Development and application of a next-generation-sequencing (NGS) approach to detect known and novel gene defects underlying retinal diseases.

0.84 MB·English

by  Audo, Isabelle

Audoetal.OrphanetJournalofRareDiseases2012,7:8
http://www.ojrd.com/content/7/1/8
RESEARCH Open Access
Development and application of a next-generation-
sequencing (NGS) approach to detect known and
novel gene defects underlying retinal diseases
Isabelle Audo1,2,3,4,5*, Kinga M Bujakowska1,2,3, Thierry Léveillard1,2,3, Saddek Mohand-Saïd1,2,3,4,
Marie-Elise Lancelot1,2,3, Aurore Germain1,2,3, Aline Antonio1,2,3,4, Christelle Michiels1,2,3, Jean-Paul Saraiva6,
Mélanie Letexier6, José-Alain Sahel1,2,3,4,7,8, Shomi S Bhattacharya1,2,3,5,9 and Christina Zeitz1,2,3*
Abstract
Background: Inherited retinal disorders are clinically and genetically heterogeneous with more than 150 gene
defects accounting for the diversity of disease phenotypes. So far, mutation detection was mainly performed by
APEX technology and direct Sanger sequencing of known genes. However, these methods are time consuming,
expensive and unable to provide a result if the patient carries a new gene mutation. In addition, multiplicity of
phenotypes associated with the same gene defect may be overlooked.
Methods: To overcome these challenges, we designed an exon sequencing array to target 254 known and
candidate genes using Agilent capture. Subsequently, 20 DNA samples from 17 different families, including four
patients with known mutations were sequenced using Illumina Genome Analyzer IIx next-generation-sequencing
(NGS) platform. Different filtering approaches were applied to identify the genetic defect. The most likely disease
causing variants were analyzed by Sanger sequencing. Co-segregation and sequencing analysis of control samples
validated the pathogenicity of the observed variants.
Results: The phenotype of the patients included retinitis pigmentosa, congenital stationary night blindness, Best
disease, early-onset cone dystrophy and Stargardt disease. In three of four control samples with known genotypes
NGS detected the expected mutations. Three known and five novel mutations were identified in NR2E3, PRPF3, EYS,
PRPF8, CRB1, TRPM1 and CACNA1F. One of the control samples with a known genotype belongs to a family with
two clinical phenotypes (Best and CSNB), where a novel mutation was identified for CSNB. In six families the
disease associated mutations were not found, indicating that novel gene defects remain to be identified.
Conclusions: In summary, this unbiased and time-efficient NGS approach allowed mutation detection in 75% of
control cases and in 57% of test cases. Furthermore, it has the possibility of associating known gene defects with
novel phenotypes and mode of inheritance.
Keywords: NGS, retinal disorders, diagnostic tool.
Background impairmentsuchasinrod-conedystrophies,alsoknownas
Inherited retinaldisordersaffectapproximately 1in 2000 retinitis pigmentosa (RP) or cone and cone-rod dystro-
individualsworldwide[1].Symptomsandassociatedphe- phies.Theheterogeneityofthesediseasesisreflectedinthe
notypes are variable. In some groups the disease can be numberofunderlyinggenedefects.Todatemorethan150
mildandstationarysuchasincongenitalstationarynight geneshavebeenimplicatedindifferentformsofretinaldis-
blindness (CSNB) or achromatopsia (ACHM), whereas ordershttp://www.sph.uth.tmc.edu/Retnet/home.htmand
other disorders are progressive leading to severe visual yetinasignificantproportionofpatientsthediseasecaus-
ingmutationcouldnotbeidentified,suggestingadditional
*Correspondence:[email protected];[email protected] novel genes that remain to be discovered. Furthermore,
1INSERM,U968,Paris,F-75012,France recentstudieshave outlined that distinct phenotypescan
Fulllistofauthorinformationisavailableattheendofthearticle
©2012Audoetal;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons
AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin
anymedium,providedtheoriginalworkisproperlycited.
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page2of17
http://www.ojrd.com/content/7/1/8
be related to the dysfunction of the same gene [2-4]. (Qiagen, Courtaboeuf,France).DNAsamplesfrom some
Furthermore,theremaybeadditionalphenotype-genotype patients with a diagnosis of RP were first analyzed and
associationsthatarestillnotrecognized.Thestate-of-the- excludedforknownmutationsbyapplying commercially
art phenotypic characterization including precise family available microarray analysis (arRP and adRP ASPER
historyandfunctionalaswellasstructuralassessment(i.e. Ophthalmics, Tartu, Estonia). Insome cases, pathogenic
routine ophthalmic examination, perimetry, color vision, variantsinEYS, C2orf71, RHO,PRPF31,PRPH2 andRP1
fullfieldandmultifocalelectroretinography(ERG),fundus wereexcludedbydirectSangersequencingofthecoding
autofluorescence (FAF) imaging and optical coherence exonic and flanking intronic regions of the respective
tomography(OCT))allowstargetedmutationanalysisfor genes[13-17]. ConditionsusedtoamplifyPRPH2 canbe
somedisorders.However,inmostcasesofinheritedretinal providedonrequest.
