Molecular Genetics and Metabolism xxx (2014) xxx–xxx

YMGME-05803; No. of pages: 6; 4C:

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

j ourna l homepage: www.e lsev ie r .com/ locate /ymgme

Novel pathogenic COL11A1/COL11A2 variants in Stickler syndromedetected by targeted NGS and exome sequencing

Frederic R. Acke a,b, Fransiska Malfait b, Olivier M. Vanakker b, Wouter Steyaert b, Kim De Leeneer b,Geert Mortier c, Ingeborg Dhooge a, Anne De Paepe b, Els M.R. De Leenheer a, Paul J. Coucke b,⁎a Department of Otorhinolaryngology, Ghent University Hospital, De Pintelaan 185, B-9000 Ghent, Belgiumb Center for Medical Genetics, Ghent University Hospital, De Pintelaan 185, B-9000 Ghent, Belgiumc Department of Medical Genetics, Antwerp University Hospital, University of Antwerp, Wilrijkstraat 10, 2650 Edegem, Belgium

Abbreviations:MLPA, multiplex ligation-dependent pgeneration sequencing; T-NGS, targeted next-generation sknown significance;WES,whole-exome sequencing.⁎ Corresponding author.

E-mail address: Paul.Coucke@ugent.be (P.J. Coucke).

http://dx.doi.org/10.1016/j.ymgme.2014.09.0011096-7192/© 2014 Published by Elsevier Inc.

Please cite this article as: F.R. Acke, et al., Novsequencing, Mol. Genet. Metab. (2014), http

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 25 July 2014Received in revised form 1 September 2014Accepted 1 September 2014Available online xxxx

Keywords:COL11A1COL11A2Stickler syndromeNext-generation sequencingTargeted NGSExome sequencing

Introduction: Stickler syndrome is caused bymutations in genes encoding type II and type XI collagens. About 85%of the pathogenic variants is found in COL2A1 (Stickler type 1), whereas aminority ofmutations has been report-ed in COL11A1 (Stickler type 2) and COL11A2 (Stickler type 3). Beside the typical skeletal and orofacial manifes-tations, ocular anomalies are predominantly present in type 1 and type 2, while hearing loss ismore pronouncedin type 2 and type 3.Methods:We performed COL11A1mutation analysis for 40 type 2 Stickler patients and COL11A2 mutation anal-ysis for five type 3 Stickler patients, previously all COL2A1 mutation-negative, using targeted next-generationsequencing (NGS) whereas whole-exome sequencing (WES) was performed in parallel for two patients. Threepatients were analyzed for both genes due to unclear ocular findings.Results: In total 14 COL11A1 and two COL11A2mutations could be identified, seven of which are novel. Splice sitealterations are the most frequent mutation type, followed by glycine substitutions. In addition, six variants ofunknown significance (VUS) have been found. Identical mutations and variants were identified with both NGS

techniques.Conclusion:Weexpand themutation spectrum of COL11A1 and COL11A2 in Stickler syndrome patients and showthat targeted NGS is an efficient and cost-effective molecular tool in the genetic diagnosis of Stickler syndrome,whereas the more standardized WES might be an alternative approach.

© 2014 Published by Elsevier Inc.

1. Introduction

Stickler syndrome comprises a clinically and genetically heteroge-neous group of heritable connective tissue disorders, characterized byjoint hypermobility, premature joint degeneration, myopia, retinal de-tachment, conductive and/or sensorineural hearing loss, midfacial hy-poplasia, micrognathia and palatal defects [1,2]. Three main types canbe distinguished based on the presence/absence of vitreous anomaliesand severity of the hearing loss [3,4]. Type 1 Stickler syndrome, charac-terized by a ‘membranous’ vitreous and mild-to-moderate high-frequency sensorineural hearing loss, is the most common type (85%)and is caused by mutations in the COL2A1 gene [5,6]. The features oftype 2 Stickler syndrome, present in about 10% of Stickler patients andin which a COL11A1 mutation can be found [7–17], include a ‘beaded’

robe amplification; NGS, next-equencing; VUS, variants of un-

el pathogenic COL11A1/COL11://dx.doi.org/10.1016/j.ymgm

vitreous and pronounced sensorineural hearing loss affecting allfrequencies. The rare type 3 Stickler syndrome, caused by a COL11A2mutation [18–23], is similar to type 2 with respect to hearing loss, butcan be distinguished by the absence of ocular anomalies.

