. The untested eye was covered with a black plastic ocular shield as described above. The left pupil was dilated with tropicamide 1% ophthalmic solution (Akorn Inc., Lake Forest, IL) prior to electroretinography. Rod and mixed cone-rod mediated responses were recorded after 20 min of dark-adaptation with scotopic single white flash stimuli of three different light luminances (0.01, 3, and 10 cds/m 2 ). Multiple responses (25 for the 0.01 cds/m 2 single white flash and 10 for the 3, and 10 cds/m 2 single white flash) were averaged. Following 10 min of light-adaptation to white background illumination of 30 cd/m 2, 1 Hz single white flash (3 and 10 cds/m 2 ) and 30 Hz white flicker (3 cds/m 2 ) cone-mediated signals were recorded. Multiple responses (10 for the 1 Hz single flash and s sweeps for the 30 Hz flicker) were averaged. 2.6 Sequencing and Cloning Normal canine retina was collected from a wt dog and flash frozen immediately following euthanasia. Total RNA was isolated from the retina using TRIzol reagent (TRIzol; Invitrogen, Carlsbad, CA) after homogenization. First stranded cdna was synthesized from 2 µg of RNA using High Capacity Reverse Transcription Kit (Applied 40

53 Biosystem; Waltham, MA) according to the manufacturer s instructions. The resultant cdna was used in a 50 µl PCR reaction using the FailSafe PCR System (Epicentra; Madison, WI) using custom designed primers Mel_cDNAF (forward 5 - ACCACCCCCAGGATGAAC-3 ) and Mel_cDNAR (reverse 5 - CTGCAGGCTTGTCCCTGT-3 ). Primers were designed using Primer3 ( and the predicted canine sequence (CanFam 2.0) assembly; Each 50 µl reaction contained PCR grade water, FailSafe PCR enzyme mix, 1 µl (100ng/ µl) DNA template, 1 µl of both forward and reverse primers, and one of the 12 different FailSafe PCR 2X PreMixes (A-L). PCR reaction was performed under the following conditions: an initial denaturation step at 94 C for 3 min, then 94 C for 30 s, 58 for 1 min, and 72 C for 2 mins for 35 cycles, followed by a final extension at 72 C for 5min. PCR products were electrophoresed on 1.5% agarose gel containing ethidium bromide and a single DNA band of the correct size was observed for FailSafe PCR 2X PreMix G. The DNA was purified using the QIAquick Gel Extraction Kit (QIAGEN Inc., USA; Valencia, CA) according to manufacturer s protocol. The purified 1,446 bp PCR product was cloned into the pcr 4-TOPO TA vector using the TOPO TA cloning kit (Thermo Fisher Scientific; Waltham, MA) according to the provided protocol. The plasmid was prepared from an overnight culturing using the PureYield Plasmid Miniprep System (Promega; Madison, WI) and 5-10 individual clones were sequenced using the Applied Biosystems Automated 3730 DNA analyzer (Applied Biosystems, Foster City, CA, USA), and aligned to the predicted sequence (CanFam 2.0 assembly 41

54 ( using Sequencher software (Gene Codes Corporation, Ann Arbor,MI, USA) (GeneBank KU341721). The predicted amino acid sequence of the canine melanopsin was aligned with other species using the program Clustal Omega ( 2.7 qrt-pcr Analysis of Retinal Gene Expression Primers and Taqman Probes were designed for canine Opn4 using the Primer Express Software Version 2.0 (Applied Biosystems, Foster City, CA). Primer pair and probe were designed to neighboring exon sequences to ensure that only cdna and not genomic DNA was amplified. Using cdna synthesized as described in the above section, qrt-pcr was performed with the custom designed primer pairs MelDegE67F (forward GCGGCTACAGAGAGAGTGGAA) and MelDegE67R (reverse - ACATGAACTCCGTGCCAGC) and probe (CTTTCGCTGGGTATT) that spans the junction of exons 6 and 7. All the qrt-pcr reactions were performed in 96 well plates using an ABI 7500 real-time PCR machine with the 7500 detection software (v2.0.1, Applied Biosystems, Foster City, CA). Cycling parameters were as follows 50 C for 2 min, then 95 C for 10 min, followed by 40 cycles of 95 C for 15 s and 60 C for 1 min. Relative melanopsin/opn4 gene expression was compared with the GAPDH product and calculated as 1/[2^(Ct Gene Ct GAPDH )]. The ratio of relative gene expressions was calculated between the diseased eye and wt control using the CT method. 131 Ratio that was different from 1 was relevant and indicated either an increase (>1) or decrease (<1) of gene expression. 42