diseases,similarphenotypicfeaturescanbeduetoalarge
numberofdifferentgenedefects. Molecular genetic analysis using NGS
Various methods can be used for the identification of A custom-made SureSelect oligonucleotide probe library
thecorrespondinggeneticdefect.Allthesemethodshave was designed to capture the exons of 254 genes for dif-
advantages and disadvantages. Sanger sequencing is still ferent retinal disorders and candidate genes according
thegold-standardindeterminingthegenedefect,butdue to Agilent’s recommendations (Table 1). These genes
totheheterogeneityofthedisordersitistimeconsuming include 177 known genes underlying retinal dysfunction
andexpensivetoscreenallknowngenes.Mutationdetec- (http://www.sph.uth.tmc.edu/retnet/sum-dis.htm, Octo-
tionbycommerciallyavailableAPEXgenotypingmicroar- ber 2010, Table 1) and 77 candidate genes associated
rays (ASPER Ophthalmics, Estonia) [5,6] allows the with existing animal models and expression data (Table
detectionofonlyknownmutations.Inaddition,aseparate 2). The eArray web-based probe design tool was used
microarrayhasbeendesignedforeachinheritancepattern, for this purpose https://earray.chem.agilent.com/earray.
which tends to escalate the costs especially in simplex The following parameters were chosen for probe design:
cases, for which inheritance pattern cannot be predeter- 120 bp length, 3× probe-tiling frequency, 20 bp overlap
mined. Indirect methods with single nucleotide poly- in restricted regions, which were identified by the imple-
morphism (SNP) microarrays for linkage and mentation of eArray’s RepeatMasker program. A total of
homozygositymappingarealsopowerfultools,whichhas 27,430 probes, covering 1177 Mb, were designed and
proven itsreliabilityinidentifyingnovelandknowngene synthesized by Agilent Technologies (Santa Clara, CA,
defects[7-12].However,incaseofhomozygositymapping USA). Sequence capture, enrichment, and elution were
the method can only be applied to consanguineous performed according to the manufacturer’s instructions
families or inbred populations. To overcome these chal- (SureSelect, Agilent). Briefly, three μg of each genomic
lenges, we designed a custom sequencing array in colla- DNA were fragmented by sonication and purified to
boration with a company (IntegraGen, Evry, France) to yield fragments of 150-200 bps. Paired-end adaptor oli-
target all exons and part of flanking sequences for 254 gonucleotides from Illumina were ligated on repaired
knownandcandidateretinalgenes.Thisarraywassubse- DNA fragments, which were then purified and enriched
quently applied through NGS to a cohort of 20 patients by six PCR cycles. 500 ng of the purified libraries were
from 17 families with different inheritance pattern and hybridized to the SureSelect oligo probe capture library
clinicaldiagnosisincludingRP,CSNB,Bestdisease,early- for 24 h. After hybridization, washing, and elution, the
onsetconedystrophyandStargardtdisease. eluted fraction underwent 14 cycles of PCR-amplifica-
tion. This was followed by purification and quantifica-
Methods tion by qPCR to obtain sufficient DNA template for
Clinical investigation downstream applications. Each eluted-enriched DNA
The study protocol adhered to the tenets of the Declara- sample was then sequenced on an Illumina GAIIx as
tion of Helsinki and was approved by the local Ethics paired-end 75 bp reads. Image analysis and base calling
Committee (CPP, Ile de France V). Informed written was performed using Illumina Real Time Analysis
consent was obtained from each study participant. Index (RTA) Pipeline version 1.10 with default parameters.
patients underwent full ophthalmic examination as Sequence reads were aligned to the reference human
described before [13]. Whenever available, blood sam- genome (UCSC hg19) using commercially available soft-
ples from affected and unaffected family members were ware (CASAVA1.7, Illumina) and the ELANDv2 align-
collected for co-segregation analysis. ment algorithm. Sequence variation annotation was
performed using the IntegraGen in-house pipeline,
Previous molecular genetic analysis which consisted of gene annotation (RefSeq), detection
