The Spectrum of Subclinical Best Vitelliform Macular Dystrophy in Subjects with Mutations in BEST1 Gene

Abstract

Purpose.: To describe the morphologic and functional characteristics of subclinical Best vitelliform macular dystrophy (VMD) in subjects with mutation in the BEST1 gene.

Methods.: Best-corrected visual acuity (BCVA), funduscopic appearance, fundus autofluorescence (FAF), spectral-domain optical coherence tomography (SD-OCT), and electro-oculography (EOG) were assessed in 23 consecutive subjects from nine unrelated families with known mutations in the BEST1 gene (eight distinct BEST1 mutations).

Results.: Six subjects were identified with BEST1 mutations (three male, three female; aged 8 to 30 years) without clinically detectable (subclinical) Best VMD (absence of both symptoms and funduscopic lesions). All six subjects showed 20/20 BCVA and normal FAF findings. In these 6 of 26 subjects from five different families, we found five distinct mutations in the BEST1 gene. In three (six eyes) out of these six subjects with BEST1 gene mutations (two families: p.G15D; p.A243V), SD-OCT showed overall normal findings. In the other three subjects (six eyes) with BEST1 gene mutations (three families: p.V9A; p.R92C; p.I230T), we found, on SD-OCT, a thicker and more reflective appearance of the layer between the retinal pigment epithelium and the interface of inner segments and outer segments of the photoreceptor (the Verhoeff's membrane). EOG showed a reduced light-peak:dark-trough ratio in 5 of 12 eyes. Changes on SD-OCT were present in the absence of EOG abnormalities (two of six eyes), and vice versa (one of six eyes).

Conclusions.: Subclinical Best VMD (absence of both symptoms and funduscopic lesions) in subjects with BEST1 mutation may vary from the absence of any morphologic and functional abnormalities to the presence of specific SD-OCT and EOG changes.

Vitelliform macular dystrophy (VMD) (OMIM 153,700), also called Best disease, ¹ has an autosomal dominant pattern of inheritance with very variable penetrance and expressivity. BEST1 (chromosome 11q12-q13) ² is the only gene virtually involved in all Best VMD cases. It encodes a 68 kDa protein called bestrophin-1, ³ which is localized to the basolateral plasma membrane of the retinal pigment epithelium (RPE) and appears to exhibit properties of Ca++–activated Cl- channels. ⁴ Nearly all BEST1 mutations causing Best VMD affect single amino acids, at one of 66 different positions in bestrophin-1. 

Best VMD is a clinically heterogenous and pleomorphic disease, having a bimodal onset distribution with one maximum peak before puberty and a second following puberty and extending through the fifth decade of life. ^{5 –7}  

The onset of Best VMD is characterized by symptoms of metamorphopsia, blurred vision, and decrease of central vision. At the fundus a well-circumscribed 0.5- to 2-disc–diameter “egg-yolk” lesion within the macula may be observed. ⁸ This represents the vitelliform stage of the disease, the second of five progressive stages defined on the basis of fundus examination. This early stage of the disease may be followed by the pseudohypopyon third stage, the vitelliruptive fourth stage, and the atrophic/fibrotic fifth stage. The subclinical form of the disease, or previtelliform stage, represents the first stage and is characterized by absence of symptoms and normal macula or subtle RPE alterations on fundus examination. 

An abnormal electro-oculogram (EOG) ^{9 –11} with a reduced or nondetectable light-peak:dark-trough ratio (≤ 1.55), a blockage effect of fluorescein in the choroid, ¹² and evidence of increased autofluorescence by the yellow vitelliform lesions ⁷ are considered the main diagnostic criteria in the clinical diagnosis of Best VMD. ^11,12 Several authors have recently highlighted the usefulness of optical coherence tomography (OCT) in the diagnosis of Best VMD to be the demonstration of the vitelliform material that accumulates in the subretinal space and on the outer retinal surface. ^{13 –16} However, no data have been published so far on the different functional and morphologic aspects of the subclinical form of the disease (the previtelliform stage). 

