A digenic contribution of RPGRIP1 and SLC4A4 to juvenile-onset open-angle glaucoma phenotype with concomitant corneal dystrophy

Juvenile open-angle glaucoma is an uncommon form of glaucoma, generally associated with high IOP and a significant visual field loss.

A 15-year-old boy presented with decreased vision in both eyes (BE). His visual acuity with corrective glasses was 6/9 in the right eye (RE) and 6/6 in the left eye (LE). He had a myopic refractive error of −9 D(s) in the RE and −10.5 D(s) in the LE. His intraocular pressure (IOP) in BE was recorded as 50 mmHg, for which he eventually underwent trabeculectomy in BE. He had normal open angles on gonioscopy.

There was no abnormality in the anterior segment, except for a mild corneal haze that was initially attributed to his high IOP [Fig. 1a]. A careful slit-lamp examination revealed the presence of hyperreflective granularity in the anterior stroma [Fig. 1b]; this was confirmed on anterior segment optical coherence tomography, which showed diffuse areas of hyperreflectivity in the Bowman’s layer and anterior stroma [Fig. 1c]. On confocal microscopic examination of the corneal layers, we found a granular, hyperreflective appearance of the keratocytes in the corneal stroma, with multiple differently oriented dark striae, which were especially prominent, dark, thick, and multilayered at 100–150 μm depth [Fig. 1d and e]. On fundus examination, he had advanced cupping in BE [Fig. 1f], with peripapillary atrophy and myopic chorioretinal degeneration. His central corneal thickness (CCT) was 685 µ in RE and 689 µ in LE, and axial length was 25.9 mm and 26.75 mm respectively.

Figure 1.

Anterior segment photo of the proband showing corneal haze on diffuse illumination (a). Hyperreflectivity can be seen in the anterior stroma on focal illumination (b) and anterior segment optical coherence tomography (c). Confocal microscopy revealed longitudinal striae (white arrows) among hyperreflective keratocytes (black arrowheads) at 168 µ (d), as well as at a deeper level (235 µ) of the cornea (e). Disc photograph of the proband showed the presence of advanced glaucomatous cupping and myopic chorioretinal atrophy (f)

The child was born of a third-degree consanguineous marriage and was the second of three siblings [Fig. 2]. The mother and his siblings were examined, and none had disc changes suggestive of glaucoma. His father, however, had raised IOP of 24 mmHg in BE, with normal open angles on gonioscopy. His CCT was 578 µ in RE and 576 µ in LE. Neither the father nor the proband had any evidence of retinitis pigmentosa [Suppl Figs. 1 ^{(1.1MB, tif)} and 2 ^{(1MB, tif)} ], with both having a normal peripheral retina, as seen on wide-field fundus evaluation and fundus autofluorescence. The son had severely constricted visual fields (VF), with central 10° remaining in BE due to advanced glaucoma. The father had a normal VF in RE; however, an early inferior defect could be seen in the LE [Suppl Fig. 3 ^{(3MB, tif)} ].

Figure 2.

Pedigree of the family showing RPGRIP1 c.931-3_935delCAGAATCA and SLC4A4 c.2058T>A variants. The affected status of the proband with JOAG and corneal dystrophy is shown in black, while grey (father) represents POAG. +/− denotes a heterozygous genotype, and +/+ denotes a homozygous genotype. Y indicates age in years (at the time of the study)

Paired-end whole-exome sequencing (WES) for the parents and the proband was carried out after informed consent. The WES (DNA fragmentation, library preparation, and solution-phase hybrid capture) was performed according to the manufacturer’s instructions (Agilent Technologies). The enriched library was sequenced using the Illumina HiSeq 2500 platform with 150-bp paired-end reads. The read depth was at least 100×. The Burrows–Wheeler Aligner software tool[1] was used to align the sequencing reads with hg19/GRCh37. Bioinformatic analysis was performed using Genome Analysis Toolkit (GATK),[2] and variants were annotated and interpreted using Varseq Ver 2.0 (www.goldenhelix.com/products/VarSeq/index.html). Nonsynonymous missense variants, frameshift variants, and indels in exonic and splice regions that were either likely pathogenic or pathogenic were selected from among known glaucoma genes with a minor allele frequency of <0.001. For genetic evaluation, we focused our search on already known genes associated with JOAG namely, MYOC, CYP1B1, LTBP2, OPTN, ASB10, WDR36, EFEMP1, NTF4, TBK1, FOXC1, and PAX6. We also looked for genes associated with corneal stromal dystrophies in the family.

