Electrophysiologic and phenotypic features of an autosomal cone-rod dystrophy caused by a novel CRX mutation

Electrophysiologic and Phenotypic Features of an Autosomal Cone–Rod Dystrophy Caused by a Novel CRX Mutation Matthew A. Lines, BSc, Marc He´bert, PhD, Kerry E. McTaggart, MSc, Sarah J. Flynn, BSc, Matthew T. Tennant, MD, Ian M. MacDonald, MD, CM Purpose: To reexamine a large Albertan family previously reported with a progressive cone dystrophy with variable phenotype and to map the disorder using molecular genetic techniques. Design: Observational case series. Participants: Twenty-nine subjects (10 affected) from four generations of a large kindred were clinically examined. Twenty-three of these individuals, as well as two unaffected spouses, were included in the molecular genetic study. Subject ages ranged from 17 to 91 years of age. Methods: Disease status and associated ocular abnormalities were assessed primarily by measurement of visual acuity, color vision, fundus photography, and both full-field and multifocal electroretinography (ERG and mfERG). Linkage of the disorder to the rhodopsin gene was studied using microsatellites. A mutational screen of the CRX gene was performed to identify coding sequence changes. Main Outcome Measures: Visual acuity and color discrimination were reduced in clinically affected individuals; full-field flash ERG was used to measure function of both cones and rods. mfERG and fundus photography allowed documentation of the observed macular changes. Results: We noted a variable, adult-onset macular dystrophy, progressing in some cases to a retinitis pigmentosa–like phenotype. Both photopic and scotopic full-field ERG amplitudes were reduced by approximately 50%, demonstrating involvement of both photoreceptor systems. A reduced b-wave amplitude with a relatively preserved a-wave was observed at both cone and rod levels. Macular involvement was confirmed by mfERG. The rhodopsin locus was excluded by haplotype analysis. A novel frameshift mutation was detected in exon III of the CRX retinal homeobox gene. ERG and molecular genetic findings were consistent with the reclassification of this disease as an autosomal dominant cone–rod dystrophy (CRD) Conclusions: We report a novel CRX mutation causing autosomal dominant CRD. Observed ERG changes suggest that this mutation primarily impairs inner retinal function. Because retinal expression of CRX is limited to photoreceptors, this dysfunction may be the result of faulty photoreceptor communication with second-order retinal neurons. We propose misexpression of gated cation channels caused by altered CRX activity as one putative mechanism by which a sole photoreceptor defect may selectively impair neurotransmission without disrupting the upstream events of phototransduction. Ophthalmology 2002;109:1862–1870 © 2002 by the American Academy of Ophthalmology. Retinal dystrophies are a genetically heterogeneous and clinically variable family of disorders. Cone–rod dystrophy (CRD), in particular, may present as clinically distinct retinal dystrophies even within a single kindred. When examined individually, cases of CRD often resemble bull’s-eye or pattern macular dystrophy, which may lead to the diagnosis of cone or macular dystrophy. In contrast, advanced Originally received: August 22, 2001. Accepted: April 18, 2002. Manuscript no. 210729. From the Department of Ophthalmology, University of Alberta Faculty of Medicine and Dentistry, Edmonton, AB, Canada. Preliminary findings of this study were previously presented at the Macula Society, Phoenix, Arizona, March 2001, and at the International Society for Clinical Electrophysiology in Vision, Magog, Quebec, June 2001. Supported by the University of Alberta Hospitals Foundation, Edmonton, Alberta, Canada. Reprint requests to Dr. Ian M. MacDonald, 8-32 Medical Sciences Building, University of Alberta, Edmonton AB, Canada T6G2H7.

1862

© 2002 by the American Academy of Ophthalmology Published by Elsevier Science Inc.

cases may produce tapetoretinal dystrophic changes more consistent with retinitis pigmentosa. However, electrophysiologic features demonstrating both diffuse cone and rod dysfunction may clinically distinguish CRD from cone dystrophies and other forms of macular degeneration. In 1975, Pearce1 reported a family with several cases of retinal degeneration, with a variable phenotype consistent with a dominant progressive cone dystrophy. Clinical follow-up using contemporary electrodiagnostic techniques has supported the reclassification of this disorder as an autosomal dominant progressive CRD. In light of the variable and newly expanded phenotype encountered in this particular family, we attempted to confirm or exclude linkage of the disorder to a panel of candidate loci previously associated with similar retinal photoreceptor dystrophies. We chose the human rhodopsin (RHO) gene2 as a starting point for our candidate linkage analysis because of the resemblance between the tapetoretinal dystrophy phenotype presented in this study (see Results) and that of RHO-linked ISSN 0161-6420/02/$–see front matter PII S0161-6420(02)01187-9