Our purpose was to analyze the functional and morphologic characteristics in asymptomatic subclinical subjects issuing from Best VMD families carriers of mutations in the BEST1 gene. 

Methods

Nine Best VMD families segregating BEST1 mutations, ¹⁷ ascertained at the Créteil University Eye Clinic and at the Foggia University Eye Clinic, were included in this study. Informed consent was obtained according to a Paris XII University and a Foggia University Institutional Review Board–approved protocol. This study has been performed in accordance with the ethical standards set forth in the 1964 Declaration of Helsinki. 

In each family, affected and unaffected relatives who agreed to participate to the study were screened for the BEST1 mutation identified in the BEST VMD proband using direct sequencing as previously described. ¹⁷  

All individuals included in this study underwent a complete ophthalmologic examination, including assessment of best-corrected visual acuity (BCVA) measured at 4 m with standard Early Treatment Diabetic Retinopathy Study charts, fundus biomicroscopy, color photography of the fundus (TRC-50 retinal camera; Topcon, Tokyo, Japan), fundus autofluorescence (FAF) frames (Heidelberg Retina Angiograph II; Heidelberg Engineering, Heidelberg, Germany), and red free and fluorescein angiography frames (Topcon TRC-50; Heidelberg Retina Angiograph II). Recordings of EOG were done according to the International Society for Clinical Electrophysiology of Vision standard. OCT examination was performed with time-domain (TD)-OCT (OCT 3000 Stratus; Humphrey-Zeiss, San Leandro, CA) or spectral domain (SD)-OCT (HD-OCT, OCT 4000 Cirrus, Humphrey-Zeiss; Spectralis SD-OCT, Heidelberg Engineering). All scans were positioned within the macular area and throughout the vitelliform lesions, based on color fundus photography and FAF. Fundus pictures and SD-OCT scans were analyzed and interpreted independently by three retinal specialists (GQ, JZ, and EHS). Disagreement regarding interpretation of the different features was resolved by open adjudication. 

The clinical diagnosis of Best VMD (> second stage) was based on the presence, on fundus examination, of one or multiple subfoveal vitelliform lesions in at least one eye and on the evidence of autofluorescence by the yellow vitelliform lesions. The subclinical form of the disease represents the first stage of Best VMD (the previtelliform stage), which is defined by absence of symptoms and normal macula or subtle RPE alterations on fundus examination. In this study, subjects were diagnosed with the subclinical form of the disease if they were asymptomatic and did not show subfoveal lesions or FAF changes. 

On EOG, a reduced light-peak:dark-trough ratio ≤ 1.55 was considered abnormal (as per our laboratory protocol). Ratios > 1.55 were considered as normal values. 

Results

A total of 23 subjects (11 male and 12 female; mean age, 26.9 years; range, 3 to 70 years), from nine unrelated families with known mutations in the BEST1 gene and presenting at least one family member affected with Best VMD, were included in this study (Fig. 1; Table 1). Many affected and unaffected relatives of Best VMD patients (for a total of 63 relatives) were excluded from the study because they were not screened for mutations in the BEST1 gene; of note, only few (12 out of 63 unscreened relatives) were diagnosed with Best VMD (Fig. 1). Two unrelated individuals screened for mutations were also excluded from the study because they did not carry any BEST 1 mutation; they had no clinically detectable Best VMD with normal SD-OCT and EOG findings. 

Figure 1.

 

Pedigree of the studied families. Here all subjects with the subclinical form of Best vitelliform macular dystrophy are presented in white because of the absence of symptoms. In our previous analysis ¹⁷ subjects CT03 and CT14 were presented in black because of exploration abnormalities (OCT and EOG).

Table 1.