55 2.8 Immunohistochemistry (IHC) Retinal IHC was performed as previously described Following enucleation, the globes were fixed in 4% paraformaldehyde for 3 hours and then in 2% paraformaldehyde for 21 hours, both in 0.1 mol/l PBS at 4 C. Subsequently, the tissue was processed for embedding in optimal cutting temperature medium Cryosections were then cut at 10 µm thickness. Immunohistochemical staining was performed using the antibodies outline in Table 5. A custom made rabbit anti-canine melanopsin N- terminal (NH 2 - MNPPSGPGAQEPGC-amide) and C-terminal (CAKAPLRPRGQAVETPGKV-amide) antibody (21 st Century Biochemicals; Marlboro, MA) was generated and affinity purified. Alexa Fluor -labeled chicken anti-rabbit IgG, goat anti-rabbit IgG, or goat anti-mouse IgG was used as secondary antibodies. DAPI stain was used to detect cell nuclei. Slides were mounted with Gelvatol and were imaged using the Olympus FluoView FV1000 confocal laser scanning microscope (Olympus America, Inc, Center Valley, PA). 43

56 Table 5. Antibodies used in the study Antigen Cone alpha transducin (GNAT2) Human cone arrestin (hcar) NeuN clonea60 Neurofilament 200 (NF200) Host Rabbit polyclonal Rabbit polyclonal Mouse polyclonal Rabbit polyclonal Catalogue No./Source Santa Cruz sc- 390 Cheryl Craft (University of Southern California) Millipore MAB377 Working Dilution Sigma N4142 1:1000 Normal Retina Localization 1:5000 Cone outer segments 1:5000 Cone photoreceptors 1:2000 Ganglion cells Nerve fiber, outer and inner pexiform layer L/M opsin Rabbit polyclonal Millipore AB5405 1:500 Outer segments of L/M - cones Melanopsin 1 Rabbit 21st Century 1:1000 iprgcs Rhodopsin Mouse monoclonal Millipore MAB5316 1:1000 Rod outer segments/axons and pedicles Santa Cruz Biotechnology Inc., Santa Cruz, CA; Millipore Corporation, Temecula, CA; Sigma -Aldrich Co. LLC., St. Louis, MO; 21 st Century Biochemicals, Marlboro, MA 44

57 CHAPTER 3 - RESULTS 3.1 Normal Chromatic Pupillometry Recorded From Wt Dogs The testing parameters developed based on human protocol 77 were performed on 5 wt dogs (Figure 7). After a 10-minute dark adaptation, the mean pupil diameter of the 5 wt dogs was 10.8 ± 2.4 mm. The mean constriction amplitudes induced by the dim blue/rod testing, bright blue/iprgc testing, and bright red/cone testing stimuli in wt dogs were 21.3 ± 10.6%, 50.0 ± 17.5%, and 19.4 ± 7.4% respectively. The mean latency induced by the dim blue/rod testing, bright blue/iprgc testing, and bright red/cone testing stimuli in wt dogs were 0.26 ± 0.05 s, 0.28 ± 0.10 s, and 0.32 ± 0.03 s respectively. The PLRs to the bright blue stimuli were characterized by a sustained response, which is thought to be melanopsin driven based on previous studies. 77 The melanopsin response was sustained for several minutes (mean ± SD: 7.7 ± 4.6 min) after the offset of the bright blue stimulus. A small light artifact (movement of the pupil detector) that lasted for several milliseconds after the light stimuli onset was observed for cone testing conditions. A considerable variation in pupil constriction amplitude was noted between individual dogs (Figure 7B). There was a significant positive correlation between baseline pupil size and constriction amplitude for the iprgc testing protocol (r=0.72, p<0.05) but not for the rod (r=0.46, p=0.19) or cone (r=-0.27, p=0.45) testing protocol. No other obvious causes for this variation could be identified, including age, gender, and anesthesia level. 45