Copy number variant (CNV) analysis was performed to identify CNVs in the known glaucoma genes. Homozygosity mapping was used to identify long stretches of homozygous haplotypes in genes associated with glaucoma.

We found a novel deletion (NM_020366.4:c. 931-3_935delCAGAATCA, chr14:21779980delCAGAATCA, and p. N311Gfs*8) of the last three base pairs of intron 8 and the first five base pairs of exon 9 in RPGRIP1 gene, in both the proband and his father. This deletion is predicted to be likely pathogenic according to the American College of Medical Genetics and Genomics (ACMG) criteria. Phylogenetic conservation assessment of p.N311, using the protein sequence of RPGRIP1 (NP_065099.3) as an input file in the National Center for Biotechnology Information (NCBI) Multiple Sequence Alignment Viewer, (Version 1.22.0), revealed it to be highly conserved in over 100 species. Homozygosity mapping did not reveal any large regions among genes known for glaucoma. We also found a novel homozygous missense variant in the SLC4A4 gene, NM_003759.4:c. 2058T>A (Chr 4:72397802, p.Phe686Leu), in the proband. Both his parents carried a heterozygous allele. This variant is predicted to be likely pathogenic according to ACMG criteria.

Considering that RPGRIP1 co-segregated with the high IOP in this family and that no other known glaucoma gene could be found, we are inclined to consider the association of RPGRIP1 with raised IOP in the proband and his father. Biallelic RPGRIP1 variants are generally associated with Leber congenital amaurosis or cone-rod dystrophy. The mechanism of how RPGRIP1 variants could cause a rise in IOP is unknown, though we can hypothesize that, since it encodes scaffold protein isoforms, it might affect the binding of other proteins in the trabecular meshwork. The RPGRIP1 gene lies in the 14q11.2 locus, which has been associated with POAG.[3] Fernandez-Martinez et al. implicated the RPGRIP1 gene as a susceptibility gene in various forms of glaucoma, including Juvenile onset open angle glaucoma (JOAG).[4] Another report describing the association of the RPGRIP1 gene with glaucoma was by Shah et al., who reported six members from a large Primary open angle glaucoma (POAG) family of 84 subjects with a heterozygous RPGRIP1 variant.[5]

While the father had early glaucoma, the proband had a much severe disease, with higher IOP and advanced glaucomatous optic atrophy. Bi-allelic mutations in SLC4A4 have been known to be associated with band-shaped corneal dystrophy and congenital glaucoma without renal tubular acidosis.[6] However, diffuse corneal haze with myopia, as seen in our case, has also been described.[7] We believe the homozygous mutation in the SLC4A4 gene in the proband was the primary cause of glaucoma, with the RPGRIP1 variant contributing to the severity of the glaucoma phenotype. In our case, variants were present in the two genes already known for their association with glaucoma; however, as this is only a single small family, in which statistical evidence of phenotype–genotype linkage cannot be ascertained, we can only hypothesize the existence of a digenic contribution of both RPGRIP1 and SLC4A4 genes leading to a severe JOAG phenotype.

Data availability

Data supporting the findings of the study are available from the corresponding author (VG) on request.

Author contributions

Concept and design: VG, VM; Data Collection: SG, AP, MAM, AS; Writing: VG, MAM, and AP; Overall responsibility: Viney Gupta.

Ethics approval

The study was approved by the Institute Ethics Committee, AIIMS, New Delhi, India.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Suppl Figure 1

Wide-field fundus photography (a and b) and fundus autofluorescence (c and d) of the proband revealed no signs of retinal dystrophy. His dark-adapted 0.01 (e) full-field electroretinogram showed reduced b-wave amplitude in BE due to advanced glaucoma

Suppl Figure 2

Wide-field fundus photographs (a and b) and fundus autofluorescence (c and d) of the father revealed no peripheral retinal abnormality; although a wedge-shaped nerve fiber defect could be seen superotemporal to the disc in the LE. His dark-adapted 0.01 (e) full-field electroretinogram, was diminished, but less so compared to his son

Suppl Figure 3

Visual fields of the proband (a and b) and his father (c and d), showing BE severely depressed visual fields in the proband and an early inferior scotoma in the LE of the father (d)