Lines et al 䡠 Autosomal Dominant Cone–Rod Dystrophy Caused by a CRX Mutation retinitis pigmentosa (RP).3–5 The human RHO locus encodes an opsin that forms the major protein component of rod outer segment membranes and that initiates the events of photon capture and phototransduction in retinal rods. RHO-linked mutations, primarily autosomal dominant, are observed in roughly one of four cases of human RP.3–5 A second candidate gene, the cone–rod homeobox transcription factor (CRX), has reported mutant phenotypes diagnosed as autosomal dominant (adCRD2),6 – 8 Leber congenital amaurosis,6,9 –13 and late-onset dominant RP.6 The CRX gene’s three exons encode a paired-type homeodomain transcription factor (TF) homologous to human OTX1 and OTX2.8,14,15 CRX is expressed abundantly in retinal photoreceptors8,14,15 and in pineal gland pinealocytes, where it regulates pineal-specific gene expression in the circadian modulation of melatonin production.16 In photoreceptor cells, CRX has been implicated in the activation of a retinaspecific transcriptional network that contains multiple phototransduction pathway genes, with products including RHO, the cone opsins, transducin, phosducin, rod and cone arrestins, recoverin, and cyclic guanosine monophosphate (cGMP)–specific phosphodiesterase ␤.14,15,17,18 Two other significant CRX targets in retina include an outer segment structural protein (retinal degeneration, slow peripherin), and a soluble chromophore carrier, interphotoreceptor retinoid-binding protein. Not surprisingly, similar retinal dystrophies result from mutations within such downstream targets. RDS/peripherin mutations are themselves common causes of autosomal dominant macular dystrophy and RP,19 –21 whereas RHO mutations produce either an RP 3–5, 22 phenotype or cause complete congenital stationary night blindness.23,24 In addition, CRX affects the expression of at least two cGMP-gated cation channel subunits in retinal rods (CNGA1 and CNGA3)17 and may regulate at least one such gene in cones (CNGB1),25 alluding to a role in modulating ion balance and membrane potentials in both types of photoreceptors. Although we initially hoped to analyze other relevant CRD candidate genes such as RDS/peripherin and GUC2D, our molecular genetic analysis has mapped the CRD phenotype of the studied family to a novel, heterozygous frameshift mutation in exon III of the CRX gene. This mutation is predicted to truncate vital transactivation and protein–protein interaction domains in the C-terminal half of the protein. A presumed decrease in CRX-mediated transcriptional activation is likely to produce compound effects on the expression of numerous photoreceptor-specific genes. In addition, we report that this mutation causes a specific reduction of the electroretinogram (ERG) b-wave in CRDaffected subjects of this kindred, suggesting inner retinal dysfunction that may be explained by impaired neurotransmission at the level of retinal photoreceptor cells.

Materials and Methods Subjects This article is based on a clinical and molecular reexamination of four generations of a large Albertan kindred whose phenotype was

initially described by Pearce1 in 1975 as a progressive autosomal dominant cone dystrophy. Ongoing ophthalmic examination has since allowed better characterization of the disease course in this family. Twenty-nine individuals were clinically reexamined in this study, 23 of whom also participated in the DNA mapping analysis. An additional two spouses with normal vision (III-6 and III-11) were involved in the molecular study as controls. Both the genotype and ocular phenotype of 10 affected and 15 clinically unaffected persons were studied in total. Criteria for the assignment of subjects to the affected group included poor central vision, color vision loss, and funduscopic signs of maculopathy. Because this disorder is progressive and has a highly variable adult age of onset, assignment of at-risk individuals as unaffected is tentative and dependent on ongoing clinical verification by the methods outlined in the following. Human subject participation and receipt of informed consent from each individual were approved through the University of Alberta Health Ethics Review Board.

Electrophysiologic Testing All subjects underwent identical laboratory ERG and multifocal ERG (mfERG). Electrical signals were recorded using a DTL electrode, which was positioned deep within the conjunctival sac during ERG and placed riding the lower eyelid for mfERG. Pupils were dilated (tropicamide 1%) during both tests. Corneas were anesthetized (proparacaine 1%) before the mfERG only. ERG recordings were obtained using a LKC UTAS ⫺3000 biologic system (LKC Technologies Inc., Gaithersburg, MD) set at a signal bandwidth of 0.3 to 500 Hz or 75 to 500 Hz for determination of oscillatory potentials (OPs). Four to 16 single responses were averaged before amplitude measurements were obtained. Photopic responses were obtained using a standard flash set at 4 cd/m2s.sec and a rod-desensitizing background (Ganzfeld) set at a luminance of 30 cd/m2. Scotopic mixed cone–rod ERG responses were obtained after 20 minutes of dark adaptation using bright single flashes presented in a Ganzfeld stimulator. At the standard flash intensity, a strong a-wave and b-wave ERG component can be observed. For the pure rod response, the standard flash was lowered by 2.5 log units. At that intensity only the b-wave is observed. ERG parameters analyzed included the b-wave amplitude, calculated from baseline or trough of a-wave (if present) to peak of b-wave voltage. A-wave (when present) and b-wave latencies were also measured. Both eyes were recorded in all patients. Because our observation seemed symmetrical, for simplicity, we are presenting only data from the right eye. The mfERG was performed using a VERIS Science 4 system (EDI Inc., San Mateo, CA). The stimulus matrix consisted of 61 hexagons scaled with eccentricity and was presented with a 7-inch monitor. This particular stimulus allowed us to record 61 localized cone ERGs from the macular and paramacular regions up to 20° from central. The stimulus was viewed through an eye-camera (fitted to the monitor), which allowed us to monitor eye movement and correct for refractive errors. The luminance of hexagons in the white state was set at 500 cdm⫺2, the background was set at 250 cdm⫺2, and the fixation target (a cross) was set at 50 cdm⫺2. Signal was amplified 100,000 times, and the bandwidth was set at 10 to 100 Hz. We recorded a total of 4 minutes of data split into 16 segments of 15 seconds each. Only first-order ERG waveforms, which represent activity generated by cone photoreceptors and bipolar/Mu¨ ller cells, are presented to demonstrate both the macular and paramacular deficits that were observed in CRD-affected subjects. These (first order) focal ERGs display a negative deflection (N1), followed by a positive deflection (P1). Similar to conventional a-wave and b-wave components of the ERG, these wave-

