|
|
||||||||
1From the IRO (Institute of Research in Ophthalmology), Sion, Switzerland; the 2Department of Ophthalmology, Jules Gonin Eye Hospital, University of Lausanne, Lausanne, Switzerland; the 3II Department of Ophthalmology Aristotle University of Thessaloniki, Thessaloniki, Greece; 4Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland; 5Service dOphtalmologie, CHU (Centre Hospitalier Universitaire) dHussein-Dey, Alger, Algeria; 6Service dOphtalmologie, CHU Tlemcen, Tlemcen, Algeria; 7CHU Beni Messous, Alger, Algeria; and the 8Department of Ophthalmology, University of Bern, Inselspital, Bern, Switzerland.
| Abstract |
|---|
|
|
|---|
METHODS. All accessible family members underwent a complete ophthalmic examination, and blood was obtained for DNA extraction. Homozygosity mapping was performed with markers flanking 12 loci associated with LCA. The 15 exons of TULP1 were sequenced.
RESULTS. Seven of 30 examined family members were affected, including five with EORD and two with LCA. All patients had nystagmus, hemeralopia, mild myopia, and low visual acuity without photophobia. Fundus features were variable among EORD patients: typical spicular retinitis pigmentosa or clumped pigmented retinopathy with age-dependent macular involvement. A salt-and-pepper retinopathy with midperipheral retinal pigment epithelium (RPE) atrophy was present in the older patients with LCA, whereas the retina appeared virtually normal in the younger ones. Both scotopic and photopic electroretinograms were nondetectable. Fundus imaging revealed a perifoveal ring of increased fundus autofluorescence (FAF) in the proband, and optical coherence tomography disclosed a thinned retina, mainly due to photoreceptor loss. Linkage analysis identified a region of homozygosity on chromosome 6, region p21.3, and mutation screening revealed a novel 6-base in-frame duplication, in the TULP1 gene.
CONCLUSIONS. Mutation in the TULP1 gene is a rare cause of LCA/EORD, with only 14 mutations reported so far. The observed intrafamilial phenotypic variability could be attributed to disease progression or possibly modifier alleles. This study provides the first description of FAF and quantitative reflectivity profiles in TULP1-related retinopathy.
LCA, early-onset retinal degeneration (EORD), and RP may be considered a continuum of entities, in which LCA represents the extreme end of severity, as far as age of onset and functional outcome are concerned.3 4 5 Based on behavior toward light during the first 2 years of life, Hanein et al.3 divided patients with LCA/EORD into two main groups: (1) congenital or very early cone–rod dystrophy, and (2) severe yet progressive rod–cone dystrophy.
Poor vision may either be present at birth or develop progressively during childhood or adolescence. Fundus examination is usually normal at birth and retinopathy does not become evident for several years. Extinguished ERGs before the age of 1 year can be used as a criterion to distinguish LCA from EORD.6 Associated clinical signs such as posterior polar cataract and keratoconus may be part of the phenotype. Refractive error also seems to be a distinct feature, as patients with mutation in certain genes are very hyperopic, whereas other mutated genes are associated with emmetropia or high myopia.3
To date, nine genes (GUCY2D, CRB1, RPE65, RPGRIP1, AIPL1, CRX, RDH12, IMPDH1, and CEP290)3 4 5 6 7 8 9 and three loci (LCA3, LCA5, and LCA9)3 have been reported to be associated with LCA. LRAT and TULP1 have also been implicated in LCA,3 although controversy remains regarding their phenotypic associations, and other researchers consider them to be a cause of juvenile or early-onset RP rather than LCA.4 5 6 It has been shown, that all LCA genes except RPGRIP1 are also associated with other retinal dystrophies, and that different mutations in one gene can cause alternate phenotypes.4 Moreover, some patients with LCA have been found to have inherited mutations in more than one of these LCA-associated genes, suggesting modifier alleles.10
To add to the complexity, the reported genes participate in a wide variety of retinal pathways, are expressed in separate retinal cells, and have diverse putative functions. The exact pathophysiological role of most of them is as yet unknown.4 11
Efforts have been made to define clinical subtypes and to establish gene-specific phenotypes as well as LCA carrier phenotypes, so as to guide the genetic testing strategies and counseling.3 12 13 Recognizing distinct clinical subtypes and correlating them with a specific genetic background is a necessary step into the understanding of this heterogeneous group of disorders.
