Macular Dystrophy, Vitelliform, 2

A number sign (#) is used with this entry because vitelliform macular dystrophy-2 (VMD2), also known as Best disease, is caused by heterozygous mutation in the bestrophin gene (BEST1; 607854) on chromosome 11q12.

Description

Best vitelliform macular dystrophy is an early-onset autosomal dominant disorder characterized by large deposits of lipofuscin-like material in the subretinal space, which creates characteristic macular lesions resembling the yolk of an egg ('vitelliform'). Although the diagnosis of Best disease is often made during the childhood years, it is more frequently made much later and into the sixth decade of life. In addition, the typical egg yolk-like lesion is present only during a limited period in the natural evolution of the disease; later, the affected area becomes deeply and irregularly pigmented and a process called 'scrambling the egg' occurs, at which point the lesion may appear as a 'bull's eye.' The disorder is progressive and loss of vision may occur. A defining characteristic of Best disease is a light peak/dark trough ratio of the electrooculogram (EOG) of less than 1.5, without aberrations in the clinical electroretinogram (ERG). Even otherwise asymptomatic carriers of BEST1 mutations, as assessed by pedigree, will exhibit an altered EOG. Histopathologically, the disease has been shown to manifest as a generalized retinal pigment epithelium (RPE) abnormality associated with excessive lipofuscin accumulation, regions of geographic RPE atrophy, and deposition of abnormal fibrillar material beneath the RPE, similar to drusen. Occasional breaks in the Bruch membrane with accompanying neovascularization have also been reported, although Best disease is not noted for extensive choroidal neovascularization. Many of these features are also found in age-related macular degeneration (see 603075) (summary by Braley, 1966; White et al., 2000; Marmorstein et al., 2000; Leroy, 2012).

For a discussion of genetic heterogeneity of vitelliform macular dystrophy, see VMD1 (153840).

Clinical Features

Best (1905) described a family in which 8 persons were affected with hereditary vitelliform macular dystrophy. Follow-up of this family by Vossius (1921) and Jung (1936) increased the number of affected individuals to 22. Friedenwald and Maumenee (1951) observed affected mother and daughter. Davis and Hollenhorst (1955) described a kindred containing at least 24 affected persons in 5 generations. The age of onset of manifest visual disability varied from very early childhood to adolescence. Cystoid macular degeneration was described in a dominant pedigree pattern by Falls (1949) and Sorsby et al. (1956). Vail and Shoch (1965) followed up on an extensively affected kindred and reported histologic findings in a patient who died at 78 years of age.

Braley and Spivey (1964) examined 27 members of a large 4-generation Iowa family of Dutch ancestry, 10 of whom had vitelline macular degeneration. Onset of disease was before age 10 years in 3 cases, before age 20 in 3, before age 30 in 3, and in the early 30s in 2. At least 4 patients had sudden onset of visual loss that improved to some degree. Good visual acuity was present in some patients despite severe macular changes. In every patient with visual loss, a central scotoma was present that conformed to the position and size of the macular lesion. Dark adaptation was normal in all family members; color vision was normal in all unaffected family members, whereas most affected family members showed red-green deficiency. Braley and Spivey (1964) noted that not all patients exhibit the classic 'sunny side up' vitelliform lesion as the initial stage of macular degeneration.

In Sweden, Nordstrom and Barkman (1977) and Nordstrom and Thorburn (1980) traced 250 cases of Best disease to one gene source in the 17th century. An apparently homozygous father had 11 children, all of whom were affected. Age of onset varied from early childhood to the 40s and 50s. The electrooculogram (EOG) was helpful in preclinical detection. The range of severity was wide among the 11; indeed, one, aged 24, could be identified only by pathologic EOGs. The homozygotic state did not differ from the heterozygotic state.