TotalgenomicDNAwasextractedfromperipheralblood of known polymorphisms (dbSNP 131, 1000 Genome)
leucocytesaccordingtomanufacturer’srecommendations followed by mutation characterization (exonic, intronic,
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page3of17
http://www.ojrd.com/content/7/1/8
Table 1Known retinal disease genes Table 1Known retinal disease genes(Continued)
Number Genename 48 DMD
1 ABCA4 49 DPP3
2 ABCC6 50 EFEMP1
3 ADAM9 51 ELOVL4
4 AHI1 52 ERCC6
5 AIPL1 53 EYS
6 ALMS1 54 FAM161A
7 ARL6 55 FBLN5
8 ARMS2 56 FSCN2
9 ATXN7 57 FZD4
10 BBS10 58 GNAT1
11 BBS12 59 GNAT2
12 BBS2 60 GPR98
13 BBS4 61 GRK1
14 BBS5 62 GRM6
15 BBS7 63 GUCA1A
16 BBS9 64 GUCA1B
17 BEST1 65 GUCY2D
18 C1QTNF5 66 HMCN1
19 C2 67 HTRA1
20 C2orf71 68 IDH3B
21 C3 69 IMPDH1
22 CA4 70 IMPG2
23 CABP4 71 INPP5E
24 CACNA1F 72 INVS
25 CACNA2D4 73 IQCB1
26 CC2D2A 74 JAG1
27 CDH23 75 KCNJ13
28 CDH3 76 KCNV2
29 CEP290 77 KLHL7
30 CERKL 78 LCA5
31 CFB 79 LRAT
32 CFH 80 LRP5
33 CHM 81 MERTK
34 CLN3 82 MFRP
35 CLRN1 83 MKKS
36 CNGA1 84 MKS1
37 CNGA3 85 MTND1
38 CNGB1 86 MTND6
39 CNGB3 87 MT-AP6
40 CNNM4 88 MTND2
41 COL11A1 89 MTND5
42 COL2A1 90 MTND4
43 COL9A1 91 MYO7A
44 CRB1 92 NDP
45 CRX 93 NPHP1
46 CYP4V2 94 NPHP3
47 DFNB31 95 NPHP4
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page4of17
http://www.ojrd.com/content/7/1/8
Table 1Known retinal disease genes(Continued) Table 1Known retinal disease genes(Continued)
96 NR2E3 143 RP1L1
97 NRL 144 RP2
98 NYX 145 RP9
99 OAT 146 RPE65
100 OFD1 147 RPGR
101 OPA1 148 RPGRIP1
102 OPA3 149 RPGRIP1L
103 OPN1LW 150 RS1
104 OPN1MW 151 SAG
105 OPN1Sw 152 SDCCAG8
106 OTX2 153 SEMA4A
107 PANK2 154 SLC24A1
108 PAX2 155 SNRNP200
109 PCDH15 156 SPATA7
110 PCDH21 157 TEAD1
111 PDE6A 158 TIMM8A
112 PDE6B 159 TIMP3
113 PDE6C 160 TLR3
114 PDE6G 161 TLR4
115 PDZD7 162 TMEM126A
116 PEX1 163 TOPORS
117 PEX2 164 TREX1
118 PEX7 165 TRIM32
119 PGK1 166 TRPM1
120 PHYH 167 TSPAN12
121 PITPNM3 168 TTC8
122 PRCD 169 TTPA
123 PROM1 170 TULP1
124 PRPF3 171 UNC119
125 PRPF31 172 USH1C
126 PRPF8 173 USH1G
127 PRPH2 174 USH2A
128 RAX2 175 VCAN
129 RB1 176 WFS1
130 RBP3 177 ZNF513
131 RBP4
132 RD3
silent, nonsense etc.). For each position, the exomic fre-
133 RDH12
quencies (homozygous and heterozygous) were deter-
134 RDH5 mined from all the exomes already sequenced by
135 RGR IntegraGen and the exome results provided by HapMap
136 RGS9 project.
137 RGS9BP
138 RHO Investigation of annotated sequencing data
139 RIMS1 We received the annotated sequencing data in the form
of excel tables. On average 946 SNPs and 83 insertions
140 RLBP1
and deletions were identified for each sample (Figure 1).
141 ROM1
By using the filtering system, we first investigated var-
142 RP1
iants (nonsense and missense mutations, intronic
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page5of17
http://www.ojrd.com/content/7/1/8
Table 2Candidate genesfor retinal disorders
Number Genename Reason References
1 ADCY1 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
2 ANKRD33 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
3 ANXA2 Promotionofchoroidalneovascularization [36]
4 ARL13B Ciliaprotein,mutationsleadtoJoubertSyndrome [37]
5 BMP7 RegulationofPax2inmouseretina [38]
6 BSG - ThierryLeveillardpersonalcommmunication
7 CAMK2D diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
8 CCDC28B ModifierforBBS [39,40]
9 CLCN7 Cln7-/-micesevereosteopetrosisandretinaldegeneration [41]
10 COL4A3 Alportsyndrome,witheyeabnormalities [42,43]
11 COL4A4 Alportsyndrome,witheyeabnormalities [42,44]
12 COL4A5 Alportsyndrome,witheyeabnormalities [42,45]
13 CUBN - PersonalcommunicationRenataKozyraki
14 CYP1B1 glaucoma [46]
15 DOHH diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
16 DSCAML1 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
17 ESRRB diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
18 FIZ1 InteractorofNRL [47]
19 GJA9 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
20 GNAZ diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
21 GNGT1 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
22 GPR152 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
23 HCN1 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
24 HEATR5A diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
25 HIST1H1C Expressedinretina Expressiondatabases
26 IMPG1 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
27 INSL5 diff.Expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
28 KCNB1 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
29 KCTD7 Expressedinretina Expressiondatabases
30 LASS4 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
31 LRIT2 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiomRd1mouse
32 LRP2 - PersonalcommunicationRenataKozyraki
33 MAB21L1 diff.expressionRd1mouse Chalmeletal.,manuscriptinpreparatiom
34 MAP2 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
35 MAS1 DegenerationofconesduetoexpressionofMas1 [48]
36 MAST2 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
37 MPP4 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
38 MYOC glaucoma [49]
39 NDUFA12 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
40 NEUROD1 BETA2/NeuroD1-/-mouse:photoreceptordegeneration [50]
41 NOS2 glaucoma [51]
42 NXNL1 Rod-derivedconeviabilityfactor [52]
43 NXNL2 Rod-derivedconeviabilityfactor2 [53]
44 OPN1MW2 Coneopsin,medium-wave-sensitive2 [54]
45 OPTN glaucoma [55]
46 PFKFB2 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
47 PIAS3 Rodphotoreceptordevelopment [56]
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page6of17
http://www.ojrd.com/content/7/1/8
Table 2Candidate genesfor retinal disorders (Continued)
48 PKD2L1 Diff.expressioninhumanretinaldetachment Delyferetal.2011submitted
49 PLEKHA1 Age-relatedmaculardegeneratiom [57]
50 PPEF2 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
51 RAB8A InteractswithRPGR,roleinciliabiogenesisandmaintenance [58]
52 RABGEF1 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
53 RCVRN diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
54 RGS20 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
55 RNF144B diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
56 RORB Rodphotoreceptordevelopmentinmice [59]
57 RXRG Retinoicacidreceptor,highlyexpressedintheeye Expressiondatabases
58 SGIP1 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
59 SLC16A8 Alteredvisualfunctioninko-mice [60]
60 SLC17A7 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
61 STAM2 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
62 STK35 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
63 STX3 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
64 SV2B diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
65 TBC1D24 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
66 THRB EssentialforM-conedevelopmentinrodents [61]
67 TMEM216 Ciliaprotein,mutationsleadtoJoubertandMeckelsyndrome [62]
68 TMEM67 Ciliaprotein,mutationsleadtoJoubert [63]
69 TRPC1 diff.expressionrd1mouse diff.expressionRd1mouse
70 UHMK1 diff.expressionrd1mouse diff.expressionRd1mouse
71 VSX1 StimulatorforpromoterNXNL1 [64]
72 VSX2 StimulatorforpromoterNXNL1 [64]
73 WDR17 diff.expressionrd1mouse diff.expressionRd1mouse
74 WDR31 diff.expressionNxnl1-/-mouse [65]
75 WISP1 diff.expressionrd1mouse Chalmeletal.,manuscriptinpreparatiom
76 XIAP ProtectsphotoreceptorsinanimalmodelsofRP [66]
77 ZDHHC2 diff.expressionRd1mouse Chalmeletal.,manuscriptinpreparatiom
variants located +/- 5 apart from exon), which were acid residue did not change it was considered as “highly
absent in dbSNP and NCBI databases http://ncbi.nlm. conserved”. If a different change was seen in fewer than
nih.gov/. In the absence of known gene defects or puta- five species and not in the primates then it was consid-
tive pathogenic variants (see below) in the first step, we ered as “moderately conserved” and if a change was pre-
selected known genes, which were previously clinically sent in 5-7, it was considered as “weakly conserved”,
associated including variants present in dbSNP and otherwise the amino acid residue was considered as “not
NCBI databases (Figure 1). Each predicted pathogenic conserved”, 5) pathogenicity predictions with bioinfor-
variant was confirmed by Sanger sequencing. matic tools (Polyphen: Polymorphism Phenotyping,
http://genetics.bwh.harvard.edu/pph/ and SIFT: Sorting
