Inherited Retinal Degeneration: Genetics, Disease Characterization, and Outcome Measures

Lead Guest Editor: Naheed Khan Guest Editors: Benedetto Falsini, Mineo Kondo, and Anthony G. Robson Inherited Retinal Degeneration: Genetics, Disease Characterization, and Outcome Measures Journal of Ophthalmology

This is a special issue published in “Journal of Ophthalmology.” All articles are open access articles distributed under the Creative Com- mons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Editorial Board

Monica L. Acosta, New Zealand Ian Grierson, UK David P. Piñero, Spain Hee B. Ahn, Republic of Korea Vlassis Grigoropoulos, Greece Jesús Pintor, Spain Usha P. Andley, USA Vishal Jhanji, Hong Kong Antonio Queiros, Portugal Siamak Ansari-Shahrezaei, Austria Naoshi Kondo, Japan Anthony G. Robson, UK Taras Ardan, Czech Republic Ozlem G. Koz, Turkey Mario R. Romano, Italy Francisco Arnalich-Montiel, Spain Hiroshi Kunikata, Japan Dirk Sandner, Germany Takayuki Baba, Japan Toshihide Kurihara, Japan Ana Raquel Santiago, Portugal Paul Baird, Australia George Kymionis, Greece Patrik Schatz, Sweden Antonio Benito, Spain Neil Lagali, Sweden Kin Sheng Lim, UK Mehmet Borazan, Turkey Achim Langenbucher, Germany Wisam A. Shihadeh, USA Francis Carbonaro, Malta Van C. Lansingh, Mexico Bartosz Sikorski, Poland Chi-Chao Chan, USA Paolo Lanzetta, Italy Shivalingappa K. Swamynathan, USA Lingyun Cheng, USA Theodore Leng, USA Suphi Taneri, Germany Chung-Jung Chiu, USA Marco Lombardo, Italy Christoph Tappeiner, Switzerland Colin Clement, Australia Tamer A. Macky, Egypt Stephen Charn Beng Teoh, Singapore Miguel Cordero-Coma, Spain Edward Manche, USA Panagiotis Theodossiadis, Greece Ciro Costagliola, Italy Flavio Mantelli, USA Biju B. Thomas, USA Vasilios F. Diakonis, USA Enrique Mencía-Gutiérrez, Spain Lisa Toto, Italy Priyanka P. Doctor, India Marcel Menke, Switzerland Manuel Vidal-Sanz, Spain Michel E. Farah, Brazil Lawrence S Morse, USA Gianmarco Vizzeri, USA Giulio Ferrari, Italy Majid M. Moshirfar, USA Suichien Wong, UK Paolo Fogagnolo, Italy Ramon Naranjo-Tackman, Mexico Victoria W Y Wong, Hong Kong Joel Gambrelle, France Neville Osborne, UK Terri L. Young, USA M.-A. Gamulescu, Germany Ji-jing Pang, USA Hyeong Gon Yu, Republic of Korea Santiago Garcia-Lazaro, Spain Enrico Peiretti, Italy Vicente Zanon-Moreno, Spain Contents

Inherited Retinal Degeneration: Genetics, Disease Characterization, and Outcome Measures Naheed W. Khan, Benedetto Falsini, Mineo Kondo, and Anthony G. Robson Volume 2017, Article ID 2109014, 2 pages

Correlation of Macular Focal Electroretinogram with Ellipsoid Zone Extension in Stargardt Disease Edoardo Abed, Giorgio Placidi, Luigi Calandriello, Marco Piccardi, Francesca Campagna, Matteo Bertelli, Angelo Maria Minnella, Maria Cristina Savastano, and Benedetto Falsini Volume 2017, Article ID 3643495, 7 pages

ABCC6 Gene Analysis in 20 Japanese Patients with Angioid Streaks Revealing Four Frequent and Two Novel Variants and Pseudodominant Inheritance Satoshi Katagiri, Yuya Negishi, Kei Mizobuchi, Mitsuyoshi Urashima, Tadashi Nakano, and Takaaki Hayashi Volume 2017, Article ID 1079687, 7 pages

PRPF3-Associated Autosomal Dominant Retinitis Pigmentosa and CYP4V2-Associated Bietti’s Crystalline Corneoretinal Dystrophy Coexist in a Multigenerational Chinese Family Xiaohong Meng, Qiyou Li, Hong Guo, Haiwei Xu, Shiying Li, and Zhengqin Yin Volume 2017, Article ID 4156386, 10 pages

Genotype-Phenotype Characterization of Novel Variants in Six Italian Patients with Familial Exudative Vitreoretinopathy Giancarlo Iarossi, Matteo Bertelli, Paolo Enrico Maltese, Elena Gusson, Giorgio Marchini, Alice Bruson, Sabrina Benedetti, Sabrina Volpetti, Gino Catena, Luca Buzzonetti, and Lucia Ziccardi Volume 2017, Article ID 3080245, 10 pages

Physical Activity and Quality of Life in Retinitis Pigmentosa Joshua D. Levinson, Ethan Joseph, Laura A. Ward, Joe R. Nocera, Machelle T. Pardue, Beau B. Bruce, and Jiong Yan Volume 2017, Article ID 6950642, 6 pages Hindawi Journal of Ophthalmology Volume 2017, Article ID 2109014, 2 pages https://doi.org/10.1155/2017/2109014

Editorial Inherited Retinal Degeneration: Genetics, Disease Characterization, and Outcome Measures

1 2 3 4,5 Naheed W. Khan, Benedetto Falsini, Mineo Kondo, and Anthony G. Robson

1Department of Ophthalmology & Visual Science, University of Michigan, Ann Arbor, MI 48105, USA 2Department of Ophthalmology, Fondazione Policlinico Universitario A. Gemelli, Università Cattolica del S. Cuore, Rome, Italy 3Department of Ophthalmology, Mie University Graduate School of Medicine, Tsu, Japan 4Moorﬁelds Eye Hospital, 162 City Road, London EC1V 2PD, UK 5Institute of Ophthalmology, University College London, 11-43 Bath Street, London EC1V 9EL, UK

Correspondence should be addressed to Naheed W. Khan; [email protected]

Received 13 September 2017; Accepted 14 September 2017; Published 25 September 2017

Copyright © 2017 Naheed W. Khan et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Inherited retinal degenerations (IRDs) lead to incurable cause of disease. Development of effective treatment strate- vision loss and affect 1 in 2000 to 3000 individuals. These gies requires robust clinical characterization to establish disorders may cause blindness associated with dysfunction genotype-phenotype correlations and to determine the ther- or death of photoreceptor cells, but prognosis is difficult to apeutic window and optimal stage for intervention. Katagiri determine due to variable disease expression. There have and colleagues demonstrate in 20 patients with angioid been significant advances in understanding the pathogenesis streaks that ABCC6 variants play a significant role in affected and genetics of IRDs with more than 250 genes being impli- Japanese individuals. Retinal degenerations may express cated to date. This has led to the development of treatments variability in phenotype even within the same family. Iarossi aimed at restoring function or delaying progression. et al. report genetic heterogeneity and associated variable The design of therapeutic interventions depends on the phenotypes in familial exudative vitreoretinopathy (FEVR) phenotype and the molecular characteristics of the IRD. which is a complex disorder characterized by incomplete With advances in genetic testing methodology, there is an development of the retinal vasculature. The phenotypic vari- increased detection of variants in multiple genes in the same ability in such disorders adds a further degree of complexity family or even in the same individual, which makes the deter- when establishing study protocols. mination of the primary genetic cause of disease more diffi- Another major consideration is the development of tests cult to assess. Using next generation sequencing, Meng to monitor the natural history of the disease to best measure et al. show in a multigeneration Chinese family that PRPF3- outcomes and to evaluate therapeutic efficacy in human clin- associated autosomal dominant retinitis pigmentosa coexists ical trials. Newer imaging technologies have made it possible with CYP4V2-associated Bietti’s crystalline corneoretinal to study microstructural details of the retina. Quantitative dystrophy. This further complicates the characterization of assessment of the structural and functional integrity of the phenotypes of an autosomal dominant condition with an retina and the correlation between the functional and struc- autosomal recessive condition which results in a more severe tural measures will improve our understanding of the disease disease phenotype. and help in the design of appropriate outcome measures and Due to genetic heterogeneity of these retinal conditions, therefore in determining which patients might be the best patients may have very similar clinical phenotypes but differ- candidates for treatments. There is a battery of methods to ent genetic diagnoses, requiring for example gene-specific test visual function and retinal function, which poses a therapy or gene-editing treatments to address the underlying challenge when deciding which test or which combination 2 Journal of Ophthalmology of tests to choose for a particular phenotype. Abed and colleagues evaluate the relationship between cone photoreceptor function assessed by visual acuity and the focal electroretinogram (FERG) and the integrity and structure of photoreceptors using optical coherence tomography in Stargardt disease. Their study demonstrates that FERG amplitude can be reli- ably used to monitor macular cone function and that visual acuity is a useful indicator of foveal function. Neuroprotection is a largely nonspecific strategy to provide a protective environment for slowing the process of photoreceptor degeneration, thus extending the therapeutic window to later stages of disease. Exercise has been shown to have neuroprotective effects on photoreceptors in mouse models of retinal degeneration. Levinson et al. did a retrospective case-control study to report baseline physical activity levels in individuals with retinitis pigmentosa. They used three quality of life questionnaires to determine the relationship between physical activity and visual function and showed that increased physical activity is associated with greater self-reported visual function and quality of life.

Naheed W. Khan Benedetto Falsini Mineo Kondo Anthony G. Robson Hindawi Journal of Ophthalmology Volume 2017, Article ID 3643495, 7 pages https://doi.org/10.1155/2017/3643495

Research Article Correlation of Macular Focal Electroretinogram with Ellipsoid Zone Extension in Stargardt Disease

Edoardo Abed, Giorgio Placidi, Luigi Calandriello, Marco Piccardi, Francesca Campagna, Matteo Bertelli, Angelo Maria Minnella, Maria Cristina Savastano, and Benedetto Falsini

Ophthalmology Department, Università Cattolica del Sacro Cuore, Largo Agostino Gemelli 8, 00168 Rome, Italy

Correspondence should be addressed to Edoardo Abed; [email protected]

Received 19 January 2017; Revised 2 July 2017; Accepted 26 July 2017; Published 20 August 2017

Academic Editor: Biju B. Thomas

Copyright © 2017 Edoardo Abed et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Stargardt disease (STGD1) is the most common cause of inherited juvenile macular degeneration. This disease is characterized by a progressive accumulation of lipofuscin in the outer retina and subsequent loss of photoreceptors and retinal pigment epithelium. The aim of this study was to evaluate the relationship between cone photoreceptor function and structure in STGD1. Macular function was assessed by visual acuity measurement and focal electroretinogram (FERG) recording while spectral domain optical coherence tomography (SD-OCT) imaging was performed to evaluate the integrity of photoreceptors. FERG amplitude was signiﬁcantly reduced in patients with Stargardt disease (p <00001). The amplitude of FERG showed a negative relationship with interruption of ellipsoid zone (EZ) (R2 =054, p <00001) and a positive correlation with average macular thickness (AMT). Conversely, visual acuity was only weakly correlated with central macular thickness (CMT) (R2 =012, p =004). In conclusion, this study demonstrates that FERG amplitude is a reliable indicator of macular cone function while visual acuity reﬂects the activity of the foveal region. A precise assessment of macular cone function by FERG recording may be useful to monitor the progression of STGD1 and to select the optimal candidates to include in future clinical trials to treat this disease.

1. Introduction pigment epithelium (RPE) with subsequent RPE apoptosis and photoreceptor degeneration [5]. Autosomal recessive Stargardt disease (STGD1) is the most In the past 20 years, pattern electroretinogram (PERG) frequent childhood hereditary macular dystrophy, affecting has been extensively used to assess macular function in 1 per 10000 individuals [1]. patients with STGD1 [6, 7]. PERG is not recordable in almost STDGD1 is caused by the mutation of the ATP-binding all patients with STGD1 even in the case of preserved visual cassette, subfamily A, member 4 (ABCA4) gene, which acuity and in the absence of clear signs of macular degenera- encodes a transport protein localized in photoreceptor outer tion at fundus examination. As a consequence, PERG is a segments [2, 3]. The main role of ABCA4 protein is to valuable tool to establish the diagnosis of STDG1, especially remove potentially toxic retinoids, such as N-retinylidene- in the earlier phases of the disease. However, for the same phosphatidylethanolamine (NRPE) and phosphatidyletha- reason, PERG may be inappropriate to estimate the degree nolamide (PE), which originated from the phototransduction of macular dysfunction and to assess disease progression. process [4]. Furthermore, PERG signal reflects the activity of retinal gan- Mutations of ABCA4 gene determine an increase glion cells and the inner retina [8] while STDG1 primarily of NRPE and PE with generation of N-retinylidene-N- affects the RPE and photoreceptor cells. retinylethanolamine (A2E), a lipofuscin fluorophore. Flicker focal electroretinogram (FERG) is a diagnostic The shedding and phagocytosis of photoreceptors outer tool that selectively assesses macular cone photoreceptor segments lead to lipofuscin accumulation in the retinal and bipolar cell activity [9]. In previous clinical studies, this 2 Journal of Ophthalmology technique showed good test-retest repeatability, allowing a of Cirrus software. In the case of foveal sparing, the ampli- long term follow-up of macular dysfunction in retinitis tude of EZ zone interruption nasally and temporally to the pigmentosa [10] and cone-rod dystrophy [11]. Furthermore, fovea was measured and summed. it has been demonstrated that, in cone-rod dystrophies, FERG amplitude decline may anticipate visual acuity loss of 2.2. Focal Electroretinography. FERG recording was per- several years [12]. formed in accordance to a previously described technique Spectral domain optical coherence tomography (SD- [9–19]. Before examination, pupils were pharmacologically OCT) is a noninvasive imaging technique that provides dilated with tropicamide 1% eye drops to at least 8 mm diam- information about the morphology of all distinct retinal eter. FERGs were recorded monocularly with an Ag-AgCl layers and enables reliable and repeatable measurements of electrode taped on the skin over the lower eyelid. A similar macular thickness. In patients with STDG1, SD-OCT has electrode was placed over the eyelid of the contralateral allowed clinicians to accurately visualize the extent and the patched eye and was used as reference. degree of the degeneration of both inner and outer retinal Stimuli consisted of flickering uniform fields generated layers [13–15]. More specifically, ellipsoid zone extension ° by an array of 8 red light-emitting diodes covering a 18 can be considered an important imaging parameter that diameter with a mean luminance of 80 cd/m2 and a temporal allows clinicians to evaluate to the extent of damage of frequency of 41 Hz. The dominant wavelength of the stimu- macular photoreceptors [16]. lus was 630 nm. The flickering stimulus was sinusoidally The aim of our study was to assess macular cone function driven by a custom-made digital frequency generator and by FERG recording in STDG1 and to explore the correlation presented on the rear of a modified Ganzfeld bowl (Primus; between photoreceptor function and structure as determined LACE Elettronica, Pisa, Italy) illuminated at the same mean by FERG and SD-OCT, respectively. luminance as the stimulus. Adiffusing filter was placed in front of the LED array in 2. Methods order to make it appear as a circle of uniform red light. A steady DC signal maintained the mean luminance of the Patients with STDG1 evaluated at the Department of stimulus. A small square marker was placed in the center of Ophthalmology of the Catholic University of Rome between the Ganzfeld bowl to allow the investigator maintaining June 2012 and June 2015 were retrospectively enrolled. The steady fixation on the foveal region. The examined patients diagnosis of STDG1 was clinically established and then con- were placed at 30 cm distance from the stimulus. firmed, in all patients, by genetic testing with next generation FERG signals were amplified, filtered (band pass filter sequencing (NGS) technology. The examination protocol between 1 and 250 Hz), and averaged (12-bit resolution, included best-corrected visual acuity (BCVA) measurement 2 kHz sampling rate, 1600 repetitions in 8 blocks). Signals with Snellen charts, slit-lamp biomicroscopy, indirect oph- exceeding the threshold voltage (25 mV) were rejected to thalmoscopy, spectral domain optical coherence tomography minimize noise coming from blinks or eye movements. After (SD-OCT), and FERG recording. Twenty age-matched the recording, a Fourier discrete analysis was performed to healthy patients were also enrolled and served as controls isolate the FERG’s first harmonic (1F), and its peak-to- for FERG amplitude. peak amplitude was measured. Averaging and Fourier anal- The study followed the tenets of the Declaration of ysis were also performed on signals sampled asynchronously Helsinki and was approved by the Ethics Committee of the at 1.1 times the temporal frequency of the stimulus, to esti- Catholic University of Sacred Heart of Rome. All patients mate the background noise at the fundamental component. signed a written informed consent before the enrollment. Under these conditions, the FERGs recorded were above the noise level (noise amplitude <0.08 mV in all cases) and 2.1. Imaging. SD-OCT scans were performed using the sufficiently reliable (the variation coefficient in amplitude Cirrus OCT (Carl Zeiss Meditec Inc., Dublin, California, was 20%). USA). The examination protocol consisted of a 6 × 6mm macular cube, centered on the fovea, composed of 128 horizontal b-scans of 512 a-scans each. Retinal thickness values 2.3. Statistical Analysis. Data from the right eye of both were automatically calculated by Cirrus OCT software for patients with STGD1 and controls were used for the analysis each of the nine areas corresponding to the Early Treat- to reduce the risk of data redundancy. Two-tailed unpaired ment Diabetic Retinopathy Study Research Group (ETDRS) t-test was used to compare FERG 1F amplitude between [17, 18]. Both the average macular thickness (AMT) and the two groups and to compare BCVA and FERG amplitude the central macular thickness (CMT) were recorded. The between STGD1 patients with or without foveal involve- AMT and CMT correspond to the mean retinal thickness ment. Analysis of variance and Sidak post hoc test was per- in the circular zones of 6 and 1 mm diameter, respectively, formed to compare macular functional and anatomical centered in the fovea. impairment between patients with different mutation sever- Cirrus software was also used to navigate the acquired ity and patient with early (<17 years) and late (>17 years) macular cube in order to identify the horizontal b-scan with age of onset. Receiving operating characteristic (ROC) the largest interruption of photoreceptor ellipsoid zone (EZ). curves were performed to assess whether FERG alterations The maximum extent of EZ interruption was then manually may be useful to diagnose STGD1 and to predict photore- measured by two independent examiners using the calipers ceptors loss in these patients. Journal of Ophthalmology 3

Table 1: Demographics and clinical characteristics of patients with Stargardt macular dystrophy.

Patient Age at BCVA RE BCVA LE FERG 1F FERG 1F Sex Age First allele Second allele number onset (Snellen decimals) (Snellen decimals) RE (μV) LE (μV) 1 F 22 15 p.R1098C — 0.1 0.1 0.51 0.28 2 F 33 10 p.V767D p.R2030X 0.05 0.1 0.17 0.2 p.Gly1961Glu, 3 M 25 15 p.Val2026ThrfsX52 0.4 0.4 0.52 0.43 p.Gln1332_cys1339dup 4 F 63 35 p.Gly1961Glu — 0.15 0.15 0.68 0.38 p.Gly991Arg, 5 M 58 32 p.Arg1300Gln 0.02 0.02 0.52 0.52 p.Glu1087Lys 6 M 33 14 p.Asn78Ser p.Trp880Cys 0.02 0.01 0.21 0.15 7 M 50 24 p.R1098C — 0.4 0.8 0.52 0.59 8 F 16 9 p.Q2220X p.R943Q 0.1 0.1 0.08 0.06 9 M 41 16 c.634C>T c.1497G>C 0.2 0.2 0.74 0.39 10 F 37 21 c.4771G>A — 0.1 0.1 0.2 0.22 11 M 11 8 c.247_250dup c.4139C>T 0.8 0.8 0.26 0.34 12 M 41 5 p.His423Arg p.Arg943Gln 0.1 0.02 0.18 0.15 13 F 18 10 IVS35+2T>C IVS40+5G>A 0.2 0.2 0.63 0.3 14 M 48 43 c.5714+5G>A — 1.0 1.0 0.33 0.46 15 F 39 9 p.Arg18Trp p.Val767Asp 0.05 0.1 0.33 0.35 16 M 15 9 p.Gly1961Glu p. Ser2255Ile 0.1 0.1 0.79 0.7 17 M 47 19 p.Thr897Ile — 0.01 0.05 0.1 0.09 p.Gln21Ter, p. Arg212His, 18 M 36 31 0.1 0.1 0.37 0.46 p.Arg943Gly p.Gly1961Glu p.Trp821Arg 19 M 16 13 p. Tyr850Cys 0.9 0.9 0.1 0.2 p.Gly1961Glu 20 F 31 21 p.Thr977Pro c.IVS40+5G>A 0.2 0.1 0.46 0.15 21 F 22 15 p.Gly1961Glu c.4709_4711delA 0.1 0.1 0.57 0.45 22 M 48 13 p.Gly1961Glu p.His1406ProfsX29 0.1 0.1 0.19 0.25 23 M 16 8 p.Val931Met p.Val931Met 0.1 0.1 0.57 0.4 24 M 42 8 p.Val256splice p.Trp1479X 0.01 0.005 0.14 0.16 25 M 21 14 IVS13+1g>a IVS40+5g>40 0.1 0.15 0.2 0.17 p.Tyr850Cys, 26 F 23 17 p.Gly1961Glu 0.3 0.3 0.07 0.09 p.Thr959Ala 27 F 56 23 p.Gly1961Glu p.Gly1961Glu 0.8 0.1 0.16 0.31 28 M 52 20 p.Arg1108His p.Gly1961Glu 0.1 0.4 0.48 0.93 29 F 31 14 p.P1486L p.W700X 1.0 1.0 0.13 0.03 30 F 12 8 p.Tyr1400X p.Val931Met 0.2 0.2 0.45 0.46 31 F 77 13 p.Ala1598Asp p.Val2062fsX2113 0.1 0.1 0.24 0.33 32 F 11 10 IVS6-1G>T p.Asn1436Ile 0.2 0.2 0.66 0.95 33 M 44 10 IVS45+1G>A p.Gly1961Glu 0.1 0.1 0.14 0.15

Pearson’s correlation test was used to correlate macular Mean age at observation was 34.4 ± 16.5 years while functional and anatomical parameters. A p value < 0.05 was mean age of symptoms onset was 15.8 ± 8.9 years. Eight considered statistically signiﬁcant. patients had one missense mutation in each allele (24.2%), and 16 patients (48.5%) had two or more missense muta- 3. Results tions in at least one allele while the remaining 9 patients (27.3%) had a null mutation in at least one allele. Mean Thirty-three patients (18 males and 15 females) with STGD1 CMT, AMT, and EZ interruption were 141 ± 40, 217 ± 35, were included in this study. Patients’ demographics and and 3911 ± 1423 μm, respectively. molecular and functional characteristics are reported in FERG was recordable in all patients of both study Table 1, while the results of OCT examination are reported groups. FERG 1F amplitude was signiﬁcantly reduced in in Table 2. patients with STGD1 compared to controls (0.35 ± 0.22 4 Journal of Ophthalmology

Table 2: OCT data of individual patients with Stargardt disease.

