A 16-year-old male with juvenile-onset BVMD (Patient 1) and a 43-year-old female with adult-onset BVMD (Patient 2), both from southern China, were diagnosed at Zhongshan Ophthalmic Center (Guangzhou, P.R. China). Visual acuity was examined using the Early Treatment Diabetic Retinopathy Study chart (Precision Vision, La Salle, IL, USA) (28). Images of the anterior segment were captured using a BX 900 Slit Lamp (Haag-Streit, Bern, Switzerland). Measurements of the anterior segment were recorded with Pentacam HR version 70700 (Oculus VR, LLC, Wetzlar, Germany). Optical Coherence Tomography (OCT) was performed by Cirrus HD-OCT (Carl Zeiss Meditec, Inc., Dublin, CA, USA). Fundus photography and fundus fluorescein angiography (FFA) imaging was performed using a Heidelberg Retina Angiograph (Heidelberg Engineering, Heidelberg, Germany). Multifocal electroretinography (mfERG) was performed to assess the amplitudes of the rod and cone responses using the Espion electrophysiology system (Diagnosys LLC, Littleton, MA, USA). Physical examinations were performed to exclude systemic diseases.

Venous blood samples were collected from the study subjects, their family members and 200 subjects without BVMD within the same population (15–48 years old; sex ratio: Male/female=108/92). Genomic DNA was extracted from peripheral blood leukocytes using standard protocols. Briefly, a total amount of 1 ml of blood sample was collected from each subject, lysed by red blood cell lysis buffer (Sigma Aldrich, Merck KGaA, Darmstadt, Germany), and centrifuged at 2,000 × g for 5 min at room temperature. Genomic DNA was extracted from peripheral blood leucocytes using a DNA extraction kit (Qiagen GmbH, Hilden, Germany) (29,30). Exons of the BEST1 gene were amplified by polymerase chain reaction (PCR) with primers as previously described (31–34). The primer sequences are listed in Table I. PCR was conducted in a 50 µl reaction system. using the PCR amplification kit (Takara Bio, Inc., Otsu, Japan). The amplification included a single 5 min step at 94°C, followed by 40 cycles of 94°C for 45 sec, 58–61°C for 45 sec, 72°C for 45 sec and a final 10 min step at 72°C. The PCR products were sequenced from both directions with an ABI3730 Automated Sequencer (PE Biosystems Inc., Foster City, CA). Sequenced products were analyzed using Seqman (version 2.3; Technelysium Pty Ltd., Brisbane, Australia), and compared with reference sequences in the database at the National Center for Biotechnology Information (NC_000011.10).

To analyze the effect of missense variants, polymorphism phenotyping (PolyPhen)and the sorting intolerant from tolerant (SIFT) algorithms were used to predict the possible impact of an amino acid substitution on the protein structure and function, using straightforward physical and comparative considerations (35–39). Variants were considered to be pathogenic when at least one of the two programs predicted a deleterious effect of the amino acid substitution on the protein structure and function. The Human Gene Mutation Database (http://www.hgmd.cf.ac.uk/ac/index.php) was used to screen mutations reported in published studies. HomoloGene (https://www.ncbi.nlm.nih.gov/homologene) was used to check whether the mutated amino acid residues were conserved across different species.

All experimental protocols and the methods were carried out in accordance with the guidelines approved by the ethics committee of Zhongshan Ophthalmic Center of Sun Yat-sen University (Guangzhou, P.R. China). Written informed consent was obtained from each subject in accordance with The Declaration of Helsinki. All participants provided informed consent for the publication of their data, including images and examination results.

Patient 1 had no known familial history of ocular disease. Refractive error was +4.50 diopter sphere (DS) in both eyes, with best corrected visual acuity (BVCA) at 1.0 in the right eye and 0.1 in the left eye. The cornea and the lens were transparent. Fundus examination revealed that the right eye had prominent yellow-white sub-retinal scarring with pigmented borders, surrounded by a serous retinal detachment. The left eye exhibited fragmented vitelliform lesions (Fig. 1A and B). FFA revealed a mild hyperfluorescence with moderate leakage in the fovea of the macula in both eyes (Fig. 1C and D). OCT scans revealed that the foveal region in both eyes was abnormally thick, due to neuroretinal detachment from the RPE. In the right eye, the neuroretinal detachment was likely triggered by the abnormal accumulation of hyperreflective materials beneath the retina (Fig. 2A), whereas in the left eye, it was likely due to the accumulation of sub-retinal fluid (Fig. 2B). mfERGs (response of the posterior fundus) revealed a mild decrease in the amplitude of the foveal response in both eyes, although the peripheral mfERG amplitudes were within normal limits (Fig. 2C and D).

Patient 2 had myopia and their decline in vision had occurred over the last 3 years. Refractive error was −10.0 DS in the right eye and −9.0 DS in the left eye. The BVCA was 0.4 in the right eye and counting fingers at 50 cm away from the left eye. The cornea and the lens were transparent. Fundus examination revealed atrophic lesions in both eyes (Fig. 3A and B). FFA revealed significant early hyperfluorescence that had increased intensity at the late stage of the angiographic sequence, with mild leakage in the right eye (Fig. 3C). The macular lesions exhibited a dystrophic pattern in the left eye (Fig. 3D). OCT revealed that the foveal regions in both eyes were abnormally thin due to atrophy of the retina and RPE (Fig. 4A and B). Similar to Patient 1, mfERGs of Patient 2 revealed a significant decrease in the amplitude of the foveal response in both eyes, although most of the peripheral mfERG amplitudes were within normal limits (Fig. 4C and D).

A heterozygous mutation c.903T>G (p.D301E) in exon 8 of the BEST1 gene was identified in Patient 1, but not in the unaffected family members or the normal controls in the same population. A c.1608C>T (p.T536T) single nucleotide polymorphism (SNP) in exon 10 of the BEST1 gene was identified in Patient 2 (Fig. 5A). Multiple sequence alignment performed using the HomoloGene database indicated that the residues at position 301 and 536 of bestrophin-1 are highly conserved (Fig. 5B). PolyPhen and SIFT predicted that the D301E amino acid substitution in bestrophin-1 was potentially damaging (Fig. 5C). SIFT predicted that the T536T substitution is tolerated.