 

Summary of Clinical Findings and Probands BEST1 Mutations

Subjects	Mutation	Position	Missense Effect	Age, Sex	Age of Onset (y)	Lesion Type		BCVA		Complications
						RE	LE	RE	LE
FG01 (Family FG I)	C>T728 heterozygous	Exon 7	A243V	49, M	41	Atrophy	Atrophy	20/125	20/160	—
FG02 (Family FG I)	C>T728 heterozygous	Exon 7	A243V	45, F	37	Vitelliruptive	Vitelliruptive	20/25	20/25	—
FG03 (Family FG I)	C>T728 heterozygous	Exon 7	A243V	75, M	67	Pseudohypopion	Vitelliruptive	20/50	20/125	—
FG04 (Family FG I)	C>T728 heterozygous	Exon 7	A243V	13, F	—	None	None	20/20	20/20	—
FG05 (Family FG I)	C>T728 heterozygous	Exon 7	A243V	17, F	—	None	None	20/20	20/20	—
FG06 (Family FG II)	G>A275 heterozygous	Exon 4	R92H	16, F	11	Fibrosis	Fibrosis	20/160	20/160	CNV RLE
FG07 (Family FG II)	G>A275 heterozygous	Exon 4	R92H	3, M	2	Vitelliform	Vitelliform	20/32	20/32	—
FG08 (Family FG III)	G>A44 heterozygous	Exon 2	G15D	3, F	2	Vitelliform	Vitelliform	20/25	20/25	—
FG09 (Family FG III)	G>A44 heterozygous	Exon 2	G15D	30, M	—	None	None	20/20	20/20	—
CT01 (Family CT I)	C>T274 heterozygous	Exon 4	R92C	8, M	8	Pre-vitelliform	Pre-vitelliform	20/20	20/20
CT02 (Family CT I)	C>T274 heterozygous	Exon 4	R92C	14, M	8	Fibrosis	Fibrosis	20/50	20/40	CNV RLE
CT03 (Family CT II)	T>C689 heterozygous	Exon 7	I230T	11, M	10	Pre-vitelliform	Pre-vitelliform	20/20	20/25	—
CT04 (Family CT II)	T>C689 heterozygous	Exon 7	I230T	42, F	41	Multifocal	Multifocal	20/32	20/25	—
CT05 (Family CT II)	T>C689 heterozygous	Exon 7	I230T	9, M	6	Vitelliruptive	Vitelliruptive	20/125	20/125	—
CT06 (Family CT III)	C>T272 heterozygous	Exon 4	T91I	44, M	36	Atrophy	Atrophy	20/125	20/40	—
CT07 (Family CT III)	C>T272 heterozygous	Exon 4	T91I	19, F	11	Fibrosis	Fibrosis	20/200	20/40	CNV RE
CT08 (Family CT IV)	A>G10 heterozygous	Exon 2	T4A	27, F	20	Atrophy	None	20/50	20/25	—
CT09 (Family CT IV)	A>G10 heterozygous	Exon 2	T4A	23, F	16	Pseudohypopion	Atrophy	20/32	20/50	CNV LE
CT10 (Family CT V)	C>T73 heterozygous	Exon 2	R25W	10, F	9	Vitelliruptive	Fibrosis	20/20	20/200	—
CT11 (Family CT V)	C>T73 heterozygous	Exon 2	R25W	36, F	30	Vitelliruptive	Vitelliruptive	20/63	20/63	—
CT12 (Family CT V)	C>T73 heterozygous	Exon 2	R25W	70, M	60	Pseudohypopion	None	20/50	20/20	—
CT13 (Family CT VI)	T>C26 heterozygous	Exon 2	V9A	44, M	7	Atrophy	Fibrosis	20/50	20/200	—
CT14 (Family CT VI)	T>C26 heterozygous	Exon 2	V9A	12, F	12	Pre-vitelliform	Pre-vitelliform	20/20	20/20	—

 BCVA, best corrected visual acuity; CNV, choroidal neovascularization; F, female; LE, left eye; M, male; RE, right eye.

Among the 23 patients harboring a BEST1 mutation (Table 1), 17 patients presented with either bilateral (15 of 17) or unilateral (2 of 17) clinically detectable Best VMD, while six patients had 20 of 20 BCVA and normal funduscopic and FAF findings (subclinical Best VMD). These six carrier individuals (three male, three female; aged 8 to 30 years) belonged to five families, each of which segregated a different BEST1 mutation (Table 2). 

Table 2.

 

Summary of Clinical Findings in Patients with Subclinical Best Vitelliform Macular Dystrophy

Case	Age, Sex	Family	Follow-up (mo)	Missense Mutation	Eye	BCVA	OCT	EOG (Arden)
Case 1 (CT14)	12, F	CT VI	18	Exon 2; V9A	RE	20/20	Thickened VM	1.08
					LE	20/20	Thickened VM	1.39
Case 2 (CT01)	8, M	CT I	12	Exon 4; R92C	RE	20/20	Thickened VM	0.98
					LE	20/20	Thickened VM	1.12
Case 3 (CT03)	11, M	CT II	–	Exon 7; I230T	RE	20/20	Thickened VM	3.10
					LE	20/20	Thickened VM	1.96
Case 4 (FG09)	30, M	FG IV	12	Exon 2; G15D	RE	20/20	Normal	5.22
					LE	20/20	Normal	2.08
Case 5 (FG04)	13, F	FG I	24	Exon 7; A243V	RE	20/20	Normal	1.58
					LE	20/20	Normal	1.67
Case 6 (FG05)	17 F	FG I	24	Exon 7; A243V	RE	20/20	Normal	1.66
					LE	20/20	Normal	1.36

 BCVA, best corrected visual acuity; EOG, electro-oculogram; F, female; M, male; LE, left eye; RE, right eye.

Three out of the six patients (subject FG04 and subject FG05, family FG I; subject FG09, family FG III) presented with overall normal SD-OCT findings in both eyes (Figs. 2 and 3). EOG light-peak:dark-trough ratios (Arden ratio) were within normal ranges for five of six eyes (from 1.58 to 5.22), and one eye revealed an Arden ratio ≤ 1.55 (subject FG06, left eye) (Fig. 3). 

Figure 2.

 

Color fundus photograph of the left eye of Case 4 (subject FG09, 30 years old [p.G15D]) showing overall normal macula (top left panel). Spectral-domain optical coherence tomography (SD-OCT) scan shows normal findings at the fovea (bottom left panel). Arden ration was normal (>1.55). Color fundus photograph (top right panel) and OCT (bottom right panel) scan of the left eye of subject FG08 (daughter of Case 4, 30 years old [p.G15D]) show the typical vitelliform lesion within the macula. Arden ratio was abnormal (≤1.55).

Figure 3.

 

FAF of the left eye of Case 6 (subject FG05, 17 years old [p.A243V]) showing overall normal macula (top left panel). SD-OCT shows normal findings at the fovea (bottom left panel). Arden ration was abnormal (≤1.55). FAF of the left eye (top right panel) and OCT (bottom right panel) of the left eye of subject FG02 (mother of Case 6, 45 years old [p.G15D]) show the vitelliruptive lesion within the macula. Arden ratio was abnormal (≤1.55).

In the other three carrier subjects (six eyes) (subject CT01 from family CT I; subject CT03 form family CT II; subject CT14 form family CT VI), we found, on SD-OCT, a thicker and more reflective appearance of the layer between the RPE and the interface of inner segments (ISs) and outer segments (OSs) of the photoreceptor (the Verhoeff's membrane) (Figs. 4 and 5). ^{14 –16} In all eyes but two (four of six) from a single patient (subject CT03) (Fig. 5), EOG showed a reduced light-peak:dark-trough ratio (Arden ratio) ≤ 1.55. 

Figure 4.