Patient number EZ interruption RE (μm) EZ interruption LE (μm) AMT RE (μm) AMT LE (μm) CMT RE (μm) CMT LE (μm) 1 2922 3665 257 256 121 118 2 6000 6000 289 289 266 273 3 2092 2728 255 257 138 147 4 2872 3428 262 255 155 133 5 3384 3401 237 249 99 80 6 6000 6000 134 145 97 95 7 3398 3145 207 199 186 253 8 5866 5916 154 135 112 98 9 3444 4272 215 198 135 112 10 4156 4453 198 204 121 125 11 3857 3688 185 176 117 113 12 6000 6000 195 187 147 132 13 3600 3445 225 234 145 138 14 5261 4241 224 235 241 251 15 5421 5578 212 203 144 143 16 2353 2535 253 243 166 153 17 3906 4453 248 241 183 157 18 3214 2896 225 231 149 158 19 4370 3999 243 239 160 159 20 3827 4972 212 208 145 134 21 2741 2895 223 243 138 146 22 3585 3352 244 250 104 108 23 2936 3336 243 254 173 171 24 6000 6000 135 147 74 81 25 3575 3936 221 236 125 144 26 5308 6000 212 201 139 127 27 3315 2980 194 197 131 124 28 2867 2100 258 260 100 108 29 5370 5227 218 223 189 209 30 3557 3715 192 185 89 91 31 6000 6000 174 188 95 101 32 1868 1632 227 234 154 161 33 6000 6000 194 187 103 112 and 1.68 ± 0.27, resp., p <00001). The diagnostic accuracy FERG 1F amplitude was negatively correlated with the of FERG amplitude to distinguish STGD1 and healthy extension of EZ interruption (R2 =054, p = 0 < 0001) patients, assessed by ROC analysis, was 100% (area under (Figure 3) and positively correlated with AMT (R2 =016, the curve 1.0). The corresponding ROC curve is shown p =002). Conversely, no correlation was noted between in Figure 1. FERG amplitude and CMT (R2 =0003, p =076). BCVA Interestingly, while FERG 1F amplitude was significantly did not correlate with anatomical alterations, except a weak < μ 2 reduced ( 1 V) in all patients with STGD1, six patients negative relationship with CMT (R =012, p =004). (18%), due to foveal sparing, had a good visual acuity Interestingly, no differences were noted regarding FERG ≥ ( 0.8) in one or both eyes despite a remarkable interruption 1F amplitude, BCVA, and OCT parameters between patients fi of EZ (Figure 2). Patients with foveal sparing had signi - with early or late onset or with different molecular mutation cantly better BCVA than those with foveal involvement severity (p >005 for all analyses). (0.90 ± 0.05 and 0.13 ± 0.02, resp., p <00001) while no differences were noted in FERG 1F amplitude between 4. Discussion these two groups of patients (0.20 ± 0.04 and 0.38 ± 0.04, resp., p =007). The aim of our study was to explore the correlation between Correlations between macular functional and structural functional and anatomical photoreceptor alterations in parameters are summarized in Table 3. patients with STDG1, as evaluated by FERG and SD-OCT, Journal of Ophthalmology 5

ROC curve: ROC of unpaired t-test data Table 3: Results of correlation analyses between macular functional 150 and anatomical alterations in patients with Stargardt macular dystrophy.

2 BCVA FERG 1 F R =004, p =027 2 100 BCVA EZ interruption R =001, p =056 2 BCVA CMT R =012, p =004 2 BCVA AMT R =00001, p =095 2 Sensitivity (%) 50 FERG 1F EZ interruption R =054, p =00001 2 FERG 1F CMT R =0003, p =076 2 FERG 1F AMT R =016, p =002 0 R2 =014 p =003 0 50 100 150 EZ interruption CMT , 2 100% ‒ specificity (%) EZ interruption AMT R =057, p <00001 2 CMT AMT R =030, p =0001 Figure 1: ROC curve showing diagnostic accuracy of FERG 1F amplitude in detecting Stargardt macular dystrophy.

8000

6000

4000

2000 Ellipsoid zone interruption ( 𝜇 m) 0 Figure 2: SD-OCT scans of the right and left eye of a patient with 0.0 0.2 0.4 0.6 0.8 1.0 significant disruption of macular photoreceptors and foveal fERG 1F amplitude (𝜇V) sparing. While FERG 1F amplitude was severely reduced, BCVA was 1.0 in both eyes. Figure 3: Scatterplot illustrating the significant correlation between FERG first harmonic amplitude and ellipsoid zone interruption. respectively. Compared to controls, FERG 1F amplitude was reduced in patients with STDG1 and ROC analysis showed relationship between the extent of ellipsoid zone disruption that focal electroretinogram had a diagnostic accuracy of and visual acuity. 100% in discriminating healthy and STDG1 patients. These This apparent discrepancy may be related to a relatively results indicate that, similarly to PERG, FERG recording high number of patients with foveal sparing in our study. may be useful to establish the diagnosis of STDG1. Interest- In these cases, the integrity of photoreceptor layers in the ingly, despite the significant reduction of macular function, foveal region lead to a preservation of visual acuity despite FERG 1F amplitude was recordable in all STDG1 patients, a substantial loss of EZ in the surrounding retina. This suggesting that this test may be a valuable tool also to assess assumption seems to be confirmed by the observation that the deterioration of macular dysfunction in these subjects. A while no relationship was evident between AMT and BCVA, recent study demonstrated that FERG fundamental har- a weak even if significant positive correlation was found monic alterations anticipate by several years the deteriora- between CMT and visual acuity. tion of visual acuity in patients affected by cone-rod Conversely, FERG 1F amplitude showed a significantly dystrophies [12]. Longitudinal studies are warranted to negative correlation with the extent of EZ loss and was determine whether FERG alterations may predict visual markedly altered also in cases with foveal sparing acuity decrease in STDG1. whereas, in these patients, BCVA was largely preserved. Surprisingly, no significant correlation was noted These data would suggest that BCVA measurement between BCVA and the entity of photoreceptor loss, deter- reflects only roughly the function of the foveal region, ° mined as the maximal linear EZ interruption. This finding while FERG signals evocated by an 18 flickering stimulus is in contrast with a previous observational case series of may provide an objective and reliable quantification of 14 patients by Ergun et al. [13] reporting a negative macular cone function. 6 Journal of Ophthalmology

The exact quantification of macular cone residual activity [5] J. Weng, N. L. Mata, S. M. Azarian, R. T. Tzekov, D. G. Birch, may play an important role for the optimal enrollment of and G. H. Travis, “Insights into the function of Rim protein in patients in future clinical trials. photoreceptors and etiology of Stargardt’s disease from the In particular, this assessment may be critical for therapies phenotype in abcr knockout mice,” Cell, vol. 98, no. 1, – aimed at preservation or restoration of photoreceptor func- pp. 13 23, 1999. tion such as nutritional treatments [20, 21], growth factor [6] N. Lois, G. E. Holder, C. Bunce, F. W. Fitzke, and A. C. Bird, “ administration [22–25], and gene augmentation [26, 27]. Phenotypic subtypes of Stargardt macular dystrophy-fundus fi flavimaculatus,” Archives of Ophthalmology, vol. 119, no. 3, Hence, these ndings suggest that FERG recording may – be a useful adjunct tool to increase the accuracy of patient’s pp. 359 369, 2001. [7] N. Lois, G. E. Holder, F. W. Fitzke, C. Plant, and A. C. Bird, selection in promising future clinical trials. “ Interestingly, even if FERG fundamental harmonic ampli- Intrafamilial variation of phenotype in Stargardt macular dystrophy-fundus flavimaculatus,” Investigative Ophthalmol- tude was also correlated with AMT, this correlation was ogy and Visual Science, vol. 40, no. 11, pp. 2668–2675, 1999. notably weaker. “ Our study failed to demonstrate a correlation between [8] G. B. Arden and C. R. Hogg, Clinical and experimental evidence that the pattern electroretinogram (PERG) is generated early onset and disease severity and the decay of retinal func- fi in more proximal retinal layers than the focal electroretino- tion in patients with STDG1. This nding is apparently in gram (FERG),” Annals of the New York Academy of Sciences, contrast with the results of a previous study by Fujinami vol. 388, pp. 580–607, 1982. et al. [28], showing that childhood onset is associated with [9] P. Errico, B. Falsini, V. Porciatti, and F. M. Cefalá, “The human a more severe alteration of retinal function. However, that focal electroretinogram as a function of stimulus area,” Docu- study found only an increased proportion of patients with menta Ophthalmologica, vol. 75, no. 1, pp. 41–48, 1990. ff di use impairment of retinal function which was noted in [10] B. Falsini, L. Galli-Resta, A. Fadda et al., “Long-term decline of patients with early onset of STDG1 disease compared with central cone function in retinitis pigmentosa evaluated by focal adults while visual acuity was not significantly different in electroretinogram,” Investigative Ophthalmology and Visual the two study groups. This data would suggest that macular Science, vol. 53, no. 12, pp. 7701–7709, 2012. function is immediately altered in STDG1 causing an early [11] L. Galli-Resta, B. Falsini, G. Rossi et al., “Bilateral symmetry of decay of focal ERG and visual acuity in both adult and child- visual function loss in cone-rod dystrophies,” Investigative hood onset patients while retinal-wide involvement develops Ophthalmology and Visual Science, vol. 57, no. 8, pp. 3759– more frequently in patients with early disease onset. 3768, 2016. In summary, FERG recording is significantly altered in [12] L. Galli-Resta, M. Piccardi, L. Ziccardi et al., “Early detection of STDG1 and has an optimal diagnostic accuracy in detecting central visual function decline in cone-rod dystrophy by the Stargardt disease. The reduction of FERG 1F amplitude is use of macular focal cone electroretinogram,” Investigative closely related to the extent of macular photoreceptor disrup- Ophthalmology and Visual Science, vol. 54, no. 10, pp. 6560– tion. These findings suggest that FERG recording may be a 6569, 2013. useful tool to monitor the progression of macular dysfunc- [13] E. Ergun, B. Hermann, M. Wirtitsch et al., “Assessment of cen- tion in Stargardt disease. Furthermore, an accurate estima- tral visual function in Stargardt’s disease/fundus flavimacula- ” tion of residual cone function will be important to optimize tus with ultrahigh-resolution optical coherence tomography, Investigative Ophthalmology and Visual Science, vol. 46, the design and inclusion criteria of future clinical trials to – treat STDG1. no. 1, pp. 310 316, 2005. [14] T. R. Burke, S. Yzer, J. Zernant, R. T. Smith, S. H. Tsang, and “ Conflicts of Interest R. Allikmets, Abnormality in the external limiting membrane in early Stargardt disease,” Ophthalmic Genetetics, vol. 34, – The authors declare that there is no conflict of interest no. 1-2, pp. 75 77, 2013. “ regarding the publication of this paper. [15] W. C. Huang, A. V. Cideciyan, A. J. Roman et al., Inner and outer retinal changes in retinal degenerations associated with ABCA4 mutations,” Investigative Ophthalmology and Visual References Science, vol. 55, no. 3, pp. 1810–1822, 2014. [1] P. P. Bither and L. A. Berns, “Stargardt’s disease: a review of [16] S. Saxena, K. Srivastav, C. M. Cheung, J. Y. Ng, and T. Y. Lai, “ the literature,” Journal of the American Optometric Associa- Photoreceptor inner segment ellipsoid band integrity on ” tion, vol. 59, no. 2, pp. 106–111, 1988. spectral domain optical coherence tomography, Clinical – [2] M. Illing, L. L. Molday, and R. S. Molday, “The 220-kDa Rim Ophthalmology, vol. 8, pp. 2507 2522, 2014. protein of retinal rod outer segments is a member of the [17] L. A. Paunescu, J. S. Schuman, L. L. Price et al., “Reproducibil- ABC transporter superfamily,” Journal of Biological Chemistry, ity of nerve fiber thickness, macular thickness, and optic nerve vol. 272, no. 15, pp. 10303–10310, 1997. head measurements using StratusOCT,” Investigative Ophthal- – [3] H. Sun and J. Nathans, “Stargardt’s ABCR is localized to mology and Visual Science, vol. 45, no. 6, pp. 1716 1724, 2004. the disc membrane of retinal rod outer segments,” Nature [18] Early Treatment Diabetic Retinopathy Study Research Group, Genetics, vol. 17, no. 1, pp. 15-16, 1997. “Photocoagulation for diabetic macular edema: early treat- ” [4] F. Quazi, S. Lenevich, and R. S. Molday, “ABCA4 is an ment diabetic retinopathy study report number 1, Archives – N-retinylidene-phosphatidylethanolamine and phosphatidyl- of Ophthalmology, vol. 103, pp. 1796 1806, 1985. ethanolamine importer,” Nature Communications, vol. 925, [19] A. Scupola, A. Mastrocola, P. Sasso et al., “Assessment of reti- no. 3, 2012. nal function before and after idiopathic macular hole surgery,” Journal of Ophthalmology 7

American Journal of Ophthalmology, vol. 156, no. 1, pp. 132.e1–139.e1, 2013. [20] L. Ma, Y. Kaufman, J. Zhang, and I. Washington, “C20-D3- vitamin A slows lipofuscin accumulation and electrophysio- logical retinal degeneration in a mouse model of Stargardt disease,” Journal of Biological Chemistry, vol. 286, pp. 7966– 7974, 2011. [21] T. S. Aleman, A. V. Cideciyan, E. A. Windsor et al., “Macular pigment and lutein supplementation in ABCA4-associated retinal degenerations,” Investigative Ophthalmology and Visual Science, vol. 48, no. 3, pp. 1319–1329, 2007. [22] W. M. Zein, B. G. Jeﬀrey, H. E. Wiley et al., “CNBG3-achroma- topsia clinical trial with CNTF: diminished rod pathway responses with no evidence of improvement of function,” Investigative Ophthalmology and Visual Science, vol. 55, no. 10, pp. 6301–6308, 2014. [23] B. Falsini, G. Iarossi, A. Chiaretti et al., “NGF eye-drops topical administration in patients with retinitis pigmentosa, a pilot study,” Journal of Translational Medicine, vol. 14, p. 8, 2016. [24] D. G. Birch, R. G. Weleber, J. L. Duncan, G. J. Jaﬀe, W. Tao, and Ciliary Neurotrophic Factor Retinitis Pigmentosa Study Groups, “Randomized trial of ciliary neurotrophic factor deliv- ered by encapsulated cell intraocular implants for retinitis pigmentosa,” American Journal of Ophthalmology, vol. 156, no. 2, pp. 283–292, 2013. [25] E. Abed, G. Corbo, and B. Falsini, “Neurotrophin family members as neuroprotectants in retinal degenerations,” BioDrugs, vol. 29, no. 1, pp. 1–13, 2015. [26] K. Binley, P. Widdowson, J. Loader et al., “Transduction of photoreceptors with equine infectious anemia virus lentiviral vectors: safety and biodistribution of StarGen for Stargardt disease,” Investigative Ophthalmology and Visual Science, vol. 54, no. 6, pp. 4061–4071, 2013. [27] J. Kong, S. R. Kim, K. Binley et al., “Correction of the disease phenotype in the mouse model of Stargardt disease by lentiviral gene therapy,” Gene Therapy, vol. 15, no. 19, pp. 1311– 1320, 2008. [28] K. Fujinami, J. Zernant, R. K. Chana et al., “Clinical and molecular characteristics of childhood-onset Stargardt disease,” Ophthalmology, vol. 122, no. 2, pp. 326–334, 2015. Hindawi Journal of Ophthalmology Volume 2017, Article ID 1079687, 7 pages https://doi.org/10.1155/2017/1079687

Research Article ABCC6 Gene Analysis in 20 Japanese Patients with Angioid Streaks Revealing Four Frequent and Two Novel Variants and Pseudodominant Inheritance

1 1,2 1 2 Satoshi Katagiri, Yuya Negishi, Kei Mizobuchi, Mitsuyoshi Urashima, 1 1,3 Tadashi Nakano, and Takaaki Hayashi

1Department of Ophthalmology, The Jikei University School of Medicine, Tokyo, Japan 2Division of Molecular Epidemiology, The Jikei University School of Medicine, Tokyo, Japan 3Department of Ophthalmology, Katsushika Medical Center, The Jikei University School of Medicine, Tokyo, Japan

Correspondence should be addressed to Takaaki Hayashi; [email protected]

Received 18 May 2017; Accepted 6 August 2017; Published 20 August 2017

Academic Editor: Anthony G. Robson

Copyright © 2017 Satoshi Katagiri et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Purpose. To report the spectrum of ABCC6 variants in Japanese patients with angioid streaks (AS). Patients and Methods. This was a single-center cohort study. The medical records of 20 patients with AS from 18 unrelated Japanese families were retrospectively reviewed. Screening of the ABCC6 gene (exons 1 to 31) was performed using PCR-based Sanger sequencing. Results. Eight ABCC6 variants were identified as candidate disease-causing variants. These eight variants included five known variants (p.Q378X, p.R419Q, p.V848CfsX83, p.R1114C, and p.R1357W), one previously reported variant (p.N428S) of unknown significance, and two novel variants (c.1939C>T [p.H647Y] and c.3374C>T [p.S1125F]); the three latter variants were determined to be variants of significance. The following four variants were frequently identified: p.V848CfsX83 (14/40 alleles, 35.0%), p.Q378X (7/40 alleles, 17.5%), p.R1357W (6/40 alleles, 15.0%), and p.R419Q (4/40 alleles, 10.0%). The ABCC6 variants were identified in compound heterozygous or homozygous states in 13 of 18 probands. Two families showed a pseudodominant inheritance pattern. Pseudoxanthoma elasticum was seen in 15 of 17 patients (88.2%) who underwent dermatological examination. Conclusions. We identified disease-causing ABCC6 variants that were in homozygous or compound heterozygous states in 13 of 18 families (72.2%). Our results indicated that ABCC6 variants play a significant role in patients with AS in the Japanese population.