We report a large Algerian family exhibiting a novel TULP1 mutation and its variable phenotypic expression, ranging from typical LCA to EORD.
| Methods |
|---|
|
|
|---|
All accessible members of the family had a complete clinical examination, including refraction, best corrected visual acuity, Goldmann perimetry, color vision testing (Ishihara plates or Farnsworth D-15 hue test), pupillary responses, biomicroscopy, tonometry, and funduscopy after pupillary dilatation. Fluorescein angiography was performed on the propositus (subject VII3) and two other affected members (subjects VI25 and VI26). Macular optical coherence tomography (OCT; Stratus OCT 3; Carl Zeiss Meditec Inc., Dublin, CA), fundus autofluorescence (FAF) measurement with a confocal scanning laser ophthalmoscope (Heidelberg Retina Angiogram [HRA 2], Heidelberg Engineering, Heidelberg, Germany), and photopic and scotopic full-field electroretinography (ISCEV standards14 ) were available only for the proband investigated in Lausanne, the remainder of the family being examined in Algeria.
For analysis of retinal lamination irregularities raw data of each line scan were exported from the OCT device. Light reflection profiles (LRPs) of each line scan were analyzed (IGOR Pro 5.05a; Wavemetrics Inc., Lake Oswego, OR).15 16 17 A 6-year-old normal subject was used as the control.
DNA Analysis
Genomic DNA was extracted from peripheral blood of the family members by using a standard procedure. Fifty-one unrelated healthy individuals of Algerian ancestry were used as control subjects. Linkage to the 12 autosomal recessive loci that were known at the time of investigation was tested by using polymorphic genetic markers flanking the region of each gene by approximately 2.5 Mb (Table 1a) . Homozygosity mapping was performed because of consanguinity and our hypothesis that this family shared a unique mutation by descent.
|
| Results |
|---|
|
|
|---|
|
All tested eyes had a myopic refractive error, except for patient VI10 who presented mild hyperopia (Table 2) . Visual acuity at examination ranged from perception of light to 20/100. The proband (VII3) was examined on two occasions (8 months apart), and during this interval, his visual acuity declined from 20/100 to 20/250 in the better eye. Visual fields were moderately to severely restricted in all subjects who managed to undergo the test, and only the V4e stimulus was seen. Color vision was severely disturbed without a specific axis of confusion. The pupillary reflex was extremely sluggish in most of the patients. None presented signs of keratoconus.
|
|
|
|
Pigmentary retinopathy and retinal vessel attenuation were present in all but the youngest subject (VII4). Patients VI25, VI26, and VI9 presented severely sclerosed arteries and multiple occluded or ghost vessels in the periphery and occasionally at the posterior pole (Fig. 4d , subject VI9). Pigment spicules and clumps varied in intensity among patients. In sibs VI9 and VI10, pigment migration was predominantly paravenous (Fig. 4e , subject VI10). Pinpoint yellow-white deposits were also present in the peripheral retina in subjects VI25, VI26, VI9, and VII3 (Fig. 4d , subject VI9).
Fluorescein angiography confirmed the presence of peripapillary retinal pigment epithelial atrophy (subjects VI25 and VI26) and the absence of macular edema.
Fifteen of the 17 examined asymptomatic carriers of the mutation (mean age, 35 years; range, 14–85) had an unremarkable fundus appearance. Two heterozygous subjects (V13 and V10, age 66 and 64 years, respectively) presented a few drusenlike macular deposits, and V10 additionally presented age-related pigmentary macular changes.
Molecular Analysis
Homozygosity mapping excluded all the tested loci with the exception of TULP1 located at 6p21.3 for which the parents were heterozygous, and all affected individuals were homozygous for the two analyzed markers. TULP1 sequencing revealed five sequence modifications.
