In this study we used DNA from 81 patients with clinical diagnosis of retinitis pigmentosa (RP). Supposedly, 9 patients had autosomal dominant RP (adRP), 65 had autosomal recessive RP (arRP) and 7 patients had either adRP or arRP. Peripheral blood samples were collected in EDTA tubes and DNA was extracted as described elsewhere. The assignment of inheritance pattern was based on a family history provided by the patients. However, some of the elderly patients did not have details about visual impairment of their siblings and parents and most of the cases were simplex. DNA from siblings or parents was available only in three families. The study was approved by the Swedish Ethical Review Authority (Etikprövningsmyndigheten) and conducted in accordance with ethical principles for medical research involving human subjects as stated in Declaration of Helsinki. An informed consent was obtained from all patients or their parents before inclusion to the study. During first examination of patient RP6795 an informed consent was obtained from her parents.
Molecular genetic analyses
The arrayed primer extension assay (APEX) was performed via Asper Biogene (Tartu, Estonia) (https://www.asperbio.com/asper-ophthalmics). DNA from several patients (RP18, RP26, RP74, RP103, RP154, RP160, RP165, RP6795 and VC101) was analysed in 2006-2007 using a panel for arRP-associated 501 known variants in 16 genes (CERKL Gene ID - ENSG00000188452, OMIM 608381; CNGA1 Gene ID - ENSG00000198515, OMIM 123825; CNGB1 - ENSG00000070729, OMIM 600724; MERTK Gene ID - ENSG00000153208, OMIM 604705; PDE6A– Gene ID ENSG00000132915, OMIM 180071; PDE6B– Gene ID ENSG00000133256, OMIM 180072; PNR/NR2E3– Gene ID ENSG00000278570, OMIM 604485; RDH12– Gene ID ENSG00000139988, OMIM 608830; RGR – Gene ID ENSG00000148604, OMIM 600342; RLBP1– Gene ID ENSG00000140522, OMIM 180090;SAG– Gene ID ENSG00000130561, OMIM 181031; TULP1– Gene ID ENSG00000112041, OMIM 602280; CRB1– Gene ID ENSG00000134376, OMIM 604210;RPE65– Gene ID ENSG00000116745, OMIM 180069; USH2A– Gene ID ENSG00000042781, OMIM 608400; and USH3A/CLRN1 – Gene ID ENSG00000163646, OMIM 606397).
Next generation sequencing (NGS) of 56 arRP-associated genes was performed at Asper Biogene (Tartu, Estonia). Samples were analysed using the BWA Enrichment app in BaseSpace (Illumina). Alignment was performed with BWA and variants were called with GATK41. Mean coverage was 87 reads for the whole gene panel and mean coverage of EYS gene was 99 reads. Variants that had an alternative variant frequency less than 20% and/or had a call quality score lower than 20 were automatically filtered out. Confirmation of NGS findings and subsequent screening of EYS gene was done with Sanger sequencing using BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA). Primer sequences for the screening of the EYS gene are available upon request. The products of sequencing reactions were analysed on ABI 3500xL Dx Genetic Analyser (Applied Biosystems). Sequences were aligned and evaluated using the Sequencher software version 4.9 (Gene Codes Corporation, Ann Arbor, MI, USA). All changes were assigned according to the GenBank Reference Sequence Version FJ416331; GI: 212675237; Transcript Reference Sequence: NM_001142800.1 and described according the HGVS recommendations42.
For detection of copy number changes in exonic sequences of the EYS gene we used Multiplex Dependent Probe Amplification (MLPA) with probe mix P328-A1-0811 or P328-A2-0217 lacking probes for exon 9 or exons 2, 7, 9 and 27, respectively (MRC Holland, Amsterdam, Netherlands). Additionally, three MLPA probes were designed aiming to detect breakpoints of the duplication detected in family 501. The sequences of these MLPA probes are available in Supplementary Figure S3. The MLPA reactions were run on ABI 3500xL Dx Genetic Analyser (Applied Biosystems) and the data was evaluated in GeneMarker v.2.7.0 (SoftGenetics, State College, PA, USA) with deletions set to be detected at quote values below 0.75 and duplications set to be detected at quote values above 1.30.
In vitro splice assay using EYS minigenes
Two intronic EYS variants, c.2992_2992+6delinsTG and c.3877+1G>A were tested with pSPL3 exon trapping vector (Invitrogen, Carlsbad, CA) as described previously by Jonsson et al43. Genomic DNA from RP103 and RP15 patients was used for PCR amplification for fragments with EYS c.3877+1G>A and c.2992_2992+6delinsTG, respectively.