1. Introduction The ATP-binding cassette subfamily C member 6 (ABCC6) gene was first reported in 2000 as the cause of Angioid streaks (AS; OMIM #607140) are a hereditary retinal PXE [3–5]. Screening of the ABCC6 gene has been performed disease involving irregular streaks that radiate from the optic in cases with PXE in several ethnic groups [6–10]. In the disc due to cracking of Bruch’s membrane. There are numer- Japanese population, there are two case reports of ABCC6 ous systemic associations with AS, including pseudox- gene analysis in patients with PXE/A [11, 12]. Subsequently, anthoma elasticum (PXE; OMIM #264800), Paget’s disease two large-scale studies for ABCC6 gene analysis have been of bone, sickle cell anemia, and Ehlers-Danlos syndrome reported in 54 patients with AS [13] and in 76 patients with [1]. PXE is an autosomal recessive disease characterized by PXE [14] and have revealed differences in the ABCC6 muta- the accumulation of mineralized and fragmented elastic tion/variant spectrum between the two studies in Japanese fibers in the skin. The disease is the most common disease patients [13, 14]. For example, although the p.Q378X variant associated with AS and was reported to be found in 59% of was frequently seen in PXE patients [14], it was never patients with AS [2]. detected in AS patients [13]. In addition, two variants 2 Journal of Ophthalmology

(p.V848CfsX83 and p.Q378X) are frequently found in PXE purified with a QIAquick PCR Purification Kit (Qiagen) patients [14], leading to the hypothesis that frequently found and used as templates for sequencing. Both strands were variants may also exist in Japanese patients with AS. analyzed on an automated sequencer (ABI Prism 3700 The purpose of this study was to determine whether DNA Analyzer, Applied Biosystems, Tokyo, Japan). We frequent ABCC6 variants or hotspots exist in Japanese used BAC clone CIT987SK-A-962B4 (accession number patients with AS. We screened the ABCC6 gene in 20 patients U91318) for genomic DNA and ABCC6 mRNA (accession with AS from 18 unrelated Japanese families, and we number NM_001171.5) sequences from the National Center identified four frequent variants, two novel variants, and for Biotechnology Information. a pseudodominant inheritance pattern. 2.3. Evaluation of the Identified ABCC6 Variants. To predict 2. Materials and Methods the functional impact of the ABCC6 variants, we performed three different in silico analyses using the Polyphen-2 (http:// 2.1. Medical Records Used in the Current Study. This was a genetics.bwh.harvard.edu/pph2/), SIFT (http://sift.jcvi.org), single-center cohort study. The medical records of 20 and PROVEAN (http://provean.jcvi.org) programs. To inves- patients (11 males and 9 females, aged 16 to 81 years, mean: tigate the frequency of the identified variants, we used the 53 years) with AS from 18 unrelated Japanese families and Human Genetic Variation database (http://www.hgvd.genome. eight unaffected family members of these patients were retro- med.kyoto-u.ac.jp/index.html) and the Exome Aggregation spectively reviewed at the Jikei University Hospital from June Consortium database (http://exac.broadinstitute.org). We also 2006 to February 2010. There was no consanguinity of referred to the Leiden Open Variation Database 3.0 (https:// parents of all probands (Table 1). The protocol used for this databases.lovd.nl/shared/genes/ABCC6) and the Human study was approved by the Institutional Review Board of the Gene Mutation Database (http://www.hgmd.cf.ac.uk) for Jikei University School of Medicine. The protocol adhered to determining the pathogenicity of the identified variants. the tenets of the Declaration of Helsinki, and informed consent was obtained from each participant. Ophthalmic 3. Results and dermatological examinations were performed for the diagnosis of AS with or without choroidal neovascularization 3.1. Pathogenicity of the Identified ABCC6 Variants. In total, (CNV) and PXE; the results are summarized in Table 1. eight ABCC6 variants were identified as candidate disease- Among the 20 patients with AS, 15 were diagnosed with causing variants. The data of these eight ABCC6 variants PXE, two were diagnosed not having PXE, and three were are summarized in Tables 1 and 2. Five of these eight ABCC6 not examined for PXE; eight AS patients were complicated variants are known variants that have previously been with CNV. Only two patients (cases 3 and 6) with CNV were reported (c.1132C>T [p.Q378X], c.1256G>A [p.R419Q], 50 years old or younger. Fundus and fluorescein angiography c.2542delG [p.V848CfsX83], c.3340C>T [p.R1114C], and images of a representative proband (case number 6, c.4069C>T [p.R1357W]). c.1283A>G (p.N428S) is a variant JU#0451) are shown in Figure 1. of unknown significance as the p.N428S variant was only found in controls but not in patients with AS in a previous 2.2. Molecular Genetic Studies. Blood samples were obtained study [13]. However, in all three in silico programs, the from all affected cases and some of their family members. p.N428S variant was predicted to cause severe damage to Genomic DNA was isolated from peripheral white blood cells protein function. The remaining two variants, c.1939C>T using a Gentra Puregene Blood Kit (Qiagen, Hilden, (p.H647Y) and c.3374C>T (p.S1125F), have not been Germany); the DNA was used as a template for amplifying previously reported. The SIFT and PROVEAN programs human ABCC6 genomic sequences. To analyze the ABCC6 predicted p.H647Y to cause severe damage, whereas the gene, all 31 exons including the exon/intron boundaries Polyphen-2 program predicted the variant to be benign. were amplified by PCR with the primer pairs shown in Regarding the variant p.S1125F, the in silico programs Supplemental Table 1 available online at https://doi.org/ predicted it to cause severe damage to protein function. 10.1155/2017/1079687. The genomic nucleotide sequences Neither of the p.H647Y and p.S1125F variants was found in of exons 1 to 9 of the ABCC6 gene were very similar to those the Single Nucleotide Polymorphism Database, the Human of ABCC6 pseudogenes. In particular, we confirmed that the Genetic Variation database, the Exome Aggregation Consor- pseudogenes contained the nonsense variant c.1132C>T tium database, the Leiden Open Variation Database 3.0, or (p.Q378X) in exon 9 even in the Japanese population, as the Human Gene Mutation Database. Interestingly, segrega- was previously reported [6]. For exon 9, to differentiate the tion analysis in family 7 (JU#0481) revealed that two variants, real ABCC6 gene from the pseudogenes, we performed p.N428S and p.H647Y, were on the same allele (Figure 2). long-range PCR to amplify the 4165 bp region from intron 8 to exon 10 using the following primer pair: forward primer 3.2. Genotypes of the Patients with Angioid Streaks. Among ABCC6-LR3F in intron 8 and reverse primer ABCC6-LR2R the 18 families, ABCC6 variants were identified in compound in exon 10, which is considered to be a unique exon of the heterozygous or homozygous states in 13 probands and in ABCC6 gene. All PCR reagents, except for the primers, were heterozygous states in four probands, while no variants were supplied by Takara-Bio (Shiga, Japan). The PCR amplifica- identified in one proband (Table 1). Among the 13 families tions were performed in a DNA Thermal Cycler (PTC-200, with disease-causing ABCC6 variants, results were consistent MJ Research, Waltham, MA, USA). The PCR products were with autosomal recessive inheritance in 11 families, while two ora fOphthalmology of Journal Table 1: Clinical profiles and ABCC6 variants of the patients with angioid streaks.

Case Age at Amino acid Family number Gender AS CNV PXE Consanguinity Genotype Exon Nucleotide changes Notes number examination changes 19 c.2542delG p.V848CfsX83 1 (JU#0376) 1 F 70 + + + − Homo 19 c.2542delG p.V848CfsX83 2 (JU#0390) 2 F 49 + − + − Hetero 9 c.1132C>T p.Q378X Compound 9 c.1132C>T p.Q378X 3 M 46 + + + − Father of case 4 hetero 19 c.2542delG p.V848CfsX83 3 (JU#0446) Compound 9 c.1132C>T p.Q378X 4 F 16 + − + − Daughter of case 3 hetero 10 c.1256G>A p.R419Q Compound 19 c.2542delG p.V848CfsX83 4 (JU#0448) 5 M 57 + − + − hetero 29 c.4069C>T p.R1357W Compound 9 c.1132C>T p.Q378X Described in 5 (JU#0451) 6 F 45 + + + − hetero 29 c.4069C>T p.R1357W Figure 1 9 c.1132C>T p.Q378X 6 (JU#0458) 7 F 60 + + + − Homo 9 c.1132C>T p.Q378X 10, c.1283A>G, Compound p.N428S, p.H647Y 7 (JU#0481) 8 M 44 + −− − 15 c.1939C>T hetero 19 c.2542delG p.V848CfsX83 8 (JU#0483) 9 M 51 + − ND −−−− − 9 (JU#0489) 10 F 23 + − + − Hetero 24 c.3340C>T p.R1114C Compound 19 c.2542delG p.V848CfsX83 11 M 71 + + + − Father of case 12 hetero 24 c.3374C>T p.S1125F 10 (JU#0554) 19 c.2542delG p.V848CfsX83 12 M 36 + − + − Homo Son of case 11 19 c.2542delG p.V848CfsX83 Compound 9 c.1132C>T p.Q378X 11 (JU#0602) 13 M 73 + + + − hetero 19 c.2542delG p.V848CfsX83 Compound 19 c.2542delG p.V848CfsX83 12 (JU#0476) 14 F 49 + −− − hetero 29 c.4069C>T p.R1357W Compound 19 c.2542delG p.V848CfsX83 13 (JU#MD196) 15 M 77 + + ND − hetero 29 c.4069C>T p.R1357W 14 (JU#0627) 16 F 81 + − ND − Hetero 19 c.2542delG p.V848CfsX83 Compound 10 c.1256G>A p.R419Q 15 (JU#0630) 17 M 28 + − + − hetero 29 c.4069C>T p.R1357W Compound 19 c.2542delG p.V848CfsX83 16 (JU#0639) 18 M 57 + − + − hetero 29 c.4069C>T p.R1357W Compound 10 c.1256G>A p.R419Q 17 (JU#0651) 19 F 51 + + + − hetero 19 c.2542delG p.V848CfsX83 18 (JU#0653) 20 M 77 + ND + − Hetero 10 c.1256G>A p.R419Q

F = female; M = male; AS = angioid streaks; CNV = choroidal neovascularization; PXE = pseudoxanthoma elasticum; ND = not determined; homo = homozygous; hetero = heterozygous. 3 4 Journal of Ophthalmology

(a) (b)

(c)

Figure 1: Fundus and fluorescein angiography images of the right eye. Images from a 45-year-old female proband (case 6, JU#0451). The color image (a) shows irregular streaks radiating from the optic disc and a small retinal hemorrhage in the macula. A peau d’orange appearance is also seen temporal to the macula. Fluorescein leakage is observed from the early (b) to the late (c) phases due to choroidal neovascularization. families showed a pseudodominant inheritance pattern. with AS, and six pathogenic variants (p.R419Q, p.E422K, Figure 2 shows the pedigrees of the three families (families p.V848CfsX83, a deletion of exon 23, c.3774_3775insC, and 3, 7, and 10) with the pseudodominant inheritance pattern p.E1427K) were found [13]. Among those 54 patients, 17 and/or novel ABCC6 variants (p.H647Y and p.S1125F). patients (32%) carried homozygous or compound heterozy- Twenty patients with AS in the current study had overlap- gous variants, 17 patients (32%) carried heterozygous ping ABCC6 variants, and some of these variants were found variants, and 20 patients (36%) did not carry any of the at a high frequency, including p.V848CfsX83 in 14 of 40 variants [13]. Notably, only two variants (p.R419Q and alleles (35.0%), p.Q378X in 7 of 40 alleles (17.5%), p.V848CfsX83) overlapped with our case series study. Our p.R1357W in 6 of 40 alleles (15.0%), and p.R419Q in 4 of results showed high frequencies for some variants, including 40 alleles (10.0%). 14/40 alleles (35.0%) for p.V848CfsX83, 7/40 alleles (17.5%) for p.Q378X, 6/40 alleles (15.0%) for p.R1357W, and 4/40 4. Discussion alleles (10.0%) for p.R419Q (Table 2); of note, high frequencies for p.Q378X and p.R1357W were not found in the previ- In the current study, we identified eight disease-causing ous study [13]. Recently, Iwanaga et al. [14] reported an variants (p.Q378X, p.R419Q, p.V848CfsX83, p.R1114C, ABCC6 mutation analysis study in 76 Japanese patients with p.R1357W, p.N428S, p.H647Y, and p.S1125F) in the ABCC6 PXE revealing that 56 (80%) of 76 patients had eye complica- gene, two of which (p.H647Y and p.S1125F) were novel tions that were likely attributed to AS, while our study variants. Thirteen of 18 AS probands (72.2%) exhibited showed that PXE was seen in 15 of 17 AS patients (88.2%) homozygous or compound heterozygous states for the who underwent dermatological examination. The patho- ABCC6 variants, supporting the idea that AS is an autosomal genic variant spectrum revealed five frequent variants: recessive disorder. p.V848CfsX83 (34/152 alleles, 22.4%), p.Q378X (30/152 Only one large-scale ABCC6 gene analysis study has been alleles, 19.7%), the deletion of exons 2 and 4 (15/152 alleles, previously reported; it was performed in 54 Japanese patients 9.9%), p.Q199X (11/152 alleles, 7.2%), and p.R419Q (9/152 ora fOphthalmology of Journal

Table 2: Variants in the ABCC6 gene identiﬁed in the 20 patients with angioid streaks.

HGVD ExAC East ExAC total Polyphen-2 Total SIFT PROVEAN Exon Nucleotide Amino acid Homozygous Heterozygous dbSNP ID (allele Asian (allele (allele results Pathogenicity References (alleles) results results frequency) frequency) frequency) (HumVar) Le Saux et al. 0.002% 0.000% 0.000% [2001], 9 c.1132C>T p.Q378X 1 5 7/40 rs72650699 Pathogenic (2/427) (0/8646) (1/121188) Pulkkinen et al. [2001] 0.742 0.001% 0.001% 0.000% 0.00 −3.27 Iwanaga et al. c.1256G>A p.R419Q 0 4 4/40 rs772434460 (possibly Pathogenic (2/1077) (10/8204) (10/111520) (damaging) (deleterious) [2017] damaging) 10 0.999 Sato et al. [2009] 0.009% 0.001% 0.000% 0.00 −4.69 c.1283A>G p.N428S 0 1 1/40 rs201880691 (probably Pathogenic (reported in the (20/1123) (12/8524) (13/117386) (damaging) (deleterious) damaging) controls) Not Not Not 0.127 0.01 −3.32 N/A (novel 15 c.1939C>T p.H647Y 0 1 1/40 Not reported Pathogenic reported reported reported (benign) (damaging) (deleterious) variant) 0.003% 0.000% 19 c.2542delG p.V848CfsX83 2 10 14/40 rs67867306 Inconclusive Pathogenic Sato et al. [2009] (27/8654) (27/121324) 0.998 Not 0.000% 0.000% 0.00 −7.13 Gheduzzi et al. c.3340C>T p.R1114C 0 1 1/40 rs63749794 (probably Pathogenic reported (0/8608) (12/120184) (damaging) (deleterious) [2004] damaging) 24 0.992 Not Not Not 0.00 −4.85 N/A (novel c.3374C>T p.S1125F 0 1 1/40 Not reported (probably Pathogenic reported reported reported (damaging) (deleterious) variant) damaging) 1.000 0.006% 0.001% 0.000% 0.00 −6.56 Miksch et al. 29 c.4069C>T p.R1357W 0 6 6/40 rs63750428 (probably Pathogenic (14/1099) (5/7944) (6/107190) (damaging) (deleterious) [2005] damaging) N/A = not applicable; HGVD = Human Genetic Variation database (http://www.hgvd.genome.med.kyoto-u.ac.jp/index.html); ExAC = Exome Aggregation Consortium database (http://exac.broadinstitute.org); Polyphen-2 (http://genetics.bwh.harvard.edu/pph2/); SIFT (http://sift.jcvi.org); PROVEAN (http://provean.jcvi.org). 5 6 Journal of Ophthalmology

Family 3 (JU#0446) Family 7 (JU#0481) Family 10 (JU#0554)

1 22211 I II

Q378X/V848CfsX83 R419Q/+ N428S, H647Y/+ S1125F/V848CfsX83 V848CfsX83/+

1 2 1 2 1 II II II

Q378X/R419Q Q378X/+ N428S, H647Y/V848CfsX83 N428S, H647Y/+ V848CfsX83/V848CfsX83

Figure 2: Pedigrees of three Japanese families with angioid streaks (AS). Two novel ABCC6 variants (p.H647Y and p.S1125F) are found in families 7 and 10. Families 3 and 10 exhibit a pseudodominant inheritance pattern. The identified ABCC6 variants cosegregated with the patients with AS. alleles, 5.9%) [14]. Of note, three variants (p.V848CfsX83, patients may have had a deletion variant, for example, the p.Q378X, and p.R419Q) were also frequently seen in our deletion of exon 23 [13] or the deletion of exons 2 and 4 study. The ABCC6 variants were identified to be in homozy- [14], which cannot be detected using our Sanger-based gous or compound heterozygous states in 13 of 18 families methodology. A large cohort study with comprehensive (72.2%) (Table 1). Two families (families 3 and 10) of the clinical and genetic examinations will be necessary to clarify 18 families exhibited a pseudodominant inheritance pattern, the genotype-phenotype correlations in patients with AS. and all four patients (cases 3, 4, 11, and 12) carried the In conclusion, our results suggest that ABCC6 variants p.V848CfsX83 variant (homozygously or compound hetero- play a significant role in patients with AS in the Japanese zygously) that was detected the most frequent (Figure 2). population and that four variants (p.Q378X, p.V848CfsX83, Taken together, these findings indicate that the ABCC6 p.R1357W, and p.R419Q) are highly correlated with AS. variants are likely a major cause of not only PXE but also AS in the Japanese population. Conflicts of Interest Our study revealed four frequent ABCC6 variants (p.Q378X, p.V848CfsX83, p.R1357W, and p.R419Q) occu- The authors declare that there are no conflicts of interest. pying 77.5% (31/40 alleles) of the total alleles. In contrast, two common recurrent variants (p.R1141X and del23_29) [7, 15] accounted for up to 40% of all variants in the Acknowledgments European population [16, 17], but they were not found in The authors thank the patients and their family members for the Chinese and Japanese populations [10, 13, 14], including participating in this study. This study was supported by JSPS in our study. These findings suggest that the ABCC6 variant KAKENHI Grant nos. 19592042 (Takaaki Hayashi) and spectrum differs between Asian and European populations. 17K11434 (Takaaki Hayashi). Two novel ABCC6 variants (p.H647Y and p.S1125F) were identified in the current study. The p.H647Y and p.N428S variants were located on the same allele (Tables 1 References and 2) in family 7 (JU#0481). The pathogenicity of the p.N428S variant had not been determined in a previous study [1] J. G. Clarkson and R. D. Altman, “Angioid streaks,” Survey of – because the variant was identified only in the control eyes Ophthalmology, vol. 26, no. 5, pp. 235 246, 1982. [12]. In the present study, the in silico programs predicted [2] R. O. Scholz, “Angioid streaks,” Archives of Ophthalmology, – that both variants (p.N428S and p.H647Y) cause severe vol. 26, no. 4, pp. 677 695, 1941. “ damage to protein function. We concluded that both the [3] A. A. Bergen, A. S. Plomp, E. J. Schuurman et al., Mutations in ” p.N428S and p.H647Y variants on the same allele were vari- ABCC6 cause pseudoxanthoma elasticum, Nature Genetics, fi vol. 25, no. 2, pp. 228–231, 2000. ants of signi cance. The other novel variant, p.S1125F, was “ predicted by the in silico programs to cause severe damage [4] O. L. Saux, Z. Urban, C. Tschuch et al., Mutations in a gene encoding an ABC transporter cause pseudoxanthoma elasti- to protein function, and case 11 (I-2 in family 10) carried cum,” Nature Genetics, vol. 25, no. 2, pp. 223–227, 2000. the p.S1125F and p.V848CfsX83 variants and exhibited AS [5] F. Ringpfeil, M. G. Lebwohl, A. M. Christiano, and J. Uitto, with PXE (Table 1), indicating the pathogenicity of p.S1125F. “Pseudoxanthoma elasticum: mutations in the MRP6 gene The current study had several limitations, including case encoding a transmembrane ATP-binding cassette (ABC) selection bias (as a single-center cohort), the small number of transporter,” Proceedings of the National Academy of Sciences n =20 cases ( ), and our genetic analysis method in which of the United States of America, vol. 97, no. 11, pp. 6001– deletion variants (such as deletion of exon 23 [13]) cannot 6006, 2000. be detected. In this study, four patients (cases 2, 10, 16, and [6] L. Cai, A. Lumsden, U. P. Guenther et al., “A novel Q378X 20) carried heterozygous variants, and one patient (case 9) mutation exists in the transmembrane transporter protein did not carry any of the variants (Table 2); however, these five ABCC6 and its pseudogene: implications for mutation analysis Journal of Ophthalmology 7

in pseudoxanthoma elasticum,” Journal of molecular medicine (Berlin, Germany), vol. 79, no. 9, pp. 536–546, 2001. [7] X. Hu, R. Peek, A. Plomp et al., “Analysis of the frequent R1141X mutation in the ABCC6 gene in pseudoxanthoma elasticum,” Investigative Ophthalmology & Visual Science, vol. 44, no. 5, pp. 1824–1829, 2003. [8] N. Chassaing, L. Martin, P. Calvas, M. L. Bert, and A. Hovnanian, “Pseudoxanthoma elasticum: a clinical, patho- physiological and genetic update including 11 novel ABCC6 mutations,” Journal of Medical Genetics, vol. 42, no. 12, pp. 881–892, 2005. [9] S. Miksch, A. Lumsden, U. P. Guenther et al., “Molecular genetics of pseudoxanthoma elasticum: type and frequency of mutations in ABCC6,” Human Mutation, vol. 26, no. 3, pp. 235–248, 2005. [10] L. Jin, Q. Jiang, Z. Wu et al., “Genetic heterogeneity of pseudoxanthoma elasticum: the Chinese signature profile of ABCC6 and ENPP1 mutations,” The Journal of Investigative Dermatology, vol. 135, no. 5, pp. 1294–1302, 2015. [11] Y. Noji, A. Inazu, T. Higashikata et al., “Identification of two novel missense mutations (p.R1221C and p.R1357W) in the ABCC6 (MRP6) gene in a Japanese patient with pseudoxanthoma elasticum (PXE),” Internal medicine (Tokyo, Japan), vol. 43, no. 12, pp. 1171–1176, 2004. [12] S. Yoshida, M. Honda, A. Yoshida et al., “Novel mutation in ABCC6 gene in a Japanese pedigree with pseudoxanthoma elasticum and retinitis pigmentosa,” Eye (London, England), vol. 19, no. 2, pp. 215–217, 2005. [13] N. Sato, T. Nakayama, Y. Mizutani, and M. Yuzawa, “Novel mutations of ABCC6 gene in Japanese patients with angioid streaks,” Biochemical and Biophysical Research Communica- tions, vol. 380, no. 3, pp. 548–553, 2009. [14] A. Iwanaga, Y. Okubo, M. Yozaki et al., “Analysis of clinical symptoms and ABCC6 mutations in 76 Japanese patients with pseudoxanthoma elasticum,” The Journal of Dermatology, vol. 44, no. 6, pp. 644–650, 2017. [15] O. L. Saux, K. Beck, C. Sachsinger et al., “A spectrum of ABCC6 mutations is responsible for pseudoxanthoma elasticum,” American Journal of Human Genetics, vol. 69, no. 4, pp. 749– 764, 2001. [16] E. G. Pfendner, O. M. Vanakker, S. F. Terry et al., “Mutation detection in the ABCC6 gene and genotype-phenotype analysis in a large international case series affected by pseudoxanthoma elasticum,” Journal of Medical Genetics, vol. 44, no. 10, pp. 621–628, 2007. [17] J. Uitto, A. Varadi, L. Bercovitch, P. F. Terry, and S. F. Terry, “Pseudoxanthoma elasticum: progress in research toward treatment: summary of the 2012 PXE international research meeting,” The Journal of Investigative Dermatology, vol. 133, no. 6, pp. 1444–1449, 2013. Hindawi Journal of Ophthalmology Volume 2017, Article ID 4156386, 10 pages https://doi.org/10.1155/2017/4156386