Traditional Sanger sequencing, which has routinely been used toidentify disease-causing mutations, is laborious, expensive and time-consuming, especially for large genes like COL11A1 and COL11A2 thatconsist of more than 130 exons all together. The recent introduction ofnext-generation sequencing (NGS) as a tool for the identification ofmu-tations has been proven to be very powerful, efficient and cost-effective[24]. The choice of the enrichment strategy preceding NGS sequencingitself, is still a matter of debate. Either a targeted approach by PCR orarray-capturing limited to the genes of interest (targeted NGS, T-NGS),or a whole-exome capturing approach (whole-exome sequencing,WES) can be applied, the former resulting in a better coverage reachingalmost 100% of the targeted region and the latter also enabling the in-vestigation of genes other than the known candidate genes. Neverthe-less, the relatively high cost of WES and the incomplete coverageshould be considered, and handling of incidental findings should bediscussed.

A2 variants in Stickler syndrome detected by targetedNGS and exomee.2014.09.001

2 F.R. Acke et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx

In the current study, our aim was to further expand the mutationspectrum and refine the genotype–phenotype correlation of Sticklersyndrome by analyzing the COL11A1 and COL11A2 genes in a COL2A1-negative Stickler syndrome cohort and to evaluate the use of the PCR-based T-NGS and WES in the molecular diagnostic work-up of Sticklersyndrome.

2. Materials and methods

2.1. Patient selection

All clinical records of patients previously referredwith a suspicion ofStickler syndrome who were COL2A1 mutation-negative by Sangersequencing and multiplex ligation-dependent probe amplification(MLPA) analysis, were reviewed. Minimal clinical criteria for COL11A1analysis in the current study were signs of a beaded vitreous, indicativefor type 2 Stickler syndrome, and/or the presence of moderate to severesensorineural hearing loss (N40 dB) indicative for type 2 or type 3Stickler syndrome. If only one of these was observed, at least twoother symptoms of Stickler syndrome (high myopia defined as moresevere than −6 diopters, glaucoma/cataract, retinal detachment, jointhypermobility/premature joint degeneration, midfacial hypoplasia/micrognathia, palatal defect) had to be present for inclusion. In pro-bands where presence of some of these clinical characteristics wasuncertain, a positive family history for these characteristics was alsotaken into account. If hearing loss and at least two other clinicalhallmarks of Stickler syndrome were observed in the absence of ocularinvolvement, COL11A2 analysis was initially performed. In total 48 pa-tients were selected accounting for 18% of our COL2A1-negative Sticklerpatient cohort. When analysis of one gene was negative, the other genewas also investigated.

2.2. Next-generation sequencing analysis

For the T-NGS approach, specific primers were designed for all 67COL11A1 and 66 COL11A2 exons and their flanking intronic sequences(sequences available on request). Primers were tailed at their 5′ endwith anM13 universal sequence. In case of limited amounts or lowqual-ity DNA, DNA amplification using GenomiPhi (GE Healthcare, LittleChalfont, UK)was applied. The PCR amplification reactionswere carriedout on a GeneAmp PCR system (Applied Biosystems, Foster City, CA,USA). PCR products were sequenced on a MiSeq instrument (Illumina,San Diego, USA). The bioinformatics pipeline included the CLC Geno-mics Workbench 6.0.2 (CLC Bio, Aarhus, Denmark) followed by an in-house developed software package and tools for variant interpretation.The obtained sequenceswere comparedwith theCOL11A1 and COL11A2reference sequences NM_001854.3 and NM_080680.2, respectively.Single nucleotide variants and small insertions and deletions werecalled using the quality-based variant calling in the CLCGenomicswork-bench (CLC Bio).

WES was performed in parallel for two affected individuals. Exomecapture was carried out using the TruSeq Exome Enrichment Kit(Illumina) and sequencing was performed on the Illumina HiSeq 2000platformwith paired-end 100-bp reads. The CLC Genomics Workbenchv6.0.4 (CLC Bio) software was used for duplicate read removal, readmapping against the human genome reference sequence (NCBI,GRCh37.p5/hg19), coverage analysis and variant calling and annotation.Data analysis was restricted to variants in the coding regions of COL2A1,COL11A1 and COL11A2, excluding the identification of incidentalfindings.