58 Figure 7. Average (A) and individual (B) PLRs from the wt dogs. The gray shaded area and black bar represents the 1 second light stimulus presentation. Open circle indicates the light artifact. 46

59 3.2 Primary Rod Disease Leads to Loss of Pupil Responses to Low-intensity Blue Stimulus Figure 8 shows the normalized PLRs to the 1-second stimuli for each of the dogs affected by mutations in PDE6A, PDE6B, and CNGB1 compared to the normalized PLRs averaged for the five wt dogs. The low-intensity blue light stimuli in the dark that tested the rods did not induce measurable PLRs in any of the dogs affected by the three diseases. A sustained response to the bright blue light stimuli represents the melanopsin driven response and was still present in all affected dogs. The brightintensity red light stimuli with a bright blue light background that tested the cones resulted in measurable PLRs in most of the affected dogs confirmed also by the video observation of pupil constriction, with the exception of the 8 month old PDE6A- and CNGB1- mutant dogs. The chromatic pupillometry results correspond with the previously described functional phenotype of these diseases as shown by the ERG. One of the younger 5 month old PDE6B-mutant dogs had an ERG performed during the same session as the chromatic pupillometry. ERGs from a representative wt dog and the younger PDE6B-mutant dog are illustrated (Figure 8B). In the PDE6B-mutant dog, there was absent rod-mediated ERG response and preservation of reduced cone photoreceptor responses. As mentioned previously, the 8 month old PDE6A-mutant dog and the 2 CNGB1-mutant dogs were expected to still have some remaining cone function but this was not found with chromatic pupillometry testing. The bright-intensity red light stimuli with a blue background light did not result in measureable PLRs (Figure 8A), which was also 47

60 confirmed by observing the real time video recording. However, a small amplitude deflection in the pupil tracing was often seen, which could have resulted from loss of tracking of the true pupil borders or a reflection onto the pupil area causing an artifact. The 8 month old PDE6A-mutant dog also had an ERG performed during the same chromatic pupillometry testing session and the ERG showed nonrecordable rodmediated response but a reduced, yet recordable cone mediated ERG response (Figure 8B). It was difficult to ascertain with certainty whether a small cone mediated pupil response was present or due to an artifact in the pupil tracing. The mean constriction amplitude induced by the bright blue for all the affected dogs and bright red stimuli for the affected dogs excluding the non-responders were 49.1 ± 13.4% and 24.2 ± 23.6%, respectively. Compared to wt dogs, there was no significant difference in constriction amplitude for both the bright blue (p=0.89) and bright red stimuli (p=0.55). The mean latency induced by the bright blue/iprgc testing and bright red/cone testing stimuli for all the affected dogs were 0.32 ± 9.56 s and 0.27 ± 9.32 s, respectively. Compared to wt dogs, there was no significant difference in latency for both the bright blue (p=0.37) and bright red stimuli (p=0.89). A variation in pupil constriction amplitude was also noted between individual PDE6Bmutant dogs similar to what was found for the wt dogs and no obvious causes for this variation could be identified, including age, gender, and anesthesia level. No significant correlation between baseline pupil size and constriction amplitude for the iprgc (p=0.14) or cone (p=0.84) testing protocol was found. 48