1863

Ophthalmology Volume 109, Number 10, October 2002 Table 1. Phenotypes of Cone–Rod Dystrophy Affected Subjects by Age of Last Examination Generation IV-7 IV-5 IV-6 V-2 IV-8 IV-3 IV-2 IV-1 IV-11 IV-10 III-1 IV-9 IV-4* II-3 III-7 III-5

Age

Acuity (Right Eye)

Acuity (Left Eye)

Ishihara Plates Correct

23 27 28 33 34 46 47 51 52 57 62 62 67 81 87 90

20/240 20/100 20/240 20/20 20/400 20/400 HM 20/300 20/250 20/60 20/400 20/160 20/25 NLP 20/400 20/200

20/240 20/200 20/240 20/20 20/400 20/400 HM 20/30 20/250 20/40 20/50 20/160 20/25 NLP 20/400 20/70

8/16 OU 0/15 OU 0/15 OU 16/16 OU 5/15 OU 1/15 OU 0/15 OU 1/15 OD; 8/15 OS 0/16 OU 15/15 OU Not tested 10/15 OD; 4/15 OS 6/15 OD; 7/15 OS Not tested 2/15 OU 0/15 OU

HM ⫽ perceived hand motion only; NLP ⫽ no light perception; OD ⫽ right eye; OS ⫽ left eye; OU ⫽ both eyes. * Individual IV-4 showed a color discrimination deficit not attributable to reduced acuity.

forms demonstrate activity generated by cone photoreceptors and by bipolar/Mu¨ ller cells, respectively.

DNA Extraction DNA for molecular genetic analyses was obtained by standard organic extraction from peripheral blood lymphocyte lysates. Mutational screening of each individual for the diagnostic CRX mutation was performed twice from duplicate samples to minimize the possibilities of error and cross-contamination.

quencing was carried out as per instructions using ␣33P-ddNTP terminators in dGTP-based dNTP mixture to obtain an unambiguous bidirectional sequence for each of the five fragments. Sequencing products were subjected to 6% denaturing polyacrylamide gel electrophoresis and visualized by exposure to KODAK BIOMAX-MR autoradiographic film. A single 1-bp insertion was identified at codon 149 of exon III of the gene. No other sequence variations were observed.

Results Linkage Mapping to Rhodopsin We examined linkage of the CRD phenotype to the dinucleotide repeat markers AMF199xd6 (GDB#188278), AFMb020zb9 (#609402), and AFMa064vb1 (#614178),26,27 which encompass a 5.9-Mb region on 3q23 containing the RHO gene. Each chosen marker has an estimated maximum heterozygosity greater than 0.8. Leukocyte DNA from each examined individual was amplified by polymerase chain reaction with random incorporation of ␣-35S deoxyadenosine triphosphate label into product fragments. Polymerase chain reaction products were subjected to 6% denaturing polyacrylamide gel electrophoresis and visualized by exposure to KODAK BIOMAX-MR autoradiographic film at ⫺80° C. Haplotypes were constructed as definitively as possible by inspection. No internal recombination was observed within marker haplotypes, suggesting that RHO alleles remained in-phase with the markers through each generation.

Reexamination of affected individuals within this family presents a redefined clinical picture of progressive adCRD. Measures of bilateral visual acuity and color discrimination are presented along with the pedigree and individual disease status assignments in Table 1 and Figure 1. Age of onset varied significantly in the cases studied, ranging approximately between the second and sixth decades. Acuity changes and macular degeneration were preceded by nyctalopia and/or reduced color discrimination in several cases. Although it was noted in the earlier clinical description of the family1 that one individual (here designated III-5) seemed nonpenetrant, he has since developed vision loss caused by maculopathy (Fig 2, Table 1). Apparent nonpenetrance in such cases may

CRX Exon Amplification and Sequencing To determine any CRX coding sequence difference between affected and unaffected individuals, we amplified and sequenced the three exons of the gene, including exon–intron boundaries, in five overlapping fragments using oligonucleotide polymerase chain reaction primers reported by Sohocki et al.6 Product bands to be used as templates for cycle sequencing were excised and extracted by QIAquick Spin gel extraction kit (QIAGEN, Hilden, Germany) column purification as per manufacturer instructions. Manual Thermo Sequenase (Amersham: Buckinghamshire, UK) cycle se-

1864

Figure 1. Pedigree demonstrating autosomal dominant inheritance of cone–rod dystrophy in a large kindred. Examined individuals are denoted by symbols bearing horizontal bars.