2.3. Confirmation with Sanger sequencing

Assays lacking sufficient sequencing depth following NGS (b20×coverage) were verified by Sanger sequencing. Primer sequences areavailable on request.

Please cite this article as: F.R. Acke, et al., Novel pathogenic COL11A1/COL11sequencing, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgm

All likely pathogenic variants identified with NGS (T-NGS as well asWES) were confirmed with Sanger sequencing of the involved exon/flanking sequence. DNA samples of relatives of patients inwhoma path-ogenic variant was found, were tested for this variant by means ofSanger sequencing as well, if available.

2.4. Variant reporting

Causality of the variants was based on previous reporting in litera-ture, in type XI collagen mutation databases (LOVD, HGMD, Ensembl),or in the 1000 genome project [25]. For all novel variants, causalitywas assessed using in silico prediction software (Alamut, InteractiveBiosoftware, Rouen, France; Splice Site Prediction by Neural Network,http://www.fruitfly.org/seq_tools/splice.html) and segregation of thevariant in the family when samples were available. No additional splic-ing analyseswere performed because of a lack of patients' tissue, such asskin fibroblasts.

3. Results

3.1. Screening strategy

In total, 48 patients were included in the study. The initial T-NGSsequencing for the most likely gene, based on the presence or absenceof ocular signs, included 43 COL11A1 (Stickler type 2) and eightCOL11A2 (Stickler type 3) analyses (Fig. 1, three patients were initiallytested for both genes due to unclear ocular findings). Patients negativefor the most likely gene were consequently sequenced for the othergene. The two patients analyzed by both T-NGS and WES had a type 2Stickler phenotype. Detailed clinical information is presented inTable 1 and Table S1.

3.2. Next-generation sequencing analysis

The overall mean coverage depth in our initial T-NGS approach was669× for COL11A1 and 346× for COL11A2, for 43 and eight patientsrespectively. Minimal sequencing depth was set at 20×. For COL11A1,1% of the total number of exons in all patients scored below the targeteddepth, while this was only 0.6% for COL11A2 (Table S2). Those specificexons were re-analyzed by Sanger sequencing.

TheWES analysis (2 patients) generated 8.5 and 12.7 Gb of uniquelymapped sequence data, respectively, with 94.5% and 96.8% of thetargeted exonic regions covered at least 20×. When we focus on thetype II and type XI collagen genes, 93.4% of COL2A1, 99.2% of COL11A1and 100% of COL11A2 on average were covered sufficiently (≥20) forboth individuals using the TruSeq Exome enrichment kit.

3.3. Variant detection and interpretation

Wedetected 14 COL11A1 and two COL11A2 pathogenic variantswiththe T-NGS technique, together with six variants of unknown signifi-cance (VUS) (Table 2). In addition, all likely non-pathogenic exonicand intronic (within 20 bp of the exon/intron boundary) variants inCOL11A1 and COL11A2 identified by the initial T-NGS analysis are listedin Table S3.

For the WES approach, we could confirm the pathogenic COL11A1variant in both patients (Table 2), and all 17 non-pathogenic COL11A1variants that were also detected by T-NGS. Moreover, all non-pathogenic COL2A1 variants previously detected by Sanger sequencingin these individuals, could be confirmed. The WES analysis of theCOL11A2 gene did not reveal any additional pathogenic variant inthese patients.

All 14 pathogenic COL11A1 variants, six of which are novel, are locat-ed within the region encoding the helical domain. Nine of these aresplice site alterations, four are missense variants and one is a nonsensevariant. Of the nine splice site alterations, seven are located in the fully

A2 variants in Stickler syndrome detected by targetedNGS and exomee.2014.09.001

Fig. 1. Inclusion scheme of patients, divided for the different analyses based on the presenting phenotype.