61 Figure 8. Chromatic pupillometry and ERG results for dogs with primary rod disease. (A) PLRs from dogs affected by mutations in CNGB1, PDE6A, and PDE6B. Dark, bold blue and red lines represent averaged PLRs from wt dogs. Lighter blue and red lines represent individual PLRs from affected dogs. The represents the non-recordable PLR from the 8 month old PDE6A-mutant dog and the CNGB1-mutant dogs. The gray shaded area and black bar represents the 1 second light stimulus presentation. (B) Representative full-field ERGs recorded from a young PDE6B- and PDE6Amutant dog. Responses were recorded from a wt dog for comparison. In the young PDE6B- and PDE6A-mutant dogs, rod-mediated response was non-recordable while the cone-mediated function was recordable but the amplitude was reduced compared to wt. 49

62 Figure 8 (cont d). 50

63 3.3 Primary Cone Disease Leads to Reduced Pupil Responses to Bright Red Stimulus In the 6 dogs affected by the CNGB3 mutation, response to bright red stimulus was mostly absent; however, there was interference due to the light artifact (Figure 9A). A well-preserved response to both dim and bright blue stimulus was noted. The results of the chromatic pupillometry correspond with the ERG (Figure 9B). There was normal rod function and loss of cone function in the affected dogs compared with a wt dog. Although most chromatic pupillometry results were as expected based on the phenotypic description of the disease, there were, however, 2 CNGB3-mutant dogs that did have pupil responses to bright red light stimulus, but which were very reduced (Figure 9A). A small recordable photopic ERG was also observed in these 2 dogs, indicating incomplete ACHM (Figure 9B). These two dogs underwent behavioral vision testing in the obstacle avoidance course as outlined in a previous study and their performance was the same as other CNGB3-mutant dogs, indicating that their remaining cone function was not enough to support day vision (results not shown). 134 The mean constriction amplitude induced by the dim blue and bright blue stimuli were 31.2 ± 6.2% and 63.4 ± 8.7%. Compared to wt dogs, the constriction amplitude for the dim blue stimuli was significantly greater (p=0.02) and for the bright blue stimuli was not significantly different (p=0.16). The mean latency induced by the dim blue stimuli and bright blue stimuli were 0.29 ± 0.06 s and 0.27 ± 0.07s. Compared to wt dogs, there was no significant difference in latency for both the bright blue (p=0.26) and bright blue stimuli (p=0.78). 51

64 For the 2 CNGB3-mutant dogs that had pupil constriction with the bright red light, the mean constriction amplitude and median latency induced by the bright red stimuli were 9.1 ± 2.4% and 0.26 ± 0.05 s, respectively. The mean latency was similar compared to normal dogs but the mean constriction amplitude was considerably lower. Sample size was too small for quantitative, statistical analysis. A variation in pupil constriction amplitude was also noted between individual CNGB3- mutant dogs similar to what was found for the wt dogs and no obvious causes for this variation could be identified, including age, gender, and anesthesia level. There was a significant positive correlation between baseline pupil size and constriction amplitude for the iprgc testing protocol (r=0.72, p<0.05) but not for the rod (r=-0.34, p=0.34) testing protocol. 52

65 Figure 9. Chromatic pupillometry and ERG for dogs with primary cone disease. (A) PLRs from dogs affected by mutations in CNGB3. (a) Represents the PLRs from the CNGB3-mutant dogs with no response to bright red stimuli with blue background, as expected. (b) Represents the PLRs from the CNGB3-mutant dogs with reduced response to bright red stimuli with blue background. Dark, bold blue and red lines represent averaged PLRs from wt dogs. Lighter blue and red lines represent individual PLRs from affected dogs. The gray shaded area and black bar represents the 1 second light stimulus presentation. (B) Representative full-field ERGs recorded from CNGB3-mutant dogs. (a) Responses recorded from a wt dog for comparison. (b) Normal rod-mediated function and loss of cone-mediated function in the CNGB3-mutant dog. Although a lower amplitude was noted for the rod ERG, it was still within normal reference range. (c) Normal rodmediated function and small cone-mediated function (black arrows) in the CNGB3-mutant dog with a response to bright red stimuli with blue background recorded via chromatic pupillometry. The black star indicates an artifact. 53

