In silico 3D structural modeling of the ADGRV1 protein identifies a previously undetected cysteine-rich domain and additional Calxβ domains compared to previous 2D-based models

(A) Schematic representation of the protein domain structure of the large isoform (isoform B) of human and zebrafish ADGRV1, based on 2D-SMART predictions. Both proteins exhibit a repetitive domain architecture, comprising a signal peptide, 35 Ca²⁺-binding calcium exchanger β domains (Calxβ), a laminin G-like domain (LamG-like), 6 epilepsy-associated repeats (EAR), a G-protein-coupled receptor (GPCR) proteolytic site, a seven-transmembrane region, and an intracellular region with a C-terminal class I PDZ-binding motif. Numbers indicate amino acid positions. (B) In silico 3D modeling of zebrafish (magenta) and human (blue) ADGRV1, based on AlphaFold2 predictions, suggests a more complex domain architecture than 2D-SMART-based models. 3D models revealed additional Calxβ domains as compared to the SMART-based predictions (39 instead of the previously predicted 35 Calxβ domains) and the presence of a previously unidentified cysteine-rich domain. Additionally, the 3D models reveal that the LamG-like domain, the cysteine-rich domain, and the EAR repeats protrude from the Calxβ domains, rather than existing as solitary domain structures that are separated by other (types of) domains. (C) Updated 2D schematic representation of human and zebrafish ADGRV1 (isoform B) based on novel 3D protein modeling.

In silico modeling of the ADGRV1 protein domain architecture after the excision of ADGRV1 exon 9 and exons 40–42

(A) Schematic representation of the protein domain structure of the large isoform (isoform B) of human and zebrafish ADGRV1, based on 2D-SMART protein predictions. Regions encoded by exon 9 and exons 40–42 are highlighted with dashed boxes. (B) In silico 3D modeling of human ADGRV1, based on AlphaFold2. The Calxβ domains encoded by ADGRV1 exon 9 are depicted in cyan. (C) In silico 3D modeling of the protein domains encoded by exons 6–12 in man and zebrafish. The part encoded by exon 9 is depicted in cyan (human ADGRV1) or green (zebrafish Adgrv1). Removal of exon 9 is predicted to result in a discontinuity in the Calxβ domain structure. (D) In silico 3D modeling of human ADGRV1, based on AlphaFold2 predictions. The parts of Calxβ domains encoded by ADGRV1 exons 40–42 are depicted in cyan. (E) In silico 3D modeling of the protein domains encoded by exons 35–45 in man and zebrafish. The part encoded by exons 40–42 is depicted in cyan (human ADGRV1) or green (zebrafish Adgrv1). Removal of exons 40–42 results in the production of a single-hybrid Calxβ domain, with a 3D structure highly similar to native Calxβ domains. (F) Structural comparison of native human and zebrafish Calxβ domains 21, along with the hybrid Calxβ domain resulting from the removal of exons 40–42 in both man and zebrafish. The superimposed image shows that the 3D structures of the hybrid human (blue) and zebrafish (magenta) Calxβ domains closely resemble the conformations of the native human Calxβ domain 21 (blue–cyan) and zebrafish Calxβ domain 21 (magenta–green).

Design and characterization of the genetically modified adgrv1^Δexon9 and adgrv1^Δexon40-42 zebrafish lines

(A) Schematic representation illustrating the exon-excision approach. The anticipated excision of the target exons in injected embryos (1 day post-fertilization [dpf]) was confirmed by Sanger sequencing analysis. Excision of the genomic segment encompassing adgrv1 exon 9 led to the incorporation of three additional nucleotides (GTT) at the repair site. (B) RT-PCR results demonstrate the absence of adgrv1 exon 9 in adgrv1^Δexon9 larvae, as well as the absence of exons 40–42 in adgrv1^Δexon40-42 larvae (5 dpf). Sanger sequencing of RT-PCR amplicons confirmed the successful removal of the target exons from the adgrv1 transcripts. PCR (−): negative PCR control.

Analysis of Adgrv1 expression in retinal sections of wild-type, adgrv1^rmc22, adgrv1^Δexon9, and adgrv1^Δexon40-42 zebrafish

(A) Schematic representation of a zebrafish photoreceptor cell. OS, outer segment; CC, connecting cilium; IS, inner segment; ONL, outer nuclear layer. (B) Retinal cryosections of wild-type, homozygous adgrv1^rmc22, adgrv1^Δexon9, and adgrv1^Δexon40-42 zebrafish larvae (5 dpf) labeled with antibodies directed against Adgrv1 (red) and centrin (green). Nuclei are counterstained with DAPI (blue). No Adgrv1 signal was detected in adgrv1^rmc22 and adgrv1^Δexon9 zebrafish retinae. In contrast, Adgrv1 was present and localized adjacent to centrin, a marker of the basal body and connecting cilium, in the retinas of wild-type and adgrv1^Δexon40-42 zebrafish larvae (n = 14 larvae per line, from two biological replicates for wild-type, adgrv1^rmc22, adgrv1^Δexon9, and adgrv1^Δexon40-42 larvae). Magnified images are shown at the right side. Scale bars, 10 μm.