3F.R. Acke et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx

conserved GT or AG consensus splice site sequences (positions +/−1and +/−2), essential for correct splicing. The intron 34 variantc.2754+5GNA has previously been shown to result in skipping ofexon 35 [16] andwas shown to segregatewith thedisease in the currentstudy. The intron 50 variant c.3816+2dupT has also been reportedbefore as a de novo mutation [8]. Three COL11A1 missense variants,leading to glycine substitutions located in the helical domain, are pre-dicted to result in misfolding of the helix [26]. We also detected anarginine-to-cysteine substitution p.(Arg1076Cys) in the helical domainof the α1(XI) collagen chain. Cysteine residues are normally not foundin the triple helical domain of fibrillar collagens and their introductionis supposed to result in abnormal disulfide bond formation alteringhelix integrity [27], a mechanism which has already been proven to bedisease-causing in other fibrillar collagen genes [28,29]. Finally, onenonsense variant p.(Gln561*) resulting in a premature stopcodon, wasidentified. This variant is expected to result in nonsense-mediatedmRNA decay given its location in exon 15.

The two pathogenic COL11A2 variants include one glycine substitu-tion and one non-glycine substitution, both in the helical domain.There is no doubt about the disease-causing nature of the glycine substi-tution p.(Gly1489Ser) destabilizing theα-helical structure [26]whereasthe p.(Arg539Trp) amino acid change has previously been reported [30].

The VUS include three COL11A1 splice site alterations and threeCOL11A2 variants (Table 2). The COL11A1 variant c.3816+5GNC(found in two probands) is suggestive for impaired splicing based onsplice site prediction programs, and is located in intron 50 in which dif-ferent pathogenic variants have already been detected. The remainingCOL11A1 splice site alteration is the silent variant p.(Lys597Lys), locatedat the very end of exon 17 and predicted to be disease-causing by insilico software as well by familial segregation (Fig. 2). The COL11A2splice site alterations reported as VUS are the silent variantsp.(Gly1258Gly) and p.(Pro1422Pro), predicted to create a novel splicedonor and acceptor site respectively. Moreover, a missense variantp.(Pro1534Ser) was also reported to be of unknown significance(replacement of a highly conserved amino acid, Grantham distance 74).

In the initial T-NGS approach, we analyzed the COL11A1 or COL11A2gene depending on the presence or absence of ocular symptoms,respectively. If no pathogenic variant was identified, the other genewas also investigated, however no additional pathogenic variantswere identified. All families with a pathogenic COL11A1 variant had aclear ocular phenotype, except for one with inconclusive ocular exami-nation due to young age (proband S42). In none of theCOL11A2-positivepatients, an ocular phenotype was observed whereas hearing loss wasreported in all these families.

Please cite this article as: F.R. Acke, et al., Novel pathogenic COL11A1/COL11sequencing, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgm

4. Discussion

Weperformed targetedNGS for type XI collagen genes in 48unrelat-ed probands with a clinical diagnosis suggestive for a type 2 or type 3Stickler syndrome inwhom a COL2A1mutation had previously been ex-cluded. Pathogenic variants could be detected in 16 patients, 14 in theCOL11A1 gene and two in the COL11A2 gene. In addition, six variantsof unknown significance have been identified, which are predicted tobe disease-causing although functional analysis could not be performed.

The detected pathogenic COL11A1 variants are predominantly splicesite alterations and glycine substitutions across the entire α-helical do-main, consistent with previously reported COL11A1 variants in Sticklersyndrome. Mainly based on previous findings, pathogenic splice site al-terations result in in-frame exon skipping and thus exert a dominantnegative effect [15]. However, we identified the first pathogenicCOL11A1 nonsense mutation in Stickler syndrome, suggesting that alsohaploinsufficiency of COL11A1 might result in a Stickler phenotype. Ofinterest, our results suggest that the exon 50 donor splice site ofCOL11A1 is a mutational hot spot as three out of the 14 pathogenicvariants and two VUS are located in this specific region. Mutations inthis region have been associated with Marshall syndrome, which isdistinguished from Stickler syndromebased on slightly diverse dysmor-phic features [8,11]. However, clinical distinction is subtle and the allelicMarshall and type 2 Stickler syndrome therefore seem to be slightlydiverse variants of the same entity. All pathogenic COL11A1 variantswere detected in patients with an ocular phenotype, which is in agree-ment with previous findings [3,18]. In contrast, no eye symptoms werefound in the Stickler patients with a pathogenic COL11A2 variant ofwhichwe found two in our cohort, a glycine and a non-glycine substitu-tion. No other symptom seems capable to differentiate between type 2and type 3 Stickler syndrome. Whereas hearing impairment in type 1Stickler syndrome is mainly a mild-to-moderate high-frequency senso-rineural hearing loss, the type 2 and type 3 forms show a more pro-nounced hearing loss affecting all frequencies [4].