 

FAF of the right eye of Case 2 (subject CT01, 8 years old [p.R92C]) showing overall normal macula (top left panel). SD-OCT shows a thicker and more reflective appearance of the Verhoeff's membrane (bottom left panel). Arden ratio was abnormal (≤1.55). FAF of the right eye (top right panel) and SD-OCT (bottom right panel) of the right eye of subject CT02 (brother of Case 2, 14 years old [p.R92C]) show the fibrotic lesion within the macula. Arden ratio was abnormal (≤1.55).

Figure 5.

 

Color fundus photograph of the right eye of Case 3 (subject CT03, 11 years old [p.I230T]) showing overall normal macula (top left panel). SD-OCT scan shows a thicker and more reflective appearance of the Verhoeff's membrane (bottom left panel). Arden ratio was normal (>1.55). Color fundus photograph (top right panel) and SD-OCT (bottom right panel) scan of the right eye of subject CT05 (brother of Case 4, 9 years old [p.I230T]) show the vitelliruptive lesion within the macula. Arden ratio was abnormal (≤1.55).

Follow-up evaluation was available for all patients but one (subject CT03). After a mean of 18 months (range 12 to 24 months), BCVA and both funduscopic and FAF findings remained unchanged in all eyes. Also, based on OCT findings, we did not find any sign of disease progression. 

A summary of main clinical findings and BEST1 mutations is presented in Table 1. Even with the same mutation, the age of onset and the stage of the disease were highly variable interfamilially and intrafamilially. In family FG I, two family members (subject FG04 and subject FG05) showed overall normal funduscopic and SD-OCT findings (at the age of 13 and 17 years, respectively), and one family member with the same heterozygous BEST1 mutation p.A243V (subject FG03) was diagnosed at a very late age (67 years of age). The heterozygous p.R92H an p.G15D mutations accounted for the earliest disease manifestations (at two years of age, subject FG07, family FG II, and subject FG08, family FG III, respectively) or either a later onset (at the age of 11 years, subject FG06, family FG II [p.R92H]), or even no disease manifestation (at the age of 30 years, subject FG09, family FG III [p.G15D]), respectively. In family CT I, the heterozygous p.R92H mutation resulted in bilateral CNV development in subject CT02 (14 years old) and in previtelliform Best VMD in subject CT01 (8 years old). Subject CT04 (family CTII), heterozygous for the p.I230T, presented bilateral multifocal Best VMD features on fundus examination. Interestingly, one of her sons with the same heterozygous BEST1 mutation (subject CT05) presented the vitelliruptive stage of the disease in both eyes at the age of nine years, and another of her sons with the same heterozygous BEST1 mutation (subject CT03) presented the previtelliform stage of the disease in both eyes at the age of 11 years. In family CT VI, the heterozygous p.V9A mutation resulted in bilateral end stage in both eyes in subject CT13 (44 years old) and in previtelliform Best VMD in subject CT14 (12 years old). Two unrelated subjects carrying different mutations of exon 2 presented unilateral Best VMD (subject CT08 [p.T4A] and subject CT12 [p.R25W] from family CT IV and family CT V, respectively). 

Discussion

Incomplete penetrance and expressivity are well-known features in Best VMD disease. The large variability within and between families of the clinical expression of BEST1 mutations ranging, in our series, from severe Best VMD to subclinical disease is consistent with previous reports. ^{17 –32} Recently, Lacassagne et al. ²⁸ found not only a considerable intrafamilial phenotypic variability in patients carrying the p.S144G mutation, but even absence of pathologic phenotype in a patient carrying the isolated p.Y5X mutation in the BEST1 gene. Age of onset may also vary greatly among patients with Best VMD ²⁹ ; however, a recent report highlighted how age of onset may be considered a major criterion to distinguish Best VMD from other similar macular dystrophies not strictly related to mutations in the BEST1 gene. ³⁰ In this context of broad phenotypic variability, even with a single BEST1 mutation, fundus autofluorescence, optical coherence tomography, and EOG are valuable noninvasive techniques for phenotyping and follow-up of Best VMD patients. ^31,32  