Lines et al 䡠 Autosomal Dominant Cone–Rod Dystrophy Caused by a CRX Mutation Exclusion of the Rhodopsin Locus To examine the role of rhodopsin in the heritable pathogenesis of this disease, we conducted a haplotype-based linkage study of three polymorphic microsatellite markers26,27 spanning a 5.9-Mbp region on chromosome 3 tightly linked to RHO. Through this analysis we were able to exclude RHO as a candidate gene for the disease. Several meioses showed recombination between the CRD phenotype and transmitted parental haplotypes (Fig 5, Table 3).

The CRX Gene Is Mutated in CRD-Affected Individuals

Figure 2. Fundus photos of cone–rod dystrophy–affected patients show a progressive macular to general retinal dystrophy. All eyes shown are right eyes. A, Individual III-5, age 62. B, Individual III-5, age 90, showing macular dystrophy; acuity 20/200. C, Individual IV-9, age 34, displaying macular chorioretinal atrophy. D, Individual II-3, age 81, displaying retinitis pigmentosa–like chorioretinal dystrophy; no light perception.

reflect the variable age of onset of the CRD phenotype in this disorder, because individual V-2, who seemed clinically normal at age 33, carried the causative mutation (detailed later). The phenotype of this family was variable, with some affected members showing only mildly reduced visual acuity, color loss, and minimal macular changes. Fundus photographs (Fig 2) of several cases showed significant areas of macular chorioretinal atrophy. In the most severe case, an RP-like generalized retinal atrophy and loss of light perception represents the phenotypic endpoint of progressive CRD.

ERG and mfERG Findings Full-field ERG demonstrated a diffuse cone and rod dysfunction, supporting the reclassification of the family’s retinal cone dystrophy as a CRD. A consistent reduction of both photopic and scotopic b-wave amplitudes of approximately 60% and 40%, respectively, were observed in all affected members (Fig 3, Table 2). The photopic ERG demonstrated delayed a-wave and b-wave peak times and a relatively severe reduction of the second oscillatory potential (OP2) in three of four affected individuals (Fig 3B). In addition, in both the photopic and scotopic bright-flash (standard flash) ERGs (Figs 3A, 3D), the amplitude of the b-wave was selectively reduced, whereas the a-wave remained relatively well preserved (Table 2). Inspection of the 61 focal ERG responses (Fig 4) derived from the mfERG (first-order kernel) indicated dramatic central retinal dysfunction in all patients. Paramacular dysfunction was also observed, albeit to a lesser extent, consistent with the observation of both macular and diffuse cone system damage. Note that the protocol used in the present mfERG does not allow for the assessment of rod function.

To the best of our knowledge, all reported pathogenic mutations in the human CRX gene fall in coding regions of its three exons. This consideration allowed us to conduct a simple mutational screen of the gene’s coding sequences by means of direct sequencing of five overlapping polymerase chain reaction products. Sequence comparisons between affected and unaffected subjects presented a novel heterozygous insertion of an additional cytosine residue into a (C)4 stretch between ⫹444 and ⫹447 in exon III of the CRX gene in two affected individuals (Fig 6). Mutational analysis, when extended to the entire kindred, confirmed a clear pattern of cosegregation of the CRD phenotype in phase with this frameshift insertion, with the single exception of individual V-2, who inherited the causative mutation but was not affected at age 33 (Fig 7). This apparent nonpenetrance may be related to the variable age of onset of this disorder. No additional mutations were present in exons or at normal splice junctions of the gene. The predicted effect of this mutation is a shift in reading frame beginning at residue 150, resulting in the truncation of the Cterminal 150 amino acids of the polypeptide and introduction of an out-of-frame 23 residue sequence ([150]LRLPNHGSGHCVHLEPSLRVPFA) followed by a premature nonsense codon. This large deletion abolishes several functionally significant domains of the CRX TF, while leaving the DNA-binding homeodomain and nuclear localization motifs intact.