A small 6-base in-frame duplication, c.1593 to 1594dupTTCGCC (HSTULPS01-5, Retina International Mutation Database www.retina-international.org), was identified in exon 15 (Figs. 5a 5b) .18 This change was homozygous in all seven affected members, heterozygous in their parents and 15 collaterals, and absent in 51 ethnic-matched control subjects. This mutation leads to a predicted duplication of two amino acids in the highly conserved C-terminal region of the protein (FA531-532dup; Fig. 5c ). None of the 102 control chromosomes presented this change. In addition, this mutation was not observed in more than 500 index cases with various forms of RP.
|
| Discussion |
|---|
|
|
|---|
Previous linkage analysis in a large Dominican pedigree with autosomal recessive (ar)EORD mapped the disease to 6p21.3.19 This locus, known as RP14, was later shown to harbor mutations in the TULP1 gene.20 21 TULP1 belongs to a highly expressed retinal family of tubby genes, including TULP2 and TULP3. TULP1 mutations are known to cause retinal degeneration with a wide range of severity, from congenital blindness/LCA to adult-onset RP.3 22 The most commonly described phenotype is severe, early-onset retinal degeneration.
Although the specific function of TULP1 is not yet known, recent studies have shown that the protein is expressed in both rods and cones.23 It may act as a transcription factor24 or be implicated in protein trafficking,25 such as rhodopsin transport.26
TULP1 appears to be a rare cause of LCA, as mutations of this gene are reported in only 1.7% of patients with LCA.3 A total of only 14 disease-causing alleles of TULP1 have been reported in 11 index patients, including seven missense mutations (Arg400Trp, Gly368Trp, Arg420Pro, Phe491Leu, Ile459Lys, Phe382Ser, and Lys489Arg),3 20 22 27 five splice-site mutations (c.99+1G>A, IVS2DS G-A,+1, IVS14-6C-A, IVS4-2delAGA, and IVS14DS G-A,+1),3 20 21 22 28 29 one nonsense mutation (Glu402X),3 and one frame-shift deletion (c.937delC).28 These changes were observed in four simplex and seven multiplex pedigrees of autosomal recessive retinitis pigmentosa (arRP) or EORD and were present in populations of different origins, including Spanish. Of note, the present Algerian family is of Andalusian origin and immigrated to the region of Tlemcen in the 15th century at the time of the Reconquista.
A body of evidence supports a causative role of the FA531-532dup mutation in this family. The mutation segregates in all affected members, is absent in the control population, and compromises the structure of the highly conserved carboxyl terminal end of the protein.
The heterozygous change A496T has previously been described as a polymorphism of uncertain pathogenicity,20 22 since it was only found heterozygously in patients in whom an undetected compound heterozygosity could not be excluded. The coexistence of A496T with FA531-532dup in a normal subject (VI17) provides evidence that A496T is not a disease-causing mutation.
The prevalence of TULP1 mutations has not been specifically investigated in patients with RP. Gu et al.22 found TULP1 double mutations in 2 (1.17%) of 171 patients with arRP. In 1.75% of the same cohort, single heterozygous, potentially pathogenic mutations were observed.22 In their study of 536 patients with RP, Hagstrom et al.20 found pathogenic TULP1 mutations in <0.5% of the cases (2/536 or 0.37%). Even if these statistics underestimate the actual prevalence, since the mutation detection with current techniques is <100%, it is reasonable to say that TULP1 mutations are a rare cause of RP or EORD.