Research Article PRPF3-Associated Autosomal Dominant Retinitis Pigmentosa and CYP4V2-Associated Bietti’s Crystalline Corneoretinal Dystrophy Coexist in a Multigenerational Chinese Family

1,2 1,2 3 1,2 1,2 1,2 Xiaohong Meng, Qiyou Li, Hong Guo, Haiwei Xu, Shiying Li, and Zhengqin Yin

1Southwest Hospital and Southwest Eye Hospital, Third Military Medical University, Chongqing 400038, China 2Key Lab of Visual Damage and Regeneration & Restoration of Chongqing, Chongqing 400038, China 3Department of Medical Genetics, Third Military Medical University, Chongqing 400038, China

Correspondence should be addressed to Shiying Li; [email protected] and Zhengqin Yin; [email protected]

Received 16 May 2017; Accepted 2 July 2017; Published 7 August 2017

Academic Editor: Mineo Kondo

Copyright © 2017 Xiaohong Meng et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Purpose. To characterize the clinical and molecular genetic characteristics of a large, multigenerational Chinese family showing different phenotypes. Methods. A pedigree consisted of 56 individuals in 5 generations was recruited. Comprehensive ophthalmic examinations were performed in 16 family members affected. Mutation screening of CYP4V2 was performed by Sanger sequencing. Next-generation sequencing (NGS) was performed to capture and sequence all exons of 47 known retinal dystrophy-associated genes in two affected family members who had no mutations in CYP4V2. The detected variants in NGS were validated by Sanger sequencing in the family members. Results. Two compound heterozygous CYP4V2 mutations (c.802- 8_810del17insGC and c.992A>C) were detected in the proband who presented typical clinical features of BCD. One missense mutation (c.1482C>T, p.T494M) in the PRPF3 gene was detected in 9 out of 22 affected family members who manifested classical clinical features of RP. Conclusions. Our results showed that two compound heterozygous CYP4V2 mutations caused BCD, and one missense mutation in PRPF3 was responsible for adRP in this large family. This study suggests that accurate phenotypic diagnosis, molecular diagnosis, and genetic counseling are necessary for patients with hereditary retinal degeneration in some large mutigenerational family.

1. Introduction last updated November 16, 2016), of which ﬁve genes have been reported in Chinese adRP patients [3–7]. Retinitis pigmentosa (RP) (MIM 268000) is the most com- Bietti’s crystalline corneoretinal dystrophy (BCD) (MIM mon form of hereditary retinal degeneration (HRD), with a 210370) is an autosomal recessive retinal dystrophy that is worldwide prevalence of 1 in 4000 [1]. The disease can be characterized by numerous tiny glistening yellow-white inherited in an autosomal recessive (AR), autosomal domi- crystals that are scattered at the posterior pole of the ret- nant (AD), or X-linked manner [2]. Autosomal dominant ina, progressive atrophy of the retinal pigment epithelium RP (adRP) is the most common form of RP and typically (RPE), and choroidal sclerosis. Patients with BCD are usu- begins with night blindness in the early teens, followed ally present in the 2nd or 3rd decade of life and progress by progressive loss in the peripheral visual ﬁeld, subse- to legal blindness by the 5th or 6th decade [8]. Mutations in quent loss of vision, and eventually legal blindness. To the CYP4V2 gene (MIM 608614) are associated with BCD date, mutations in 22 genes have been associated with adRP [9]. BCD is relatively common in the East Asian populations, (RetNet: http://www.sph.uth.tmc.edu/retnet/sum-dis.htm, especially in Chinese and Japanese populations [9–18]. 2 Journal of Ophthalmology

I 12 II 12 34 III 12 3 45 6 78 910

IV 12 3 4 5 6 7 8 9 10 11 12 13 1415 1617 18 19 20 21 2223 24 25

V 12 3 456789 10 1112 13 14 1516 171819 20 2122 23 2425 2627 28 29 30 31

VI 12 3456789 10111213 1415161718 19 20 VII 12 34

Figure 1: Pedigree plot. The proband is indicated by an arrow. One affected patient with red solid box showed clinical findings compatible to the diagnosis of Bietti’s crystalline corneoretinal dystrophy (VI:1), and the other affected members with black solid box presented clinical features of retinitis pigmentosa. Males and females are represented by squares and circles, respectively. Filled symbols: affected members; open symbols: unaffected members.

BCD and RP are considered as two different types of twenty-six unaffected family members (Table 1) underwent retinal dystrophies with distinct clinical courses and features examination, including best-corrected visual acuity testing during its early stage. However, the fundus features at the with the Snellen vision chart, fundoscopy, slit-lamp biomi- later stage of BCD are occasionally similar to a severe form croscopy, spectral domain optical coherence tomography of RP. The interaction or coexistence of the two clinical phe- (SD-OCT, Spectralis OCT, Version 6.0; Heidelberg Engi- notypes thus requires further elucidation. In the present neering, Germany), full-field electroretinogram (FERG), study, we distinguish the inheritance patterns, clinical phe- and multifocal electroretinogram (mfERG). For the ages 6 notype, and molecular genetic characteristics of the patients months, 1, 2, 3, and 4 years old, the visual acuity was assessed in a large, multigeneration Chinese family with RP and BCD. using Teller acuity cards and then converted into Snellen vision chart. 2. Methods 2.2. Mutation Screening. Genomic DNA was extracted from 2.1. Pedigree. A pedigree consisted of 56 individuals in 5 peripheral blood samples of 39 family members (Table 1) generations was recruited. The Ethics Review Board of the using a QIAamp DNA Blood Midi Kit (Qiagen, Hilden, Ger- Southwest Hospital (Chongqing, China) approved the many) following the manufacturer’s standard procedure. All research protocol (number 2012-11), which adhered to the coding exons and intron-exon boundaries of the CYP4V2 tenets of the Declaration of Helsinki, and informed consent gene were amplified by polymerase chain reaction (PCR) was obtained from all participants. using primers described by Li et al. [9]. The PCR products The proband (Figure 1, VI:1) was initially presented to were subsequently purified with a TIANgen Mini Purifica- our medical institution for genetic counseling based on the tion Kit (Tiangen Biotech Co. Ltd., Shanghai, China) and observation that most of the family members developed night sequenced by Sanger sequencing with an ABI BigDye Termi- blindness and visual loss and even complete blindness, nator Cycle Sequencing Kit v3.1 (Applied Biosystems (ABI), resulting in an inability to work. This five-generation family Foster City, CA). CYP4V2 sequencing was performed in from Southwest of China was assessed in terms of RP and eight patients (III:9, IV:1, IV:6, IV:12, V:1, V:12, VI:1, and BCD. Thirty-nine participants were ascertained at the VI:2), and the detected mutation was further screened in 12 Southwest Eye Hospital, Southwest Hospital, Chongqing, affected family members and 11 unaffected members. China (Figure 1). No consanguineous marriage in the family Next-generation sequencing (NGS) was then applied to was declared. two affected family members with RP (III:9 and IV:1), who The proband presented clinical features that were did not have CYP4V2 mutations, then to identify disease- compatible with a diagnosis of BCD (VI:1), and the family causing variants in 47 RP-related genes including the members were subsequently evaluated. Twenty-two living PRPF31, CRB1, PRPF8, CA4, TULP1, PRPF3, ABCA4, individuals in the family had the clinical features of RP and RPE65, EYS, CERKL, NRL, FAM161A, FSCN2, TOPORS, presented similar symptoms of night blindness and progres- SNRNP200, SEMA4A, PRCD, NR2E3, MERTK, USH2A, sive reduction in their field of vision. The RP phenotype PDE6B, PROM1, KLHL7, PDE6A, RGR, CNGB1, IDH3B, followed an autosomal dominant pattern of inheritance in SAG, GUCA1B, CNGA1, BEST1, TTC8, C2orf71, ARL6, this pedigree (Figure 1). IMPG2, PDE6G, ZNF513, DHDDS, PRPF6, CLRN1, MAK, Thirteen affected individuals (III:9, IV:1, IV:3, IV:5, IV:6, CDHR1, FLVCR1, RLBP1, SPATA7, AIPL1, and LRAT genes. IV:10, IV:12, IV:14, IV:20, V:3, V:9, V:12, and VI:1) and The detected variants in NGS were validated by Sanger ora fOphthalmology of Journal Table 1: Demographic and clinical features and genotypes of PRPF3 and CYP4V2 mutations in a family with adRP and arBCD.

BCVA PRPF3 mutation CYP4V2 mutations Number Gender Age First symptom Phenotype Eye complications Systemic diseases OD OS c.1481C>G c.802-8_810del17insGC c.992A>C III:8 Female 76 None Normal 20/250 20/250 Cataract ND ND ND III:9 Male 80 Night blindness RP LP LP Cataract MT Wild Wild III:10 Female 80 None Normal 20/50 20/50 Cataract ND ND ND IV:1 Female 82 Night blindness RP LP LP Cataract MT Wild Wild IV:3 Male 57 Night blindness RP LP LP Cataract ND Wild Wild IV:5 Male 57 Night blindness RP LP LP Cataract ND ND ND IV:6 Male 57 Night blindness RP LP LP Cataract MT Wild Wild IV:7 Female 54 None Normal 20/30 20/30 None Wild Wild Wild IV:10 Male 39 Night blindness RP 20/200 20/200 None MT Wild Wild IV:12 Male 35 Night blindness RP 20/400 LP None ND Wild Wild IV:14 Male 55 Night blindness RP 20/100 20/200 Cataract MT Wild Wild IV:15 Female 50 None Normal 20/35 20/35 None ND ND ND IV:16 Female 51 None Normal 20/35 20/20 None ND ND ND IV:18 Male 49 None Normal 20/20 20/20 None ND ND ND IV:20 Male 45 Night blindness RP 20/100 20/200 None MT Wild Wild V:1 Female 57 None Normal 20/20 20/25 None Breast cancer ND MT Wild V:2 Male 60 None Normal 20/30 20/30 Cataract Fatty liver disease ND Wild MT V:3 Female 55 Night blindness RP 20/10 20/200 None Breast cancer MT Wild Wild V:7 Male 45 None Normal 20/30 20/30 Myopia Wild Wild Wild V:8 Female 46 None Normal 20/18 20/18 None Wild Wild Wild V:9 Male 43 Night blindness RP 20/200 LP Myopia MT Wild Wild V:11 Male 24 None Normal 20/18 20/18 None Wild Wild Wild V:12 Male 22 Night blindness RP 20/100 20/100 Myopia MT Wild Wild V:13 Male 13 None Normal 20/30 20/20 None ND Wild Wild V:15 Female 8 None Normal 20/20 20/20 None ND Wild Wild V:17 Male 37 None Normal 20/25 20/25 None ND Wild Wild V:18 Female 35 None Normal 20/18 20/18 High myopia ND ND ND V:19 Female 35 None Normal 20/25 20/20 Myopia ND ND ND V:20 Male 33 None Normal 20/18 20/18 None Wild Wild Wild V:31 Male 15 None Normal 20/18 20/18 Myopia ND ND ND VI:1 Male 33 Decreased vision BCD 20/25 20/30 Myopia, uveitis Wild MT MT VI:2 Female 29 None Normal 20/40 20/50 High myopia ND ND ND VI:10 Female 22 None Normal 20/18 20/18 Myopia ND Wild Wild VI:12 Male 15 None Normal 20/20 20/20 None ND Wild Wild 3 4

Table 1: Continued.

BCVA PRPF3 mutation CYP4V2 mutations Number Gender Age First symptom Phenotype Eye complications Systemic diseases OD OS c.1481C>G c.802-8_810del17insGC c.992A>C VI:14 Male 13 None Normal 20/20 20/20 None ND Wild Wild VI:15 Male 10 None Normal 20/20 20/20 None ND Wild Wild VI:16 Male 9 None Normal 20/20 20/20 None ND Wild Wild VII:1 Male 5 None Normal 20/30 20/30 None ND MT Wild VII:2 Female 1 None Normal 20/30 20/200 Hyperopia ND Wild Wild MT: mutation; ND: not detected; LP: light perception; NLP: nonlight perception. ora fOphthalmology of Journal Journal of Ophthalmology 5

1000 100

800 80

600 60 Depth

400 40 Coverage (%)

200 20

0 0 1×1 1×21×31×41×51×61×71×8 1×91×10 1×11 1×12 1×13 1×14 1×15 1×16 Exon & intron

Mean depth Median depth Coverage (%) (a)

CCAAG A CCCACG AAGGT AG AA

PRPF3 (NM_004698), c.1481C > G (p.Thr494Met) (b)

Figure 2: The depth and coverage of next-generation sequencing of the PRPF3 gene and the chromatogram obtained by Sanger sequencing (patient III:9). (a) The rectangle shows the averaged sequencing depth and coverage of the family for all 16 exons of the PRPF3 gene as screened by next-generation sequencing. (b) Sanger sequencing detected a heterozygous mutation (c.1481C>T, p.Thr494Met) in PRPF3. sequencing and screened in other 6 affected and 7 unaffected congenital night blindness, and seven affected members individuals in the family. already presented legal blindness. The proband’s mother (V:1) and aunt (V:3) had breast cancer, and his father (V:2) had fatty liver disease. 3. Results 3.1. Clinical Features. The demographic and clinical features 3.2. Mutations in the CYP4V2 Gene. Two previously reported of the living affected members and mutation carriers are CYP4V2 mutations (c.802-8_810del17insGC and c.992A>C summarized in Table 1. The age of enrollment ranged from (p.H331P)) were detected in this family. The proband (VI:1) 1 to 82 years. The visual acuity ranged from 20/30 to nonlight was compound heterozygous for both mutations. The c.802- perception (NLP). Seven family members had refractive 8_810del17insGC mutation was maternally derived (V:1), errors, including myopia (ranging −0.75 to −8 diopters) whereas the c.992A>C mutation was paternally inherited and astigmatism, and twelve members presented with cata- (V:2). Two other family members (V:9 and VII:1) were ract. All affected individuals except for the proband had heterozygous for the c.802-8_810del17insGC mutation. 6 Journal of Ophthalmology

(a) (b)

(e) (f)

(g)

(h)

Figure 3: Continued. Journal of Ophthalmology 7

CCCGGAACC TTTTCCA T AA CCTT

CYP4V2 c.802-8_810del17insGC

GGC AATTT T GG G GGCCC A

CYP4V2 c.992A>C, p.H331P

(i)

Figure 3: Fundal images and chromatograms of the proband with a clinical diagnosis with Bietti’s crystalline corneoretinal dystrophy and harboring compound heterozygous mutations in the CYP4V2 gene (patient VI:1). Fundal photographs (a, b), autoﬂuorescence images (c, d), and near-infrared images (e, f) of both eyes are shown on the left, and chromatograms of two mutations are demonstrated on the right (g, h).

3.3. Mutations in the PRPF3 Gene. Targeted NGS of two pathogenic CYP4V2 mutations were detected in the appar- affected members (IV:1 and III:9) revealed one common mis- ently normal daughter (VII:2). Notably, no PRPF3 mutations sense mutation in the PRPF3 gene (c.1481C>T) (Figure 2), were detected in the proband. which was then screened by Sanger sequencing in 8 affected (III:9, IV:1, IV:3, IV:6, IV:14, V:9, V:12, and VI:1) and seven 3.4.2. Type 2. In addition to the proband, other family unaffected family members (IV:9, V:1,V:7, V:11, V:16, V:20, members affected with adRP presented with night blindness and VI:13) for cosegregation analysis. since birth. Best-corrected visual acuity was from 200/400 to NLP. Fundus examination showed severe features of RP, 3.4. Clinical and Molecular Manifestations of Affected Family with a mass of bone-spicule pigmentation depositions, more Members. Two types of clinical and molecular manifestations severe RPE atrophy involving the macular and choroidal were observed in this family: (i) a BCD phenotype that was sclerosis extending to the midperipheral retina, whereas related to the compound heterozygous CYP4V2 mutations partial attenuation of the retinal blood vessels, slight waxy and (ii) a RP phenotype that was associated with the PRPF3 pallor of the optic disc, was presented (Figure 4). FERG mutation and followed an autosomal dominant pattern demonstrated undetectable responses both in scotopic and of inheritance. photopic conditions and extinguished mfERG. One PRPF3 mutation, c.1481C>T (p.T494M), was 3.4.1. Type 1 (Proband VI:1). The proband was a 33-year-old detected in 13 family members, including 11 males and 2 man referred to us for genetic counseling based on a signifi- females. No novel mutation and previously reported muta- cant decrease in visual acuity starting at the age of 17 years. tions were detected in the other 45 genes in the panel. The The patient developed night blindness in his early 30s. He identified mutation (c.1481C>T) cosegregated with the RP had high myopia (−7.00 D) in both eyes, and best-corrected phenotype in 11 affected family members tested and was Snellen visual acuity was 20/30 in his both eyes. There was not observed in 9 unaffected family members (Figure 1). This a history of chronic uveitis in his left eye since age 28. He mutation was observed across four generations. Taken was diagnosed with BCD based on clinical findings that together, the c.1481C>T mutation was considered to be the included numerous tiny glistening yellow-white crystals scat- main cause of adRP in this family. tered at the posterior pole of the retina, RPE atrophy (Figure 3), and decreased responses in FERGs and mfERGs. 4. Discussion Two previously reported disease-causing mutations in CYP4V2 (c.802-8_810del17insGC in exon 7 and c.992A>C PRPF3 (MIM 607301) is a precursor mRNA-processing (p.H331P) in exon 8) were identified in the proband [19]. factor gene that was first identified for adRP in 2002 [20]. The compound heterozygosity was confirmed by screening In the present study, a pathogenic mutation (c.1481C>T, his unaffected parents; his mother (V:1) carried the c.802- p.T494M) in the PRPF3 gene was identified in 11 individ- 8_810del17insGC variant, and his father (V:2) harbored the uals presenting an adRP phenotype in a five-generation c.992A>C mutation. The proband’s unaffected son (VII:1) Chinese family. The molecular genetic features of a Chinese had the c.802-8_810del17insGC mutation, whereas no pedigree with a PRPF3 mutation have been previously 8 Journal of Ophthalmology

(a) (b)

CAAG ACCCCAC G AAGGT AG AA

PRPF3 (NM_004698), c.1481C >G (p.Thr494Met) (e)

Figure 4: Fundal photographs and chromatogram of a patient with severe phenotype of retinitis pigmentosa and harboring PRPF3 mutations (patient IV:12). Fundal photographs of the right eye (a, b) and the left eye (c, d), and chromatogram of the PRPF3 mutation (c.1481C>T, p.T494M) is shown (e). reported. The c.1481C>T mutation is considered to be one study, two compound heterozygous mutations in CYP4V2 of the most common mutations in PRPF3 [20–25]. Previous (c.802-8_810del17insGC and c.992A>C) were identified in reports have shown that patients harboring the c.1481C>T the proband who presented typical BCD. In our previous mutation develop early-onset night blindness, visual field loss, study, CYP4V2 mutation screening among 92 Chinese and visual acuity loss between the ages of 30 and 40, as well as patients with BCD showed that c.802-8_810del17insGC loss of ERG responses after the age of 30. Compared to those in and c.992A>C are common pathogenic mutations in previously reported Japanese, Spanish, Korean, Swiss, and Chinese with BCD [26]. The parents of the proband are North American families, members of this Chinese family not a consanguineous marriage couple. So we speculate that with the c.1481C>T mutation presented a more severe disease these heterozygous mutations in Chinese population may be phenotype, which included congenital blindness, severe visual universal. This phenomenon may be related to the common acuity loss, extended RPE atrophy, and completely extin- ancestor based on the huge population of China. The hetero- guished ERG responses. zygous state of the same gene carried by parents is consistent Mutations in the CYP4V2 gene (MIM 608614) are the with the autosomal recessive inheritance pattern. This will only known causative factor for BCD to date. The CYP4V2 be important for prenatal testing for family planning, early gene consists of 11 exons and encodes a 525 amino acid pro- finding carrier status, and determining risk of inheritance tein belonging to the CYP450 family. CYP4V2 is widely in Chinese. expressed in tissues, including the retina, RPE, lymphocytes, Coexistence of variants in two or three genes associated heart, brain, placenta, lung, liver, skeletal muscle, kidney, with retinal degeneration has rarely been reported in a family and pancreas, which has been thought to play a crucial role [3]. In the present study, we identified the coexistence of two in fatty acid and corticosteroid metabolism. In the present distinct phenotypes in one family, namely, BCD and RP, Journal of Ophthalmology 9 which were caused by the pathogenic variants in the CYP4V2 [6] H. Guo, Y. Qin, Q. Meng, H. Zhang, H. Jin, and Y. Chen, and PRPF3 genes, respectively. The mode of inheritance of “Linkage analysis and mutation screening of the rhodopsin the two diseases was maintained in this family, in which gene in a Chinese Bai family with autosomal dominant retinitis BCD demonstrated an autosomal recessive trait and RP pigmentosa,” Journal of Human Genetics, vol. 55, no. 9, – showed an autosomal dominant trait. pp. 571 576, 2010. Two types of clinical and molecular manifestations iden- [7] C. Zhao, D. L. Bellur, S. Lu et al., “Autosomal-dominant retini- tified in this study include (i) a BCD phenotype related to tis pigmentosa caused by a mutation in SNRNP200, a gene required for unwinding of U4/U6 snRNAs,” American Journal CYP4V2 mutations and (ii) an RP phenotype related to – PRPF3 variants. Clinical features for (i) BCD and (ii) RP of of Human Genetics, vol. 85, no. 5, pp. 617 627, 2009. [8] M. I. Kaiser-Kupfer, C. C. Chan, T. C. Markello et al., “Clinical this family were similar to those in previous reports. The ’ proband affected with BCD in this family had a later onset biochemical and pathologic correlations in Bietti s crystalline dystrophy,” American Journal of Ophthalmology, vol. 118, for night blindness and relatively slow progression, with a – ff no. 5, pp. 569 582, 1994. predominantly a ected choroid at the posterior pole. On “ the other hand, family members with RP caused by the [9] A. Li, X. Jiao, F. L. Munier et al., Bietti crystalline corneoretinal dystrophy is caused by mutations in the novel gene PRPF3 mutation showed a more severe phenotype. Our CYP4V2,” American Journal of Human Genetics, vol. 74, study provides a better understanding of the genotype- no. 5, pp. 817–826, 2004. phenotype correlation in a family with two independent [10] T. Gekka, T. Hayashi, T. Takeuchi, S. Goto-Omoto, and pathogenic gene mutations and may be used in clinics for “ ff K. Kitahara, CYP4V2 mutations in two Japanese patients the di erential diagnosis of retinal degenerations. with Bietti ’s crystalline dystrophy,” Ophthalmic Research, fi In summary, this is the rst report on PRPF3-associated vol. 37, no. 5, pp. 262–269, 2005. adRP and CYP4V2-associated arBCD in a large multigener- [11] K. Y. Lee, A. H. Koh, T. Aung et al., “Characterization of Bietti ational Chinese family. The inheritance pattern of each crystalline dystrophy patients with CYP4V2 mutations,” gene mutation is independent. Our study provides an Investigative Ophthalmology & Visual Science, vol. 46, no. 10, insight into the clinical effects of two independent gene pp. 3812–3816, 2005. mutations in a large family to facilitate accurate diagnosis [12] J. Lin, K. M. Nishiguchi, M. Nakamura, T. P. Dryja, E. L. Berson, and disease counseling. and Y. Miyake, “Recessive mutations in the CYP4V2 gene in East Asian and Middle Eastern patients with Bietti crystalline Conflicts of Interest corneoretinal dystrophy,” Journal of Medical Genetics, vol. 42, no. 6, article e38, 2005. The authors declare that they have no conflicts of interest. [13] M. Shan, B. Dong, X. Zhao et al., “Novel mutations in the CYP4V2 gene associated with Bietti crystalline corneoretinal Acknowledgments dystrophy,” Molecular Vision, vol. 11, pp. 738–743, 2005. [14] Y. Wada, T. Itabashi, H. Sato, M. Kawamura, A. Tada, and M. The authors are grateful to Chen Sun, Mingfang Zhang, Gang Tamai, “Screening for mutations in CYP4V2 gene in Japanese Wang, and Min Wang for their great support. This work was patients with Bietti’s crystalline corneoretinal dystrophy,” supported by the National Basic Research Program of China American Journal of Ophthalmology, vol. 139, no. 5, pp. 894– (973 Program, 2013CB967002 and 2013CB967003) and the 899, 2005. National Natural Science Foundation of China (81200710). [15] Z. B. Jin, S. Ito, Y. Saito, Y. Inoue, Y. Yanagi, and N. Nao-i, “Clinical and molecular findings in three Japanese patients with crystalline retinopathy,” Japanese Journal of Ophthal- References mology, vol. 50, no. 5, pp. 426–431, 2006. “ [1] J. A. Boughman, P. M. Conneally, and W. E. Nance, “Popula- [16] T. Y. Lai, T. K. Ng, P. O. Tam et al., Genotype phenotype ’ tion genetic studies of retinitis pigmentosa,” American Journal analysis of Bietti s crystalline dystrophy in patients with ” of Human Genetics, vol. 32, no. 2, pp. 223–235, 1980. CYP4V2 mutations, Investigative Ophthalmology & Visual – [2] C. Ayuso and J. M. Millan, “Retinitis pigmentosa and allied Science, vol. 48, no. 11, pp. 5212 5220, 2007. conditions today: a paradigm of translational research,” [17] J. C. Zenteno, R. Ayala-Ramirez, and F. Graue-Wiechers, “ Genome Medicine, vol. 2, no. 5, p. 34, 2010. Novel CYP4V2 gene mutation in a Mexican patient with Bietti’s crystalline corneoretinal dystrophy,” Current Eye [3] K. P. Lim, S. P. Yip, S. C. Cheung, K. W. Leung, S. T. Lam, and – To CH, “Novel PRPF31 and PRPH2 mutations and co- Research, vol. 33, no. 4, pp. 313 318, 2008. occurrence of PRPF31 and RHO mutations in Chinese [18] D. N. Liu, Y. Liu, X. H. Meng, and Z. Q. Yin, “The characteriza- patients with retinitis pigmentosa,” Archives of Ophthalmol- tion of functional disturbances in Chinese patients with Bietti’s ogy, vol. 127, no. 6, pp. 784–790, 2009. crystalline dystrophy at different fundus stages,” Graefe’s [4] B. V. V. E. Alvarez, Z. Yang, A. H. Koh et al., “Identification and Archive for Clinical and Experimental Ophthalmology, – characterization of a novel mutation in the carbonic anhydrase vol. 250, no. 2, pp. 191 200, 2012. IV genethatcauses retinitis pigmentosa,”Investigative Ophthal- [19] G. Giuffre, “Progression of Bietti’s crystalline dystrophy,” Jour- mology & Visual Science, vol. 48, no. 8, pp. 3459–3468, 2007. nal Francais d’Ophtalmologie, vol. 14, no. 4, pp. 249–254, 1991. [5] L. Wang, M. Ribaudo, K. Zhao et al., “Novel deletion in the pre- [20] C. F. Chakarova, M. M. Hims, H. Bolz et al., “Mutations in mRNA splicing gene PRPF31 causes autosomal dominant reti- HPRP3, a third member of pre-mRNA splicing factor genes, nitis pigmentosa in a large Chinese family,” American Journal of implicated in autosomal dominant retinitis pigmentosa,” Medical Genetics Part A, vol. 121A, no. 3, pp. 235–239, 2003. Human Molecular Genetics, vol. 11, no. 1, pp. 87–92, 2002. 10 Journal of Ophthalmology

[21] V. Vaclavik, M. C. Gaillard, L. Tiab, D. F. Schorderet, and F. L. Munier, “Variable phenotypic expressivity in a Swiss family with autosomal dominant retinitis pigmentosa due to a T494M mutation in the PRPF3 gene,” Molecular Vision, vol. 16, pp. 467–475, 2010. [22] Y. Wada, T. Itabashi, H. Sato, and M. Tamai, “Clinical features of a Japanese family with autosomal dominant retinitis pigmentosa associated with a Thr494Met mutation in the HPRP3 gene,” Graefe’s Archive for Clinical and Experimental, vol. 242, no. 11, pp. 956–961, 2004. [23] C. Kim, K. J. Kim, J. Bok et al., “Microarray-based mutation detection and phenotypic characterization in Korean patients with retinitis pigmentosa,” Molecular Vision, vol. 18, pp. 2398–2410, 2012. [24] M. Martínez-Gimeno, M. J. Gamundi, I. Hernan et al., “Muta- tions in the pre-mRNA splicing-factor genes PRPF3, PRPF8, and PRPF31 in Spanish families with autosomal dominant retinitis pigmentosa,” Investigative Ophthalmology & Visual Science, vol. 44, no. 5, pp. 2171–2177, 2003. [25] F. Blanco-Kelly, M. García-Hoyos, M. Cortón et al., “Genotyp- ing microarray: mutation screening in Spanish families with autosomal dominant retinitis pigmentosa,” Molecular Vision, vol. 18, pp. 1478–1483, 2012. [26] X. H. Meng, H. Guo, H. W. Xu et al., “Identiﬁcation of novel CYP4V2 gene mutations in 92 Chinese families with Bietti’s crystalline corneoretinal dystrophy,” Molecular Vision, vol. 20, pp. 1806–1814, 2014. Hindawi Journal of Ophthalmology Volume 2017, Article ID 3080245, 10 pages https://doi.org/10.1155/2017/3080245

Research Article Genotype-Phenotype Characterization of Novel Variants in Six Italian Patients with Familial Exudative Vitreoretinopathy

1 2 3 4 Giancarlo Iarossi, Matteo Bertelli, Paolo Enrico Maltese, Elena Gusson, 4 3 3 5 1 Giorgio Marchini, Alice Bruson, Sabrina Benedetti, Sabrina Volpetti, Gino Catena, 1 6 Luca Buzzonetti, and Lucia Ziccardi

1Department of Ophthalmology, Bambino Gesù IRCCS Children's Hospital, Rome, Italy 2MAGI-Human Medical Genetics Institute, Bolzano, Italy 3MAGI-Human Medical Genetics Institute, Rovereto, Italy 4Eye Clinic, Department of Neurosciences, Biomedicine and Movement, University and AOUI (Azienda Ospedaliera Universitaria Integrata) of Verona, Verona, Italy 5Dipartimento Anestesia e Rianimazione Materno Infantile, Ospedale San Filippo Neri, Rome, Italy 6“G.B. Bietti” Foundation, IRCCS, Rome, Italy

Correspondence should be addressed to Paolo Enrico Maltese; [email protected]

Received 3 April 2017; Accepted 4 June 2017; Published 5 July 2017

Academic Editor: Naheed Khan

Copyright © 2017 Giancarlo Iarossi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Familial exudative vitreoretinopathy (FEVR) is a complex disorder characterized by incomplete development of the retinal vasculature. Here, we report the results obtained on the spectrum of genetic variations and correlated phenotypes found in a cohort of Italian FEVR patients. Eight probands (age range 7–19 years) were assessed by genetic analysis and comprehensive age-appropriate ophthalmic examination. Genetic testing investigated the genes most widely associated in literature with FEVR: FZD4, LRP5, TSPAN12, and NDP. Clinical and genetic evaluations were extended to relatives of probands positive to genetic testing. Six out of eight probands (75%) showed a genetic variation probably related to the phenotype. We identified four novel genetic variants, one variant already described in association with Norrie disease and one previously described linked to autosomal dominant FEVR. Pedigree analysis of patients led to the classification of four autosomal dominant cases of FEVR (caused by FZD4 and TSPAN12 variants) and two X-linked FEVR probands (NDP variants). None of the patients showed variants in the LRP5 gene. This study represents the largest cohort study in Italian FEVR patients. Our findings are in agreement with the previous literature confirming that FEVR is a clinically and genetically heterogeneous retinal disorder, even when it manifests in the same family.

1. Introduction Insights into genetic causes of this disorder have emerged in the last few years. Four genes involved in the Norrin/ Familial exudative vitreoretinopathy (FEVR, OMIM 133780) β-catenin signaling pathway are linked to the disease: NDP, is a complex disorder characterized by incomplete develop- FDZ4, LRP5, and TSPAN12, and are responsible for about ment of the retinal vasculature. Typical signs of this disorder 50% of FEVR cases [3]. include avascular areas in the peripheral retina detectable by Norrin/Frizzled signaling is known to be critical for fluorescein angiography due to abnormal congenital angiogen- retinal angiogenesis by controlling retinal vascular growth esis that can cause early-onset neovascularization, falciform and architecture both in the developing eye and in adult vas- folds, lipid exudation, and tractional retinal detachment [1, 2]. culature; impairment of the signaling system has profound 2 Journal of Ophthalmology effects on retinal vascular organization and leads to severe spherical equivalent. All examinations were performed at vascular abnormalities [4, 5]. the referring clinic, and the results were collected at the end Depending on which gene is defective, FEVR may have of the study. different patterns of inheritance. Autosomal dominant inher- Based on clinical data, each patient was classified accord- itance, due to FZD4, LRP5, and TSPAN12 gene variations ing to the FEVR clinical staging system [16]. [6–8], is prevalent. Autosomal recessive and X-linked All patients received genetic counseling to explain the inheritances have also been reported due to variations in risks and benefits of genetic testing, and informed consent the LRP5 and NDP genes, respectively [9, 10]. Recessive was obtained from their parents. The study followed the variants in TSPAN12 have also been described in a family tenets of the Declaration of Helsinki. with retinal dysplasia and severe FEVR [11]. Demographic details and information on personal and Clinical features are extremely variable, ranging from family medical history were recorded and used to determine absence of symptoms to complete blindness, even in the same the inheritance pattern and to draw pedigrees according to family [12]. the recommendations of the National Society of Genetic This evidence has led researchers to suspect the Counselors [17] using the HaploPainter software [18]. involvement of other unknown genes [3, 8]. Two other genes, Genetic testing was performed at MAGI’s Laboratories ZNF408 encoding zinc finger protein 408 and KIF11 encod- (MAGI - Human Medical Genetics Institute, Rovereto and ing kinesin family member 11, are linked to FEVR, although Bolzano, Italy). Extracted DNA (Blood DNA Kit E.Z.N.A.; the evidence of their association needs to be clarified. When Omega Bio-Tek Inc., Norcross, GA, USA) underwent genetic testing was performed on the patients enclosed in polymerase chain reaction (PCR) to amplify all coding the present study, the above-mentioned novel genes had regions and the intron/exon junctions of the FZD4 not yet been discovered; therefore, we only focused on testing (OMIM ∗604579, NM_012193), LRP5 (OMIM ∗603506, for the NDP, FDZ4, LRP5, and TSPAN12 genes. NM_002335), TSPAN12 (OMIM ∗613138, NM_012338), In literature on Italian FEVR patients, only three genetic and NDP (OMIM ∗300658, NM_000266) genes. Purified association studies describing a total of six probands are products were sequenced with a Beckman Coulter CEQ reported [13–15]. Here, we present the results of clinical 8000 sequencer (Beckmann Coulter, Milano, Italy). All and genetic characterization of the largest Italian FEVR laboratory protocols are available on request. population, consisting of six probands and ten relatives, as The electropherograms of amplified fragments were a contribution to the FEVR disease molecular epidemiology analyzed using ChromasPro 1.5 (Technelysium Pty. Ltd., in our country. Australia) and Sequencher 5.0 (Gene Codes®; Ann Arbor, We identified four new variants, one variant already MI, USA) software and compared to GenBank reference described in association with Norrie disease and a variant sequences with the Basic Local Alignment Search Tool previously described in association with autosomal dominant (BLAST; http://blast.ncbi.nlm.nih.gov). All detected genetic FEVR. The genotype-phenotype correlation of each variant variations were sequenced bidirectionally at least twice was evaluated by a family segregation study, and new variants for confirmation. were further characterized for their putative pathogenic To identify variants previously reported as pathogenic, potential by in silico evaluation. the Human Gene Mutation Database (HGMD; http:// www.biobase-international.com/product/hgmd) was con- 2. Materials and Methods sulted. To evaluate Minor Allele Frequencies (MAF) in populations, all discovered genetic variants were also Probands (age ranging from 7 to 19 years) and their fam- looked up in the public database of single-nucleotide variants ilies were examined in two different eye clinics: the Eye (dbSNP; www.ncbi.nlm.nih.gov/SNP/) and the Exome Clinic, Department of Neurosciences, Biomedicine and Variant Server (EVS; http://evs.gs.washington.edu/EVS/) Movement, (University and Azienda Ospedaliera Univer- database. New nucleotide variations were assessed for patho- sitaria Integrata of Verona, Italy) and the Department of genicity using the PolyPhen 2 algorithm (Polymorphism Ophthalmology (Bambino Gesù IRCCS Children’s Hospital, Phenotyping v2; http://genetics.bwh.harvard.edu/pph2) Rome, Italy). considering the HumVar-trained model [19], the SIFT Probands and relatives underwent comprehensive algorithm (Sorting Intolerant From Tolerant; http://sift. age-appropriate ophthalmic examination, including best- bii.a-star.edu.sg/) [20] and MutationTaster (http://www. corrected visual acuity (BCVA) measurement with the Early mutationtaster.org) [21]. Defective splicing was evaluated Treatment Diabetic Retinopathy Study (ETDRS) charts, using the Human Splicing Finder online software version expressed as a logarithm of the minimum angle of resolution 3 (http://www.umd.be/HSF3/HSF.html) [22]. When possi- (logMAR), slit-lamp biomicroscopy, indirect ophthalmos- ble, wild-type amino acid properties were compared with copy with 15 D noncontact lens (Volk), fluorescein angiogra- the variations (http://www.russelllab.org/aas/aas.html) [23]. phy or color fundus photos (obtained by RetCam in younger Variants were reported following the current nomencla- patients requiring examination under anaesthesia and by ture as described by the Human Genome Variation Society Daytona wide-field retinography in compliant older sub- (http://www.hgvs.org/mutnomen/recs.html). jects), optical coherence tomography (OCT), and full-field To classify sequence variants, we used the criteria by electroretinogram (ERG) recorded according to the ISCEV the American College of Medical Genetics and Genomics standards. Refractive errors ranged between +2 and −2 (ACMG) Standards and Guidelines [24]. Journal of Ophthalmology 3

Table 1: Features of genetic variations found in FEVR families.

Family ID Genotype Nucleotide change Amino acid change SIFT Polyphen Mutation taster Classiﬁcation References Gene Fam. 1 FZD4 Het c.277C>T p.(Gln93*) —— — Pathogenic Novel variant NM_012193 Fam. 2 FZD4 Het c.542G>A p.(Cys181Tyr) T PrD DC Pathogenic [31] NM_012193 Fam. 3 FZD4 Het c.611G>T p.(Cys204Phe) D PrD DC Likely pathogenic Novel variant NM_012193 Fam. 4 NDP Hemi c.362G>A p.(Arg121Gln) D PrD DC Pathogenic [25] NM_000266 Fam. 5 NDP Hemi c.313G>C p.(Ala105Phe) D PrD DC Likely pathogenic Novel variant NM_000266 Fam. 6 TSPAN12 Het c.67-2A>G Defective splicing —— — Likely pathogenic Novel variant NM_012338

(i)

1 2 ⁎ E− E+ (FZD4 p.[(Gln93 )];[=])

⁎

(ii)

1 ⁎ E+ (FZD4 p.[(Gln93 )];[=])

Figure 1: Pedigree and ocular features of family 1. Color fundus photographs by RetCam and ﬂuorescein angiograms from the 9 y/o male proband (II:1) of family 1. The RE presented normal posterior pole; however, the ﬂuorescein angiography showed a large area of avascular retina in the peripheral retina (FEVR stage 1). No signs of neovascularization or peripheral exudation were noticed. The LE presented a falciform retinal fold anchored to the temporal retinal sector without signs of peripheral exudation (FEVR stage 3A). ∗Documented clinical evaluation; E+ and E−, positive and negative to genetic test, respectively; RE, right eye; LE, left eye.