66 Figure 9 (cont d). 54

67 3.4 Response to Bright Blue Light is Maintained with Advanced Outer Retinal Disease With retinal disease progression, there was an absent pupillary response to dim blue and bright red stimuli, indicating loss of rod and cone function (Figure 10A). The melanopsin response was maintained in all dogs even when rod- and cone-mediated retinal function was not detectable, providing strong evidence that this stimulus is specific for inner retina. The melanopsin post-illumination sustained pupil response was maintained and there was a significant increase in the mean latency in the retinal disease dogs (0.61 ± 0.26 s) by 0.33 s compared to the wt dog (p=0.0003), consistent with a melanopsin mediated pupil response without rod or cone influence. The mean initial, transient constriction amplitude induced by the bright blue stimuli was 38.3 ± 13.9% and when compared to wt dogs, there was a significant decrease in constriction amplitude by 11.7% (p=0.03), indicating loss of contribution of rods and cones to the initial constriction amplitude. One of the older PRCD dogs had both rod and cone function preserved that was detected by both the chromatic pupillometry and ERG (Figure 10). One of the older STK38L-mutant dogs had absent pupil response to all light stimuli including bright blue. The loss of inner retinal response is due to advanced retinal degeneration involving the inner retina. 107, 116 The function of STK38L is not known to cause changes in the inner retinal response. 55

68 A variation in pupil constriction amplitude was also noted between individual affected dogs similar to that found for the wt dogs and no obvious causes for this variation could be identified, including age, gender, and anesthesia level; however individual variations in disease severity could be possible. No significant correlation between baseline pupil size and constriction amplitude for the iprgc (p=0.14) or cone (p=0.43) testing protocol was found. 56

69 Figure 10. Chromatic pupillometry and ERG results for dogs with advanced outer retinal disease. (A) PLRs from dogs affected by mutation in STK38L, IQCB1, RPGR, RPE65, and PRCD, and older dogs affected by mutations in PDE6B, RD3, and PDE6A. Dark, bold blue and red lines represent averaged PLRs from wt dogs. Lighter blue and red lines represent individual PLRs from affected dogs. The gray shaded area and black bar represents the 1 second light stimulus presentation. The represents the non-recordable bright blue light PLR from the older STK38L-mutant dog. The black arrow represents the preservation of rod and cone function in the older PRCD-mutant dog. (B) Representative full-field ERGs recorded from an older PDE6B-, STK38L-, and PRCD-mutant dog. Responses were recorded from a wt dog for comparison. For the PDE6B- and STK38K- mutant dogs, the rod- and cone- mediated function was non-recordable. For the PRCD-mutant dog, the cone-mediated function was recordable but the amplitude was reduced compared to the wt. 57

70 Figure 10 (cont d). 58

71 Figure 10 (cont d). 59

72 Figure 10 (cont d). 3.5 iprgc Function Loss The dog with severe optic nerve head coloboma associated with the NEHJ1 mutation was also affected with RD3-rcd2 and therefore, all pupillary responses were absent, including to the bright blue stimulus (Fig. 11). 60

73 Figure 11. PLRs from the dog affected by a mutation in RD3 and concurrent severe optic nerve head coloboma associated with the NEHJ1 mutation. Dark, bold blue and red lines represent averaged PLRs from wt dogs. Lighter blue and red lines represent the PLRs from the affected dog. The gray shaded area and black bar represents the 1 second light stimulus presentation. 61