Assessment of the pathogenicity of variants Intolerant From Tolerant, http://blocks.fhcrc.org/sift/
Following criteria were applied to evaluate the patho- SIFT.html) if at least one of the program predicted the
genic nature of novel variations identified by NGS: 1) variant to be possibly damaging, it was considered to be
stop/frameshift variants were considered as most likely pathogenic; 6) presence of the second mutant allele in
to be disease causing; 2) co-segregation in the family; 3) the case of autosomal recessive inheritance. Mutations
absence in control samples; 4) for missense mutations were described according to the HGVS website http://
amino acid conservation was studied in the UCSC Gen- www.hgvs.org/mutnomen. In accordance with this
ome Browser http://genome.ucsc.edu/ across species nomenclature, nucleotide numbering reflects cDNA
from all different evolutionary branches. If the amino numbering with +1 corresponding to the A of the ATG
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page7of17
http://www.ojrd.com/content/7/1/8
(cid:47)(cid:374)(cid:410)(cid:286)(cid:336)(cid:396)(cid:258)(cid:39)(cid:286)(cid:374)(cid:3)
(cid:115)(cid:258)(cid:396)(cid:349)(cid:258)(cid:374)(cid:410)(cid:3)(cid:282)(cid:286)(cid:410)(cid:286)(cid:272)(cid:410)(cid:349)(cid:381)(cid:374)(cid:855)
(cid:1013)(cid:1008)(cid:1010)(cid:3)(cid:94)(cid:69)(cid:87)(cid:400)(cid:3)(cid:258)(cid:374)(cid:282)(cid:3)(cid:1012)(cid:1007)(cid:3)(cid:47)(cid:374)(cid:24)(cid:286)(cid:367)(cid:400)
•(cid:282)(cid:271)(cid:94)(cid:69)(cid:87)(cid:3)(cid:258)(cid:374)(cid:282)(cid:3)(cid:69)(cid:18)(cid:17)(cid:47)(cid:3)(cid:296)(cid:349)(cid:367)(cid:410)(cid:286)(cid:396)(cid:349)(cid:374)(cid:336)
•(cid:47)(cid:374)(cid:272)(cid:367)(cid:437)(cid:282)(cid:349)(cid:374)(cid:336)(cid:3)(cid:381)(cid:374)(cid:367)(cid:455)(cid:3)(cid:286)(cid:454)(cid:381)(cid:374)(cid:349)(cid:272)(cid:3)(cid:258)(cid:374)(cid:282)(cid:3)(cid:1085)(cid:876)(cid:882)(cid:1009)(cid:3)(cid:349)(cid:374)(cid:410)(cid:396)(cid:381)(cid:374)(cid:349)(cid:272)(cid:3)(cid:448)(cid:258)(cid:396)(cid:349)(cid:258)(cid:374)(cid:410)(cid:400)
•(cid:47)(cid:374)(cid:272)(cid:367)(cid:437)(cid:282)(cid:349)(cid:374)(cid:336)(cid:3)(cid:381)(cid:374)(cid:367)(cid:455)(cid:3)(cid:373)(cid:349)(cid:400)(cid:400)(cid:286)(cid:374)(cid:400)(cid:286)(cid:3)(cid:258)(cid:374)(cid:282)(cid:3)(cid:374)(cid:381)(cid:374)(cid:400)(cid:286)(cid:374)(cid:400)(cid:286)(cid:3)(cid:373)(cid:437)(cid:410)(cid:258)(cid:410)(cid:349)(cid:381)(cid:374)(cid:400)
•(cid:28)(cid:454)(cid:272)(cid:346)(cid:258)(cid:374)(cid:336)(cid:286)(cid:3)(cid:374)(cid:286)(cid:448)(cid:286)(cid:396)(cid:3)(cid:393)(cid:396)(cid:286)(cid:400)(cid:286)(cid:374)(cid:410)(cid:3)(cid:349)(cid:374)(cid:3)(cid:1005)(cid:1004)(cid:1004)(cid:1004)(cid:3)(cid:336)(cid:286)(cid:374)(cid:381)(cid:373)(cid:286)(cid:400)(cid:855)
(cid:1005)(cid:1005)(cid:3)(cid:448)(cid:258)(cid:396)(cid:349)(cid:258)(cid:374)(cid:410)(cid:400)
•(cid:60)(cid:374)(cid:381)(cid:449)(cid:374)(cid:3)(cid:336)(cid:286)(cid:374)(cid:286)(cid:3)(cid:282)(cid:286)(cid:296)(cid:286)(cid:272)(cid:410)(cid:845) •(cid:69)(cid:381)(cid:3)(cid:282)(cid:271)(cid:94)(cid:69)(cid:87)(cid:3)(cid:258)(cid:374)(cid:282)(cid:3)(cid:69)(cid:18)(cid:17)(cid:47)(cid:3)(cid:296)(cid:349)(cid:367)(cid:410)(cid:286)(cid:396)(cid:349)(cid:374)(cid:336)
•(cid:373)(cid:381)(cid:282)(cid:286)(cid:3)(cid:381)(cid:296)(cid:3)(cid:349)(cid:374)(cid:346)(cid:286)(cid:396)(cid:349)(cid:410)(cid:258)(cid:374)(cid:272)(cid:286) •(cid:38)(cid:381)(cid:396)(cid:3)(cid:258)(cid:396)(cid:3)(cid:272)(cid:258)(cid:400)(cid:286)(cid:400)(cid:855)(cid:3)(cid:28)(cid:454)(cid:272)(cid:346)(cid:258)(cid:374)(cid:336)(cid:286)(cid:400)(cid:3)(cid:272)(cid:258)(cid:374)(cid:3)(cid:271)(cid:286)(cid:3)
(cid:1004)(cid:3)
•(cid:393)(cid:396)(cid:381)(cid:410)(cid:286)(cid:349)(cid:374)(cid:3)(cid:296)(cid:437)(cid:374)(cid:272)(cid:410)(cid:349)(cid:381)(cid:374) (cid:346)(cid:286)(cid:410)(cid:3)(cid:393)(cid:396)(cid:286)(cid:400)(cid:286)(cid:374)(cid:410)(cid:3)(cid:349)(cid:374)(cid:3)(cid:1005)(cid:1004)(cid:1004)(cid:1004)(cid:3)(cid:336)(cid:286)(cid:374)(cid:381)(cid:373)(cid:286)(cid:400)(cid:3)
(cid:448)(cid:258)(cid:396)(cid:349)(cid:258)(cid:374)(cid:410)(cid:3)
•(cid:393)(cid:381)(cid:367)(cid:455)(cid:393)(cid:346)(cid:286)(cid:374)(cid:853)(cid:3)(cid:400)(cid:349)(cid:296)(cid:410)(cid:855) •(cid:60)(cid:374)(cid:381)(cid:449)(cid:374)(cid:3)(cid:336)(cid:286)(cid:374)(cid:286)(cid:3)(cid:282)(cid:286)(cid:296)(cid:286)(cid:272)(cid:410)(cid:845)
(cid:1005)(cid:882)(cid:1009)(cid:3)(cid:448)(cid:258)(cid:396)(cid:349)(cid:258)(cid:374)(cid:410)(cid:400)(cid:3) (cid:1005)(cid:882)(cid:1009)(cid:3)(cid:448)(cid:258)(cid:396)(cid:349)(cid:258)(cid:374)(cid:410)(cid:400)(cid:3)
•(cid:18)(cid:381)(cid:374)(cid:296)(cid:381)(cid:396)(cid:373)(cid:258)(cid:410)(cid:349)(cid:381)(cid:374)(cid:3)(cid:271)(cid:455)(cid:3)(cid:94)(cid:258)(cid:374)(cid:336)(cid:286)(cid:396)(cid:3)(cid:400)(cid:286)(cid:395)(cid:437)(cid:286)(cid:374)(cid:272)(cid:349)(cid:374)(cid:336)
•(cid:18)(cid:381)(cid:882)(cid:400)(cid:286)(cid:336)(cid:396)(cid:286)(cid:336)(cid:258)(cid:410)(cid:349)(cid:381)(cid:374)(cid:3)(cid:258)(cid:374)(cid:258)(cid:367)(cid:455)(cid:400)(cid:349)(cid:400)
•(cid:28)(cid:454)(cid:272)(cid:367)(cid:437)(cid:400)(cid:349)(cid:381)(cid:374)(cid:3)(cid:349)(cid:374)(cid:3)(cid:272)(cid:381)(cid:374)(cid:410)(cid:396)(cid:381)(cid:367)(cid:3)(cid:400)(cid:258)(cid:373)(cid:393)(cid:367)(cid:286)(cid:400)
Figure1Flowchartofvariantanalysis.IntegraGenprovidedtheresultsinformofexceltables.Foreachsampleonaverage946SNPsand83
inDelsweredetected,ofwhich11representmissense,nonsenseorputativesplicesitemutations,whichwereabsentindbSNB,NCBIand1000
genomedatabases.Ofthose1-5variantswerepredictedtobepathogenic.Incasewherenoneofthevariantswerepredictedtobepathogenic,
dbSNB,NCBIand1000genomedatabaseswereincludedtodetectmutationsreferencedwithanrs-number.Co-segregationanalysiswas
performedinfamilieswithputativepathogenicvariants.
translation initiation codon in the reference sequence. Validation of the novel genetic testing tool for retinal
The initiation codon is codon 1. The correct nomencla- disorders
ture for mutation was checked applying Mutalyzer To validate the novel genetic testing tool for retinal dis-
http://www.lovd.nl/mutalyzer/. orders, we used four DNA samples from families, in
which we had previously identified different types of
Results mutations by Sanger sequencing: one 1 bp duplication
Theoverallsequencingcoverageofthecapturedregions and one 1 bp deletion in PRPF31 and missense muta-
was98.4%and90.4%fora1×anda10×coveragerespec- tions in TRPM1 and BEST1 (Table 3). Three of the four
tively. The overall sequencing depth was > 120×. The mutations were detectable by NGS, whereas the deletion
number of reference and variant sequences detected by in PRPF31 was not identified. To validate if this was due
NGS, reflected the correct zygosity state of the variant; to a technical problem of deletion detection in general
on average if 50% of the sequences represented the var- or low coverage at this position, the sequencing depth
iant, then a heterozygous state was called, while if 100% was investigated in detail. Indeed the coverage at this
ofthesequencesrepresentedthevariant,thenahomozy- position reflected by the mean depth was only ~1-6 for
gousorhemizygousstatewasannotatedbyIntegraGen. all samples. This indicates that although the coverage in
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page8of17
http://www.ojrd.com/content/7/1/8
Table 3Patients withknown mutations used to validate the novel genetic approach forretinal disorders
Index Phenotype Gene Mutation Allele Readreference Readvariant Mutationdetectedby Mean
State NGS NGS NGS depth
CIC00034, adRP PRPF31 c.666dup het 11 13 yes 21.3-22.5
F28 p.
I223YfsX56
CIC00140, adRP PRPF31 c.997delG het - - no 5.0-5.2
F108 p.
E333SfsX5
CIC00238, arCSNB TRPM1 c.1418G> homo 0 38 yes 36.7
F165 C
p.R473P
CIC00707, BestandadCSNBsee BEST1 c.73C>T het 40 38 yes 99.4
F470 Table5 p.R25W
general was very good, specific probes used here need to detected by NGS was a novel heterozygous TRPM1
be redesigned to improve the capture for specific exons. mutation (Table 4, Figure 3).