3. Results (Hemi); SIFT score system: tolerated (T), deleterious (D); Polyphen 2 score system: benign (B), possibly damaging Genetic testing revealed that 6 out of 8 (75%) probands had a (PoD), probably damaging (PrD); Mutation Taster score genetic variation in one of the genes analyzed. Consequently, system: polymorphism (P), disease causing (DC); VUS: a segregation study was performed by extending clinical and variation of unknown significance. genetic analyses to a total of 10 relatives. Pedigree analysis of patients with inherited forms of the 3.1. Family 1. The male proband (II:1) came to our attention disease led to the classification of four autosomal dominant only recently at the age of nine years. cases of FEVR (FZD4 and TSPAN12 variations) and one Previously performed fluorescein angiography with X-linked FEVR proband (NDP variation) (see Table 1). RetCam showed a large avascular area in the peripheral None of the patients showed variations in the LRP5 gene. retina without signs of neovascularization or peripheral No MAF are reported for the identified variants in all the exudation in his right eye and a falciform retinal fold questioned database. anchored to the temporal retinal sector without signs of In silico evaluation of genetic variations identified in peripheral exudation in the left eye (Figure 1). The patient FEVR families; genotype heterozygous (Het), hemizygous was diagnosed with stage 1 and stage 3A FEVR in the right 4 Journal of Ophthalmology

⁎

(i) 12 E+ (FZD4 p.[(Cys181Tyr)];[=])

⁎

(ii) 1 2 E+ (FZD4 p.[(Cys181Tyr)];[=])

Figure 2: Pedigree and ocular features of family 2. Red-free images and fluorescein angiograms from the proband’s father (I:1) showing similar findings in both eyes consisting of slight abnormalities of the peripheral retinal vasculature as indicated by the arrows. The posterior pole and the mid periphery were normal. Daytona wide-field color fundus photos and OCT scan of the proband (II:1) at the age of 2 years showing the presence of a falciform fold and peripheral avascular retina in the RE. The wide-field angiography of the LE (early and late phases) revealed a normal posterior pole and a severe leakage from undiscovered peripheral neovascular networks that were treated promptly to avoid exudation and visual impairment. ∗Documented clinical evaluation; E+ and E−, positive and negative to genetic test, respectively; |asymptomatic/presymptomatic carrier (stage 1 FEVR); RE, right eye; LE, left eye. and left eyes, respectively. Clinical conditions are currently Genetic testing revealed that the proband and his father stable, and visual acuity is 0.0 LogMAR in the right eye and were heterozygous for a variation of the gene FZD4, 1.0 LogMAR in the left eye. It was not possible to perform a p.(Cys181Tyr); the mother was not assessed. Figure 2 shows complete clinical examination on the proband’s mother the autosomal dominant transmission of the variant (I:2) and father (I:1). The last fundus examination of the p.(Cys181Tyr) in the FZD4 gene and the clinical features of mother showed normal retinal features; however, we could the family. not proceed with any instrumental examinations due to the patient’s refusal to undergo further testing. 3.3. Family 3. The proband (II:2), a 19 y/o male, presented Genetic testing revealed that the proband and his mother disease onset at the age of 13 years, with bilateral exudative were heterozygous for the new p.(Gln93∗) variation in the retinopathy associated with slight visual acuity reduction. FZD4 gene. In detail, the patient showed exotropia of the right eye from early infancy and visual acuity of 0.1 LogMAR in both eyes. 3.2. Family 2. This male proband (II:1) presented exotropia At the age of 13 years, he suffered a sharp decline in visual of the right eye at the age of 2 years. Ophthalmological acuity (0.3 LogMAR in both eyes), associated with hard examination under general anaesthesia showed inoperable retinal exudates in the temporal periphery at the fundus falciform fold in the right eye and vascular alterations in examination. Based on the presence of posterior hyaloid con- the left eye. The patient was diagnosed with stage 3 and stage traction, macular edema, macular dragging, and accentuated 2 FEVR in the right and left eyes, respectively. At the age peripheral exudation in both eyes, bilateral stage 4A FEVR of 3 years, scleral buckle, cryo treatment, and laser treat- was postulated. A few months after diagnosis, the patient ment for retinal detachment were performed in the right underwent vitrectomy, cryo treatment, and injection of eye and laser treatment of the retinal periphery in the left humanized anti-VEGF monoclonal antibody (Bevacizumab) eye. After a period of clinical stationary conditions, at the in both eyes. age of 5 years, the patient required further laser treatment The patient was monitored every 6 months for the in the left eye due to reactivation of the disease. Visual following three years by visual acuity measurements, OCT acuity is currently 0.1 and 0.2 LogMAR in the right and scans, and fluorescein angiographies. Since the peripheral left eye, respectively. exudations remained active, with angiographic evidence The fundus examination of the father (I:1) showed slight of leakage, the patient underwent multiple sessions of abnormalities of the peripheral retina suggesting a condition Argon laser, cryo treatment, and Bevacizumab injections of FEVR healthy carrier. (Figure 3). At the time of this report, visual acuity was Journal of Ophthalmology 5

⁎⁎

(i) 1 2 E− E+(FZD4 p.[(Cys204Phe)];[=])

⁎⁎

(ii) 1 2 E− E+ (FZD4 p.[(Cys204Phe)];[=])

Figure 3: Pedigree and ocular features of family 3. Daytona wide-field color fundus photos and OCT scans of the proband (II:2) at the age of 19 years with bilateral exudative manifestation of FEVR stage 4A. Peripheral retinal signs of cryo treatment and Argon laser barrage, performed after evidence of fluorangiographic leakage, can be seen in the inferior temporal sectors in both eyes. Daytona wide-field color fundus photographs of the proband’s mother (I:2), showing treatment with Argon laser in the retinal periphery of both eyes. Angiograms from the posterior pole and from the midperiphery do not present signs of disease’s activity after laser treatment. ∗Documented clinical evaluation; E+ and E−, positive and negative to genetic test, respectively; RE, right eye; LE, left eye.

⁎ ⁎

(i) 1 2 I:1 I:2

⁎⁎

(ii) 1 2 E+ (NDP p.[(Arg121Gln)]) E+ (NDP p.[(Arg121Gln)])

Figure 4: Pedigree and ocular features of family 4. Fluorescein angiograms by RetCam from the 7 y/o male proband (II:2) of family 4, showing FEVR stage 2B with slight optic nerve pallor, areas of chorioretinal atrophy due to previous laser treatment and a persisting small area of leakage in the peripheral retina of the RE. Color fundus photographs by RetCam of the LE with FEVR stage 3B clearly show the presence of a falciform fold arising from the optic nerve head, with partial traction on the surrounding retina. Stage 3A FEVR of the proband’s older brother (II:1) is shown in the color fundus photograph from his LE. Macular OCT images of proband’s father and mother are also shown in upper left and upper right quadrants of ﬁgure, respectively. The mother’s exam shows a slight reduction of macular thickness. ● ∗Documented clinical evaluation; E+ and E−, positive and negative to genetic test, respectively; obligate carrier; RE, right eye; LE, left eye.

0.1 LogMAR in both eyes without signs of pathological 3.4. Family 4. This 7 y/o male proband (II:2) came to our activity for the last three years. attention at the age of one year presenting congenital sensory The proband’s father (I:1) and sister (II:1) are healthy, nystagmus. Ophthalmological examination under sedation whereas his mother (I:2), 52 years old, was diagnosed revealed the presence of a falciform fold arising from the with stage 2B FEVR and related mild macular edema at optic nerve head, with partial traction on the surrounding the age of 46. Both eyes were treated with Argon laser, retina in the left eye and slight optic nerve pallor with areas and her current visual acuity is 0.0 LogMAR in both of chorioretinal atrophy due to previous laser treatment in eyes (Figure 3). the peripheral retina in the right eye. Consequently, stages Autosomal dominant inheritance was suspected before 3B and 2B FEVR were diagnosed in the left and right eyes, genetic evaluation (Figure 3). Genetic evaluation showed that respectively. A ﬂuorescein angiography exam performed the proband and his mother were heterozygous for a new with RetCam showed a small area of hyperﬂuorescence next p.(Cys204Phe) variant in the FZD4 gene, while the father to the laser-treated areas, which required further laser and the sister were negative to the test. treatment (Figure 4). At subsequent examinations, both eyes 6 Journal of Ophthalmology

⁎

(i) 1 2 E+ (NDP p.[(Ala105Phe)];[=])

I:2 ⁎

(ii) 1 E+ (NDP p.[(Ala105Phe)])

Figure 5: Pedigree and ocular features of family 5. Color fundus photographs by RetCam from the 3 y/o male proband (II:1) of family 5 showing a closed funnel total retinal detachment in the right eye and peripheral exudation with vascular abnormalities before and after laser treatment in the left eye. Fluorescein angiograms performed with RetCam before and after laser treatment showing peripheral ischemic areas with leakage mainly in temporal and inferior peripheral retina in the left eye. Top right, OCT macular scans from the proband’s mother (I:2) showing normal retinal features. ∗Documented clinical evaluation; E+, positive to genetic test; RE, right eye; LE, left eye; LE∗, images taken at a subsequent examination after the laser treatment. were stable. At the orthoptic examination, torticollis due to respectively. At subsequent examinations, the right eye nystagmus with the face rotated to the right to favour fixation progressed into a closed funnel total retinal detachment of the right eye in adduction was found. Visual acuity was whereas the left eye was stable. The child’s father could 0.1 LogMAR in the left eye and 0.4 LogMAR in the right not be examined while the mother showed a visual acuity eye, respectively. of 0.0 LogMAR in both eyes. Her fundus was normal present- The proband’s 10 y/o brother (II:1) was diagnosed with ing only a slight mottling of the retinal pigmented epithelium bilateral retinal dystrophy and falciform retinal fold at the in the peripheral retina. Full-field electroretinography and age of three, suggesting a stage 3A FEVR (Figure 4). The left macular OCT scans were within normal limits in both eyes. eye also showed a slight posterior opacity of the lens while the (Figure 5). Genetic testing revealed that the proband and ectopic pupil was visible in the right eye. Visual acuity was his mother (I:2) have a p.(Ala105Phe) variation in the lower than 0.1 LogMAR in both eyes. NDP gene. The children’s parents have visual acuities of 0.0 Log- MAR in both eyes. The father’s fundus was normal, whereas 3.6. Family 6. The 7 y/o male proband (II:1) was periodically the mother showed slight mottling of the retinal pigmen- monitored for visual loss due to FEVR since the age of one. ted epithelium, more evident at the periphery. Full-field The fundus examination under sedation showed a falciform electroretinography was within normal limits in the par- retinal fold arising from the optic nerve head, involving the ents, though the mother’s scotopic ERG responses were macula and anchored in the temporal periphery in his left borderline and her macular OCT scans showed a reduction eye (Figure 6). No exudation was evident near the fold or of foveal thickness suggesting a condition of asymptomatic the temporal or nasal periphery. Based on aforementioned carrier (Figure 4). clinical data, stage 3A FEVR was postulated. No significant Genetic testing revealed the p.(Arg121Gln) variant in the changes were found at the subsequent follow-up. The NDP gene, already associated in literature with Norrie disease ophthalmological examination performed at the age of three (MIM ∗310600) [25] (Figure 4). showed residual visual acuity of 1.0 LogMAR in both eyes. The proband’s father did not present any sign of ocular 3.5. Family 5. The 3 y/o male proband (II:1) came to our pathology while the mother was diagnosed with stage 1 attention at the age of four months presenting leukocoria FEVR in both eyes. in the right eye. The fundus examination under sedation Genetic testing revealed the heterozygous IVS2 c.67- showed an inoperable macula-involving retinal detachment 2A>G variant in the TSPAN12 gene, and the same variant with subretinal exudate in the right eye and peripheral was found in the proband’s mother (I:2), thus configuring exudation with vascular abnormalities in the left eye. A an autosomal dominant transmission (Figure 6). fluorescein angiography exam performed with RetCam showed peripheral ischemic areas with leakage mainly in 4. Discussion temporal and inferior peripheral retina which required laser treatment in the left eye. Consequently, stages 4B In the present work, we reported the clinical findings and and 1B FEVR were diagnosed in the right and left eyes, the genetic analysis from a cohort of 6 unrelated Italian Journal of Ophthalmology 7

⁎ factor (NDP gene), FZD4 receptor (FZD4 gene), LRP5 coreceptor (LRP5 gene), and tetraspanins (TSPAN12 gene) ff (i) are described as associated with di erent pathological 1 2 phenotypes, ranging from mild to severe, even in the same E+ (TSPAN12 c.[67-2A > G];[=]) family, or with clinically indistinguishable patterns caused by variations in different genes. Our genetic testing investigated on the four genes most ⁎ widely associated in literature with FEVR: NDP, FDZ4, LRP5, and TSPAN12 [3]. (ii) In the present report, families 1, 2, and 3 had three different heterozygous variants in the FZD4 gene. FZD4 P 1 E+ (TSPAN12 c.[67-2A > G];[=]) codes for FZD4, a member of the Frizzled family of seven- transmembrane Wnt-binding receptors. The binding of Wnt or Norrin ligands with FZD4 in conjunction with LRP5 coreceptors results in the activation of canonical β- catenin-dependent signaling. The N-terminal extracellular cysteine-rich domain (CRD), conserved among Frizzled family members, is necessary for binding to ligands [29]. The p.(Gln93∗) variation in the FZD4 gene found in family 1 introduces a premature stop codon, and therefore, it is predicted to be pathogenic. In families 2 and 3, the two variants are found in the extracellular N-terminal domain of the receptor, in amino acid positions 181 and 204, respectively. Both are presumably downstream of the CRD required for ligand binding, according to Smallwood et al. (i.e., the 114-amino acid region Figure 6: Pedigree and ocular features of family 6. RetCam color extending from the first to the tenth conserved CRD cyste- fundus photograph from the LE of the 7 y/o male proband (II:1) ine) [30], but their importance is relevant because variations of family 6, showing FEVR stage 3A for the presence of a in both codons, associated with a pathological phenotype, falciform retinal fold arising from the optic nerve head, involving have already been described in literature and both can the macula and anchored in the temporal periphery without signs ∗ − therefore be considered mutational hot spots. of exudation. Documented clinical evaluation; E+ and E , The variant p.(Cys181Tyr) has already been described positive and negative to genetic test, respectively. |Asymptomatic/ by Drenser et al. and associated with autosomal dominant presymptomatic carrier (stage 1 FEVR); LE, left eye. FEVR [31]. The author concluded that this cysteine residue is not known to form an intracellular disulphide bond; it is nevertheless the 11th of 13 cysteine residues families made up of 6 probands and 10 family members, that are conserved in vertebrates and may be required thus presenting the results from the largest cohort of Italian for receptor dimerization [32], a function that may be FEVR families and data from four novel genetic variants relevant to the mechanism of Wnt binding and signaling linked to FEVR. [33]. In family 2, the healthy father, who carries this var- Retinal vascularization during eye development and iant, showed a subclinical phenotype (stage 1 FEVR). It is maintenance of its normal architecture are processes finely well known that variations in this gene have complete orchestrated by the Norrin/β-catenin signaling pathway. (100%) penetrance but variable expression: members of This pathway is distinguished from the Wingless (Wnts) the same family may have the same variant and show signaling pathways, which include a large family of ligands different severities of the disorder or may have retinal and are involved in many cell processes, and is specific for changes detectable only by diagnostic tools such as fluo- the development and maintenance of the retinal vasculature. rescein angiography. Moreover, the disease phenotype in The bond in NDP is in fact highly specific for the receptor this family showed interesting aspects of FEVR pathology: complex FZD4-LRP5 [26] while TSPAN12 is required for (1) an asymmetrical impairment of the retina, (2) the FZD4/β-catenin signaling induced by Norrin. Moreover, importance of a follow-up in patients by fluorescein angi- TSPAN12 expression is restricted to the vasculature within ography, even in case of a normal (left eye) posterior pole, the retina [27]. (3) the necessity of a different treatment approach for the Aberrations and malfunctioning of this signaling system lesions depending on the stages of the disease, and (4) the affect the development of the retinal vasculature, and this evidence that inactive lesions may later on constitute an translates into a variety of eye diseases such as Norrie disease, ophthalmological emergency (retinal detachment in the FEVR, and retinopathy of prematurity [5, 6, 28]. All these right eye). disorders are clinically and genetically heterogeneous. Similarly, variations in the 204 codon, as we found in Nucleotide alterations in the genes’ coding for fundamental family 3, have already been associated with autosomal dom- components of this signaling pathway, such as NDP growth inant FEVR in two reports, one describing the amino acid 8 Journal of Ophthalmology change p.(Cys204Arg) [34] and the other, the p.(Cys204Tyr) Poulter et al. described the c.67-1G>C variant in change [35]. both homozygous and heterozygous states in two FEVR A functional study by Zhang et al. showed that FZD4 Indian cousins; the c.67-1G>G splice variant causes exon binding to NDP is disrupted by Cys204Arg, suggesting that 3 deletion, resulting in a frameshift and a premature termina- the CRD may be beyond the previously predicted region, tion codon (p.Leu23GlyfsX66) [11]. Considering that the as described by Smallwood et al. [30], or that NDP bind- c.67-2A>G variant reported in our work involves the same ing to FZD4 requires the CRD plus additional residues, acceptor splicing site, it is reasonable to expect the same C-terminal to the CRD [29]. The autosomal dominant pathogenetic effect. feature of FZD4 variants could be due either to haploin- The severe clinical expression of the disease observed sufficiency or to a dominant-negative effect, as previous in our patient is consistent with previous studies where findings suggest [36]. variations in TSPAN12 gene are reported to be associated In family 4, the proband and the brother are affected by with very severe disease phenotypes [11, 27]. an X-linked form of FEVR due to a variant in the NDP gene. NDP is a protein ligand not belonging to the Wnt family which recognises and binds with high affinity and specificity 5. Conclusions to the CRD of FZD4 (and not to CRDs of the other 14 mammalian Frizzled and secreted Frizzled-related proteins) To our knowledge, this is the largest cohort study of Italian [30] and activates the canonical signaling pathway [26]. FEVR patients. In the present retrospective study involving The p.(Arg121Gln) variant in the NDP gene has already different clinical sites, the probands and their family mem- been reported in literature, and the amino acid residue 121 bers were clinically examined and identified by standard seems to be a mutational hot spot. Indeed, the same amino diagnostic tests, albeit performed by different operators; this, acid change from arginine to glutamine has been associated together with the limited observed population, impedes with Norrie disease [25]; however, variants in the same critical analysis of the phenotypic differences associated codon, specifically p.(Arg121Gly) [37] and p.(Arg121Trp) with the variations in different genes of the Norrin/ [38], have been associated both with Norrie disease and Frizzled signaling pathway in our group of FEVR patients. with X-linked FEVR, that is, the p.(Arg121Leu) [39] and Variations of the analyzed genes were found in 6 out of 8 p.(Arg121Trp) variants [40]. (75%) patients, and four novel variants responsible for the Similarly, the p.(Ala105Phe) found in family 5 lies in a phenotype were identified. mutational hot spot since two other variants in the same The variants in the FZD4 gene found in families 1, 2, and codon, namely, the p.(Ala105Thr) and the p.(Ala105Glu), 3 were associated with the mildest phenotype: in families 1 were described as associated with Norrie Disease [13] and and 2, FEVR manifested with late onset and vision was FEVR [41], respectively. preserved in the affected family members, whereas in family Both variants found in families 4 and 5 are located in the 3, patients showed earlier disease onset with relatively pre- C-terminal end of the Norrie protein, probably affecting the served vision in the proband and mild loss of visual acuity secondary structure and function of the protein. in the mother. Norrie disease is a complex disorder in which blindness The X-linked FEVR variants involving the NDP gene in in early childhood may be accompanied by sensorineural families 4 and 5 are considered severe because they determine deafness and progressive mental retardation, inherited by an early onset of the disease with relevant retinal alterations. X-linked recessive transmission. Heterozygous carriers rarely Minimal signs of retinal abnormalities were detected also in manifest clinical features of the disorder, though some cases the mother of the two affected children who carried the same have been described [42, 43]. The absence of a syndromic variant in family 4, while in family 5, the variant exerted no pattern with extraocular manifestations and preservation of effects in the proband’s mother. some vision suggested a diagnosis of FEVR rather than The variant in the TSPAN12 gene in family 6 caused a Norrie disease, in both families. very severe phenotype that manifested from the first months Furthermore, the description of the same variant in sub- of life with almost complete vision loss. jects with different clinical patterns may indicate the involve- Our results are in agreement with the previously ment of other factors that might modify the phenotype. described literature confirming that familial exudative In family 6, the proband and his mother harbour the vitreoretinopathy presents a penetrance close to 100% [16] same heterozygous variant IVS2 c.67-2A>G in the TSPAN12 but is clinically and genetically heterogeneous [12], even in gene. TSPAN12 is a member of the tetraspanin superfamily the same family [44]. Indeed, the two new heterozygous characterized by four transmembrane domains. It is a key variants found in the FZD4 gene and those in NDP and component of the NDP-FZD4-LRP5 signaling complex that TSPAN12 genes, respectively, were correlated with variable cooperatively promotes multimerization of FZD4 and its phenotypical presentation ranging from relatively mild to associated proteins to elicit physiological levels of signaling severe anatomical and functional impairment. Severity of [27]. Both dominant and recessive variants in TSPAN12 have phenotype was dependent on gene involvement and site of been described in FEVR [8, 11]. nucleotide variations. The novel variant reported here is predicted to interfere Analysis of new genes recently found to be associated with the consensus sequence for the splice acceptor site of with FEVR will make it possible to improve the understand- intron 2, as confirmed using HSF. ing of the pathogenesis of the disease. Journal of Ophthalmology 9