Visual function in TULP1-related retinal degeneration has been described as severely reduced when compared with that of other patients with arRP,20 28 or much better than in other groups of patients with LCA.29 Lack of severe central visual dysfunction and absence of nystagmus was described in two 8-year-old Phe382Ser homozygotes, who presented acuities of 0.3 and 0.5, suggesting a less severe phenotype specific to this TULP1 mutation.27
When comparing our patients to the numerous IVS14+1G>A homozygotes from the large Dominican kindred,29 a similar course of severity and progression of visual handicap was noted. Relative preservation of visual acuity (20/100) and visual field (near-normal V4e isopter, undetectable I4e target) in the first decade of life, although nystagmus appeared as early as in the first year. Common clinical findings were a myopic refractive error in most of the patients, minimal peripheral pigmentary retinopathy, and vessel attenuation (early aspect of RP) present from the first decade, initial normal disc and macular appearance, eventual development of ophthalmoscopically visible maculopathy, and optic disc pallor. An annulus of yellow deposits in the central 10° of the macula was present in many patients in the second and third decades.29 This finding was also reported in one of two Phe382Ser homozygotes.27 Macular geographic atrophy, present in our eldest patient, was described in a 42-year-old IVS14+1G>A homozygote.29
Only nuclear sclerotic cataract was reported in a 42-year-old IVS14+1G>A homozygote,29 whereas bilateral subcapsular opacities, similar to the ones in VI24, VI25 and VI26, have been described in a 26-year-old patient carrying the IVS4-2delAGA and c.937delC mutations.28 Pigmentary retinopathy is referred to as typical RP or bone-spicule–like pigment changes mainly in the peripheral retina, in most patients with TULP1-related disease.3 28 29 Apart from the vascular attenuation that is usually regarded as a typical feature of EORD or RP, we observed multiple severely sclerosed and occluded or ghost vessels in subjects VI25 and VI9. This feature, as well as peripapillary RPE atrophy, has not been reported in patients with TULP1-related LCA or EORD.
We found a totally normal clinical appearance in most of the FA531-532dup mutation carriers (15/17) of all ages. The presence of macular drusenlike deposits in two heterozygotes is in accordance with the findings in a reported Dominican kindred, in which such carriers were also found to have normal ERGs.29 So far, no clinical feature has been identified to raise the suspicion of a TULP1 mutation carrier state that would guide genotyping in families affected by LCA/EORD.
Quantitative analysis of retinal layers in the macular area in the proband (VII3) and in the age-matched control subject (Fig. 3) reveals comparable distances between the nerve fiber layer and the inner plexiform layer. On the contrary, VII3 exhibited missing reflective layers corresponding to the outer retinal structures, and the reduced signal intensity suggested an already affected outer plexiform layer. This in vivo biomicroscopic characterization of the diseased retina suggests a retrograde process of degeneration as a result of the primary photoreceptor decay.
Of interest is the simultaneous presence of a perifoveal ring of increased AF in the same patient. This finding appears not to be uncommon in patients with RP with preserved central vision30 and has also been described in some patients with LCA, although not in correlation with a specific genotype.31 FAF is an index of lipofuscin accumulation in the RPE, its preservation reflects the level of metabolic activity and thus indicates at least a partial presence of structurally intact photoreceptors and photoreceptor/RPE complex.31 In our patient, the AF image revealed that the homozygous FA531-532dup mutation in TULP1 does not halt the formation of lipofuscin fluorophores and causes macular photoreceptor cell dysfunction as opposed to cell death, at the corresponding stage of the disease. This finding can be helpful in directing mutation screening in patients with LCA/EORD3 —for example, in differentiating from RPE65 mutations—as they have been found to lead to lack of FAF.32 FAF can also be useful in monitoring possible photoreceptor cell rescue therapies in the future.31
Evaluation of autofluorescence by HRA and retinal thickness by OCT may help to discriminate between RPE65 (and possibly LRAT), TULP1, and CRMB1-related rod-cone LCAs. Further characterization of patients with TULP1 mutations with the help of these devices may shed some light on the pathogenesis of this rare retinopathy.
| Acknowledgements |
|---|
| Footnotes |
|---|
Submitted for publication August 27, 2006; revised February 28, 2007; accepted September 12, 2007.
Disclosure: A. Mataftsi, None; D.F. Schorderet, None; L. Chachoua, None; M. Boussalah, None; M.T. Nouri, None; D. Barthelmes, None; F.-X. Borruat, None; F.L. Munier, None
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C.
1734 solely to indicate this fact.
Corresponding author: Francis L. Munier, Jules Gonin Eye Hospital, Avenue de France 15, 1004 Lausanne, Switzerland; francis.munier{at}ophtal.vd.ch.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. D. Fischer, J. C. Fleischhauer, M. C. Gillies, F. K. Sutter, H. Helbig, and D. Barthelmes A New Method to Monitor Visual Field Defects Caused by Photoreceptor Degeneration by Quantitative Optical Coherence Tomography Invest. Ophthalmol. Vis. Sci., August 1, 2008; 49(8): 3617 - 3621. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |