Identification Of Disease-Causing Mutationsin Chinese Families With Autosomal Dominant Retinitis Pigmentosa

Objectives:To survey42families with clinical evidence of autosomaldominant retinitis pigmentosa (ADRP) for mutations in13genesknown to cause ADRP and to characterize the clinical features of those patients who were found to have mutations in these genes.1. Probands from each family werescreened by direct sequence for mutations in13genes known to cause ADRP;2. To screen any substantial deletion mutation in the RHO、IMPDH1、 RP1andPRPF31gene in the remaining families by multiplex ligation-dependent probe amplication(MLPA)reaction;3. Three large families without mutations in13known genes were tested for genome-widelinkage using the Illumina Infinium HumanLinkage-12panel;4. To characterize the clinical features of those patients who were found to have mutations in these genes;5. To elucidate the exact pathogenic effects of novel mutations identified in this study.Methods:1. Clinical investigation.(1) Routine ophthalmologic examinations. A full medical and family history was taken and an ophthalmological examination was performed. Each proband underwent a standard ophthalmic examination: best correct visual acuity, slit-lamp biomicroscopy, dilated direct and indirect ophthalmoscopy, fundus photography if possible.(2) Special ophthalmologic examinations. Visual field tests, optical coherence tomography (OCT) and fullfield electroretinogram (ERG) were performed on selected patients from these42families.2. Molecular genetic studies.(1) DNA preparation. Human genomic DNA was extracted from peripheral blood leukocytes using standard phenol-chloroform extraction method;(2) Mutation survey on13genes. Probands from each family werescreened for mutations in13genes known to cause ADRP by direct sequence;(3) MLPA testing. Wetested the remaining families for deletions of one or more sequencesin RHO、IMPDH1、RP1and PRPF31genes using MLPA;(4) Linkage analysis.①Genome-widelinkage analysis.Agenome-wide linkage analysis was conducted in three large families without mutations in13known genes by using the Illumina Infinium HumanLinkage-12panelcontaining5913SNPs. Multipoint linkage analysis was done using the MLINK v1.01package;②Sequence analysis. Based upon the genome wide linkage analysis, candidate genes were selected from linked regions and direct sequencing was performed using ABI3730Genetic Analyzer to evaluate all the exons and flanking introns sequences of the candidate genes.(5) Clinical assessment. To characterize the clinical features of those probands who were found to have mutations in the known genes by analyzing their clinical data, including best correct visual acuity, fundus photography, visual field tests, OCTand ERG tests.(6) Functional analysis of mutant protein. To functionally characterize one novel missense mutation L59R (RHO gene) identified in this study. Firstly, site-directed mutagenesis, using RHO-pCMV6-AC-GFP(wildtype) as a template, polymerase chain reaction(PCR) based site-specific mutagenesis was performed to constructRHO-pCMV6-AC-GFP(L59R mutant); Secondly, localization of the wild and mutant protein in SK-N-SH cells, SK-N-SH neuroblastoma cells were cultured and transfected withRHO-pCMV6-AC-GFP (wildtype) andRHO-pCMV6-AC-GFP (L59R mutant). The subcellular distribution ofRHO (L59R) in SK-N-SH cells wascompared to wild-type by visualizing GFPwith confocal microscopy.Results1. Mutation testing on13genes and MLPA testing. Upon complete analysis in13genes known to cause ADRP, a total of12distinct disease-causing were detected in the42families. Four families (9.5%) in the cohort have four different disease-causing missense mutations in RHO. Five families (11.9%) of the cohort have disease-causing mutations in PRPF31gene. Three are deletions leading to a frameshift, one is amino acid substitution and the other is a splice-site mutation. One missense mutation (2.4%) that reported earlier was found in the RP1gene. One family (2.4%) in the cohort had a novel deletion in PRPF8. One family (2.4%) in the cohort had a recurrent missense mutation in PRPF3. In addition, we identified gross deletions in thePRPF31gene in two (4.8%,2/42) additionalfamilies through MLPA testing.2. Linkage analysis.The genomewide linkage analysis of3ADRP pedigrees did not achieve significantly LOD scores. Pedigrees ADRP-RLY achieved maximum LOD scores (1.7953) between markers rs2033108and rs2870775on chromosomel7. Among all genes residing at this gene locus, CA4gene were selected as promising candidate gene and screened for mutation, however, no disease-causing mutation was indentified in this family.3. Clinical assessment. Totally we identified disease-causing mutations in14probands in this study. Age at the time of testing ranged from24to58years. The symptoms that led to the diagnosis were dominated by night blindness in all patients. Central visual acuity was ranged from hand movement to1.0. Fundus examination with indirect ophthalmoscopy after full dilatation showed relatively varying amounts of pigment deposits and different degree of retinal pigment layer atrophy. OCT examinations showed the disappearance of the IS/OS layer as well as the retinal pigment epithelium layer. All patients showed severe visual field constriction.Full-field ERG were non-recordable in scotopic conditions (rods), and the cone responses were markedly hypovolted or non-recordable in all these patients.4. Functional analysis of mutant proteinSK-N-SH cells were transfected with RHO-pCMV6-AC-GFP (wildtype), RHO-pCMV6-AC-GFP (L59R mutant) encoding wild-type (WT) opsin-GFP, L59R opsin-GFP, respectively.WT opsin-GFP localized predominantly to the plasmamembraneindicating the normal biogenesis and transportation of theprotein to the membrane. By contrast, the L59R opsin-GFP mutant was nottrafficked to the plasma membrane and accumulatedintracellularly. The mutant also formed intracellularinclusion bodies.Conclusion1. Together, mutation screening of13known ADRP genes account for retinal disease in approximately28.6%of the families in thissurvey. Among the ADRPgenes, PRPF31andRHO accounts for11.9%and9.5%of the total, respectively; RP1, PRPF3, and PRPF8each accounts for roughly2.4%. Disease-causing mutations werenot found in FSCN2, PRPH2, IMPDH1, CRX, RPE65, RDH12or KLHL7. In addition, gross deletions in PRPF31gene were identified through MLPA testing in two (4.8%) ADRP pedigrees.2. No disease-associated mutations were detected upon genomewide linkage analysis in three selected ADRP families, indicating the existence of novelADRP genes.3. Patients carrying mutations in the known gene in this study demonstrated classic RP with different degree of severity; no genotype and phenotype correlation was identified in this study.4. Expression analyses of mutant RHO protein confirmed that the missense mutationL59R led to abnormalprotein retention in the cytoplasm and intracellular inclusion formation.