Retinitis pigmentosa (RP) is the most common kind of inherited retinal dystrophy, which is a genetically heterogeneous group associated with progressive photoreceptor degeneration [1]. The typical clinical presentation of RP includes night blindness and progressive visual field loss at a young age [1]. The presence of X-linked inherited RP (XLRP) is estimated to occur in 4%-22% of all RP cases [2]. Although female carriers show various degrees of clinical characteristics of RP, severe clinical features tend to occur in affected male individuals [3]. The RP2 gene mutation is known to be a representative cause of XLRP [4]. The RP2 gene is composed of five exons and encodes 350 amino acids. RP2 is also known as the ARL3 GTPase-activating protein. It is expressed in the retinal pigment epithelium (RPE) and photoreceptor cell membrane and is involved in β-tubulin folding [5].

The transforming growth factor β-induced (TGFBI) gene, which was originally named βig-h3, produces a 34,924-nucleotide transcript containing 17 exons. The TGFBI mRNA encodes the TGF β-induced protein (TGFBIp), which consists of 683 amino acids, an amino-terminal signal peptide, and four internal fasciclin-1 domain repeats [6]. In the human eye, TGFBI is expressed almost exclusively in the cornea and the RPE [7]. TGFBI mutations are a cause of corneal dystrophies, including granular corneal dystrophy, but their association with other ocular abnormalities remains unknown [8,9].

In this case report, we present the clinical features and characteristics of two patients with RP2 exon1 deletions (c.-37_c.102+581del (720bp)) occurring in a family with a single heterozygous TGFBI point mutation (c.371G>A(p.(Arg124His))).

This case report was approved by the Institutional Review Board of Asan Medical Center and was conducted in adherence with the tenets of the Declaration of Helsinki. We present the details of two male siblings (case 1: 16 years old, case 2: 14 years old) diagnosed with RP by ophthalmologists at different major medical institutions. Both brothers underwent ophthalmology examinations when they were 4 years of age due to frequent falls caused by night blindness and decreased vision, and both were diagnosed with RP at other hospitals. They have a maternal family history of granular corneal dystrophy and presented to our hospital for genetic counseling and treatment (Supplementary Fig. 1).

Analysis of genetic variants

Genomic DNA was extracted from peripheral blood samples taken from the patients. Targeted next-generation sequencing (TGS) was performed using the Ion Torrent S5XL platform (Thermo Fisher Scientific Inc.) involving a panel considering 88 genes associated with RP (Supplementary Table 1). All exons of all genes (approximately 22,000) were captured using a SureSelect kit (ver. C2; Agilent Technologies Inc.), and the captured genomic regions were sequenced using a NovaSeq platform (Illumina Inc.). Analysis of the raw genome-sequencing data included alignment to the reference sequence (U.S. National Center for Biotechnology Information genome assembly GRCh37; accessed in February 2009). For TGS, the mean depth of coverage was approximately 500-fold, and 99.2% of the coverage was > 20-fold. Variant calling, annotation, and prioritization were performed as previously described [10]. The requirement for verifying the identified variants was waived for Torrent S5XL sequencing data when the read depth was > 100 reads and the allele frequency was 40%-60%, while validation of Illumina NextSeq and whole-exome sequencing (WES) data was achieved by subsequent Sanger sequencing [11]. Whole genome sequencing (WGS) was conducted through the National Bio Big Data Project. Sequenced reads were mapped to the human reference genome (GRCh38), and the duplicated reads were removed by Illumina DRAGEN server ver. 4.03 (Illumina, Inc.). We identified base substitutions, and short indels and variants called passed were included in the subsequent analysis. Genomic rearrangements were identified according to work by Rausch et al. [12], and we visually inspected and confirmed the breakpoints of the genomic rearrangements of interest using the integrative Genomics Viewer [13].