Conflicts of Interest exudative vitreoretinopathy (FEVR),” Investigative Ophthal- mology & Visual Science, vol. 53, pp. 2873–2879, 2012. Each author states that he/she has no proprietary interest in [12] F. N. Boonstra, C. E. van Nouhuys, J. Schuil et al., “Clinical and the development or marketing of the instruments used in molecular evaluation of probands and family members with the present study. The authors declare that they do not have familial exudative vitreoretinopathy,” Investigative Ophthal- any conflict of interest. mology & Visual Science, vol. 50, pp. 4379–4385, 2009. [13] I. Torrente, M. Mangino, M. Gennarelli et al., “Two new ’ missense mutations (A105T and C110G) in the norrin gene Authors Contributions in two Italian families with Norrie disease and familial exuda- ” Giancarlo Iarossi, Paolo Enrico Maltese, and Lucia Ziccardi tive vitreoretinopathy, American Journal of Medical Genetics, vol. 72, pp. 242–244, 1997. contributed equally to this study. [14] G. de Crecchio, F. Simonelli, G. Nunziata et al., “Autoso- mal recessive familial exudative vitreoretinopathy: evidence Acknowledgments for genetic heterogeneity,” Clinical Genetics, vol. 54, pp. 315– 320, 1998. The contribution by Fondazione Bietti in this paper was [15] M. Savarese, E. Spinelli, F. Gandolfo et al., “Familial exuda- supported by the Ministry of Health and Fondazione tive vitreoretinopathy caused by a homozygous mutation in Roma. The authors are grateful to Helen Ampt for revising TSPAN12 in a cystic fibrosis infant,” Ophthalmic Genetics, the manuscript. vol. 35, pp. 184–186, 2014. [16] A. H. Kashani, K. T. Brown, E. Chang, K. A. Drenser, A. Capone, and M. T. Trese, “Diversity of retinal vascular anom- References alies in patients with familial exudative vitreoretinopathy,” Ophthalmology, vol. 121, pp. 2220–2227, 2014. [1] C. E. Van Nouhuys, “Juvenile retinal detachment as a compli- ” [17] R. L. Bennett, K. S. French, R. G. Resta, and D. L. Doyle, cation of familial exudative vitreoretinopathy, Fortschritte der “ Ophthalmologie, vol. 86, pp. 221–223, 1989. Standardized human pedigree nomenclature: update and assessment of the recommendations of the National Society of [2] W. E. Benson, “Familial exudative vitreoretinopathy,” Trans- Genetic Counselors,” Journal of Genetic Counseling, vol. 17, actions of the American Ophthalmological Society, vol. 93, pp. 424–433, 2008. pp. 473–521, 1995. [18] H. Thiele and P. Nürnberg, “HaploPainter: a tool for drawing [3] R. W. Collin, K. Nikopoulos, M. Dona et al., “ZNF408 is pedigrees with complex haplotypes,” Bioinformatics, vol. 21, mutated in familial exudative vitreoretinopathy and is crucial pp. 1730–1732, 2005. for the development of zebrafish retinal vasculature,” Proceed- “ ings of the National Academy of Sciences of the United States of [19] I. A. Adzhubei, S. Schmidt, L. Peshkin et al., A method and ” America, vol. 110, pp. 9856–9861, 2013. server for predicting damaging missense mutations, Nature [4] X. Ye, Y. Wang, H. Cahill et al., “Norrin, frizzled-4, and Lrp5 Methods, vol. 7, pp. 248-249, 2010. signaling in endothelial cells control a genetic program for [20] P. Kumar, S. Henikoff, and P. C. Ng, “Predicting the effects retinal vascularization,” Cell, vol. 139, pp. 285–298, 2009. of coding non-synonymous variants on protein function ” – [5] T. N. Masckauchán and J. Kitajewski, “Wnt/Frizzled signaling using the SIFT algorithm, Nature Protocols, vol. 4, pp. 1073 in the vasculature: new angiogenic factors in sight,” Physiology 1081, 2009. (Bethesda), vol. 21, pp. 181–188, 2006. [21] J. M. Schwarz, D. N. Cooper, M. Schuelke, and D. “ [6] J. Robitaille, M. L. MacDonald, A. Kaykas et al., “Mutant Seelow, MutationTaster2: mutation prediction for the deep- ” frizzled-4 disrupts retinal angiogenesis in familial exudative sequencing age, Nature Methods, vol. 11, pp. 361-362, 2014. vitreoretinopathy,” Nature Genetics, vol. 32, pp. 326–330, 2002. [22] F. O. Desmet, D. Hamroun, M. Lalande, G. Collod-Béroud, M. “ [7] C. Toomes, H. M. Bottomley, R. M. Jackson et al., “Mutations Claustres, and C. Béroud, Human Splicing Finder: an online ” in LRP5 or FZD4 underlie the common familial exudative bioinformatics tool to predict splicing signals, Nucleic Acids vitreoretinopathy locus on chromosome 11q,” American Jour- Research, vol. 37, article e67, 2009. nal of Human Genetics, vol. 74, pp. 721–730, 2004. [23] M. J. Betts and R. B. Russell, “Amino acid properties and [8] K. Nikopoulos, C. Gilissen, A. Hoischen et al., “Next- consequences of substitutions,” in Bioinformatics for Geneti- generation sequencing of a 40 Mb linkage interval reveals cists, M. R. Barnes and I. C. Gray, Eds., pp. 289–316, John TSPAN12 mutations in patients with familial exudative Wiley & Sons, Ltd, Chichester, UK, 2003. vitreoretinopathy,” American Journal of Human Genetics, [24] S. Richards, N. Aziz, S. Bale et al., “Standards and guidelines vol. 86, pp. 240–247, 2010. for the interpretation of sequence variants: a joint consensus [9] B. S. Shastry and M. T. Trese, “Familial exudative vitreoretino- recommendation of the American College of Medical Genetics pathy: further evidence for genetic heterogeneity,” American and Genomics and the Association for Molecular Pathology,” Journal of Medical Genetics, vol. 69, pp. 217-218, 1997. Genetics in Medicine, vol. 17, no. 5, pp. 405–424, 2015. [10] B. S. Shastry, J. F. Hejtmancik, and M. T. Trese, “Identification [25] J. J. Fuentes, V. Volpini, F. Fernández-Toral, E. Coto, and of novel missense mutations in the Norrie disease gene X. Estivill, “Identification of two new missense mutations associated with one X-linked and four sporadic cases of (K58N and R121Q) in the Norrie disease (ND) gene in familial exudative vitreoretinopathy,” Human Mutation, two Spanish families,” Human Molecular Genetics, vol. 2, vol. 9, pp. 396–401, 1997. pp. 1953–1955, 1993. [11] J. A. Poulter, A. E. Davidson, M. Ali et al., “Recessive muta- [26] Q. Xu, Y. Wang, A. Dabdoub et al., “Vascular development tions in TSPAN12 cause retinal dysplasia and severe familial in the retina and inner ear: control by Norrin and Frizzled-4, a 10 Journal of Ophthalmology

high-affinity ligand-receptor pair,” Cell, vol. 116, pp. 883– exudative vitreoretinopathy patients,” Investigative Ophthal- 895, 2004. mology & Visual Science, vol. 56, no. 3, pp. 1937–1946, 2015. [27] H. J. Junge, S. Yang, J. B. Burton et al., “TSPAN12 regulates [42] P. Lin, S. P. Shankar, J. Duncan, A. Slavotinek, E. M. Stone, and retinal vascular development by promoting Norrin- but not T. Rutar, “Retinal vascular abnormalities and dragged maculae Wnt-induced FZD4/beta-catenin signaling,” Cell, vol. 139, in a carrier with a new NDP mutation (c.268delC) that caused pp. 299–311, 2009. severe Norrie disease in the proband,” Journal of AAPOS, [28] R. F. Gariano and T. W. Gardner, “Retinal angiogenesis vol. 14, pp. 93–96, 2010. in development and disease,” Nature, vol. 438, pp. 960– [43] T. Parzefall, T. Lucas, M. Ritter et al., “A novel missense NDP 966, 2005. mutation [p.(Cys93Arg)] with a manifesting carrier in an [29] K. Zhang, Y. Harada, X. Wei et al., “An essential role of the Austrian family with Norrie disease,” Audiology & Neuro- cysteine-rich domain of FZD4 in Norrin/Wnt signaling and Otology, vol. 19, pp. 203–229, 2014. familial exudative vitreoretinopathy,” The Journal of Biological [44] R. R. Ober, A. C. Bird, A. M. Hamilton, and K. Sehmi, “Auto- Chemistry, vol. 286, pp. 10210–10215, 2011. somal dominant exudative vitreoretinopathy,” The British [30] P. M. Smallwood, J. Williams, Q. Xu, D. J. Leahy, and Journal of Ophthalmology, vol. 64, no. 2, pp. 112–120, 1980. J. Nathans, “Mutational analysis of Norrin-Frizzled4 rec- ognition,” The Journal of Biological Chemistry, vol. 282, pp. 4057–4068, 2007. [31] K. A. Drenser, W. Dailey, A. Vinekar, K. Dalal, A. Capone Jr., and M. T. Trese, “Clinical presentation and genetic correlation of patients with mutations affecting the FZD4 gene,” Archives of Ophthalmology, vol. 127, pp. 1649–1654, 2009. [32] S. Omoto, T. Hayashi, K. Kitahara, T. Takeuchi, and Y. Ueoka, “Autosomal dominant familial exudative vitreoretinopathy in two Japanese families with FZD4 mutations (H69Y and C181R),” Ophthalmic Genetics, vol. 25, pp. 81–90, 2004. [33] C. E. Dann, J. C. Hsieh, A. Rattner, D. Sharma, J. Nathans, and D. J. Leahy, “Insights into Wnt binding and signalling from the structures of two Frizzled cysteine-rich domains,” Nature, vol. 412, pp. 86–90, 2001. [34] J. Nallathambi, D. Shukla, A. Rajendran, P. Namperumalsamy, R. Muthulakshmi, and P. Sundaresan, “Identification of novel FZD4 mutations in Indian patients with familial exudative vitreoretinopathy,” Molecular Vision, vol. 12, pp. 1086– 1092, 2006. [35] K. Nikopoulos, H. Venselaar, R. W. Collin et al., “Overview of the mutation spectrum in familial exudative vitreoretinopathy and Norrie disease with identification of 21 novel variants in FZD4, LRP5, and NDP,” Human Mutation, vol. 31, pp. 656–666, 2010. [36] A. Kaykas, J. Yang-Snyder, M. Héroux, K. V. Shah, M. Bouvier, and R. T. Moon, “Mutant Frizzled 4 associated with vitreoretinopathy traps wild-type Frizzled in the endoplasmic retic- ulum by oligomerization,” Nature Cell Biology, vol. 6, pp. 52–58, 2004. [37] D. Zhu and I. Hussel-Maumanee, “Mutation analysis of the Norrie disease gene in eleven families,” Investigative Ophthal- mology & Visual Science, vol. 35(Supplement), p. 1265, 1994. [38] A. Meindl, B. Lorenz, H. Achatz, H. Hellebrand, P. Schmitz- Valckenberg, and T. Meitinger, “Missense mutations in the NDP gene in patients with a less severe course of Norrie disease,” Human Molecular Genetics, vol. 4, pp. 489-490, 1995. [39] K. Johnson, H. A. Mintz-Hittner, Y. P. Conley, and R. E. Ferrell, “X-linked exudative vitreoretinopathy caused by an arginine to leucine substitution (R121L) in the Norrie disease protein,” Clinical Genetics, vol. 50, pp. 113–115, 1996. [40] S. Fuchs, U. Kellner, H. Wedemann, and A. Gal, “Missense mutation (Arg121Trp) in the Norrie disease gene associated with x-linked exudative vitreoretinopathy,” Human Mutation, vol. 6, pp. 257–259, 1995. [41] J. Salvo, V. Lyubasyuk, M. Xu et al., “Next-generation sequencing and novel variant determination in a cohort of 92 familial Hindawi Journal of Ophthalmology Volume 2017, Article ID 6950642, 6 pages https://doi.org/10.1155/2017/6950642

Research Article Physical Activity and Quality of Life in Retinitis Pigmentosa

1 1 2 3,4 Joshua D. Levinson, Ethan Joseph, Laura A. Ward, Joe R. Nocera, 1,3,5 1,2 1 Machelle T. Pardue, Beau B. Bruce, and Jiong Yan

1Department of Ophthalmology, Emory University School of Medicine, Atlanta, GA, USA 2Rollins School of Public Health, Emory University, Atlanta, GA, USA 3Center for Visual and Neurocognitive Rehabilitation, Atlanta VA Medical Center, Decatur, GA, USA 4Department of Neurology, Emory University School of Medicine, Atlanta, GA, USA 5Department of Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA, USA

Correspondence should be addressed to Jiong Yan; [email protected]

Received 4 January 2017; Revised 12 March 2017; Accepted 28 March 2017; Published 17 May 2017

Academic Editor: Benedetto Falsini

Copyright © 2017 Joshua D. Levinson et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Purpose. Aerobic exercise has been found to be neuroprotective in animal models of retinal degeneration. This study aims to report physical activity levels in patients with RP and investigate the relationship between physical activity and vision-related quality-of-life (QOL). Materials and Methods. A retrospective study of adult patients with RP examined in 2005–2014. Physical activity levels were assessed using the Godin Exercise Questionnaire. The NEI-Visual Function Questionaire-25 (VFQ-25), SF- 36 General Health survey, and Pepper Assessment Tool for Disability (PAT-D) were administered. Results. 143 patients participated. 81 (56.6%) patients were classified as “active” and 62 (43.4%) as “insufficiently active” by Godin score. VFQ-25 revealed statistically significant differences between the active and insufficiently active patients, including overall visual function (53.3 versus 45.1, p =0010), color vision (73.8 versus 52.9, p <0001), and peripheral vision (34.3 versus 23.8, p =0021). The physical component of the SF-36 and the PAT-D survey also demonstrated statistically significant differences (47.2 versus 52.9, p =0002; 24.3 versus 30.0, p =0010). Active patients had a higher initial Goldmann visual field (GVF) score (74.8 versus 60.1 degrees, p =0255) and final GVF score (78.7 versus 47.1 degrees, p =0069) but did not reach statistical significance. Conclusions. In RP, increased physical activity is associated with greater self-reported visual function and QOL.

1. Introduction shown to enhance memory and promote hippocampal neu- roregeneration [3] suggesting a neuroprotective effect. Previ- Retinitis pigmentosa (RP) is an inherited retinal dystrophy ous studies have suggested a beneficial effect of exercise on affecting more than 1 million individuals worldwide [1]. common ocular conditions such as age-related macular Early symptoms include night blindness and peripheral degeneration (AMD) [4, 5], glaucoma [6], and cataract [7]. visual field constriction. Advanced disease is characterized Vision loss in RP is due to progressive loss of rod and cone by a small central tunnel visual field that may progress to loss photoreceptors [1]. Recently, exercise has been shown to of central vision. In a small number of patients, RP leads to have neuroprotective effects on photoreceptors in mouse total blindness. Currently, there are no proven treatments models of retinal degeneration [8, 9]. to slow the progression of RP. Given the lack of significant While many studies have investigated exercise and its treatment options, anticipated vision loss in RP can be emo- impact on the health of the human body, the effect of exercise tionally devastating [2]. on the progression of retinal degeneration in human subjects Exercise is well known to have a positive impact on both is poorly understood. This retrospective case-control study physical and psychological health. Exercise has even been will attempt to report baseline physical activity levels in indi- 2 Journal of Ophthalmology viduals with RP and to investigate the relationship between functioning, freedom from pain, energy or fatigue, emotional physical activity and visual function reported from three well-being, and general health perceptions. Scores range quality of life (QOL) surveys. from 0 (maximum impairment) to 100 (no impairment). Subscores are used to calculate a Physical Component Sum- 2. Methods mary (PCS) and Mental Component Summary (MCS).

A retrospective review of the medical record was performed 2.4. Godin Leisure-Time Exercise Questionnaire (GLTEQ). for adult patients with retinitis pigmentosa evaluated at the An excerpt from the Godin Leisure-Time Exercise Question- Emory Eye Center between October 2004 and April 2015. naire [15] was used to determine the frequency and intensity Subjects were identified through a clinical database of patients of physical activity. Respondents report the number of times with retinal degenerative conditions. This study was approved per week that they engage in mild, moderate, or strenuous by Emory University’s Institutional Review Board and physical activity which is then multiplied by 3, 5, or 9 meta- adhered to the tenets of the Declaration of Helsinki. bolic equivalents, respectively, and summated into a total Clinical data was obtained from the medical record and score. Subjects were classified as either “active” or “insuffi- included age, gender, race, and medical history. Goldmann ciently active” using a cutoff score of 24 as suggested by prior visual fields (GVF) were reviewed. Visual field scores were studies [16]. While some studies in the literature exclude obtained for each eye by summation of the degree of pre- mild physical activity [16], we elected to include mild physi- served visual field across the horizontal and vertical merid- cal activity in our calculation with the understanding that RP ians using the III4e isopter. As quality of life of a patient is patients with advanced vision loss may have difficulty per- most affected by vision in the better-seeing eye, the eye with forming more strenuous physical activity. the larger initial visual field was analyzed for statistical pur- poses. For subjects with multiple visual fields, data was col- 2.5. Statistical Methods. Descriptive demographics and med- lected for the first and last tests within the study period. ical history statistics were calculated for the full cohort. Qual- When available, electroretinograms (ERGs) were reviewed ity of life measures were calculated overall and by Godin from each patient’s initial clinic visit. The electrophysiologist activity score category (active versus insufficiently active). reported photoreceptor function as “normal” or “mild”, The differences between the activity levels were tested using “moderate,” or “severe” photoreceptor dysfunction. As many a two-sample t-test for the overall quality of life scores, as ERG responses were indistinguishable from noise due to the well as the general vision, color vision, and peripheral vision advanced nature of the disease, a decision was made not subscores of the VFQ-25 and the physical functioning, social to include quantitative ERG measurements. The clinical- function, and general health subscores of the SF-36. intake history form asked patients if they have diabetes, Two-sample t-tests were also used to examine the differ- hypertension, or cardiovascular disease. Patients who ences between the summation of the horizontal and vertical reported a history of one or more of these conditions were GVFIII4emeasurements inthebettereyebyactivitylevel.This classified as having “vascular disease.” was done for both the initial and final measurements, as well as Subjects were contacted by telephone and four validated the difference between the initial and final GVF III4e measure- questionnaires were administered to those that consented. ments. The summed GVF III4e measurements were also Additional questions including smoking history and employ- examined with two-sample t-tests by current smoking status ment status were asked at the end of the survey. and presence of vascular disease. These relationships were also tested using a linear regression model, controlling for age. 2.1. National Eye Institute Visual Function Questionnaire-25 All statistical tests were performed using SAS v9.4 soft- (NEI VFQ-25). The NEI VFQ-25 was administered to evalu- ware and tested using an alpha of 0.05. ate subjective visual function and vision-related quality of life [10, 11]. NEI VFQ-25 consists of 25 questions designed to assess 12 aspects of daily living: general health, general vision, 3. Results near vision, distance vision, driving, peripheral vision, color 143 patients completed the telephone survey out of 496 vision, ocular pain, role limitation, dependency, social func- patients in the clinical database (28.8%). The mean age of tion, and mental health. study participants was 46.9 years. Demographic and medical 2.2. Pepper Assessment Tool for Disability (PAT-D). The history data is presented in Table 1. PAT-D is a 19-item questionnaire designed to assess mobil- The average Godin exercise score was 32.6. Eighty-one fi “ ” ity, activities of daily living (ADL), and instrumental activi- patients (56.6%) were classi ed as active while 62 (43.4%) “ ffi ” ties of daily living (IADL) [12, 13]. Participants rate their were insu ciently active. Active patients were found to have ffi level of difficulty performing each activity on a 5-point Likert a higher overall NEI VFQ-25 score than the insu ciently p =0010 scale ranging from 1 (“no difficulty”)to5(“unable to do”). active patients (53.3 versus 45.1, ) (Table 2). General vision, color vision, and peripheral vision were identified as 2.3. 36-Item Short Form Health Survey (SF-36). General subscores of particular interest. Increased physical activity health-related quality of life was assessed with the SF-36 sur- was associated with higher subscores in peripheral vision vey [14]. The SF-36 consists of 8 subscores: physical func- (p =0021) and color vision (p <0001). tioning, role limitations due to physical health problems, In evaluation of general health-related quality of life, role limitations due to emotional health problems, social active patients scored significantly higher on the physical Journal of Ophthalmology 3

