Late-Onset Slowly Progressing Cone/Macular Dystrophy in Patients With the Biallelic Hypomorphic Variant p.Arg1933Ter in RP1

Abstract

Purpose: Homozygous hypomorphic variants of the RP1 gene, including c.5797C>T, p.Arg1933Ter, have traditionally been considered non-pathogenic. This study aimed to elucidate the clinical manifestations of late-onset, slowly progressive cone/macular dystrophy in patients homozygous for p.Arg1933Ter in the RP1 gene.

Methods: Five patients with biallelic p.Arg1933Ter in RP1 were retrospectively recruited, and their clinical profiles were analyzed. Copy number variation analysis and Alu insertion assessment of genes associated with inherited retinal diseases were conducted. The results of comprehensive ophthalmological examinations, multimodal imaging, and full-field electroretinogram tests were analyzed.

Results: No specific sequencing errors or structural variations associated with the clinical phenotypes were identified. Alu element insertion in RP1 was not detected. The mean ± SD age at the first visit was 62.2 ± 9.8 years, with symptoms typically starting between 45 and 50 years of age. Two patients exhibited a mild form of cone/macular dystrophy, characterized by a relatively preserved fundus appearance and blurring of the ellipsoid zone on optical coherence tomography. Three patients had late-onset cone/macular dystrophy with significant atrophy.

Conclusions: To our knowledge, this study is the first to report that a homozygous hypomorphic variant of RP1, previously considered non-pathogenic, leads to cone/macular dystrophy.

Translational Relevance: The study introduces novel possibilities suggesting that the homozygous hypomorphic variant of RP1 may be linked to variant pathogenicity.

Introduction

RP1-related retinopathy exhibits heterogeneous characteristics in terms of inheritance patterns, age at onset, disease phenotypes, and severity.¹ Mutations in the RP1 gene account for 1.8% to 9% of cases of retinitis pigmentosa.^2–4 Monoallelic mutations in amino acids 500 to 1053 of RP1 exon 4 are frequently linked to autosomal dominant retinitis pigmentosa.⁵ Recent reports indicate that biallelic mutations in RP1 cause macular dystrophy and cone–rod dystrophy, as well as retinitis pigmentosa.^6,7 Biallelic phenotypes are frequently associated with a combination of functional hypomorphic variants and pathogenic missense or truncated variants. Notably, three such hypomorphic variants (p.Phe180Cys, p.Val190Gly, and p.Arg1933Ter) have been identified in RP1.⁶ Particularly, p.Arg1933Ter is frequently found in East Asians. Recent studies have reported normal retina findings in two individuals with homozygous hypomorphic variants (p.Arg1933Ter), and, in one individual, normal findings were observed even beyond the age of 80.⁸ Based on these examples, the concept has emerged that genetic retinal diseases could follow a quasi-Mendelian model linking both monogenic and complex inheritance.⁸ 

A notable example of this complexity is the p.Arg1933Ter variant, which occurs at a high frequency in East Asian populations. This variant challenges the conventional notion of rarity in genetic disorders, with an allele frequency of 0.2055% (41/19,950). In light of these complexities, this study aimed to delve into the implications of the homozygous hypomorphic variant p.Arg1933Ter in RP1, specifically its association with late-onset and slowly progressive cone dystrophy. 

Methods

This study complied with the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of Seoul National University Bundang Hospital (IRB no. B-2303-816-103). Informed consent for genetic testing and research was obtained from all patients. We retrospectively analyzed 830 patients with inherited retinal diseases (IRDs) who underwent genetic analysis at Seoul National University Bundang Hospital and included patients with the homozygous variant c.5797C>T encoding p.Arg1933Ter in the RP1 gene. For the genetic analysis, we used a customized exome panel consisting of 254 known and candidate genes associated with IRDs. All coding exons, 5′ and 3′ untranslated regions, and alternatively spliced regions of each gene were included. The detailed genetic analysis process and pipeline are presented in our previous study.⁹ Copy number variation analysis was performed for all IRD genes in patients with biallelic p.Arg1933Ter in RP1. To rule out incorrect mapping or incidental Alu insertion in RP1, we manually checked the read mapping status using the Integrative Genomics Viewer (IGV). We further conducted transposon detection using Mobile Element Locator Tool (MELT) software to ensure the absence of Alu insertion.¹⁰ To exclude the founder Alu insertion in exon 4 of the RP1 gene, we also conducted the targeted amplification using polymerase chain reaction (PCR) according to the previous report.¹¹ 

All patients underwent basic ophthalmic examinations, including the measurement of best-corrected visual acuity, tonometry, slit-lamp examination, and fundus examination. Optical coherence tomography (OCT), fundus autofluorescence, electroretinography (ERG), multifocal electroretinography, and visual field tests, including the Humphrey visual field 30-2 and the Goldmann visual field, were performed to diagnose and evaluate IRDs. 