Discussion The variable phenotype of this family was originally thought to consist of two discrete, coincident monogenic disorders (dominant macular dystrophy alongside a generalized tapetoretinal dystrophy).1 Here we have demonstrated RP-like retinal atrophy to be the endpoint of the phenotypic progression of this disease and have demonstrated the identity of this phenotype with adCRD2 as reported for other dominant mutations in CRX.6,7,11 In addition, this study underscores the clinical value of electrophysiology in accurate diagnosis of retinal dystrophies. Although fundus photographs of individual affected members were suggestive of a solely macular lesion, mfERG and full-field ERG observations indicated both macular and diffuse cone and rod dysfunction. Fundus reexamination over the course of clinical follow-up additionally demonstrated progressive degeneration of the peripheral and the central retina. The CRD phenotype in this family is due to a single base pair insertion in exon III of the retinal/pineal homeobox gene CRX, which introduces 23 out-of-frame amino acids and deletes the C-terminal half of the normal protein. This mutation does not affect the amino acid sequence of the

1865

Ophthalmology Volume 109, Number 10, October 2002

Figure 3. Electroretinographic records of cone–rod dystrophy. A, Photopic electroretinogram (ERG) showing decreased b-wave amplitude in affected subjects. B, Photopic oscillatory potential (OP1, OP2, OP3 indicated by arrows). Note increased a-wave and b-wave latencies (A) and reduction of the second oscillatory potential (OP2) in individuals III-5, IV-6, and IV-9 (B). C, Scotopic (dim-flash) ERG, representing pure rod response, with delayed, reduced b-wave. D, Scotopic (bright-flash) ERG, representing mixed cone–rod response, with reduced, delayed b-wave, accompanied by relatively normal a-wave amplitude. Recordings, all right eyes, were performed under International Society for Clinical Electrophysiology of Vision (ISCEV) standard conditions. Relative awave and b-wave amplitudes under these conditions are given in Table 2.

homeodomain (residues Q39 to R98),28 nor does it alter the positively charged (K113 to K120)28,29 and glutamine-rich (Q96 to Q106)28 nuclear localization signal sequences of the protein. However, truncation of one half of the normal CRX gene product may affect the protein at each of several global levels of function, including stability and general conformation of the mutant TF. Rate of proteolysis and/or solubility of the remaining polypeptide may also be relevant, whereas the loss of specific C-terminal domains may alter

transactivation and protein–protein interactions even if a stable protein is produced. The described frameshift truncates the C-terminal OTX tail and WSP transactivation domains, which together comprise the major transactivation activity of the native factor.28 In addition, these two motifs are required for association of neural retina leucine zipper TF with the CRX homeodomain.28,30 A third crudely defined region between residues 219 and 267 contains a domain that mediates productive interaction with the general TF and histone acetyltransferase p300/CPB.31,32 This func-

Table 2. Reduction of Electroretinogram Amplitudes in Cone–Rod Dystrophy–Affected Subjects (from Figure 3 Data)

Individual

Age

III-4 (normal) IV-1 (affected) III-5 (affected) IV-6 (affected) IV-9 (affected) Mean of affecteds

71 52 91 54 52

Photopic b-Wave Reduction (%)

Scotopic (Dim Flash) b-Wave Reduction (%)

a-Wave Reduction (%)

b-Wave Reduction (%)

10 64 67 60 44 59

Nil 49 47 61 39 49

1 Nil Nil 26 9 18

Nil 49 26* 55 37 41

* This degree of b-wave amplitude decrease may be attributed to the normal aging process.

1866

Scotopic (Bright Flash)

Lines et al 䡠 Autosomal Dominant Cone–Rod Dystrophy Caused by a CRX Mutation

Figure 4. Multifocal electroretinogram (ERG) of cone–rod dystrophy–affected and unaffected subjects. Trace array (all right eyes) of local responses generated by a matrix stimulus of 61 hexagons that encompass a field extending approximately 20° in radius, centered on the fovea. Note the central dysfunction observed in all affected patients.

tion is also absent in the mutant peptide, potentially preventing local chromatin opening upstream of CRX target promotors. Positions of these and other known functional domains in the CRX polypeptide, relative to the ⫹444insC mutation, are illustrated in Figure 8. However, the possibility that the mutant CRX allele does not produce a stable translation product remains significant. Although little is presently known about the pathobiol-

ogy of Leber congenital amaurosis, RP, and CRD caused by CRX mutation, alteration of the amount and function of CRX protein within retinal photoreceptors is likely to produce compound changes in the expression of numerous phototransduction genes, as is observed in Crx knockout Table 3. Haplotype Assembly of Rhodopsin-Linked Microsatellite Marker Alleles Haplotype

Figure 5. Marker haplotypes flanking the human rhodopsin locus assort independently of the cone–rod dystrophy phenotype in this kindred. Alphabetical symbols represent marker haplotypes based on three dinucleotide repeats (see Methods). Individuals III-9/10, IV-2/3, and IV-9/ 10/11 provide the most explicit evidence of recombination between cone– rod dystrophy and RHO loci. Gray symbols indicate subjects who were not included in the haplotype analysis.

A B C D E F GH* I J K L M NO* P Q

AFMb020zb9

AFM199xd6

AFM9064vb1

9 7 3 4 9 4 7,8 6 2 5 7 7 2,4 4 5

3 7 6 3 8 1 1,5 8 4 3 8 4 1,6 2 1

1 4 3 9 3 8 1,5 4 2 1 3 3 4,8 1 1

* Haplotypes ambiguous because of a lack of informative meioses.