In the current series, among BEST1 mutations carriers diagnosed with subclinical Best VMD (n = 6), half presented with and half without morphologic and/or functional alterations on SD-OCT and EOG, respectively. It is interesting to note that patients without EOG and SD-OCT alterations were young subjects (FG04 13 years old, FG05 17 years old, and FG09 30 years old). However, the absence of EOG and SD-OCT alterations in young subjects was not strictly related to the genotype. In fact, all five BEST1 mutations affected amino acid residues of distinct functional regions of bestrophin. 

Three out of six BEST1 mutations carriers presented on SD-OCT a bilateral thicker and more reflective appearance of the layer between the RPE and the IS/OS interface (Verhoeff's membrane). At first sight, this finding could seem in contrast to the normal localization of bestrophin to the basal aspect of the RPE. In fact, one would expect that the vitelliform material would accumulate in the RPE. Indeed, Arnold et al. ³³ in their clinicopathological report found that the predominant finding was a collection of extracellular material beneath the sensory retina and proposed that this material was derived internally from photoreceptor outer segments and externally from the RPE, the latter first undergoing hypertrophy and then disruption and attenuation. Moreover, in contrast to previous studies that demonstrated massive lipofuscin accumulation in the RPE, Mullins et al. ²⁶ reported one patient in whom the RPE appeared histologically healthy in some regions of the macula that exhibited loss of photoreceptors. Based on immunofluorescence studies, they proposed that the possible mistargeting of bestrophin may result in a harmful alteration of the ionic milieu of the subretinal space and contribute to the type of photoreceptor cell loss observed histologically. Therefore, we hypothesize that the lesions, at the previtelliform stage, lie beneath the sensory retina and consist of mainly photoreceptor debris, possibly as result of faulty phagocytosis by the RPE, mixed with pigment granules liberated as the RPE undergoes disruption. 

Interestingly, no strict correlation between electrophysiological alterations and morphologic changes was evidenced. However, EOG was altered in most eyes with SD-OCT alterations (four of six) and in only one eye without morphologic changes (one of six). These data suggest a trend toward an impaired function, as evaluated by EOG (light-peak:dark-trough ratio ≤ 1.55), in the presence of morphologic changes (thickening of the Verhoeff's membrane), yet subclinical (normal BCVA and fundus findings). These data also suggest an overall good sensitivity for SD-OCT to detect or exclude the presence of subclinical Best VMD. 

Even though our study was not designed to investigate disease progression, we had the opportunity to follow up five of six of the BEST1 carrier subjects with subclinical BEST VMD over a period of 18 months (mean). BCVA, funduscopic, and OCT findings remained unchanged in all eyes on examination. We acknowledge that this follow-up time is short and probably not enough with respect to disease progression. However, considering patient management and counseling, it is meaningful that subclinical form of the disease may not progress at least in a short-time period. 

Our study has several limitations. Many relatives (12 affected and 51 unaffected) of Best VMD patients were not screened for the BEST1 mutation and thus were excluded from the study: This may represent an ascertainment bias of the current analysis. Moreover, in some Best VMD patients we used TD-OCT that is not very suitable to pick up slight changes. However, given that all subclinical Best VMD patients have been investigated by means of SD-OCT, and that TD-OCT has been used only in patients with clinically detectable Best VMD (disease stage 2 to 5), the different resolutions of the two devices (TD-OCT versus SD-OCT) may have not influenced our data. Finally, we considered light-peak:dark-trough ratio > 1.55 as normal EOG, while other authors use > 1.85 as normal, 1.3–1.85 as mildly reduced, and <1.3 as severely reduced. However, also considering light-peak:dark-trough ratio > 1.85 as normal, some BEST1 carrier subjects with subclinical BEST VMD, with and without thickening of the Verhoeff's membrane, still presented normal EOG (two of six subjects, 4 of 12 eyes). 