Results

Five patients with the homozygous p.Arg1933Ter mutation in RP1 were identified. The patients had no history of uveitis or other retinal diseases, with no evidence of ocular or systemic inflammation. None of the patients had pathogenic or likely pathogenic variants in other IRD genes according to the American College of Medical Genetics and Genomics (ACMG) guidelines, and there were no reported variants in ClinVar or the Human Gene Mutation Database (HGMD). Structural variants were not detected in 254 IRD genes, including RP1. Upon manual inspection using the IGV, the sequencing reads were well mapped with high coverage (>200×), and no specific sequencing errors could be identified (Fig. 1A). The insertion of Alu elements was not detected using MELT software or PCR amplification targeting Alu insertion in RP1 (Fig. 1). Transposons were not detected in three of the five patients. In patient D, we detected five Alu insertions, whereas, in patient E, only one Alu insertion was detected in other regions (Table 1). 

Figure 1.

 

A manual inspection using the IGV and results of PCR amplification. (A) Upon manual inspection using the IGV, the sequencing reads are well mapped with high coverage (>200×), and no specific sequencing errors can be identified. (B) The insertion of Alu elements is not detected by MELT software and PCR amplification targeting Alu insertion in RP1.

Table 1.

 

In Silico Analysis Results for Detecting Alu Insertion

 

Five patients had clinical manifestations of late-onset, slowly progressive cones, or macular dystrophy (Table 2). The mean ± SD age at the first visit was 62.2 ± 9.8 years. The age of symptom onset was between 45 and 50 years. Patient E mentioned that, although the exact age of symptom onset was not clear, they maintained good vision before their 40s. All patients reported that they had no issues with their visual acuities or fields and maintained a visual acuity of 20/30 or better until their early 40s. 

Table 2.

 

Demographics and Clinical Manifestations of Patients with p.Arg1933Ter in the RP1 Gene

 

Initially, patients A and B were clinically suspected to have achromatopsia or occult macular dystrophy due to the relatively preserved photoreceptor layer, despite their advanced ages of 53 and 69 years, respectively. Fundoscopic examination and fundus autofluorescence imaging revealed a tessellated fundus appearance associated with high myopia and relatively preserved lipofuscin in the macula with minimal atrophic changes. On OCT, loss of the interdigitation zone and blurring of the ellipsoid zone, which are characteristics of achromatopsia and occult macular dystrophy, were observed. In the results of the ERG, patient A exhibited mildly decreased cone/rod response, and patient B showed normal rod response and mildly decreased cone response (Figs. 2A, 2B). Patients C, D, and E exhibited localized atrophy of the macula as observed on fundus photography and autofluorescence images. Despite their advanced ages, no abnormalities were detected in the peripheral retinas. ERG amplitudes were relatively preserved compared to the degree of macular atrophy. In both dark-adapted and light-adapted ERG, patient D showed a mild decrease in amplitude, whereas patient C demonstrated normal findings. The multifocal ERG conducted exclusively on patient A demonstrated a marked decreased in amplitude, rendering the signal nearly indiscernible. All patients who underwent visual field tests (patients A, B, C, and D) exhibited a central scotoma pattern. 

Figure 2.

 

Clinical findings of fundus photography, fundus autofluorescence, OCT, ERG, and visual field tests in patients with p.Arg1933Ter in the RP1 gene. (A, B) Patients A and B show a tessellate fundus appearance associated with high myopia. Fundus autofluorescence images show relatively preserved lipofuscin in the macula with minimal atrophic changes. Loss of the interdigitation zone and blurring of the ellipsoid zone were observed using OCT. (C–E) Patients C, D, and E show atrophy confined to the macula. Circular bands of hyper-autofluorescence are observed at the border of the atrophied hypo-autofluorescence areas and the normal retina, which correspond to the margin and atrophic area of the macula on OCT. In the results of visual field tests and ERG, a central scotoma was observed, and ERG responses showed relatively better preservation compared to the extent of macular atrophy.

Discussion

Mutations in recessive genes can either completely eliminate gene function, known as amorphic mutations, or reduce gene function, referred to as a hypomorphic mutation.¹² Hypomorphic variants are known in various IRD genes; especially in ABCA4, p.Asn1868Ile and p.Gly1961Glu have high allele frequencies of 5.6% and 0.34% in gnomAD, respectively. These frequent variants may influence the varying severity of IRDs.^13,14 The p.Arg1933Ter is a prominent hypomorphic mutation of the RP1 gene in East Asians.⁶ This mutation involves the arginine at position 1933, encoded by the CGA sequence, located at one of the CpG sites of the gene—a region susceptible to frequent methylation. During the deamination of CpG dinucleotides, methylated cytosine (C) can convert directly into thymine (T). This conversion results in a C to T mutation during DNA replication. Therefore, mutations from CGA to TGA (p.Arg1933Ter) are not uncommon compared to other types of mutations, even though mutations within an exon are closely scrutinized due to evolutionary pressures. This mutation is insufficient on its own to cause the disease and is typically reported to induce cone–rod dystrophy when it occurs in conjunction with a pathogenic mutation in the other allele of RP1.⁸ A recent report showed that homozygotes with the hypomorphic variant (p.Arg1933Ter) exhibit a normal phenotype, even at 80 years of age. However, we observed a form of late-onset and slowly progressive cone or macular dystrophy in five Korean patients who carried homozygous p.Arg1933Ter in RP1. Most patients described normal vision when they were in their 30s; however, they experienced a reduction in visual acuity starting at 40 to 50 years of age. We believe that these homozygous hypomorphic mutations may lead to milder forms of genetic disorders depending on age, environmental factors, and genetic background. Therefore, patients with homozygous hypomorphic mutations should not be assumed to be normal, and the likelihood of disease onset at an advanced age should be considered. 