1867

Ophthalmology Volume 109, Number 10, October 2002

Figure 6. Autoradiograph showing heterozygous insertion of a cytosine residue into a (C)4 tract between transcript positions ⫹444 and ⫹447 in exon III of the CRX gene. The mutation, which disrupts the reading frame of the C-terminal half of the CRX coding sequence, was initially discovered to be present in two affected and absent in three unaffected individuals.

mice.17 Misexpression of photoreceptor-specific phototransduction pathway genes may therefore be an underlying factor in this form of retinal dysfunction. However, we have observed that the a-wave amplitude of the photopic ERG is relatively well preserved in affected members of this family. In some cases the implicit time of the a-wave was prolonged, which may suggest a phototransduction deficiency, although this does not seem to be the primary pathway defect stemming from this mutation.33 The selectively reduced b-wave, which we observed both with and without delayed implicit time, is of particular interest. This may reflect lessened depolarization of second-order (bipolar and horizontal) retinal neurons or dysfunction at the level of the Mu¨ ller cells. The relative attenuation of the photopic OP2 in some affected members also implies decreased amacrine or inner plexiform neuronal activation. However, because expression of CRX is limited to the photoreceptor cell layer, the reduced b-wave and OP2 may be better explained by diminished neurotransmission caused by impairment of some aspect of photoreceptor communication with secondorder neurons. In support of this model, mutations of the photoreceptor ribbon synapse protein HRG4 (UNC119) have recently been shown to produce a strikingly similar CRD phenotype in both transgenic mice and human patients.33,34 This late-onset, clearly synaptic disorder is characterized by decreased and delayed b-wave, flattened OPs, night and color vision deficits, photoreceptor death, and subsequent transsynaptic degeneration of the inner retina. Altered abundance and/or function of gated cation channels caused by mutation of CRX is one putative mechanism by which a solely photoreceptor defect could selectively impair neurotransmission without disrupting the upstream events of phototransduction. Net efflux of Ca2⫹ and Na⫹ caused by cGMP hydrolysis and gated cation channel closure normally accompanies the hyperpolarization of stimulated photoreceptors, decreases glutamate secretion from their synaptic terminals, and causes the subsequent graded depolarization of ON-channel and hyperpolarization of OFF-channel bipolar neurons (for review, see Niemeyer35). cGMP-gated cation channels modulate photoreceptor membrane polarity and are hetero-oligomers of functional (poreforming) ␣36,37 and (Ca2⫹-calmodulin binding) modulatory ␤38,39 subunits. Functional and regulatory subunits of the

1868

channels are respectively encoded by the CNGA140,41 and CNGB125 genes in retinal rods and by CNGA337 and CNGB342 in cones. Some phenotypic overlap exists between retinal dystrophies caused by mutations of CRX and of cGMP-gated cation channel genes. Most notably, mutations in CNGA1 and CNGB1 produce RP,43,44 whereas mutations in either CNGA3 or CNGB3 produce rod monochromacy.39,42,45,46 In addition, mutations within CACNA1F (calcium channel ␣1-subunit), a retina-specific L-type voltage-gated calcium channel gene, cause incomplete congenital stationary night blindness, also accompanied by a selective reduction of the b-wave of the ERG.47– 49 Normal transcription of the CNGA1 and CNGA3 genes is altered in homozygous knockout Crx mice,17 whereas CNGB1 contains two putative CRX binding elements, although regulation by CRX remains to be demonstrated.25 From this study we conclude that the visual and electrophysiologic abnormalities associated with adCRD2 stem from a photoreceptor-specific synaptic transmission defect that may represent a channelopathy of both the rod and cone visual systems.

Figure 7. Mutational analysis demonstrates that the CRX ⫹444insC frameshift mutation cosegregates with the cone–rod dystrophy phenotype in this kindred. A ⫹ sign indicates the presence of the heterozygous frameshift mutation. A ⫺ sign denotes normal homozygotes. The mutation was inferred to be present in and transmitted by individuals I-1, II-1/2, and III-2/3, although these individuals were not sequenced or themselves included in the calculation. Individuals represented by gray symbols were not included in the mutational screen. Individual V-2 is presumed to be either nonpenetrant or presymptomatic (see Results). Excluding individual V-2 from the analysis, these data support a logarithm of odds score of log10(2)16 ⫽ 4.82 at a theta (recombination fraction) of zero, based solely on observed individual genotypes.

Lines et al 䡠 Autosomal Dominant Cone–Rod Dystrophy Caused by a CRX Mutation

Figure 8. Known functional domains within the CRX polypeptide.14,15,28,29 *S150 is the first residue altered by the ⫹444/7insC mutation.

Acknowledgments. The skillful assistance of Dr. W. G. Pearce in the original identification and investigation of this family is gratefully acknowledged. We also thank and acknowledge the patients and family members whose willing participation made this study possible.