Analysis of usherin and Whrnb expression in retinal sections of wild-type, adgrv1^rmc22, and adgrv1^Δexon40-42 zebrafish

Retinal cryosections of wild-type, mutant adgrv1^rmc22, and adgrv1^Δexon40-42 zebrafish larvae (5 dpf) labeled with antibodies directed against Whrnb (red) (A) or usherin (red) (B) and centrin (green). Nuclei are counterstained with DAPI (blue). (A) Both in wild-type and in adgrv1^Δexon40-42 zebrafish larvae, Whrnb was present at the photoreceptor periciliary region in close proximity to centrin. Quantification revealed that the Whrnb signal intensity in adgrv1^Δexon40-42 retinal sections is similar to that in wild-type sections, whereas a significant reduction of Whrnb signal was observed in retinae of adgrv1^rmc22 larvae. Magnified images are shown at the right side. Bar graphs represent the mean fluorescence intensity of anti-Whrnb staining of all photoreceptors of a single, central section of one larval zebrafish eye (n = 14 eyes). (B) Both in wild-type and adgrv1^Δexon40-42 zebrafish larvae, usherin was present at the photoreceptor periciliary region in close proximity to centrin. Quantification reveals that the usherin signal intensity in retinal sections of adgrv1^Δexon40-42 larvae is similar to that in sections of wild types, whereas a significant reduction of usherin signal was observed in retinae of adgrv1^rmc22 larvae. Bar graphs represent the mean fluorescence intensity of anti-usherin staining of all photoreceptors of a single, central section of one larval zebrafish eye, with mean ± SD (n = 14 eyes). Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparison test (∗∗p < 0.01; ∗∗∗∗p < 0.0001). Scale bars, 10 μm.

Analysis of rhodopsin localization in retinal sections of wild-type, adgrv1^rmc22, and adgrv1^Δexon40-42 zebrafish

(A) Schematic representation of a photoreceptor with rhodopsin localization in the outer segments (OS) and aberrant rhodopsin localization in the photoreceptor cell body as observed in adgrv1^rmc22 larvae. (B) Retinal cryosections of wild-type, adgrv1^rmc22, and adgrv1^Δexon40-42 zebrafish larvae (6 dpf) labeled with antibodies directed against rhodopsin (green). Nuclei are counterstained with DAPI (blue). As observed in the retinas of wild-type and adgrv1^Δexon40-42 zebrafish larvae, rhodopsin predominantly localized to the photoreceptor outer segments. Photoreceptor cells with aberrant localization of rhodopsin were regularly observed in adgrv1^rmc22 larvae (indicated with the white arrows). Scale bars, 10 μm, OS, outer segments; ONL, outer nuclear layer; INL, inner nuclear layer. (C) Total number of cells with aberrant rhodopsin localization per retinal section were plotted, with mean ± SD (n = 41 wild types, n = 42 adgrv1^Δexon40-42, and n = 40 adgrv1^rmc22 larvae). Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparison test (∗∗∗∗p < 0.0001).

ERG recordings reveal normal retinal function in adgrv1^Δexon40-42 zebrafish

(A) Representative ERG traces of wild-type, adgrv1^rmc22, and adgrv1^Δexon40-42 zebrafish larvae (7 dpf). (B) Maximum B-wave amplitudes following a light stimulus of 1,000 lux. The mean maximum B-wave amplitude of adgrv1^Δexon40-42 larvae is similar to those observed in wild types, whereas adgrv1^rmc22 larvae show a significant decrease in maximum B-wave amplitudes (∗∗∗p = 0.0001; ∗∗∗∗p < 0.0001; one-way ANOVA followed by Tukey’s multiple comparison test).

Assessment of CRISPR-Cas9 mediated editing efficiency and successful genomic excision of ADGRV1 exons 40–42 in HEK293T cells

(A) Assessment of genome-editing efficiency using TIDE analysis²⁰ following HEK293T transfection with PX459 CRISPR-Cas9 expression plasmids containing sgRNAs targeting either ADGRV1 intron 39 or intron 42 (n = 2 biological replicates per target). (B) The size of the genomic PCR fragments generated using primers in introns 39 and 42 confirmed the successful excision of ADGRV1 exons 40–42 from HEK293T cell DNA following co-transfection with both plasmids (unedited amplicon size: 3,997 bp; amplicon size after genomic exon excision: 1,950 bp). GAPDH amplification is shown as a loading control (amplicon size: 110 bp). PCR (−), negative PCR control. (C) Sanger sequencing of amplicons confirmed the successful removal of target exons.