Detection of known and novel mutations Unsolved cases
Some of the patients from the 14 families with no In six of the 14 families with Stargardt disease, adRP,
known gene defect were previously excluded for known adCD with postreceptoral defects, arRP, early onset
mutations using microarray analysis and by Sanger arCD with macrocephaly and mental retardation
sequencing in the known genes EYS, C2orf71, RHO, described in affected sister and x-linked cCSNB, the dis-
PRPF31, PRPH2 and RP1. Other samples were never ease associated mutations remain to be elucidated or
genetically investigated. In four DNA samples known validated (Table 6, Figure 5).
mutations were detected (Table 4) from three different
families with autosomal dominant (ad) or recessive (ar) Discussion
RP. All mutations co-segregated with the phenotype By using NGS in 254 known and candidate genes we
(Figure 2). In seven samples, novel mutations in known were able to detect known and novel mutations in 57%
genes were identified. These mutations co-segregated of families tested. In order to achieve this goal, we
with the phenotype from five different families with applied a rigorous protocol (Figure 1). To our knowl-
adCSNB, x-linked incomplete CSNB, adRP, arRP and x- edge, this is the first report using NGS to investigate all
linked RP (Table 5, Figures 3 and 4). One of the cases inherited retinal disorders at once. In a study restricted
from these five families was also used as a control for to adRP, Bowne and co-workers used a similar approach
Best disease carrying a known BEST1 mutation (Table including 46 known and candidate genes for adRP [18].
3). In addition to the Best phenotype, ERG-responses of All their cases had previously been screened and
this patient resembled those of complete CSNB, i.e. excluded for most of the known genes underlying adRP.
showing selective ON-bipolar pathway dysfunction. This The authors were able to identify known or novel muta-
phenotype was independent of the Best phenotype (Fig- tions in five out of 21 cases in genes not included in a
ure 3). The most likely disease causing mutation pre-screening [18]. This added five patients to their
Table 4Detection ofknown mutations by usingthe novel genetic approach for retinal disorders
Index Phenotype Pre-screening Gene Mutation Allele Read Read Reference Mutationverifiedby
State reference variant Sangerandco-
NGS NGS segregation
CIC00019, adRP Linkage,RHO, PRPF3 c.1481C>T het 25 22 [67] yes
F16 PRPF31,PRPH2, p.T494M
RP1
CIC0000893, adRP RHO,PRPF31, NR2E3 c.166G>A het 5 3 [68] yes
F574 PRPH2,RP1 p.G56R
CIC000128, arRP, - EYS c.408_423del homo - 179 [13,69] yes
F100 consang. p.N137VfsX24
CIC0000943, arRP, - EYS c.408_423delp. homo 0 193 [13,69] yes
F100 consang N137VfsX24
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page9of17
http://www.ojrd.com/content/7/1/8
Fam 16: PRPF3: M: c.1481C>T p.T494M
19 3266 843 3239 932 982 1145 1069 3251
[M]+[=] [=]+[=] [M]+[=] [=]+[=] [=]+[=][M]+[=] [M]+[=] [M]+[=][=]+[=]
3113 352 3361 1250 3455 1120 3240 910 1165 3263 1035 1142 1423 1073 3780 1084
[=]+[=] [M]+[=][=]+[=][M]+[=][=]+[=][=]+[=] [=]+[=][=]+[=] [M]+[=] [=]+[=][=]+[=] [=]+[=] [=]+[=] [=]+[=][=]+[=] [=]+[=]
? ?
909 984 983 1259 1248 1119 1023 911 913 1167 1166 1036 1037 1143 1421
[=]+[=] [M]+[=][=]+[=] [=]+[=][M]+[=][=]+[=] [=]+[=] [=]+[=][=]+[=] [M]+[=] [=]+[=][=]+[=][=]+[=] [=]+[=] [=]+[=]
Fam 100: EYS: M: c.408_423del16 p.N137VfsX24
Fam 574: NR2E3: M: c.166G>A p.G56R
128 203
[M]+[M][M]+[M]
3744 181 943 1808 893 2501 27611394 2733
[M]+[M] [M]+[M] [M]+[M] [M]+[M] [M]+ [M]+ [=]+ [M]+ [=]+
[=] [=] [=] [=] [=]
894 2502 1395
[=]+ [M]+ [M]+
[=] [=] [=]
Figure2DetectionofknownmutationsbyNGSin254retinalgenes.Theindexpatient19offamily16withadRPrevealedthep.T494M
mutationsinPRPF3,whichco-segregateswiththephenotype.Twofamilymembersneverclinicallyinvestigatedfromthelastgeneration(984
and1167carryingaquestionmark)werereportedtobenotaffectedbutcarriedthemutation.Theymaydevelopthephenotypeatalater
stage.Inadditionvariabilityofthephenotypeofthismutationwasdocumented[35].Twopatients,128and943offamily100witharRPfrom
JewishoriginrevealedtheknownEYSmutationp.N137VfsX24,whichwasfoundinallscreenedaffectedfamilymembers.Theindexpatient893
offamily574showedthepreviouslydescribedNR2E3p.G56Rmutation,whichco-segregatedwiththephenotype.
adRP cohort with known gene defects, indicating that which was not confirmed by direct Sanger sequencing,
64% of their patients show known mutations with new suggesting the possibility of false positive using the high
genes still to be discovered in the remaining 36%. The throughput screening. Verification by direct Sanger
current study provides a more exhaustive tool, since it sequencing of most likely pathogenic variants is there-
incorporates screening of 254 genes implicated in var- fore essential to validate NGS data, although the false
ious retinal disorders of different inheritance patterns positive rate is assumed to be low (in our study 1/28
and additional candidate genes for these phenotypes. verified sequence variants represented a false positive).
