Two novel mutations in PRPF3 causing autosomal dominant retinitis pigmentosa

Retinitis pigmentosa (RP) is a heterogeneous set of hereditary eye diseases, characterized by selective death of photoreceptor cells in the retina, resulting in progressive visual impairment. Approximately 20–40% of RP cases are autosomal dominant RP (ADRP). In this study, a Chinese ADRP family previously localized to the region between D1S2819 and D1S2635 was sequenced via whole-exome sequencing and a variant c.1345C > G (p.R449G) was identified in PRPF3. The Sanger sequencing was performed in probands of additional 95 Chinese ADRP families to investigate the contribution of PRPF3 to ADRP in Chinese population and another variant c.1532A > C (p.H511P) was detected in one family. These two variants, co-segregate with RP in two families respectively and both variants are predicted to be pathological. This is the first report about the spectrum of PRPF3 mutations in Chinese population, leading to the identification of two novel PRPF3 mutations. Only three clustered mutations in PRPF3 have been identified so far in several populations and all are in exon 11. Our study expands the spectrum of PRPF3 mutations in RP. We also demonstrate that PRPF3 mutations are responsible for 2.08% of ADRP families in this cohort indicating that PRPF3 mutations might be relatively rare in Chinese ADRP patients.


Results
Clinical evaluations of ADRP families. The pedigree of Family 020001 indicates a dominant inheritance pattern of three generations (Fig. 1). Medical history of all the affected individuals shows that early onset of night blindness is at age 4 to 10 years old, with subsequent loss of far peripheral vision after 20 years. Clinical details of that family were previously described 10 . Additional 95 Chinese families had a primary diagnosis of RP based on clinical descriptions provided by the referring clinicians at the time of enrollment. The inheritance pattern of those RP families is AD (Fig. 1).

Mutation analysis.
We selectively performed whole-exome sequencing (WES) on the proband (II: 4) of Family 020001 previously mapped to the region between D1S2819 and D1S2635 10 (95.53Mb-159.17 Mb). An average of 9.71 Gb of sequence data were obtained, and 99.94% of bases originated from the targeted exome, resulting in a mean coverage of 115-fold. More than 98.94% of the targeted exons were covered more than 10-fold. We employed the following filters to filter the exome results: total variants → heterozygous variants → variants that are absent in the 1000 Genomes Project, 1000G_ASN, esp6500si_all, and dbSNP137 → variants that are coding or splicing → variants that are predicted to be pathogenic. Following the filteration procedures described above, only one heterozygous missense variant in PRPF3 (NM_004698: exon10: c.1345C > G: p.R449G) was detected to be located within the linkage region between D1S2819 and D1S2635 10 , which would lead to the change of amino acid from Arginine to Glycine at residue 449 (p.R449G). Sanger sequencing confirmed this variant (Fig. 2a). The variant was predicted to be probably damaging with a score of 0.980 by PolyPhen-2, a score of 0.05 by SIFT, and a score of − 6.698 by PROVEAN. Intra-familial co-segregation analysis of Family 020001 further revealed that this heterozygous missense variant was present in all the affected and absent in normal family members, which further confirmed that the variant was a potential ADRP causing mutation for Family 020001.
To investigate the contribution of PRPF3 to ADRP in Chinese population, Sanger sequencing of PRPF3 was performed in the probands of additional 95 Chinese ADRP families. We identified another novel variant c.1532A > C in exon12 of PRPF3 gene in Family 020021 (Fig. 2a), which would lead to the change of amino acid from Histidine to Proline at residue 511 (p.H511P). This variant co-segregated in the family with RP and was predicted to be probably damaging with a score of 1.000 by PolyPhen-2, a score of 0.00 by SIFT, and a score of − 9.785 by PROVEAN.
These two mutations, c.1345C > G (p.R449G) and c.1532A > C (p.H511P), were further confirmed to be absent in 200 unrelated ethnically matched healthy controls as well as in database of probably benign variation. Both mutations are absent either in existing database of disease-causing mutations or in the reported literatures and thus considered to be novel.
Multiple orthologous sequence alignment (MSA) revealed that p.R449G and p.H511P were highly conserved from human to yeast (Fig. 2c). The residues R 449 and H 511 in human PRPF3 correspond to the residues R 245 and H 308 in Prp3 of Saccharomyces cerevisiae S288c (NP_010761.3), respectively, by MSA of PRPF3 from Homo sapiens to yeast. Prp3 structure from PDB (Protein Data Bank): 3JCM (Cryo-em Structure of the yeast spliceosomal U4/U6.U5 Tri-snrnp) shows that amino acids R 245 and H 308 of Prp3 play a key role in Prp3 binding to RNA duplex 11 suggesting that changes in those two residues may impair the structure and function of spliceosomal U4/ U6.U5 Tri-snrnp (Fig. 3a).