References 1. Pearce WG. Hereditary macular dystrophy. A clinical and genetic study of two specific forms. Can J Ophthalmol 1975; 10:319 –25. 2. Nathans J, Hogness DS. Isolation and nucleotide sequence of the gene encoding human rhodopsin. Proc Natl Acad Sci U S A 1984;81:4851–5. 3. Dryja TP, McGee TL, Hahn LB, et al. Mutations within the rhodopsin gene in patients with autosomal dominant retinitis pigmentosa. N Engl J Med 1990;323:1302–7. 4. Dryja TP, McGee TL, Reichel E, et al. A point mutation of the rhodopsin gene in one form of retinitis pigmentosa. Nature 1990;343:364 – 6. 5. Farrar GJ, McWilliam P, Bradley DG, et al. Autosomal dominant retinitis pigmentosa: linkage to rhodopsin and evidence for genetic heterogeneity. Genomics 1990;8:35– 40. 6. Sohocki MM, Sullivan LS, Mintz-Hittner HA, et al. A range of clinical phenotypes associated with mutations in CRX, a photoreceptor transcription-factor gene. Am J Hum Genet 1998;63:1307–15. 7. Swain PK, Chen S, Wang QL, et al. Mutations in the cone-rod homeobox gene are associated with the cone-rod dystrophy photoreceptor degeneration. Neuron 1997;19:1329 –36. 8. Freund CL, Gregory-Evans CY, Furukawa T, et al. Cone-rod dystrophy due to mutations in a novel photoreceptor-specific homeobox gene (CRX) essential for maintenance of the photoreceptor. Cell 1997;91:543–53. 9. Freund CL, Wang QL, Chen S, et al. De novo mutations in the CRX homeobox gene associated with Leber congenital amaurosis [letter]. Nat Genet 1998;18:311–2. 10. Swaroop A, Wang QL, Wu W, et al. Leber congenital amaurosis caused by a homozygous mutation (R90W) in the homeodomain of the retinal transcription factor CRX: direct evidence for the involvement of CRX in the development of photoreceptor function. Hum Mol Genet 1999;8:299 –305. 11. Jacobson SG, Cideciyan AV, Huang Y, et al. Retinal degen-

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

erations with truncation mutations in the cone-rod homeobox (CRX) gene. Invest Ophthalmol Vis Sci 1998;39:2417–26. Silva E, Yang JM, Li Y, et al. A CRX null mutation is associated with both Leber congenital amaurosis and a normal ocular phenotype. Invest Ophthalmol Vis Sci 2000;44: 2076 –9. Lotery AJ, Namperumalsamy P, Jacobson SG, et al. Mutation analysis of 3 genes in patients with Leber congenital amaurosis. Arch Ophthalmol 2000;118:538 – 43. Chen S, Wang QL, Nie Z, et al. Crx, a novel Otx-like pairedhomeodomain protein, binds to and transactivates photoreceptor cell-specific genes. Neuron 1997;19:1017–30. Furukawa T, Morrow EM, Cepko CL. Crx, a novel otx-like homeobox gene, shows photoreceptor-specific expression and regulates photoreceptor differentiation. Cell 1997;91:531– 41. Li X, Chen S, Wang Q. A pineal regulatory element (PIRE) mediates transactivation by the pineal/retina-specific transcription factor CRX. Proc Natl Acad Sci U S A 1998;95: 1876 – 81. Furukawa T, Morrow EM, Li T, et al. Retinopathy and attenuated circadian entrainment in Crx-deficient mice. Nat Genet 1999;23:466 –70. Livesey FJ, Furukawa T, Steffen MA, et al. Microarray analysis of the transcriptional network controlled by the photoreceptor homeobox gene Crx. Curr Biol 2000;10:301–10. Kajiwara K, Hahn LB, Mukai S, et al. Mutations in the human retinal degeneration slow gene in autosomal dominant retinitis pigmentosa. Nature 1991;354:480 –3. Wells J, Wroblewski J, Keen J, et al. Mutations in the human retinal degeneration slow (RDS) gene can cause either retinitis pigmentosa or macular dystrophy. Nat Genet 1993;3:213– 8. Kohl S, Christ-Adler M, Apfelstedt-Sylla E, et al. RDS/peripherin gene mutations are frequent causes of central retinal dystrophies. J Med Genet 1997;34:620 – 6. Olsson JE, Gordon JW, Pawlyk BS, et al. Transgenic mice with a rhodopsin mutation (Pro23His): a mouse model of autosomal dominant retinitis pigmentosa. Neuron 1992;9:815–30. Dryja TP, Berson EL, Rao VR, Oprian DD. Heterozygous missense mutation in the rhodopsin gene as a cause of congenital stationary night blindness. Nat Genet 1993;4:280 –3. Sieving PA, Richards JE, Naarendorp F, et al. Dark-light. model for nightblindness from the human rhodopsin Gly-