With this approach a cohort of both pre-screened and Overall, the study of 20 subjects from 17 families by
unscreened samples, was investigated. The mutation NGS showed that most of the targeted regions are well
detection rate of 57% is high and was never obtained covered (more than 98%). However, some of the regions
before by high throughput screening methods. Further- showed a lower coverage (GC-rich regions) or were not
more, this approach is probably less time consuming captured (repetitive regions). This was for instance the
and expensive than existing methods such as direct case for two genes underlying cCSNB, (i.e. NYX and
sequencing of all known genes or microarray analysis. GRM6) and the repetitive region of ORF15 of RPGR.
Of note however is one of the variants detected with the For GC-rich regions the capture design could be
NGS approach (i.e. p.V973L exchange in GUCY2D), improved in the future by modifying NGS chemistry, as
Audoetal.OrphanetJournalofRareDiseases2012,7:8 Page10of17
http://www.ojrd.com/content/7/1/8
Table 5Detection ofnovel mutations by usingthe novel genetic approach for retinal disorders
Index Phenotype Pre- Gene Mutation Allele Read Read Mutation Conservation Polyphen Sift
screening State reference variant verifiedby
NGS NGS Sangerand
co-
segregation
CIC00707, adCSNB RHO, TRPM1 c.1961A het 39 38 yes moderately possibly tolerated
F470 andBest PDE6B, >C conserved damaging
seeTable3 GNAT1 p.H654P
CIC000348, adRP,mild RHO, PRPF8 c.6992A het 13 10 yes moderately possibly affect
F232 PRPF31, >G conserved damaging protein
PRPH2, p. function
RP1,adRP E2331G
chip
CIC000346, adRP - PRPF8 c.6992A het 5 9 yes moderately possibly affect
F232 >G conserved damaging protein
p.E2331G function
CIC000347, as - PRPF8 c.6992A het 15 17 yes moderately possibly affect
F232 adRP >G conserved damaging protein
p.E2331G function
CIC04240, arRP, RS1 CRB1 c.2219C homo 2 194 yes highly probably affect
F2025 consang., >T conserved damaging protein
detailed p.S740F function
clinicin
[70]
CIC00199, adRPorx- RHO, RPGR c.248-2A hetero 30 22 yes conserved n.a. n.a.
F146 linkedRP PRPF31, >G splicesite
with PRPH2, splice
affected RP1,adRP defect
carrier chip
CIC04094, icCSNB - CACNA1F c.973C> hemi 0 28 yes n.a. n.a. n.a.
F1915 T
p.Q325X
it was successfully achieved for Sanger sequencing using factor gene, PRPF31 and recently for PRPF8 as well
different additives, which improved the amplification [19-22]. Interestingly, a novel TRPM1 mutation was
and subsequent sequencing. If repetitive regions like identified in a patient with adCSNB, a gene previously
ORF15 of RPGR remain problematic for sequencing by only associated with arCSNB [23-26]. This is the first
NGS, direct Sanger sequencing of these targets might be report of a TRPM1 mutation co-segregating with ad
the first screening of choice; in particular for disorders Schubert-Bornschein type complete CSNB. Since the
caused only by a few gene defects such as CSNB, and location of this mutation is not different compared to
xl-RP. other mutations leading to arCSNB, it is not quite clear
By applying NGS sequencing to our retinal panel, how TRPM1 mutations might lead to either ad or
known and novel mutations were detected in different arCSNB. Functional investigations are needed to validate
patients. We believe that our diagnostic tool is particu- the pathogenicity of this variant. Furthermore, this find-
larly important for heterogeneous disorders like RP, for ing suggests that TRPM1 heterozygous mutation carriers
which many gene defects with different prevalence have from arCSNB families should be investigated by electro-
been associated to one phenotype. It also allows the retinography to determine whether they display similar
rapid detection of novel mutations in minor genes retinal dysfunction as in affected members of the pre-
which are often not screened as a priority by direct San- sented adCSNB family. Detection of a novel RPGR splice
ger sequencing. This was the case in our study for three site mutation in family 146 presented a challenge. The
individuals from one family with adRP in which NGS actual disease causing change was concealed under a
detected a novel PRPF8 mutation in both affected and wrongly annotated rs62638633, which had previously
one unaffected family member (Table 4, Figure 4). In been clinically associated to RP by a German group
this family, the RP phenotype is mild and therefore it is http://www.ncbi.nlm.nih.gov/sites/varvu?gen-
possible that the unaffected member may develop symp- e=6103&rs=62638633, (personal communication, Mar-
toms later in life or alternatively it may be a case of kus Preising). These observations indicate that the
incomplete penetrance as reported for another splicing stringent filtering we applied initially can mask those