Case 1

The older patient had a best-corrected visual acuity (BCVA) of 10/200 in both eyes and high myopia with refractive errors of -9.25 diopters (D) (right eye) and -7.25 D (left eye). Granular corneal stromal opacity was observed in both eyes during anterior segment examination (Fig. 1A, B). Fundus photography revealed symmetric diffuse retinal dystrophy with hyperpigmented bony spicules (Fig. 1C, D).

Fig. 1. Ophthalmic imaging of case 1. Granular corneal stromal opacity can be seen in anterior segment photographs (A, B). Pseudo-color fundus images (C, D) show a bone spicule pattern of hyperpigmentation, and barrier laser scars are visible in the inferior temporal peripheral retina in both eyes. Spectral-domain optical coherence tomography images (E, F) show diffuse thinning of the entire retina and disruption of the ellipsoid zone without macular edema.

Fundus autofluorescence imaging showed macular hyperfluorescence with surrounding irregular geographic hypofluorescent areas. Optical coherence tomography (OCT) imaging of the macula revealed diffuse thinning of the neurosensory retina and disruption of the ellipsoid zone with no intraretinal or subretinal fluid (Fig. 1E, F). Full-field electroretinography showed nearly complete rod-cone system dysfunction in both eyes (Fig. 2A, C). An automated visual field test revealed concentric visual field loss sparing the macula (Fig. 2B, D).

Fig. 2. Functional test results for case 1. Full-field ERG revealed the near-complete loss of rod and cone responses (A, C; right and left eyes, respectively), while the automated visual field test revealed peripheral visual field loss sparing the central field (B, D; right and left eyes, respectively). ERG = electroretinography; OD = oculus dexter; OS = oculus sinister.

Case 2

The younger patient also presented with high myopia, with refractive errors of -10.5 D (right eye) and -9.75 D (left eye). BCVA was again 10/200 in both eyes. There was no visible corneal opacity in either eye found during anterior segment examination (Fig. 3A, B). However, fundus examination revealed retinal dystrophy with bone spicule hyperpigmentation (Fig. 3C, D). Additionally, fundus autofluorescence imaging revealed irregular areas of hyper- and hypofluorescence in the macula. The OCT (Fig. 3E, F), full -field electroretinography (Fig. 4A, C), and visual field test (Fig. 4B, D) findings were identical to those of case 1.

Fig. 3. Ophthalmic imaging of case 2. No corneal stromal opacity was seen on the anterior segment photographs (A, B). Pseudo-color fundus images (C, D) show a bone spicule pattern of hyperpigmentation, while spectral-domain optical coherence tomography images (E, F) show diffuse thinning of the entire retina and disruption of the ellipsoid zone without macular edema.

Fig. 4. Functional test results for case 2. Full-field ERG revealed almost complete loss of rod and cone responses (A, C; right and left eyes, respectively). The automated visual field test revealed peripheral visual field defects with central sparing; however, these results were of relatively low reliability due to high fixation loss in both eyes (B, D; right and left eyes, respectively). ERG = electroretinography; OD = oculus dexter; OS = oculus sinister.

In both cases, a causative mutation for RP was not identified by TGS. However, during subsequent WES, a single heterozygous TGFBI point mutation (c.371G>A(p.(Arg124His))) was confirmed in both patients. WGS was conducted to locate genetic abnormalities related to retinal dystrophy and revealed a novel RP2 exon1 deletion (c.-37_c.102+581del (720bp)).

Patients’ mother

The patients’ mother had a BCVA of 20/20 in both eyes and mild myopia with refractive errors of -2.25 D (right eye) and -2.00 D (left eye). She did not complain of night blindness, color vision disorders, or subjective visual field disturbances. However, granular corneal stromal opacity was confirmed in both eyes by anterior segment examination (Supplementary Fig. 2A, B). Fundus photography (Supplementary Fig. 2C, D) and OCT imaging (Supplementary Fig. 2E, F) showed none of the abnormalities associated with retinal dystrophy. She also underwent WES and WGS, and a single heterozygous TGFBI point mutation (c.371G>A(p.(Arg124His))) and RP2 exon1 deletion (c.-37_c.102+581del (720bp)) were confirmed.