74 3.6 Variation in Constriction Amplitude: Reproducibility and Light Intensity Series By retesting 3 dogs (1 wt and 2 CNGB3-mutants) twice within 4-5 months, we were able to show that recorded PLRs were highly reproducible, supporting the diagnostic value of chromatic pupillometry in dogs (Figure 12). We hypothesized that the variation in constriction amplitude found between the normal dogs could be due to entrance pupil size and therefore, the amount of light entering the eye. To examine the role of light stimulus intensity on pupil response both in scotopic and photopic conditions, a light intensity series was performed on 2 wt dogs (Figure 13). Under scotopic conditions (Figure 13A), saturation of peak amplitudes occurred at high light intensities ( cd/m 2 ) and there was a fast recovery with the red light stimuli. The PLRs to the blue stimuli were characterized by a sustained post-illumination response for intensities 32 cd/m 2 and this sustained response is attributed to melanopsin in the iprgc. A light artifact was noted for 0.1 s after red stimulus onset at high light intensity (400 cd/m 2 ) under scotopic conditions. The photopically matched red and blue stimuli under scoptic conditions are shown in Figure 13C. Under photopic conditions with a blue background light of 25 cd/m 2 (Figure 13B), the PLRs to the red stimuli are fairly similar to the blue stimuli of the same intensity and were thought to be cone mediated. No sustained response was noted with blue stimuli under photopic condition, which suggests that the blue background also suppressed melanopsin activity. A light artifact was noted for about 0.1 s after red and blue stimuli onset at high light intensities (100 and 400 cd/m 2 ) under photopic conditions. The 62

75 photopically matched red and blue stimuli under photopic conditions are shown in Figure 13D. Two CNGB3-mutant dogs were also test under photopic conditions. There was absence of response with all light intensities tested and a light artifact was clearly noted at a high red light intensity of 400 cd/m 2 (Figure 13E). Based on the light intensity series, it is possible that a smaller baseline pupil size would allow less light to enter the eye, contributing to the difference in constriction amplitude between dogs. Although the correlation analysis showed that a larger baseline pupil size appeared to be associated with a greater constriction amplitude for the different testing protocols, especially for the bright blue stimuli, there also appeared to be other factors that contribute to the variation in constriction amplitude. 63

76 Figure 12. PLRs from the wt and 2 CNGB3-mutant dogs that had the same chromatic pupillometry testing protocol performed within 4-5 months after the first recording. In general, there was no obvious difference in the first and second recording for any of the pupillary response parameters. ( First recording; Second recording) 64

77 Figure 13. Light intensity series performed on wt and CNGB3-mutant dogs. Scotopic (A) and photopic (B) light intensity series performed on 2 wt dogs. The average PLRs for the 2 wt dogs to the red (a) and blue (b) light at 5 intensity levels are represented. (C) Photopically matched red and blue light stimuli for scotopic conditions. (D) Photopically matched red and blue light stimuli for photopic conditions. (E) Photopic light intensity series performed on 2 CNGB3-mutant dog. The average PLRs for the affected dog to the red and blue light at 5 intensity levels are represented. The black arrow represents the light artifact noted at a high red light intensity of 400 cd/m 2 with blue background of 25 cd/m 2. For all graphs, the gray shaded area and the black bar represents the 1 second light stimulus presentation. 65