Discussion
PRPF3 gene is one of pre-mRNA splicing genes and located in the region 150.29-150.33 Mb on chromosome 1q21 6 .Though PRPF3 is essential for splicing and ubiquitously expressed in almost every cell of the body, mutations in the gene have so far been reported to cause a disease that displays retina-specific phenotype RP. So far, only three clustered PRPF3 mutations, c.1482C > T (p.T494M) 7,9,[12][13][14][15][16] , c.1478C > T (p.P493S) 7,9 and c.1466C > A (p.A489D) 17 , have been reported in RP in several populations.
T494M is a common mutation of PRPF3 in ADRP as it has also been found in English, Danish, Spanish, American, Japanese, Korean, and Swiss ADRP families. P493S occurs in a sporadic RP case from UK or Germany and in a Caucasian ADRP family from American. A489D was reported in a Spanish ADRP family (Table 1). All these three mutations are in the exon 11 (c.1427-1526) of PRPF3, for which only the region c.1427-1526 was screened for testing PRPF3 mutations in reported studies including the original linkage paper about Family 020001 of this study, because previous surveys failed to find mutations outside of this exon 8,9 . The mutation c.1345C > G (p.R449G) is in the exon 10 and c.1532A > C (p.H511P) is in the exon 12, outside of the region frequently analyzed for detecting PRPF3 mutation. In our study, the identification of two novel PRPF3 mutations expands the spectrum of PRPF3 mutations in RP and suggests that more PRPF3 mutations might be found if the region outside of the exon 11 is screened. PRPF3 mutations associated with RP are reported in different populations but there is no report about PRPF3 mutation in Chinese population. This is the first report of the identification of PRPF3 mutation in the Chinese population. Total of 96 Chinese ADRP families were recruited in our study and two PRPF3 mutations were identified in two of the families. PRPF3 mutations are responsible for 2.08% of ADRP families in this cohort indicating that PRPF3 mutations might be relatively rare in Chinese patients with ADRP.
PRPF3 gene encodes a 683aa protein 7 , which is a human homologue of the yeast U4/U6-associated splicing factor Prp3 18 . PRPF3 protein contains one PWI motif 19 , one PRP3 domain and a DUF1115 domain (Fig. 2b) 20 . Three previously reported mutations and two novel mutations in this study all locate in PRP3 domain (Fig. 2b). MSA revealed that those five residues mutated in RP patients were highly conserved across all the species (Fig. 2c). Structural analysis of the wild type Prp3 in the yeast spliceosomal U4/U6.U5 Tri-snrnp (PDB: 3JCM) showed that the yeast counterparts of human Prp3 residues R 449 and H 511 buried into two successive grooves of RNA duplex, suggesting their necessity in U4/U6 snRNA duplex interaction 11 (Fig. 3a), which suggest that two novel mutations R449G and H511P probably impair binding of PRPF3-RNA duplex because both glycine and proline have propensity to break α -helical structure of PRPF3. In contrast, yeast counterparts of human PRPF3 residuesT 494 , P 493 and A 489 are clustered in the C-terminal conserved region, which play an important role in protein-protein interaction 5 (Fig. 3b), which is mentioned the previous report. Our study may provide alternative insights into the relevant pathogenesis for RP. In summary, we, for the first time, reported PRPF3 mutation spectrum in the Chinese population and demonstrated that PRPF3 mutations are responsible for 2.08% of ADRP families in this cohort indicating that PRPF3 mutation might be relatively rare in Chinese patients with ADRP. Two novel mutations, c.1345C > G (p.R449G) and c.1532A > C (p.H511P), were identified in our study, which expands the spectrum of PRPF3 mutations in RP. Structure analysis provides alternative insights into the etiology of RP. However, future research is still warranted to establish the pathogenic mechanism underlying how novel PRPF3 mutations identified in this study would cause RP.

Study Subjects and Clinical Examinations. The Institutional Review Board (IRB) of Tongji Eye Institute
of Tongji University School of Medicine (Shanghai, China) and the Center for Gene Diagnosis, Zhongnan Hospital of Wuhan University (Wuhan, China) approved this study and the whole procedure of this study adhered to the tenets of the Declaration of Helsinki. All participating family members provided informed written consent that has been endorsed by the respective IRBs and is consistent with the tenets of the Declaration of Helsinki.
A three-generation Chinese Family 020001 with ADRP was recruited for this study. The routine and ophthalmologic examination were performed on all patients including visual acuity tests, the condition of the fundus, the extent of the visual field, and flash electroretinogram after informed consent form was obtained. Genotyping and linkage analysis were completed by Yuan et al. 10 . 7 affected and 6 unaffected individuals from this family were collected 5 ml blood samples. We also recruited another cohort of 200 unrelated and ethnically matched individuals, and an additional 95 ADRP families. Total genomic DNA was isolated with DNA extraction kits (TianGen, Beijing, China) according to the manufacturer's protocol.
Whole-exome sequencing and data analysis. Genomic DNA (5 μ g) from the proband (individual II: 4) of Family 020001 was sent to Biotechnology Corporation (Shanghai, China) for whole-exome capture followed by sequencing. The whole exome was captured using Agilent SureSelect Human All Exon Kit according to the manufacturer's instructions. Briefly, samples were first prepared as Illumina sequencing libraries, which were then enriched for the desired target according to the Illumina Exome Enrichment protocol. Then, each captured library was loaded on a HiSeq 2500 platform, where paired-end sequencing was conducted with read lengths of 100 bp, providing an average coverage depth of at least 100× for each sample. Data were aligned to UCSC Genome Browser build hg19 by the Burroughs Wheeler Aligner. Local realignment was performed with the Genome Analysis Toolkit IndelRealigner and variants were called with SAM tools. Variants were filtered against 1000 Genomes Project, 1000G_ASN, esp6500si_all, and dbSNP137. Copy-number variations (CNVs) and structural variants from WES data were also assessed and bioinformatic prediction was based on BAM files analyzed by the Integrative Genomics Viewer (IGV) and CoNIFER.
Sanger sequencing and in silico analysis. In all members of Family 020001 with available DNA samples, the mutation was validated by Sanger sequencing as previously described 15 . The primers of all coding exons including flanking sequence of RPRF3 were designed by the web-based version of the Primer3 program (Table S1). Samples from another cohort of 200 unrelated ethnically matched controls and the probands of additional 95 ADRP families were also sequenced to detect variations in PRPF3.