O'Gorman et al. (1988) described the histopathologic findings in the postmortem eyes of a 69-year-old man with this disorder. Retinal pigment epithelial (RPE) cells across the entire fundus had accumulated an excessive amount of lipofuscin as defined by ultrastructural appearance, autofluorescence, and staining properties. An accumulation of heterogeneous material located between Bruch membrane and the pigment epithelium in the fovea was interpreted as representing a previtelliform lesion. The material appeared to be derived from degenerating pigment epithelial cells and contained few intact lipofuscin granules. Foveal photoreceptor loss occurred above the lesion.

Brecher and Bird (1990) investigated the families of 12 probands who presented with foveal lesions typical of adult vitelliform macular dystrophy and found familial involvement compatible with autosomal dominant inheritance in 10 families. In the remaining 2 families, no familial involvement was detected, but both parents were not available for examination. Over half (14 of 25) of patients with abnormal fundi were asymptomatic, and most had good visual acuity, although 2 patients had visual acuities of less than 20/60 in both eyes.

Weber et al. (1994) identified a 37-year-old male who appeared to represent nonpenetrance of Best disease because he had inherited the haplotype associated in his family with the disorder, but showed no signs of the disease on repeated examination and EOG.

By optical coherence tomography (OCT) in a case of BMD, Vedantham and Ramasamy (2005) found that lipofuscin accumulated in a cystic space under the retinal pigment epithelium in the 'pseudohypopyon' stage of the disease, and that disruption of photoreceptors occurred in the 'scrambled egg' stage. The authors suggested that these findings explain the retention of good visual acuity in the pseudohypopyon stage and the loss of visual acuity in the scrambled egg stage.

Using indocyanine green angiography (ICG), Maruko et al. (2006) observed hyperfluorescent spots throughout the peripheral fundus in all 8 eyes of 4 patients with Best disease. The extensive distribution of the spots was consistent with the wide-ranging abnormalities of the retinal pigment epithelium, Bruch membrane, and choroid that have been observed histopathologically.

In an eye from a Best disease donor with a T6R mutation in the BEST1 gene, Mullins et al. (2007) found deposits containing lipid and glycoconjugates within the eye's central retinal scar. Immunohistochemical localization of bestrophin in a series of 22 unaffected eyes revealed a pattern in which macular labeling was less robust than labeling outside the macular area in 18 of the 22. Mullins et al. (2007) concluded that topographic differences in the levels of bestrophin protein might in part explain the propensity for the macula to develop lesions.

Clinical Variability

Mullins et al. (2005) restudied a male patient from the Iowa family of Dutch ancestry originally reported by Braley and Spivey (1964), in which a missense mutation in the BEST1 gene (607854.0004) was identified by Petrukhin et al. (1998). The patient had photographically documented normal maculae at age 51 years, but subsequently developed small vitelliform lesions at age 75 years, followed by widespread flecks in the midperiphery; 2 additional family members exhibited similar multifocal lesions. Histologic examination showed that the flecks represented clusters of vesicular drusen that were less eosinophilic than typical drusen but were otherwise of similar composition. Mullins et al. (2005) noted that vitelliform lesions had been documented to develop as late as 60 years of age in a patient with classic Best disease (Sorr and Goldberg, 1976), and that midperipheral flecks, while uncommon, may be present. Review of 77 consecutive photofiles of patients with a clinical and molecular diagnosis of Best disease revealed 7 patients (9.1%) from 3 unrelated families who also exhibited multifocal lesions consisting of small peripheral flecks.

Boon et al. (2007) studied 15 unrelated patients with multifocal vitelliform lesions. Age at onset was highly variable, ranging from 5 to 59 years. The peripheral lesions varied in number, size, and overall appearance, but showed similar characteristics on autofluorescence imaging and OCT compared with the central vitelliform lesion.

