Ophthalmic Molecular Genetics

Objective  To evaluate the molecular genetic defects associated with retinitispunctata albescens (RPA) in 5 patients from 3 families with this disease.

Methods  We examined 3 probands and 2 clinically affected relatives with RPA.Clinical examinations included best-corrected visual acuity, visual fieldtesting, electroretinography, dilated fundus examination, and fundus photography.Leukocyte DNA was analyzed for mutations in the exons of the genes encodingcellular retinaldehyde–binding protein 1 (RLBP1),11-cis-retinol dehydrogenase (RDH5), interphotoreceptor retinoid–binding protein (RBP3), and photoreceptor all-trans-retinoldehydrogenase (RDH8). Not all patients were evaluatedfor mutations in each gene. The exons were individually amplified and screenedfor mutations by single-stranded conformational polymorphism analysis or directgenomic sequencing.

Results  The 3 probands had similar clinical findings, including a history ofpoor night vision, the presence of punctate white deposits in the retina,and substantially reduced or absent rod responses on electroretinogram testing.One of the probands (patient 2:III:2) had 2 novel mutations in the RLBP1 gene (Arg151Trp and Gly31[2–base pair deletion], [GGA→G–]).Segregation analysis showed that the 2 mutations were allelic and that thepatient was a compound heterozygote. Both parents of the proband manifestedround white deposits in the retina. The other 2 probands had no detected pathogenicmutations in RLBP1 or in the other 3 genes evaluated.

Conclusions  The identification of novel RLBP1 mutationsin 1 of our 3 probands, all with RPA, is further evidence of genetic (nonallelic)heterogeneity in this disease. The presence of round white deposits in theretina may be observed in those heterozygous for RLBP1.

Clinical Relevance  Patients with a clinical presentation of RPA can have genetically differentmutations. Drusen-like lesions may be observed in heterozygotes in familieswith this disease and a mutation in RLBP1.

Retinitis punctata albescens (RPA) is traditionally defined as an autosomalrecessive disease characterized by nyctalopia, decreased visual acuity, multipleround white deposits in the retina, progressive attenuation of retinal arterioles,abnormal fundus pigmentation, progressive restriction of visual fields (VFs),and nondetectable or severely reduced electroretinogram (ERG) amplitudes.^1,2 Mutations in the gene encoding cellularretinaldehyde–binding protein 1 (RLBP1),^3-8 peripherin/RDS,⁹ and rhodopsin¹⁰ have been found in patients with RPA.

In 3 probands with RPA, we found 1 with 2 novel RLBP1 mutations. Because RLBP1 is involved in themetabolism of vitamin A in the retina^11-13 andbecause defects in genes encoding other proteins in this pathway have beenreported to cause retinal diseases with yellow or white retinal deposits (eg, ABCA4 and RDH5),^14,15 wesubsequently analyzed 3 additional genes that encode proteins in this pathwayfor mutations in the 2 probands with no identified RLBP1 mutations.

This project, which involved human subjects, conformed to the tenetsof the Declaration of Helsinki and was approved by the institutional reviewboard at the University of Illinois at Chicago. We evaluated 3 probands and2 clinically affected relatives with RPA. Clinical examinations included best-correctedSnellen visual acuity, dilated fundus examination by direct and indirect ophthalmoscopy,Goldmann VFs, and ERGs. Visual field examination was performed monocularlywith a Goldmann perimeter using the II-2-e, II-4-e, and V-4-e test targets.The targets were moved from a nonseeing region to a seeing region. Not allof the patients were tested with all of these targets.

An ERG was obtained by either of 2 procedures previously described.^16,17 The recording techniques adheredto the international standard for clinical electroretinography establishedby the International Society for Clinical Electrophysiology of Vision.¹⁸ All recordings were obtained with fully dilated pupilsand the use of a monopolar Burian-Allen contact lens. Scotopic responses wereobtained after 40 minutes of dark adaptation. Two patients were additionallytested after 17 hours of dark adaptation.

Three patients (1:III:11, 1:III:12, and 1:III:1) were from one Palestinianfamily (family 1, Figure 1 and Table 1), while the remaining 2 patients(2:III:2 and 3:III:3) were from separate African American and Eastern Europeanfamilies (families 2 and 3, respectively) (Figure 2, Figure 3, and Table 1). All 5 patients had night blindnessin childhood, 1 (1:III:1) as early as 3 years of age and 2 (2:III:2 and 1:III:11)by 4 years of age. The best-corrected visual acuities ranged from 20/20 to20/50 in all 5 patients. The spherical equivalent refractive errors of thesubjects ranged from +3.75 to −1.00 diopters.

Leukocyte DNA was analyzed for mutations in the 9 exons of RLBP1 and the 5 exons of RDH5 by single-strandedconformational polymorphism analysis and direct genomic sequencing accordingto previously reported methods.^7,15 Wealso searched for mutations in the genes encoding interphotoreceptor retinoid–bindingprotein (RBP3) and photoreceptor all-trans-retinol dehydrogenase (RDH8); oligonucleotideprimers and polymerase chain reaction conditions for amplifying the exonsof these genes are available on the Internet.¹⁹ Becauseour 3 families did not show an autosomal dominant mode of transmission, wedid not evaluate our patients for mutations in the rhodopsin or peripherin/RDS genes.

At age 21 years, the proband in family 1 (1:III:11, Figure 1) showed round punctate white deposits in the midperipheralretina, fewer such lesions in the posterior pole, and normal optic disc, fovea,and retinal vessels. No pigmentary abnormalities were noted (Figure 4). Her sister (1:III:12, Figure 1), at age 9 years, showed multiple white spots temporalto the macula in each eye, with normal optic disc and retinal vessels (Figure 5A). In addition, a pepper-like mottlingwas observed, most apparent in the nasal midperipheral retina (Figure 5B). A cousin (1:III:1, Figure 1), age 7 years, showed numerous white spots throughout theretina, with a normal optic disc and attenuated retinal vessels. Moderatepigment granularity, particularly in the inferior retina, was also noted (Figure 6). This patient's 47-year-old fatherand 32-year-old mother were each examined and found to have vision correctableto 20/20 in each eye. Neither showed the presence of white spots in the retinaof either eye. The father showed normal ERG cone and rod a- and b-wave amplitudes.

The fundus examination of the proband in family 2 (2:III:2, Figure 2), age 7 years, showed white depositsin the posterior pole and midperipheral retina (Figure 7). Retinal pigmentary abnormalities, including hypopigmentation,pigment mottling, and clumping, were also noted in the midperipheral retina.The retinal arterioles were not attenuated. The patient's 35-year-old motherand 36-year-old father were each examined and found to have white spots inthe posterior fundus, including nasal to the optic disc. Both parents hadvision correctable to 20/20 or better in each eye, and neither complainedof difficulty with night vision. The number of white spots was more numerousin the mother (Figure 8), who harboreda frameshift mutation, than in the father, who showed more isolated lesionsand had a missense mutation. Cone and rod ERG amplitudes were normal in eachparent.

The fundus examination of the proband in family 3 (3:III:3, Figure 3), age 49 years, showed a bluntedfoveal reflex, with mild retinal arteriole attenuation and mild waxy pallorof the optic disc. White spots were evident predominantly at and anteriorto the vascular arcades (Figure 9A).Sparse pigmentary clumping, mainly in the inferior and midperipheral retina,was observed (Figure 9B). Isolatednummular lesions of chorioretinal atrophy, which were approximately threequarters of a disc diameter in size, were found in the inferior retina ofeach eye.

Visual field testing with a Goldmann perimeter showed full VFs or onlymild restriction to the II-4-e stimulus in all 5 patients. Two of the patients(1:III:1 and 3:III:3) showed a restriction to the II-2-e target. One patient(3:III:3) had central scotomas bilaterally.

The dark-adapted isolated rod ERG response to a low-intensity blue stimuluswas nondetectable in 4 of the 5 patients (1:III:1, 1:III:12, 2:III:2, and3:III:3; Figure 10). In patient2:III:2, this was observed even after 17 hours of overnight patching. Thepatient's parents each showed normal ERG cone and rod amplitudes. One patient(1:III:11) showed a 30% increase in the isolated rod b-wave amplitude after17 hours of dark adaptation compared with the amplitude after 40 minutes ofdark adaptation (Figure 11). Thedark-adapted maximal ERG b-wave amplitude after 40 minutes of dark adaptationwas within 60 µV of the light-adapted high-intensity b-wave amplitudein 4 of the 5 patients, reflecting a marked impairment of rod function (Table 1).

The proband of family 2 (2:III:2) carried 2 mutations in RLBP1. In exon 4, there was a frameshift mutation (Gly31[2–basepair deletion], [GGA→G–], complementary DNA bases 92-93delGA),and in exon 6, there was a missense mutation (Arg151Trp, CGG→TGG, complementaryDNA sequence 451C→T) (Figure 12A).Segregation analysis in the patient's family showed that the 2 mutations wereallelic and that the patient was a compound heterozygote (Figure 12B). The other 2 probands had no likely pathogenic mutationsin RLBP1 detected by single-stranded conformationalpolymorphism analysis or direct sequencing. Proband 3:III:3 was homozygousfor a change in intron 7 (IVS7 + 20C→T) that was interpreted as likelynot to be pathogenic, because it does not appear to create or destroy a splicesite, based on splice-site prediction software.²⁰ Theprobands without identified RLBP1 mutations (1:III:11and 3:III:3) were screened for mutations in the RDH5, RDH8, and RBP3 genes. Patient 1:III:11 was homozygousfor an isocoding polymorphism (Ile141Ile, ATC→ATA, 423C→A) in exon3 of the RDH5 gene; no RDH5 sequencechanges were found in patient 3:III:3. For RDH8,patient 1:III:11 was heterozygous at both of 2 bases in codon 202 (complementaryDNA bases 116-117); depending on the phase of these changes, codon 202 hadthe allelic sequences ATG (Met) and ACA (Thr), ATA (Ile), or ACG (Thr). Thealleles ATG, ACA, and ACG have been observed in additional patients we haveevaluated in other investigations of this gene, but the ATA allele has neverbeen encountered (data not shown). Patient 3:III:3 was heterozygous at 3 sites:Leu267Leu (CTC→CTT, 743C→T), Tyr277His (TAT→CAT, 829T→C),and Met202Thr (ATG→ACG, 116T→C). Each of these 3 variations is anonpathogenic polymorphism in the RDH8 gene thatwe have found in other non-RPA individuals (data not shown). No changes inthe RBP3 gene were found in patient 1:III:11 or inpatient 3:III:3.

Maw et al²¹ were the first to describepatients with recessive mutations in the RLBP1 genein a retinal degeneration associated with diffuse "small white dots" throughoutthe fundus and the absence of bone spicule-like pigmentation. Subsequently,Burstedt et al⁵ and Morimura et al⁷ described features of RPA in patients with mutationsin RLBP1. Other patients with RPA from Saudi Arabia³ and Newfoundland⁸ havebeen described with mutations in RLBP1. In additionto the clinical findings typically found in patients with RPA, those homozygousfor the RLBP1 mutation Arg234Trp from northern Swedenhave a predilection to exhibit an atrophic-appearing macular lesion, particularlyin those older than 30 years.⁵ This form ofRPA has been referred to as Bothnia dystrophy.^5,6 A geographic atrophic macular lesionwas not a feature observed in our patients with RPA; however, circular areasof geographic atrophy in the peripheral retina, a common finding among adultpatients with Bothnia dystrophy,^5,6 wereobserved in our oldest proband (3:III:3). The lesions were similar to thosein a 52-year-old patient with RPA and an RLBP1 mutationdescribed by Morimura et al.⁷

Although certain phenotypic similarities exist between patients withRPA and fundus albipunctatus, as a group, these patients differ in certainimportant clinical and genetic features. Although younger patients with eitherof these disorders complain of nyctalopia and typically show multiple roundwhite deposits within the retina, most patients with fundus albipunctatushave a nonprogressive disease, ie, no deterioration in photoreceptor cellfunction, no attenuation of retinal vessels, and no pigment clumping in theretina. However, other patients with fundus albipunctatus and mutations in RDH5 have been described as having a cone dystrophy, leadingto loss of central visual acuity and reduced cone ERG amplitudes; these findingswere most often observed in those older than 30 years.²² Particularlyin younger patients with characteristic features of fundus albipunctatus,cone and rod function can revert to normal after a prolonged period of darkadaptation.²³ In comparison, those with RPAhave a progressive loss of photoreceptor cell function and, not infrequently,develop attenuated retinal vessels and pigmentary clumping. Furthermore, mutationsin different genes have been identified in these 2 diseases. Of interest,one of our patients with RPA showed a small partial recovery of rod functionafter prolonged dark adaptation. A partial recovery in dark adaptation hasalso been observed in younger patients with Bothnia dystrophy. In contrast,in our patient with an RLBP1 mutation, we did notobserve an improvement in dark-adapted ERG responses even after 17 hours ofdark adaptation.

Our findings suggest that the presence of a variable number of smallwhite spots in the fundus may be a diagnostic feature observed in certainheterozygotes of this disease. These lesions may have a phenotypic appearancesimilar to drusen of the Bruch membrane. Their rather haphazard distribution,absence of a glistening (calcified) appearance, and associated pigmentarychanges, in an individual younger than 40 years, are distinguishing features.Whether this finding is specific to RPA families with an RLBP1 gene mutation is yet to be verified by investigating a largernumber of RPA families with or without a demonstrable RLBP1 mutation.

The novel mutations in RLBP1 observed in oneof our patients, the absence of a demonstrable mutation in this gene in patientsfrom 2 other families, and reports on genetic studies in RPA patients fromthe existing literature further underpin the genetic heterogeneity of RPA.Genetic screening may be necessary to discriminate between some patients withRPA and those with fundus albipunctatus, especially in those with early stagesof RPA. Genetic heterogeneity exists in both of these disorders.

Corresponding author: Gerald A. Fishman, MD, Department of Ophthalmologyand Visual Sciences, UIC Eye Center, University of Illinois at Chicago, MailCode 648, 1855 W Taylor St, Chicago, IL 60612 (e-mail: gerafish@uic.edu).

Submitted for publication April 1, 2003; final revision received August15, 2003; accepted August 25, 2003.

This study was presented in part at the annual meeting of the Associationfor Research in Vision and Ophthalmology; May 6, 2002; Ft Lauderdale, Fla.

This study was supported by funds from The Foundation Fighting Blindness,Owings Mills, Md; Grant Healthcare Foundation, Chicago, Ill; grants EY01792and EY08683 from the National Institutes of Health, Bethesda, Md; and an unrestricteddepartmental grant from Research to Prevent Blindness, New York, NY.