Table 1: Demographic and medical history characteristics for 143 (Table 5). No association was seen between employment subjects with retinitis pigmentosa. and the Mental Component Summary.

n (%) 4. Discussion Age mean (SD) 46.9 (13.7) African American 26 (25%) In this survey of 143 patients with RP, approximately 4 in 10 Asian 1 (1.0%) patients were found to have “insufficiently active” lifestyles. Race∗ Others 2 (2.0%) The mean GLTEQ score of 32.6 reveals that patients with White 73 (72%) RP have a physical activity level that falls between previously published values for patients with physical disabilities [17] Female 83 (58%) Sex and healthy controls [18]. The high rate of physically inactiv- Male 60 (42%) ity seen in our study is consistent with the findings of An History of smoking 96 (67%) et al. which reported a significantly higher rate of physical Hypertension 42 (29%) inactivity in RP patients compared to healthy controls Diabetes 13 (9%) (50.8% versus 27.3%) [19]. Cardiovascular disease 9 (6%) This study revealed significantly better self-reported ∗n = 102; SD: standard deviation. overall visual function, color vision, and peripheral vision in active patients as measured by the NEI VFQ-25. Visual field scores were higher for active patients at both the initial component summary score of the SF-36 (52.9 versus 47.2, and final measurement, although these differences did not p =0002) but no significant difference was seen for the men- reach statistical significance. Quantifying visual function in tal component summary (51.1 versus 51.7, p =0731). Active RP patients is difficult. Central visual acuity is often pre- patients were found to have significantly less disabilities on served until significant visual field loss has occurred. In our the PAT-D survey (24.34 versus 30.0, p =0010). study, ERGs demonstrated severe photoreceptor dysfunction Initial visual field measurements were available for 120 in 73.2% of patients on presentation. ERG measurements in patients. The mean initial GVF summation score was 67.9 these patients are often indistinguishable from noise which degrees. The relationship between GVF scores and activity makes them difficult to follow over time. Goldmann visual level is presented in Table 3. Active subjects had a higher fields are often relied on to track visual loss, but the sensitiv- initial GVF score than the insufficiently active subjects ity is limited. GVFs have increased variability in RP patients (Table 3), but this difference did not reach statistical [20] and this variability increases with disease severity [21]. significance (74.8 versus 60.1, p =0255). Follow-up visual The use of a survey, like the NEI VFQ-25, provides a subjec- fields were available for 52 patients. In this cohort, the initial tive means to evaluate visual function and should be included mean GVF score was 90.2 degrees in the active group and as a focus of future prospective studies evaluating the effects 55.3 degrees in the insufficiently active group. The final GVF of exercise on RP. scores were also higher in the active group, and this Pardue and colleagues found aerobic exercise to have a difference approached but did not reach statistical neuroprotective effect on photoreceptors in mouse models significance (78.7 versus 47.1 degrees, p =0069). For both of retinal degeneration which they demonstrated was medi- active and insufficiently active patients, the mean GVF ated through increases in brain-derived neurotrophic factor score decreased over the follow-up period (−11.54 degrees (BDNF) [8, 9]. In humans, BDNF is one of the principle active versus −6.62 degrees insufficiently active, p =06987). growth factors known to mediate the effects of exercise on The association between age and physical activity level the brain [22]. Further research is warranted to investigate was found to approach statistical significance, with increased if BDNF may have a protective effect on photoreceptors in physical activity seen in younger patients (p =0056). How- humans. ever, age was not found to be associated with degrees of pre- While little is known regarding the effect of exercise on served initial visual field in a univariate analysis (p =01948) human retinal degeneration, several studies have demon- or in a multivariate analysis with activity level as a covar- strated a protective effect of aerobic exercise in other neuro- iate (p =08456). degenerative diseases. In a large meta-analysis, Beckett et al. Sixty-seven percent of respondents reported a history of demonstrated that physical activity is associated with a tobacco use. No significant difference in GVF scores was seen reduced risk of Alzheimer’s disease in adults over the age of between smokers and nonsmokers (65.7 versus 72.2, 65 [23]. Hernández et al. performed a systematic review of p =063) (Table 4). However, patients with a history of vas- the literature between 2003 and 2013 [24]. They found that cular disease had significantly more constricted visual fields exercise may improve cognition and performance of daily than patients without vascular disease (46.2 versus 80.0, activities in patients with Alzheimer’s disease. Exercise has p =0011). Vascular disease remained associated with a also been reported to be beneficial in delaying the onset of reduced visual field even when controlling for age (p =0032). Parkinson’s disease and Huntington’s disease [22]. Sixty-six patients reported current employment out of Less than half of the adult patients included in this survey 142 respondents (46.5%). Current employment was associ- reported current employment. Not surprisingly, patients who ated with more favorable scores on the PAT-D, VFQ-25, were employed scored more favorably on QOL and disability and the Physical Component Summary of the SF-36 surveys. This high level of unemployment is indicative of the 4 Journal of Ophthalmology

Table 2: Relationship between quality of life survey scores and exercise levels.

Overall Active group Insufficiently active group t-test Mean score (SD) Mean score (SD) Mean score (SD) ∗ p value (n = 143)(n =81)(n =62) PAT-D 26.83 (12.24) 24.35 (8.91) 30.03 (14.99) 0.0096 Overall 49.77 (19.07) 53.32 (18.85) 45.12 (18.48) 0.0104 General vision 47.69 (25.33) 50.37 (26.29) 44.19 (23.79) 0.1491 Ocular pain 85.75 (19.21) 86.88 (19.95) 84.27 (18.24) Near activities 46.50 (26.65) 51.49 (25.22) 39.98 (27.25) Distance activities 40.88 (21.55) 44.86 (22.13) 35.69 (19.77) Social functioning 54.54 (29.20) 55.86 (28.72) 52.82 (29.98) VFQ-25 Mental health 46.37 (25.59) 47.69 (25.07) 44.66 (26.37) Role difficulties 57.69 (25.08) 60.34 (26.50) 54.23 (22.85) Dependency 52.68 (30.52) 57.10 (30.45) 46.91 (29.88) Driving 17.07 (28.28) 21.48 (31.40) 11.27 (22.56) Color vision 64.78 (33.84) 73.77 (32.09) 52.87 (32.63) 0.0002 Peripheral vision 29.79 (26.88) 34.26 (28.63) 23.75 (23.21) 0.0212 Physical component Summary 50.45 (10.59) 52.91 (8.84) 47.22 (11.82) 0.0022 Mental component Summary 51.38 (10.41) 51.11 (9.84) 51.73 (11.20) 0.7313 Physical functioning 84.11 (22.19) 90.06 (15.19) 76.31 (27.14) 0.0006 Role physical 75.18 (34.13) 80.00 (30.91) 68.85 (37.26) Role emotional 86.76 (28.42) 87.08 (28.32) 86.34 (28.79) SF-36 Vitality 60.07 (21.06) 61.19 (20.32) 58.61 (22.08) Mental health 77.84 (18.15) 78.94 (17.37) 76.39 (19.17) Social functioning 85.46 (22.78) 87.50 (20.86) 82.79 (25.02) 0.2367 Bodily pain 80.14 (22.76) 85.13 (19.23) 73.61 (25.40) General health 67.61 (22.28) 71.08 (21.69) 63.11 (22.40) 0.0365 ∗p values indicate comparisons between active and insufficiently active subjects. PAT-D: Pepper Assessment Tool for Disability; SD: standard deviation; SF-36: 36-Item Short Form Health Survey; VFQ-25: Visual Function Questionnaire-25.

Table 3: Relationship between Goldmann visual ﬁeld scores and exercise levels.

Active group Insufficiently active group ∗ p value Mean GVF score (SD) Mean GVF score (SD) † Initial measurement 74.78 (70.42) (n =64) 60.07 (70.17) (n =56) 0.2552 ‡ Initial measurement 90.19 (71.94) (n =31) 55.33 (56.29) (n =21) ‡ Final measurement 78.65 (62.71) (n =31) 48.71 (47.06) (n =21) 0.0689 ‡ Difference (final–initial) −11.54 (34.47) (n =31) −6.62 (50.41) (n =21) 0.6987 † ‡ ∗p values indicate statistical comparisons between active and insufficiently active subjects based on the Godin exercise score; entire group (n = 120); only includes subjects who completed follow-up testing (n =52); GVF: Goldmann visual field; SD: standard deviation. magnitude of disability seen in patients with advanced RP, a smoking cessation is typically recommended for patients problem that is only compounded in families with multiple with retinal diseases. affected members. In our study, vascular disease was found to be a signifi- Two-thirds of patients reported a history of cigarette cant risk factor for visual field loss. While retinal vascular smoking. It is unclear why such a high rate of smoking was attenuation in RP has traditionally been thought to occur sec- found although the wording of the questionnaire, which ondary to neuronal cell loss, a role for impaired ocular blood included both current and former smoking, may influence flow in the pathogenesis of RP has been suggested [25]. it. A case-control study from Korea found a significantly This study is limited by its retrospective nature as well as lower rate of current smokers among RP patients compared several other shortcomings. The survey response rate of to healthy controls [19]. It is notable that cigarette smoking 28.8% is relatively low, although almost all patients who were was not found to be associated with lower visual field scores successfully contacted chose to participate in the survey. The in our study. Smoking promotes oxidative stress, and majority of nonresponders were patients who could not be Journal of Ophthalmology 5

Table 4: Relationship between Goldmann visual ﬁeld scores and clinical factors.

Mean GVF score (SD) Mean GVF score (SD) p value∗ History of smoking No history of smoking Initial GVF measurement 65.67 (69.14) (n =79) 72.24 (73.43) (n =41) 0.6366 History of vascular disease No vascular disease Initial GVF measurement 46.21 (45.04) (n =43) 80.04 (78.88) (n =77) 0.0110 ∗p values indicate comparisons between presence and absence of clinical factors. GVF: Goldmann visual ﬁeld; SD: standard deviation.

Table 5: Relationship between quality of life survey scores and current employment status.

Overall Currently employed Not currently employed t-test Mean score (SD) Mean score (SD) Mean score (SD) ∗ p value (n = 143)(n =66)(n =76) PAT-D 26.83 (12.24) 24.03 (7.84) 29.25 (14.78) 0.0112 VFQ-25 Overall 49.77 (19.07) 56.71 (18.77) 43.78 (17.43) <0.0001 PCS 50.45 (10.59) 52.43 (8.93) 48.79 (1.69) 0.0424 SF-36 MCS 51.38 (10.41) 50.90 (9.54) 51.78 (11.25) 0.6208 ∗p values indicate comparisons between currently employed and not currently employed subjects. MCS: Mental Component Summary of the 36-Item Short Form Health Survey; PAT-D: Pepper Assessment Tool for Disability; PCS: Physical Component Summary of the 36-Item Short Form Health Survey; SD: standard deviation; SF-36: 36-Item Short Form Health Survey; VFQ-25: Visual Function Questionnaire-25. located due to out-of-date contact information in the medical studies, we could not establish a causative relationship in record. RP is a rare disease and most studies for RP are lim- patients with RP from our study. We are, however, cau- ited by sample size. While our study includes a relatively tiously encouraged by our findings that GVF scores trended large sample size, only a small number of patients have higher in the active patients, even though they did not reach follow-up visual fields for comparison. This small number statistical significance. RP is a devastating disease with no may have limited our ability to reach statistical significance curative treatment. The promising results of exercise [8, 9] in the visual field analysis. Furthermore, GVF scores and environmental enrichment [27, 28] studies in animals decreased over time in both groups. This indicates that warrant further investigation in humans. As a result, we increased physical activity does not completely halt progres- intend to pursue a prospective, randomized controlled study sion of disease. Further studies are warranted to better quan- to investigate the causative effect of exercise on progression tify the effect of exercise on the rate of disease progression. In of RP. We encourage further investigation of lifestyle modi- this study, frequency and duration of physical activity were fications such as exercise that may allow patients to have assessed at a single point in time. However, photoreceptor improved quality of life and potentially slow progression of degeneration may begin early in life even in late onset vision loss. symptomatic patients [1, 26]. Lifelong habits of exercise would be a better indicator of a patient’s exercise pattern rather than an assessment from any single point. Age was Disclosure not found to be significantly associated with visual field scores, although the association between age and physical This study is a poster presentation at the Association for activity level approached borderline significance. Given that Research in Vision and Ophthalmology annual meeting, visual field loss in RP is progressive with age, we recommend May 1, 2016, Seattle, WA. The authors alone are responsible controlling for age in further prospective evaluations of exer- for the content and writing of the paper. cise in RP. We aimed to report baseline exercise patterns in patients with RP and investigate the relationship between physical Conflicts of Interest activity and visual function reported from three quality of life The authors report no conflicts of interest. (QOL) surveys. To our knowledge, this study represents the first clinical investigation of exercise in RP. We have demonstrated that higher physical activity levels are associated with Acknowledgments greater self-reported visual function, as measured by the NEI VFQ-25, and lower levels of disability. Due to the limitations This study was supported by the NIH Core Grant EY006360, of a retrospective study, the authors caution that when inter- Department of Veterans Affairs Rehabilitation R&D Service preting the results, one should keep association in mind Research Career Scientist Award C9257S (MTP) and Career rather than causation. While we hypothesize that exercise Development Award B8034W, and Research to Prevent may have a neuroprotective effect in RP based on animal Blindness, NY, NY. 6 Journal of Ophthalmology

References [16] S. Amireault and G. Godin, “The Godin-Shephard leisure-time physical activity questionnaire: validity evidence supporting its [1] D. T. Hartong, E. L. Berson, and T. P. Dryja, “Retinitis pigmen- use for classifying healthy adults into active and insufficiently tosa,” Lancet, vol. 368, no. 9549, pp. 1795–1809, 2006. active categories,” Perceptual and Motor Skills, vol. 120, [2] B. J. Hahm, Y. W. Shin, E. J. Shim et al., “Depression and the no. 2, pp. 604–622, 2015. vision-related quality of life in patients with retinitis pigmen- [17] D. E. Rosenberg, C. H. Bombardier, S. Artherholt, M. P. tosa,” The British Journal of Ophthalmology, vol. 92, no. 5, Jensen, and R. W. Motl, “Self-reported depression and phys- pp. 650–654, 2008. ical activity in adults with mobility impairments,” Archives [3] C. P. M. Vivar and H. van Praag, “All about running: synaptic of Physical Medicine and Rehabilitation, vol. 94, no. 4, plasticity, growth factors and adult hippocampal neurogen- pp. 731–736, 2013. ” esis, Current Topics in Behavioral Neurosciences, vol. 15, [18] E. A. Hubbard, R. W. Motl, and P. J. Manns, “The descriptive – pp. 189 210, 2013. epidemiology of daily sitting time as a sedentary behavior in [4] J. A. Mares, R. P. Voland, S. A. Sondel et al., “Healthy lifestyles multiple sclerosis,” Disability and Health Journal, vol. 8, related to subsequent prevalence of age-related macular no. 4, pp. 594–601, 2015. degeneration,” Archives of Ophthalmology, vol. 129, no. 4, “ – [19] A. R. An, D. W. Shin, S. Kim et al., Health behaviors of people pp. 470 480, 2011. with retinitis pigmentosa in the Republic of Korea,” Ophthal- “ [5] P. T. Williams, Prospective study of incident age-related mac- mic Epidemiology, vol. 21, no. 5, pp. 279–286, 2014. ular degeneration in relation to vigorous physical activity dur- [20] D. F. Ross, G. A. Fishman, L. D. Gilbert, and R. J. Anderson, ing a 7-year follow-up,” Investigative Ophthalmology & Visual “Variability of visual field measurements in normal subjects Science, vol. 50, no. 1, pp. 101–106, 2009. and patients with retinitis pigmentosa,” Archives of Ophthal- “ [6] P. T. Williams, Relationship of incident glaucoma versus mology, vol. 102, no. 7, pp. 1004–1010, 1984. physical activity and fitness in male runners,” Medicine and [21] A. K. Bittner, M. A. Ibrahim, J. A. Haythornthwaite, M. Science in Sports and Exercise, vol. 41, no. 8, pp. 1566–1572, Diener-West, and G. Dagnelie, “Vision test variability in reti- 2009. ” “ nitis pigmentosa and psychosocial factors, Optometry and [7] P. T. Williams, Prospective epidemiological cohort study of Vision Science, vol. 88, no. 12, pp. 1496–1506, 2011. reduced risk for incident cataract with vigorous physical activ- “ ity and cardiorespiratory fitness during a 7-year follow-up,” [22] C. W. Cotman, N. C. Berchtold, and L. A. Christie, Exercise builds brain health: key roles of growth factor cascades and Investigative Ophthalmology & Visual Science, vol. 50, no. 1, fl ” pp. 95–100, 2009. in ammation, Trends in Neurosciences, vol. 30, no. 9, pp. 464–472, 2007. [8] E. C. Lawson, M. K. Han, J. T. Sellers et al., “Aerobic exercise “ protects retinal function and structure from light-induced ret- [23] M. W. Beckett, C. I. Ardern, and M. A. Rotondi, A meta- inal degeneration,” The Journal of Neuroscience, vol. 34, no. 7, analysis of prospective studies on the role of physical activity ’ ” pp. 2406–2412, 2014. and the prevention of Alzheimer s disease in older adults, [9] A. M. Hanif, E. C. Lawson, M. Prunty et al., “Neuroprotective BMC Geriatrics, vol. 15, no. 1, p. 9, 2015. effects of voluntary exercise in an inherited retinal degenera- [24] S. S. Hernández, P. F. Sandreschi, F. C. Silva et al., “What are tion mouse model,” Investigative Ophthalmology & Visual the benefits of exercise for Alzheimer’s disease? A systematic Science, vol. 56, no. 11, pp. 6839–6846, 2015. review of the past 10 years,” Journal of Aging and Physical – [10] C. M. Mangione, S. Berry, K. Spritzer et al., “Identifying the Activity, vol. 23, no. 4, pp. 659 668, 2015. content area for the 51-item National Eye Institute Visual [25] Y. Murakami, Y. Ikeda, M. Akiyama et al., “Correlation Function Questionnaire: results from focus groups with visu- between macular blood flow and central visual sensitivity in ally impaired persons,” Archives of Ophthalmology, vol. 116, retinitis pigmentosa,” Acta Ophthalmologica, vol. 93, no. 8, no. 2, pp. 227–233, 1998. pp. e644–e648, 2015. [11] C. M. Mangione, P. P. Lee, P. R. Gutierrez et al., “Development [26] E. L. Berson, “Retinitis pigmentosa. The Friedenwald Lecture,” of the 25-item National Eye Institute Visual Function Ques- Investigative Ophthalmology & Visual Science, vol. 34, no. 5, tionnaire,” Archives of Ophthalmology, vol. 119, no. 7, pp. 1659–1676, 1993. – pp. 1050 1058, 2001. [27] I. Barone, E. Novelli, I. Piano, C. Gargini, and E. Strettoi, [12] W. J. Rejeski, W. H. Ettinger Jr., S. Schumaker, P. James, R. “Environmental enrichment extends photoreceptor survival Burns, and J. T. Elam, “Assessing performance-related disabil- and visual function in a mouse model of retinitis pigmentosa,” ity in patients with knee osteoarthritis,” Osteoarthritis and PloS One, vol. 7, no. 11, article e50726, 2012. – Cartilage, vol. 3, no. 3, pp. 157 167, 1995. [28] A. Sale, N. Berardi, and L. Meffei, “Enrich the environment to [13] W. J. Rejeski, E. H. Ip, A. P. Marsh, M. E. Miller, and D. F. empower the brain,” Trends in Neurosciences, vol. 32, no. 4, Farmer, “Measuring disability in older adults: the Interna- pp. 233–239, 2009. tional Classification System of Functioning, Disability and Health (ICF) framework,” Geriatrics & Gerontology Interna- tional, vol. 8, no. 1, pp. 48–54, 2008. [14] J. E. Ware Jr. and B. Gandek, “Overview of the SF-36 Health Survey and the International Quality of Life Assessment (IQOLA) Project,” Journal of Clinical Epidemiology, vol. 51, no. 11, pp. 903–912, 1998. [15] G. Godin and R. J. Shephard, “A simple method to assess exercise behavior in the community,” Canadian Journal of Public Health, vol. 10, no. 3, pp. 141–146, 1985.