Lee et al. (2012) studied 2 unrelated, initially asymptomatic male patients who exhibited incidentally discovered bilateral macular atrophic lesions at age 30 years and 51 years, respectively, with serous retinal detachment in the macula on OCT and multiple leakages around the central hypofluorescent area as well as partially dilated choroidal vessels on fluorescein angiography. The lesions were thought to represent chronic central serous chorioretinopathy but were unresponsive to treatment. Reevaluation revealed yellowish deposits at the border of serous retinal detachment areas; OCT showed hyperreflective lesions between the RPE and outer segment layers of the retina, and fundus autofluorescence (FAF) showed ring-like hyperautofluorescence around the serous retinal detachment. Both patients also had decreased Arden ratios on EOG and were found to have mutations in the BEST1 gene, resulting in a diagnosis of atypical vitelliform macular dystrophy.

Parodi et al. (2014) studied the fundus autofluorescence patterns in the eyes of 4 patients with a bilateral subclinical form of Best disease (positive testing for BEST1 gene mutation, fully preserved best-corrected visual acuity, normal fundus appearance) and the clinically unaffected eyes of 2 patients with unilateral Best disease. Short-wavelength FAF findings were consistently normal, whereas near-infrared FAF showed an abnormal pattern marked by a central hypoautofluorescence surrounded by a round area of hyperautofluorescence. Microperimetry corroborated the near-infrared FAF pattern. No changes were found in a 24- to 36-month follow-up of the patients.

Mapping

Definitive mapping of the locus for Best macular dystrophy was achieved by Stone et al. (1992), who studied a 5-generation family with 29 affected persons. Linkage analysis located the gene on chromosome 11q13. Multipoint analysis yielded a maximum lod score of 9.3 for location in the interval between INT2 (FGF3; 164950) and D11S871. Using 8 microsatellite markers, Weber et al. (1994) studied 3 multigenerational Best disease families and refined the localization of the disease gene to a 3.7-cM interval between markers at D11S903 and PYGM (608455). PCR-hybrid mapping sublocalized this interval to the pericentromeric region of chromosome 11. Identification of 3 distinct haplotypes associated with the disease in the 3 families strongly suggested independent origins of the mutations.

In a large Swedish family with more than 250 cases of Best macular dystrophy (see Nordstrom and Barkman, 1977), descended from a couple born in central Sweden in the 17th century, Forsman et al. (1992) obtained a lod score of 15.12 at a recombination fraction 0.01 for linkage with an INT2 marker on chromosome 11q13. Thus, the gene for rod outer segment protein-1 (ROM1; 180721), which maps to the same region, became the leading candidate for the site of the mutation in this disorder. Stone et al. (1992) likewise demonstrated genetic linkage of Best disease to 11q13, and Bascom et al. (1992) presented evidence that the ROM1 gene may be the site of the mutation in Best disease. Using highly polymorphic markers, Nichols et al. (1994) narrowed the genetic region that contains the Best disease gene to the 10-cM region between markers D11S871 and PYGM. Marker D11S956 demonstrated no recombinants with Best disease in 3 large kindreds and resulted in a lod score of 18.2.

Using a combination of single-strand conformation polymorphism (SSCP) analysis, denaturing gradient gel electrophoresis, and DNA sequencing to screen the entire coding region of the ROM1 gene in 11 different unrelated patients with Best disease, Nichols et al. (1994) could find no nucleotide changes in the coding sequence of any affected patient. They concluded that mutations within the coding sequence of ROM1 are unlikely to cause Best disease. Graff et al. (1994) localized the VMD2 locus to the 6-cM genetic interval between 2 DNA markers, one of which was associated with ROM1 in a large Swedish 12-generation kindred. Mutation analyses of ROM1 revealed no mutations that could explain the disease phenotype. Furthermore, one recombinant event between intragenic ROM1 polymorphisms and the Best disease phenotype was detected. Thus, ROM1 was excluded as the site of the disease-causing mutations in this kindred. Coding sequence mutations in ROM1 were also excluded by Hou et al. (1996) in 2 affected members of a large 5-generation North American pedigree with Best macular degeneration mapping to 11q.