1869

Ophthalmology Volume 109, Number 10, October 2002

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

903 Asp mutation. Proc Natl Acad Sci U S A 1995;92: 880 – 4. Ardell MD, Bedsole DL, Schoborg RV, Pittler SJ. Genomic organization of the human rod photoreceptor cGMP-gated cation channel ␤-subunit gene. Gene 2000;245:311– 8. Weissenbach J, Gyapay G, Dib C, et al. A second-generation linkage map of the human genome. Nature 1992;359:794 – 801. Dib C, Faure´ S, Fizames C, et al. A comprehensive genetic map of the human genome based on 5,264 microsatellites. Nature 1996;380:152– 4. Chau KY, Chen S, Zack DJ, Ono SJ. Functional domains of the cone-rod homeobox (CRX) transcription factor. J Biol Chem 2000;275:37264 –70. Fei Y, Hughes TE. Nuclear trafficking of photoreceptor protein Crx: the targeting sequence and pathologic implications. Invest Ophthalmol Vis Sci 2000;41:2849 –56. Mitton KP, Swain PK, Chen S, et al. The leucine zipper of NRL interacts with the CRX homeodomain. J Biol Chem 2000;275:29794 –9. Yanagi Y, Masuhiro Y, Mori M, et al. p300/CBP acts as a coactivator of the cone-rod homeobox transcription factor. Biochem Biophys Res Commun 2000;269:410 – 4. Ogryzko VV, Schiltz RL, Russanova V, et al. The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 1996;87:953–9. Kobayashi A, Higashide T, Hamasaki D, et al. HRG5 (UNC119) mutation found in cone-rod dystrophy causes retinal degeneration in a transgenic model. Invest Ophthalmol Vis Sci 2000;41:3268 –77. Kubota S, Kobayashi A, Mori, N, et al. Changes in retinal synaptic proteins in the transgenic model expressing a mutant HRG4 (UNC119). Invest Ophthalmol Vis Sci 2002;43:308 – 13. Niemeyer G. Selective rod- and cone-ERG responses in retinal degenerations. Digit J Ophthalmol 1998:4. Available from URL:http://djo.harvard.edu/meei/OA/OAhome.html. Kaupp UB, Niidome T, Tanabe T, et al. Primary structure and functional expression from complementary DNA of the rod photoreceptor cyclic GMP-gated channel. Nature 1989;342: 762– 6. Biel M, Zong X, Distler M, et al. Another member of the cyclic nucleotide-gated channel family, expressed in testis, kidney, and heart. Proc Natl Acad Sci U S A 1994;91:3505–9. Chen TY, Peng YW, Dhallan RS, et al. A new subunit of the

1870

39.

40. 41.

42.

43.

44.

45.

46.

47.

48. 49.

cyclic nucleotide-gated cation channel in retinal rods. Nature 1993;362:764 –7. Chen TY, Illing M, Molday RS, et al. Subunit 2 (or ␤) of retinal rod cGMP-gated cation channel is a component of the 240-kDa channel-associated protein and mediates Ca2⫹-calmodulin modulation. Proc Natl Acad Sci U S A 1994;91: 11757– 61. Dhallan RS, Macke JP, Eddy RL, et al. Human rod photoreceptor cGMP-gated channel: amino acid sequence, gene structure, and functional expression. J Neurosci 1992;12:3248 –56. Pittler SJ, Lee AK, Altherr MR, et al. Primary structure and chromosomal localization of human and mouse rod photoreceptor cGMP-gated cation channel. J Biol Chem 1992;267: 6257– 62. Kohl S, Baumann B, Broghammer M, et al. Mutations in the CNGB3 gene encoding the ␤-subunit of the cone photoreceptor cGMP-gated channel are responsible for achromatopsia (ACHM3) linked to chromosome 8q21. Hum Mol Genet 2000; 9:2107–16. Dryja TP, Finn JT, Peng YW, et al. Mutations in the gene encoding the ␣ subunit of the rod cGMP-gated channel in autosomal recessive retinitis pigmentosa. Proc Natl Acad Sci U S A 1995;92:10177– 81. Bareil C, Hamel CP, Delague V, et al. Segregation of a mutation in CNGB1 encoding the ␤-subunit of the rod cGMPgated channel in a family with autosomal recessive retinitis pigmentosa. Hum Genet 2001;108;328 –34. Wissinger B, Ja¨ gle H, Kohl S, et al. Human rod monochromacy: linkage analysis and mapping of a cone photoreceptor expressed candidate gene on chromosome 2q11. Genomics 1998;51:325–33. Kohl S, Marx T, Giddings I, et al. Total colourblindness is caused by mutations in the gene encoding the ␣-subunit of the cone photoreceptor cGMP-gated cation channel [letter]. Nat Genet 1998;19:257–9. Bech-Hansen NT, Naylor MJ, Maybaum TA, et al. Loss-offunction mutations in a calcium-channel ␣1-subunit gene in Xp11.23 cause incomplete X-linked congenital stationary night blindness. Nat Genet 1998;19:264 –7. Strom TM, Nyakatura G, Apfelstedt-Sylla E, et al. An L-type calcium-channel gene mutated in incomplete X-linked congenital stationary night blindness. Nat Genet 1998;19:260 –3. Boycott KM, Maybaum TA, Naylor, MJ, et al. A summary of 20 CACNA1F mutations identified in 36 families with incomplete X-linked congenital stationary night blindness, and characterization of splice variants. Hum Genet 2001;108;91–7.