The p.Arg1933Ter shows a very low allele frequency in other ethnicities; however, it is notably more prevalent in East Asians, including Koreans and Japanese. As noted earlier, the allele frequency of c.5797C>T, encoding p.Arg1933Ter, is 0.2055% (41/19,950) in East Asians according to gnomAD (https://gnomad.broadinstitute.org/). However, it is crucial to note that the homozygous form of p.Arg1933Ter has not been reported in the gnomAD database. Due to its relatively high allele frequency, this hypomorphic variant has been reported to interact with other pathogenic variants, potentially leading to cone–rod dystrophy.¹ Although no disease-causing variants in other genes were detected through genetic testing in our study, there may be undetected variants such as deep intronic variants or unknown causative genes. However, the fact that five patients exhibiting a similar clinical phenotype all shared the same homozygous mutation strongly suggests its role as the disease-causing variant, even when considering the 0.2% allele frequency. 

As shown in the Figure 3, RP1 has an unusual mRNA structure, with an untranslated exon 1, short exons 2 and 3, and a very long terminal exon 4 in which the majority of dominant mutations can be found in the middle of this exon (500–1053 amino acids). Hence, the functionally null versions tend to be those with premature stop codons in exons 2 and 3; in these, the carriers usually have no phenotype, but the homozygotes are severely affected. Interestingly, other downstream variants of p.Arg1933Ter, such as p.Glu1940Ter, p.Gln1961Ter, p.Ile1988AsnfsTer3, and p.Ile2061SerfsTer12, have been classified as pathogenic or likely pathogenic (https://www.ncbi.nlm.nih.gov/clinvar/). It is unclear why p.Arg1933Ter alone functions hypomorphically; however, it is speculated that this variation may be associated with its susceptibility to nonsense-mediated decay and, consequently, the degree of preservation of gene expression and function based on mutations. Changes in mRNA folding may obstruct ribosome access or interfere with stop codon recognition, thereby facilitating translational readthrough and leading to the extension of translation with aberrant amino acid sequences. It is speculated that p.Arg1933Ter could evade nonsense-mediated decay through such mechanisms. A recent example of quasi-Mendelian inheritance was described in which the homozygous variant of p.Arg1933Ter in RP1 exhibited a benign clinical phenotype.⁸ However, additional research is needed, as it can manifest as a relatively mild late-onset form, unlike the benign presentation in our case. 

Figure 3.

 

Genetic structure of the RP1 gene. The exon region spanning amino acids 500 to 1053 is the “dominant cluster,” primarily responsible for autosomal dominant (AD) retinitis pigmentosa (RP). The remaining exons (exons 2, 3, and 4, except the dominant cluster) are associated with variations causing autosomal recessive (AR) RP, AR macular dystrophy (MD), and AR cone dystrophy (CD).

Structural variations and Alu insertions are often not detected in a conventional next-generation sequencing pipeline; therefore, it is important to thoroughly investigate them as potential hidden causes of the disease.^15–17 In this study, copy number variation analysis was performed for 254 IRD genes, confirming the absence of structural variation. Recent studies have indicated that an Alu insertion, a type of mobile element, within exon 4 of the RP1 gene is likely to go undetected through a routine sequencing pipeline.¹⁸ This insertion of 328 additional nucleotides results in a premature termination after nine amino acid residues following Tyr1352 during the synthesis of the RP1 protein (c.4052_4053ins328, p.Tyr1352Alafs*9), and the Alu-inserted allele is predicted to function as a null allele.⁶ The Alu insertion in exon 4 of RP1 can be detected using both PCR and gel electrophoresis as well as in silico analysis. We utilized both a PCR-based approach and in silico analysis targeting p.Tyr1352Alafs*9.^11,19 Our investigation confirmed the absence of mobile element insertion within the exon region of the RP1 gene. 

In summary, we identified the homozygous hypomorphic variant p.Arg1933Ter in five patients exhibiting a form of late-onset cone or macular dystrophy, a finding that differs from previous reports where two patients displayed normal findings. We believe that biallelic hypomorphic mutations can lead to disease depending on the genetic background and various modifying factors. By unravelling the intricate genetic and phenotypic intricacies, this study sought to contribute to a deeper understanding of the enigmatic world of inherited retinal dystrophies. 

Acknowledgments

Supported by grants from the National Research Foundation of Korea funded by the Korean Ministry of Science and ICT (2022R1A2C4002114) and the New Faculty Startup Fund from Seoul National University (800-20230310). Neither the sponsor nor the funding organization had a role in the design or conduct of this research. 

Disclosure: S.W. Choi, None; S.J. Woo, None; M. Kim, None; S. Lee, None; K. Joo, None