By studying the large Swedish VMD2 family dating back to the 17th century (Nordstrom and Barkman, 1977), Graff et al. (1997) refined the VMD2 region to a span of approximately 980 kb flanked by D11S4076 and uteroglobin (UGB; 192020). Stohr et al. (1998), who gave the location of the VMD2 gene as a region of approximately 1.4 Mb on 11q12-q13.1, assembled a high-coverage YAC contig of this region. They constructed a primary transcript map that placed 19 genes within the region.

Genetic Heterogeneity

Mansergh et al. (1995) established genetic heterogeneity in this disorder by finding linkage to chromosome 11 in an Irish family and excluding linkage to chromosome 11 in a German family.

Diagnosis

Chacon-Camacho et al. (2011) performed optical coherence tomography (OCT) in symptomatic and asymptomatic individuals from 2 Mexican families segregating Best disease caused by mutation in the BEST1 gene. Symptomatic patients showed severe retinal serous retinal detachment in both families. In 1 family, an 8-year-old carrying a Q293K mutation was demonstrated to have Best disease-related retinal lesions, i.e., bilateral subfoveal lesions and unilateral serous retinal detachment. Conversely, in the other family, an asymptomatic 6-year-old carrying a W24C mutation did not demonstrate retinal abnormalities. Chacon-Camacho et al. (2011) suggested that OCT can be used during early childhood for presymptomatic diagnosis of some cases of the disease.

Molecular Genetics

In several Swedish and Dutch families with Best macular dystrophy, including the large Swedish family reported by Nordstrom and Barkman (1977) and studied by Graff et al. (1997), and the Iowa family of Dutch ancestry originally reported by Braley and Spivey (1964), Petrukhin et al. (1998) identified 5 different mutations in the VMD2 gene (607854.0001-607854.0005) that segregated with the disease.

Caldwell et al. (1999) analyzed the bestrophin gene in 13 families with Best macular dystrophy and identified heterozygous mutations in 9 families, including 6 missense mutations and a 2-bp deletion (607854.0012). In 3 of the families, there was a parent carrying the missense mutation who lacked the clinical phenotype, suggesting variable expression of the disease gene. Caldwell et al. (1999) found no mutations in the bestrophin gene in the large North American family with Best macular dystrophy previously mapped to chromosome 11q by Hou et al. (1996).

In 2 unrelated women who had vitelliform macular dystrophy diagnosed in the sixth decade of life, Allikmets et al. (1999) identified heterozygous missense mutations in the BEST1 gene (E119Q, 607854.0008 and A146K 607854.0009).

Kramer et al. (2000) identified several mutations in the VMD2 gene in German patients with macular dystrophy of juvenile and adult onset (see, e.g., 607854.0005 and 607854.0010-607854.0011) and suggested that the adult-onset patients represented a mild form of Best disease.

White et al. (2000) stated that 48 different mutations, predominantly missense mutations, had been described in the VMD2 gene in Best disease families.

Schatz et al. (2006) identified mutations in the BEST1 gene in all 6 affected members of a 3-generation Swedish family with Best macular dystrophy. One was heterozygous for an arg141-to-his (R141H; 607854.0013) mutation, 3 were heterozygous for a tyr29-to-ter (Y29X; 607854.0014) mutation, and 2 were compound heterozygous for these mutations. The 2 members who were compound heterozygous had a more severe phenotype.

In 9 (60%) of 15 unrelated patients with multifocal vitelliform lesions, Boon et al. (2007) identified heterozygosity for mutations in the BEST1 gene (see, e.g., 607854.0005).

In a 15-year-old proband with multifocal VMD, Wittstrom et al. (2011) identified compound heterozygosity for 2 mutations in the BEST1 gene: the R141H mutation and a de novo P233A substitution. The R141H mutation was present in heterozygosity in her asymptomatic mother and brother, both of whom showed delayed implicit times in a- and b-waves of combined total rod and cone full-field ERG responses.