A broad phenotypic variability may be observed in association with specific BEST1 mutations. In dominant heterozygous Best VMD, the variable phenotype is highlighted in the present study as well as in other studies. ^{17 –32} The high phenotypic variability is not unique to BEST1-related disease and may also be seen in other autosomal dominantly inherited retinal diseases, such as those caused by peripherin or RDS mutations. ³⁴  

Different mutations might cause Best VMD by different mechanisms. Some mutations in the BEST1 gene may lead to a more severely affected RPE and formation of extramacular vitelliform lesions ^26,35,36 ; however, there seems to be no clear pattern relating type of BEST1 mutation to severity of clinical expression. ³¹ Compound heterozygous, biallelic recessive, or homozygous dominant mutations in BEST1 may confer a particularly severe phenotype, featuring widespread retinal degeneration, in addition to Best VMD. ^31,37,38 Patch-clamp studies showed a reduced channel function, which was restituted after cotransfection with wild-type bestrophin, consistent with a loss of (channel) function mechanism of disease. ³⁷ However, biological events other than regulation of ion flow in the RPE, such as ceramide accumulation, may be involved in the bestrophin-associated disease process. ³⁹ Presently unknown additional genetic and environmental modifying factors may exert their influence on the phenotypic outcome. 

The well-known variability of clinical expression of BEST1 mutations within and between families, ^{17 –32} together with our findings on the extremely variable expressivity of subclinical Best VMD, are of particular interest in the era of assisted reproduction. ⁴⁰ Because prenatal diagnosis is difficult to offer in a disease with a variable expressivity and with known subclinical forms, preimplantation diagnosis is questionable, especially in individuals with subclinical forms, which have a risk to transmit the disease to descendants. 

The evidence of specific SD-OCT and EOG changes in 50% of BEST1 carriers with no manifest Best VMD symptoms or funduscopic lesions may be of help in families in which the mutation cannot be identified (for instance, mutations lying in unscreened BEST1 regions). These findings may indeed be valuable to detect asymptomatic carriers, and we propose them as a suitable follow-up. 

To date, the Human Gene Database (HGMD http://www.hgmd.cf.ac.uk/ac/all.php) reports the identification of 123 different BEST1 mutations, the very large majority of which are missense mutations or small indel that do not alter the reading frame. The nature of these mutations is consistent with the notion that Best disease-causing mutations in bestrophin-1 lead to a loss of Cl⁻ channel function with a dominant negative effect. However, few heterozygous null alleles have been reported, ^41,42 suggesting that, yet infrequent, bestrophin-1 haploinsufficiency may cause the disease. Furthermore, mutations of BEST1 splicing regulators have been demonstrated to cause dramatic phenotypes, including vitreoretinochoriopathy and nanophthalmos. ⁴³ These data, along with the very large variability of clinical presentation of Best disease, sometimes within families, suggest that both the nature of mutations and modifying factors may contribute to the phenotypic variability. It would be interesting to assess in future studies the consequence of missense mutations with variable clinical effect on the Cl⁻ channel function to precisely determine which of them are loss-of-function mutations. Moreover, when high-throughput sequencing will become available, it would be of major interest to perform association studies in Bestrophin-1 carrier patients to identify polymorphisms involved in the modulation of the phenotype. 

Footnotes

 Presented in part as a poster at the American Academy of Ophthalmology Annual Meeting, Chicago, October 14–19, 2010.

Footnotes

 Supported by the Centre de Reference Maladie Rare (CRMR) program.

Footnotes

 Disclosure: G. Querques, None; J. Zerbib, None; R. Santacroce, None; M. Margaglione, None; N. Delphin, None; L. Querques, None; J.-M. Rozet, None; J. Kaplan, None; E.H. Souied, None