Despite the significant expansion of the COL11A1 and COL11A2 mu-tation spectrum, no pathogenic variant or VUS could be identified inmore than half of the patients from our selected population (26/48,54%). Nevertheless, these patients do exhibit a phenotype suggestivefor Stickler syndrome. Several explanations might be hypothesized.First, assays capable of detecting (multi-)exon deletions or insertions,such asMLPA [17], have not been performed in this study. For this latterapproach, high quality DNA is necessary which was often not available.Second, deep intronic variants which may alter normal splicing [16],were not studied. Third, the involvement of other, perhaps still uniden-tified genes for Stickler syndrome cannot be excluded. Mutations in the

A2 variants in Stickler syndrome detected by targetedNGS and exomee.2014.09.001

Table 1Overview of included patients exhibiting a type 2 Stickler phenotype (1–40), inconclusive phenotype (41–43) and type 3 Stickler phenotype (44–48), with their most prominent clinicalsigns (not exhaustive).

Samplenumber

Highmyopia

Vitreousanomaly

Hearingloss

Palataldefect

Midfacialhypoplasia/micrognathia

Skeletalfeatures

Familial Pathogenic variantor VUS

1 + + (type I) + − + + − No2 − + (type II) + − − − No3 + + (type II) + + + + + (ocular/orofacial/skeletal) COL11A14 + + + (ocular/auditory/orofacial) COL11A15 + + (type II) + − + + + (ocular/auditory/orofacial/skeletal) COL11A16 + + + + + (ocular) No7 + − + − + + − No8 + + (type II) + + + + + (ocular) COL11A19 + + + + + − No10 + + + − − + (ocular/auditory/orofacial) No11 − + − + − + (ocular/auditory/skeletal) COL11A112 + + + (ocular/auditory/orofacial/skeletal) COL11A1 (VUS)13 + (type II) + + + (ocular/orofacial) No14 + + + + (ocular/auditory/orofacial) No15 + + − − + (ocular) No16 + + (type II) − − + (ocular) COL11A117 + + (type II) No18 + + (type II) − − + + − No19 + + + + + No20 + + + + − + (ocular/auditory/orofacial/skeletal) COL11A121 + + No22 + + (type II) − − − + No23 + + + + No24 + + + + (ocular/auditory/orofacial) COL11A1 (VUS)25 + − − + + + (ocular/auditory) No26 + + − − + No27 + + + + + − COL11A128 + + (type II) − − − No29 + + + − + − No30 + + (type II) − − + − COL11A131 + + (type I) + + + − − COL11A1 (VUS)32 + + (type II) − + + + (orofacial) No33 + − + + + (ocular/auditory/orofacial/skeletal) COL11A134 + + + (ocular/auditory) COL11A135 + − + − + + No36 + + + + COL11A137 + + + COL11A138 + + + + + (orofacial/skeletal) No39 + − − + (ocular/auditory/orofacial) No40 − + (type II) + + − + (ocular) No41 − + − + + − No42 + + + − COL11A143 + + COL11A2 (VUS)44 − − + + + − − COL11A2 (VUS)45 − − + + + − + (orofacial) COL11A2 (VUS)46 − − + + + + (orofacial) No47 − − + − − + + (skeletal) COL11A248 − − − + + + + (auditory/orofacial) COL11A2

4 F.R. Acke et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx

genes encoding type IX collagen have been identified in a few familiesexhibiting an autosomal recessive form of Stickler syndrome withclinical features partly distinct from the three main Stickler types[31–33]. Moreover, considerable phenotypical overlap between Sticklersyndrome and other connective tissue dysplasias, such as Marfan,Ehlers–Danlos and Loeys–Dietz syndrome, exists.

The T-NGS and WES approaches showed relatively high read depthand were fully consistent with Sanger sequencing regarding variant de-tection. At the same time these NGS approaches are cheaper and fastercompared to the conventional Sangermethod. As the clinicalmanifesta-tions of Stickler syndrome are strongly indicative but not absolutelyconclusive for a specific gene (e.g. membranous vitreous in Stickler pa-tients with a pathogenic COL11A1 variant) [11], T-NGS analysis for apanel of Stickler genes might become the gold standard [34], whereasWES could be performed in case T-NGS fails to demonstrate thedisease-causing mutation. WES enables to investigate other genes

Please cite this article as: F.R. Acke, et al., Novel pathogenic COL11A1/COL11sequencing, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgm

throughout the exome, which might be relevant given the genetic het-erogeneity and differential diagnosis of Stickler syndrome asmentionedabove. However, theWES approach is still more complex in view of dataanalysis, is less cost-effective compared to T-NGS, and needs an exten-sive genetic counseling concerning incidental findings in genes not re-lated to the phenotype [35].

In summary,we expand the number of COL(XI)mutations leading toStickler syndrome, and present novel types of pathogenic variants,thereby broadening the mutational spectrum of collagen XI genes.With this study, we confirm that meticulous clinical evaluation,especially based on the vitreous phenotype and the presence of consid-erable hearing loss is useful in order to select the responsible gene. Nev-ertheless, for young patients in whom the phenotype is not entirelyexpressed yet or for patients for whom limited clinical information isavailable, NGS applications, particularly targeted NGS, are valuablealternatives in the genetic evaluation of Stickler syndrome and will

A2 variants in Stickler syndrome detected by targetedNGS and exomee.2014.09.001

Table 2Identified pathogenic variants as well as variants of unknown significance.

Sample Gene Technique Nucleotide change Type of variant Exon number Reference

37 COL11A1 T-NGS c.1603GNT Missense p.(Gly535Cys) 14 Novel3 COL11A1 T-NGS c.1630-2delA Splice site alteration 15 Martin et al. [10]

Annunen et al. [8]Richards et al. [16]

5 COL11A1 T-NGS c.1630-2delA Splice site alteration 15 Martin et al. [10]Annunen et al. [8]Richards et al. [16]

30 COL11A1 T-NGS + WES c.1681CNT Nonsense p.(Gln561*) 15 Ensembl4 COL11A1 T-NGS c.1900-1GNA Splice site alteration 20 Richards et al. [16]11 COL11A1 T-NGS c.1973GNT Missense p.(Gly658Val) 21 Novel33 COL11A1 T-NGS c.2754+5GNA Splice site alteration 35 Richards et al. [16]34 COL11A1 T-NGS + WES c.3226CNT Missense p.(Arg1076Cys) 42 Novel16 COL11A1 T-NGS c.3647GNC Missense p.(Gly1216Ala) 47 Novel36 COL11A1 T-NGS c.3762+1GNA Splice site alteration 49 Novel20 COL11A1 T-NGS c.3816+1GNA Splice site alteration 50 Griffith et al. [9]27 COL11A1 T-NGS c.3816+1GNA Splice site alteration 50 Griffith et al. [9]42 COL11A1 T-NGS c.3816+2dupT Splice site alteration 50 Annunen et al. [8]8 COL11A1 T-NGS c.4554+1GNA Splice site alteration 61 Novel47 COL11A2 T-NGS c.1615CNT Missense p.(Arg539Trp) 17 Jakkula et al. [30]48 COL11A2 T-NGS c.4465GNA Missense p.(Gly1489Ser) 62 Novel

12 COL11A1 T-NGS c.1791GNA Silent p.(Lys597Lys)Aberrant splice donor site

17 Novel, VUS

24 COL11A1 T-NGS c.3816+5GNC Splice site alteration 50 Novel, VUS31 COL11A1 T-NGS c.3816+5GNC Splice site alteration 50 Novel, VUS45 COL11A2 T-NGS c.3774CNT Silent p.(Gly1258Gly)

Novel splice donor site51 Novel, VUS

44 COL11A2 T-NGS c.4266GNA Silent p.(Pro1422Pro)Novel splice acceptor site

59 Novel, VUS

43 COL11A2 T-NGS c.4600CNT Missense p.(Pro1534Ser) 63 Novel, VUS

5F.R. Acke et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx

probably become the gold standard because of their efficiency and cost-effectiveness.