Genotype/Phenotype Correlations

Boon et al. (2007) compared the clinical findings in patients with multifocal vitelliform retinal dystrophy with or without mutations in the BEST1 gene. All 9 patients with BEST1 mutations had abnormal EOG findings compared with 2 of 6 patients without BEST1 mutations; in addition, those with a mutation had a highly variable but seemingly younger age at onset and a more pronounced loss of visual acuity.

Meunier et al. (2014) reviewed 76 families with vitelliform macular dystrophy and found that 24 (53%) of 45 families with onset of disease before 40 years of age had a mutation in the BEST1 gene, whereas 3 (9.7%) of 31 families with onset after 40 years of age had a mutation in the PRPH2 gene (179605). For the remaining 49 families without a mutation in BEST1 or PRPH2, 3 (6%) had a mutation in the IMPG1 gene and 1 (2%) in the IMPG2 gene (607056). Meunier et al. (2014) stated that the IMPG1 and IMPG2 vitelliform macular dystrophies are characterized by late-onset moderate visual impairment, frequent association with drusen-like lesions, preservation of RPE reflectivity, lack of sub-RPE deposits on spectral-domain optical coherence tomography (SD-OCT), and normal or borderline results on EOG. The authors noted that although patients with a BEST1 mutation were most frequently symptomatic before the age of 40 years, there was overlap with PRPH2 patients in terms of age of onset; in addition, the presence of small satellite drusen-like lesions in the foveal area appeared to implicate the IMPG1 or IMPG2 genes.

Animal Model

In dogs with canine multifocal retinopathy (cmr), which resembles human Best disease, Guziewicz et al. (2007) identified 2 disease-specific sequence alterations in the VMD2 gene: a 73C-T stop mutation (R25X), designated cmr1, and a 482G-A missense mutation (G161D), designated cmr2. Guziewicz et al. (2007) proposed that canine cmr is a relevant animal model for Best disease.

Zhang et al. (2010) generated knockin mice carrying the Best vitelliform macular dystrophy-causing mutation W93C (607854.0001) in Best1. Both Best1(+/W93C) and Best1(W93C/W93C) mice had normal ERG a- and b-waves, but exhibited an altered light peak luminance response reminiscent of that observed in Best macular dystrophy patients. Morphologic analysis identified fluid- and debris-filled retinal detachments in mice as young as 6 months of age. By 18 to 24 months of age, Best1(+/W93C) and Best1(W93C/W93C) mice exhibited enhanced accumulation of lipofuscin in the retinal pigment epithelium (RPE), and a significant deposition of debris composed of unphagocytosed photoreceptor outer segments and lipofuscin granules in the subretinal space. The RPE cells from Best1(W93C) mice exhibited normal chloride conductances, and ATP-stimulated changes in calcium concentration in RPE cells from Best1(+/W93C) and Best1(W93C/W93C) mice were suppressed relative to Best1 +/+ littermates. The authors hypothesized that Best vitelliform macular dystrophy does not occur because of Best1 deficiency, as the phenotypes of Best1(+/W93C0) and Best1(W93C/W93C) mice are distinct from that of Best1 -/- mice with regard to lipofuscin accumulation and changes in the light peak and ATP calcium responses.

History

Rivas et al. (1986) proposed that the segment 6q25-qter contained the locus for a dominant macular degeneration. This proposal was based on the finding of macular degeneration in an 8-month-old girl with a de novo deletion of 6q25 and in another case of terminal deletion of 6q reported by Hagemeijer et al. (1977). In linkage studies in 9 kindreds, Yoder et al. (1988) found no firm evidence for linkage with 18 informative markers; the highest positive lod score was 0.57 for glutamate-pyruvate transaminase (GPT; 138200) on chromosome 8q24 at a recombination fraction of 0.30.