Expanding the genetic landscape of Usher syndrome type IV caused by pathogenic ARSG variants

Abstract

Usher syndrome (USH) is the most common cause of deafblindness. USH is autosomal recessively inherited and characterized by rod-cone dystrophy or retinitis pigmentosa (RP), often accompanied by sensorineural hearing loss. Variants in >15 genes have been identified as causative for clinically and genetically distinct subtypes. Among the ultra-rare and recently discovered genes is ARSG, coding for the lysosomal sulfatase Arylsulfatase G. This subtype was assigned as "USH IV" with a late onset of RP and usually late-onset progressive SNHL without vestibular involvement. Here, we describe nine new subjects and the clinical description of four cases with the USH IV phenotype bearing seven novel and two known pathogenic variants. Functional experiments indicated the complete loss of sulfatase enzymatic activity upon ectopic expression of mutated ARSG cDNA. Interestingly, we identified a homozygous missense variant, p.(Arg99His), previously described in dogs with neuronal ceroid lipofuscinosis. Our study expands the genetic landscape of ARSG-USH IV and the number of known subjects by more than 30%. These findings highlight that USH IV likely has been underdiagnosed and emphasize the need to test molecularly unresolved subjects with deafblindness syndrome. Finally, testing of ARSG should be considered for the genetic work-up of apparent isolated inherited retinal diseases.

Keywords: ARSG; Usher syndrome (USH); canine variant p.(Arg99His); lysosomal sulfatase; retinitis pigmentosa (RP); rod‐cone dystrophy (RCD); sensorineural hearing loss (SNHL).

FIGURE 1

Clinical findings from subject 1: (A) fundus photography. Midperipheral is a zone of atrophy resembling a bull's eye maculopathy. The retinal vasculature is mildly attenuated. Nasally and superior of the optic disc intraretinal pigmentation as bone spicules are present. The retinal vasculature was mildly attenuated, and intraretinal pigmentation as bone spicules could be observed, mostly located nasally and superior to the optic disc. The peripheral retina was relatively preserved. (B) FAF. A normal foveal autofluorescence, a hyper‐autofluorescent perifoveal border, two concentric zones with consecutive one hypo‐autofluorescent and a larger hyper‐autofluorescent area, a confluent pattern of hypo‐autofluorescence around the vascular arcades and a normal autofluorescent peripheral retina. FAF imaging showed a normal foveal autofluorescence demonstrating the foveal sparing in this subject, a hyper‐autofluorescent perifoveal border, two concentric zones with consecutive one hypo‐autofluorescent and a larger hyper‐autofluorescent area, a confluent pattern of hypo‐autofluorescence around the vascular arcades and a normal autofluorescent peripheral retina. (C) On OCT, there was a thinned retina due to the loss of the outer retinal layers in the perifoveal area without cystoid spaces. Full‐field electroretinography (ERG) showed combined scotopic and photopic dysfunction, which was more pronounced for the scotopic responses, indicating rod‐cone dysfunction. (D) A ring scotoma between 10° and 20°. Central visual field testing showed a ring scotoma between 10° and 20°. Goldmann's peripheral visual field detected only a mild reduction of the peripheral borders.

FIGURE 2

Clinical findings from subject 5: (A) fundus photographs reveal typical signs of RCD/RP, including a waxy pallor of the optic discs, narrowed retinal vessels, and pigmentary changes outside the vascular arcardes; (B) short‐wavelength fundus autofluorescence imaging shows a patchy loss of autofluorescence in the periphery, a normal autofluorescence appearance at the fovea and some patches of loss of autofluorescence in the parafoveal region; (C) horizontal SD‐OCT scans show a loss of hyperreflective bands and a thinning of the outer nuclear layer, outside the foveal region, with a well preserved outer retina at the fovea and some intraretinal cysts parafoveal on the right eye; (D) kinetic visual fields are severely constricted for both eyes; (E) audiometry shows a loss of sensitivity for the higher frequencies for both ears.

FIGURE 3

Mutational spectrum of the USH type IV cohort: (A) distribution of the identified variants within both the ARSG gene (upper panel, subject exons are depicted) and the primary amino acid sequence of the ARSG protein. (B) Distribution of the novel missense variants in AlphaFold2 (https://alphafold.ebi.ac.uk/entry/Q96EG1; left: Secondary structure elements shown as a cartoon; right: The protein surface is additionally shown transparently). The mutated amino acids are highlighted in red and depicted as spheres. The active‐site cysteine C84 is highlighted in green. (C) Multi‐species alignment of the protein sequence of ARSG for each identified missense variant and the surrounding amino acids.

FIGURE 4

Functional analysis of missense variants in ARSG found in this study. (A) Total sulfatase activity of cell lysates from untransfected cells, or cells transfected with WT ARSG and mutated variants, against the artificial substrate pNCS, represented as the percentage of turnover in untransfected cells. N = 3 biological replicates and three technical replicates. Unpaired, two‐tailed t‐test, mean ± SEM of the biological replicates are shown. (B) Immunoblot analysis of untransfected or stably transfected HT1080 cells, with plasmids coding for 3xFLAG tagged ARSG (WT), along with its form carrying the pathogenic variants (p.Pro213Leu, p.Arg384Trp, and p.Arg99His). Antibodies are against FLAG; GAPDH was used as a loading control.