PCR was performed in 25 μl volume with 0.2 μM forward primer with EcoRI- and reverse primer with NotI at 5´-endsites (Supplementary Table S1) and 1.25U of Taq-DNA polymerase with 5'-3' exonuclease activity (Ampliqon A/S, Odense, Denmark). Ligation of the amplicons was performed according to manufacturer´s instructions by adding 50 ng of pGEM-T Easy vector (Promega, Wichburg, WI, USA), 3 μl of PCR product, 3U of T4 DNA ligase and 1X Rapid Ligation buffer to a final reaction volume of 10 μl. For propagation of the pGEM-T Easy vector, XL10-Gold ultracompetent cells (Stratagene, La Jolla, USA) were transformed by adding the pGEM-T ligation reaction to 50 μl of XL10-Gold suspension as described elsewhere. Plasmid DNA was extracted from the overnight cultures (1x LB with 100 µg/ml carbenicillin) using standard protocol. The DNA concentration was measured on a NanoDrop 1000 v.3.8.1 (Thermo Fisher, Waltham, MA, USA). To identify clones with correct inserts in the pGEM-T vector, 0.5-1 μg of each plasmid DNA was treated with 3U of EcoRI and NotI (New England BioLabs, Ipswich, MA, USA) for 1-2 hours at 370C and visualized on a 1% agarose gel. Plasmids with correct size of insert were analysed by Sanger sequencing to identify wild type (wt) and mutant (mt) constructs and to rule out any sequence deviations generated by PCR. Sanger sequencing was performed using BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). For minigene construction pSPL3 was digested with EcoRI and NotI and the wildtype and mutant inserts excised from pGEM-T Easy vector were ligated into the pSPL3 vector using the same ligation protocol as for pGEM-T Easy system followed by transformation into XL10-Gold ultracompetent cells. Ratios for pSPL3 versus inserts were calculated using an online tool at www.promega.com/a/apps/biomath/?calc=ratio. Sequence confirmation was done by Sanger sequencing with pSPL3 specific primers: forward SD6-5’- TCTGAGTCACCTGGACAACC and reverse SA2-5’- ATCTCAGTGGTATTTGTGAGC (Supplementary Materail). Upon identification of one mutant and one wildtype minigene construct in the pSPL3 vector, the clones were propagated, and plasmid DNA was purified by midi-preparation using Nucleobond® Xtra Midi kit (Macherey-Nagel, Düren, Germany).
Human Embryonic Kidney cells, HEK293T Lenti-X (Clontech Laboratories, Mountain View, CA, USA) and Human Retinal Pigmented Epithelium cells, ARPE-19 (ATCC® CRL-2302TM, Manassas, VA, USA) were transfected with wt and mt pSPL3 minigenes, (EYS c.3877+1G>A and EYS c.2992_2992+6delinsTG). 2x105 HEK293T or ARPE-19 cells were seeded in 24-well plates. HEK293T cells were cultures in DMEM medium (Gibco, Gaithersburg, MD, USA) containing 1X Glutamax (Gibco) and ARPE-19 cells were grown in DMEM:F12 medium (Gibco). Both growth medium additionally contained 10% fetal bovine serum (Gibco, USA) and 1X PenStrep (Gibco). The cells were transfected after 24 hours using the Lipofectamine® 3000 system (Thermo Fisher). All transfections were done twice with HEK293T and once with ARPE-19 cells. For each well, 1 μl Lipofectamine® 3000 and 0.8 μg pSPL3 vector DNA was used. Total RNA was extracted 46-48h after transfection with NucleoSpin® RNA Plus (Macherey-Nagel) and reverse transcribed into cDNA in 20 µl reactions containing 100U Superscript® III reverse transcriptase (Invitrogen, Carlsbad, CA, USA), 200 ng random hexamers (Thermo Fischer), 0.5 mM dNTP (Roche Diagnostics, Mannheim, Germany), 5 mM DTT (Invitrogen), 1x First Strand Buffer (Invitrogen) and 128-1000 ng RNA. The cDNA reactions were incubated at 25oC for 5 min, 50oC for 60 min and 70oC for 15 min. Thereafter, PCR was performed on 5 µl cDNA (estimated 32-100 ng) using pSPL3 specific primers SD6 and SA2 (Supplementary Information), 0.75U AmpliTaq Gold DNA Polymerase, (Applied Biosystems), 1x PCR Buffer II (Applied Biosystems), 1.5 mM MgCl2 (Applied Biosystems) 0.2 mM dNTP (Roche D) with a temperature profile of 95oC for 12 min followed by 35 cycles of 95oC for 30 s, 55oC for 30 s and 72oC for 30 s and a final extension at 72oC. The PCR products were separated on 1% agarose gels and visualized under UV-light using ethidium bromide.
A full ophthalmic examination was performed at Eye Clinic of northern Sweden, and the clinical diagnosis was based on visual acuity, electroretinography (ERG) findings, fundus photography and Optical Coherence Tomography (OCT) visual fields.
Psychophysical methods and clinical examination
Visual acuity (VA) was measured using a LogMAR (Logarithm of the Minimum Angle of Resolution) chart. Slit-lamp examination, biomicroscopy and fundus examinations were performed, and fundus images were taken. Visual fields were measured with a Goldmann perimeter. Macular morphology and thickness were measured with optical coherence tomography (OCT), using a Topcon 3D OCT 2000 (Topcon Medical Systems, Oakland, NJ, USA). A pseudo-colour two-dimensional map of retinal thickness was used to confirm the retinal morphology.
Full-field electroretinograms (ERG) were recorded using Burian-Allens bipolar electrodes on an Espion Profile Ganzfeld ERG machine (Diagnosys LLC, Lowell, MA, USA) following the recommendations from the international society of Clinical Electrophysiological Vision44,45.
For the interpretation of genomic variants, prediction of splicing effect and estimation of alternative splice junctions Alamut® Visual v.2.9 (Interactive Biosoftware, Rouen, France) was used. To obtain splicing scores, the following software programs were applied: SpliceSiteFinder-like, Max- EntScan, NNSPLICE, Genesplicer, and Human Splice Finder. Frequencies of the variants present in controls were extracted from the Genome Aggregation Database (gnomAD) Version 2.1, which includes data of 125,748 exome sequences and 15,708 whole-genome sequences (https://exac.broadinstitute.org)46. We also used The Human Genomic Variant Search Engine, VarSome47. In this search engine information from more than 30 public databases is available with various annotations and predictions. Pathogenicity classification of sequence variants based on ACMG guidelines20 was automatically done in VarSome and controlled manually. According to the guidelines, variants can be classified as pathogenic, likely pathogenic, benign, likely benign, or uncertain significance using criteria for pathogenicity a) pathogenic, very strong (PVS) b) pathogenic, strong (PS); c) pathogenic, moderate (PM); and d) pathogenic, supporting (PP)20.