Conflict of interest

The authors declare that they have no conflict of interest.

exon 17 IVS17

exon 17 IVS17 exon 17 I

Fig. 2. Pedigree of proband no. 12 (arrow), in which a COL11A1 VUS (c.1791GNA, last nucleotrepresent the presence of a Stickler phenotype, while white symbols represent a normal phen

Please cite this article as: F.R. Acke, et al., Novel pathogenic COL11A1/COL11sequencing, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgm

Acknowledgments

FA holds a Ph.D. fellowship of the Research Foundation Flanders(FWO Vlaanderen), Belgium, grant number 11C4914N. FM and OMVare Senior Clinical Investigators also supported by the Research Founda-tion Flanders (FWO Vlaanderen), Belgium, grant numbers 1842313N

exon 17 IVS17

exon 17 IVS17

exon 17 IVS17VS17

ide of exon 17) has been found, probably influencing the donor splice site. Black symbolsotype. Nucleotide peaks are shown for the evaluated family members.

A2 variants in Stickler syndrome detected by targetedNGS and exomee.2014.09.001

6 F.R. Acke et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx

and 1861714N. This work was also supported by a Methusalem grant toADP from the Ghent University, Belgium, grant number 08/01M01108.We thank Dr. Patrick Willems for the critical review. We are grateful toMrs. Charlotte Opsomer, Mrs. Inge Vereecke and colleagues from theConnective Tissue Lab of the Center for Medical Genetics Ghent fortheir technical assistance.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.ymgme.2014.09.001.

References

[1] G.B. Stickler, et al., Hereditary progressive arthro-ophthalmopathy, Mayo Clin. Proc.40 (1965) 433–455.

[2] P.S. Rose, et al., Stickler syndrome: clinical characteristics and diagnostic criteria,Am. J. Med. Genet. A 138A (3) (2005) 199–207.

[3] M.P. Snead, J.R. Yates, Clinical and molecular genetics of Stickler syndrome, J. Med.Genet. 36 (5) (1999) 353–359.

[4] F.R. Acke, et al., Hearing impairment in Stickler syndrome: a systematic review,Orphanet J. Rare Dis. 7 (2012) 84.

[5] N.N. Ahmad, et al., Stop codon in the procollagen II gene (COL2A1) in a family withthe Stickler syndrome (arthro-ophthalmopathy), Proc. Natl. Acad. Sci. U. S. A. 88(15) (1991) 6624–6627.

[6] K.P. Hoornaert, et al., Stickler syndrome caused by COL2A1 mutations: genotype–phenotype correlation in a series of 100 patients, Eur. J. Hum. Genet. 18 (8)(2010) 872–880.

[7] A.J. Richards, et al., A family with Stickler syndrome type 2 has a mutation in theCOL11A1 gene resulting in the substitution of glycine 97 by valine in alpha 1 (XI)collagen, Hum. Mol. Genet. 5 (9) (1996) 1339–1343.

[8] S. Annunen, et al., Splicing mutations of 54-bp exons in the COL11A1 gene causeMarshall syndrome, but other mutations cause overlapping Marshall/Sticklerphenotypes, Am. J. Hum. Genet. 65 (4) (1999) 974–983.

[9] A.J. Griffith, et al., Marshall syndrome associated with a splicing defect at theCOL11A1 locus, Am. J. Hum. Genet. 62 (4) (1998) 816–823.

[10] S. Martin, et al., Stickler syndrome: further mutations in COL11A1 and evidence foradditional locus heterogeneity, Eur. J. Hum. Genet. 7 (7) (1999) 807–814.

[11] M. Majava, et al., A report on 10 new patients with heterozygous mutations in theCOL11A1 gene and a review of genotype–phenotype correlations in type XIcollagenopathies, Am. J. Med. Genet. A 143 (3) (2007) 258–264.

[12] M. Melkoniemi, et al., Collagen XI sequence variations in nonsyndromic cleft palate,Robin sequence and micrognathia, Eur. J. Hum. Genet. 11 (3) (2003) 265–270.

[13] A.V. Poulson, et al., Clinical features of type 2 Stickler syndrome, J. Med. Genet. 41(8) (2004) e107.

[14] L. Ala-Kokko, A.L. Shanske, Mosaicism in Marshall syndrome, Am. J. Med. Genet. A149A (6) (2009) 1327–1330.

[15] A.J. Richards, et al., Stickler syndrome and the vitreous phenotype: mutations inCOL2A1 and COL11A1, Hum. Mutat. 31 (6) (2010) E1461–E1471.