78 Figure 13 (cont d). 66

79 Figure 13 (cont d). 67

80 Figure 13 (cont d). 68

81 3.7 Presence of iprgc and Melanopsin Expression in Disease Retinas In order to design antibodies and qrt-pcr primers for canine melanopsin/opn4, the gene was cloned. The melanopsin full-length cdna contains a 1,446 bp open reading frame encoding a 482 amino acid protein (Figure 14). Alignment of the isolated canine sequences with those from mouse, chicken, lizard and human sequences, show some homology. The highest sequence similarity (89.60%) was with the cat homolog, whereas weaker (46.63%) similarity was evident with the Italian wall lizard homolog. There is about 75.4% sequence similarity to the mouse and rat homolog and about 78.9% to the human homolog. As retinal disease continues to progress, there is severe retinal degeneration leading to some loss of iprgcs and subsequently, a decrease in their melanopsin expression (Figure 15). There was a significant decrease of canine melanopsin expression in advanced stages of PRCD (p<0.05). Melanopsin expression in early onset retinal degeneration, RPGR- and PDE6B-mutant dogs, appeared to be decreased and comparable to PRCD. In contrast, there was no significant change in melanopsin expression in CNGB3-mutant dogs with achromatopsia, a non-degenerative retinal disease (p=0.193). The IHC on selected disease models corresponds with the chromatic pupillometry results in that melanopsin expressing cells with their dendritic arborizations were observed (Figure 16). Representative IHC on a young PDE6B-mutant dog showed a reduction in the rod outer segment lengths and relative preservation of cone 69

82 photoreceptors early in the disease although there is stunting of the outer segments (Figure 17). With progression of the disease, in the old PDE6B-mutant dog cones are lost. Melanopsin immunostaining of representative retinal sections from the young and old PDE6B-mutant dogs confirm the presence of iprgcs (Figure 16), which corresponds to the bright blue/iprgc chromatic pupillometry testing results. Similarly, in addition to the previously reported morphologic abnormalities and gradual loss of 93, 94, cones in CNGB3-mutant dogs, iprgc do not appear to be affected by the disease. 97 In contrast, IHC from the one STK38L-mutant dog showed loss of the normal retinal 107, 116, 131 architecture. RGCs are completely lost due to severe retinal degeneration (Figure 17). This corresponds to the negative or flat line bright blue/iprgc chromatic pupillometry testing results for this dog. 70

83 Figure 14. Alignment of the amino acid sequence of canine melanopsin with other species (mouse, chicken, lizard and human). Residues that are identical to the canine sequence are shaded. Percentage identity of canine melanopsin to other species is reported at the end of the alignment. 71

84 Figure 14 (cont d). Figure 15. qrt-pcr results for canine models of inherited retinal disease. There was a significant decrease (**) in melanopsin/opn4 expression in PRCD-mutant retinas compared to wt (p<0.05). The expression levels in the individual RPGR- and PDE6Bmutant samples were at the lower end of wt range. 72

85 Figure 16. Melanopsin immunostaining of representative retinal sections from wt and PDE6B-mutant dogs. Immunohistochemistry confirm the presence of iprgcs. Other RGCs are shown in green (NeuN). Cell nuclei are shown in blue with DAPI. Calibration bar = 20 µm. ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. 73

86 Figure 17. Immunohistochemistry results of affected dogs compared to wt dog. In young and old PDE6B-mutant dogs, rod outer segments (rho) are abnormal (young, B1) and lost (old, C1) with mislocalization of rho from the OS to the cell body. The cone OS (GNAT2, hcar, and LM) are short (young, B2) and eventually lost (old, C2). There is also progressive thinning of the ONL. In CNGB3-mutant dogs, rods (rho) are present with normal OS (D1). There fewer and abnormally shaped cone OS compared to wt (hcar, LM) and they contain no detectable GNAT2 (D1, D2). In the STK38L affected dogs, there is severe degeneration of the retina with loss of characteristic layering. No rod and cone photoreceptors are present (E1, E2). In the PRCD-mutant dogs, both rods and cones are still present with normal localization of their specific markers Rho, GNAT2, LM, and hcar despite a loss of ONL thickness (F1, F2). Except for STK38L-mutant dog (E3), dendrites/axons of secondary neurons (Neurofil) as well as RGC (NeuN) are well preserved in all the retina samples (A2, B3, C3, D3, F3). Cell nuclei are shown in blue with DAPI. Calibration bar = 20 µm. OS, outer segment; IS, inner segment; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. 74

87 Figure 17 (cont d). 75