This report is the first to present two cases of XLRP with a simultaneous TGFBI point mutation.

XLRP caused by RP2 mutations is one of the most severe forms of RP. Depending on the genetic abnormality and its penetrance, the carrier female will also show variability in phenotypic severity ranging from no visual changes to severe visual loss [3]. The RP2 protein is involved in ciliogenesis, photoreceptor integrity, and protein trafficking [4]. Most RP2 mutations destabilize the RP2 protein and cause its degradation via the proteasome pathway [14]. Such mutations result in truncated or misfolded proteins that impair the normal function of the retinas. Characteristically, most RP2-related RP patients experience early onset of macular atrophy, poor visual acuity, high myopia, and severe rodcone dysfunction [4]. Interestingly, a previous report showed that exon 4 and 5 deletion of RP2 does not seem to trigger phenotype abnormality in heterozygote female carriers [15]. Another study suggested that exon1 deletion of the RP2 gene could present as a less severe phenotype of XLRP [4]. In the current study, the patients’ mother presented with granular corneal dystrophy without retinal dystrophy. Meanwhile, her affected male offspring presented with typical clinical characteristics of RP2-related RP, with early-onset severe retinal dystrophy and high myopia [16]. A recent study reported that the majority of RP2-associated retinal dystrophy patients suffer complete loss of the foveal photoreceptor layer by the third decade of life [16]. In our cases, widespread loss of the foveal ellipsoid zone was already observed during the initial evaluation. Additional case reports will need to be reviewed in the future; however, the exon1 deletion (c.-37_c.102+581del (720bp)) of the RP2 gene does not appear to trigger phenotype alterations in female carriers but does cause the typical severe RP phenotype in affected male individuals.

The physiological role of TGFBIp is not fully understood. The ability to bind to both cell surface integrin and extracellular matrix molecules explains its role in cell adhesion and as a matricellular protein. However, the effects of TGFBI mutation on the human retina have not yet been elucidated, although it is possible to estimate the effect based on the results of experimental genetic engineering animal models. A previous study using TGFBI-knockout mice confirmed that TGFBI is expressed in the basal side of the RPE and that TGFBI knockout inhibits normal apoptosis in the inner nuclear layer by increasing the activity of the pro-survival extracellular signal-regulated protein kinase pathway, thereby interfering with retinal maturation [17]. In addition, the accumulation of lipofuscin (an indicator of aging) was confirmed in multiple organs in a transgenic mouse line overexpressing TGFBI, and obvious age-dependent retinal dystrophy was observed both histologically and functionally compared to in wild-type mice [18]. It was subsequently suggested that the disturbance of TGFBI may affect photoreceptor survival and potentially induce accelerated aging in several tissues. In the case of XLRP caused by RP2 mutation, there are reports of a spectrum of relevant conditions ranging from mild phenotypes to early-onset severe retinal dystrophies [4]. The severe phenotypes in our patients may be attributed to presumptive loss of the RP2 protein due to exon1 deletion. In addition, the experimental impact of TGFBI point mutation on photoreceptors may have implications for the severity of retinal dystrophy.

In conclusion, the management and genetic assessment of individuals presenting with unexplained RP should encompass an evaluation of potential known genetic aberrations contributing to the emerging phenotype, as well as the potential involvement of at least two distinct, infrequent, unidentified mutations, as exemplified in these cases. Retinal specialists who are responsible for the diagnosis and treatment of such patients are encouraged to formulate comprehensive strategies that incorporate appropriate genetic testing and the subsequent interpretation of its results.

The authors declare no conflicts of interest relevant to this article.

Conception (J.Y.L., B.H.L.); Design (S.U.C.); Data acquisition (S.L., Y.S.J., J.S.L.); Analysis (S.L., Y.S.J., J.S.L.); Interpretation (J.Y.L., B.H.L., S.U.C.); Writing (S.U.C.); Review (J.Y.L.); Final approval of the article (All authors)