Please cite this article as: F.R. Acke, et al., Novel pathogenic COL11A1/COL11sequencing, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgm

[16] A.J. Richards, et al., Splicing analysis of unclassified variants in COL2A1 and COL11A1identifies deep intronic pathogenic mutations, Eur. J. Hum. Genet. 20 (5) (2012)552–558.

[17] R. Vijzelaar, et al., Deletions within COL11A1 in Type 2 stickler syndrome detectedby multiplex ligation-dependent probe amplification (MLPA), BMC Med. Genet. 14(2013) 48.

[18] M. Vikkula, et al., Autosomal dominant and recessive osteochondrodysplasiasassociated with the COL11A2 locus, Cell 80 (3) (1995) 431–437.

[19] T. Avcin, et al., Early-onset osteoarthritis due to otospondylomegaepiphyseal dyspla-sia in a family with a novel splicing mutation of the COL11A2 gene, J. Rheumatol. 35(5) (2008) 920–926.

[20] T. Pihlajamaa, et al., Heterozygous glycine substitution in the COL11A2 gene in theoriginal patient with the Weissenbacher–Zweymuller syndrome demonstrates itsidentity with heterozygous OSMED (nonocular Stickler syndrome), Am. J. Med.Genet. 80 (2) (1998) 115–120.

[21] D.A. Sirko-Osadsa, et al., Stickler syndrome without eye involvement is caused bymutations in COL11A2, the gene encoding the alpha2(XI) chain of type XI collagen,J. Pediatr. 132 (2) (1998) 368–371.

[22] E. van Beelen, et al., Audiometric characteristics of two Dutch families withnon-ocular Stickler syndrome (COL11A2), Hear. Res. 291 (1–2) (2012) 15–23.

[23] M.M. Vuoristo, et al., A stop codon mutation in COL11A2 induces exon skipping andleads to non-ocular Stickler syndrome, Am. J. Med. Genet. A 130A (2) (2004)160–164.

[24] A.N. Desai, A. Jere, Next-generation sequencing: ready for the clinics? Clin. Genet. 81(6) (2012) 503–510.

[25] G.R. Abecasis, et al., An integrated map of genetic variation from 1,092 humangenomes, Nature 491 (7422) (2012) 56–65.

[26] C. Kielty, M. Grant, The collagen family: structure, assembly, and organization in theextracellular matrix, in: P.M. Royce, B. Steinmann (Eds.), Connective Tissue and ItsHeritable Disorders, Wiley-Liss, New York, 2002, pp. 159–222.

[27] P. Kannu, J. Bateman, R. Savarirayan, Clinical phenotypes associated with type IIcollagen mutations, J. Paediatr. Child Health 48 (2) (2012) E38–E43.

[28] K.P. Hoornaert, et al., The phenotypic spectrum in patients with arginine to cysteinemutations in the COL2A1 gene, J. Med. Genet. 43 (5) (2006) 406–413.

[29] F. Malfait, et al., Three arginine to cysteine substitutions in the pro-alpha (I)-collagen chain cause Ehlers–Danlos syndrome with a propensity to arterial rupturein early adulthood, Hum. Mutat. 28 (4) (2007) 387–395.

[30] E. Jakkula, et al., The role of sequence variations within the genes encoding collagenII, IX and XI in non-syndromic, early-onset osteoarthritis, Osteoarthr. Cartil. 13 (6)(2005) 497–507.

[31] G. Van Camp, et al., A new autosomal recessive form of Stickler syndrome is causedby a mutation in the COL9A1 gene, Am. J. Hum. Genet. 79 (3) (2006) 449–457.

[32] S. Baker, et al., A loss of function mutation in the COL9A2 gene causesautosomal recessive Stickler syndrome, Am. J. Med. Genet. A 155A (7) (2011)1668–1672.

[33] F. Faletra, et al., Autosomal recessive stickler syndrome due to a loss of functionmutation in the COL9A3 gene, Am. J. Med. Genet. A 164A (1) (2014) 42–47.

[34] B. Sikkema-Raddatz, et al., Targeted next-generation sequencing can replace Sangersequencing in clinical diagnostics, Hum. Mutat. 34 (7) (2013) 1035–1042.

[35] G.M. Christenhusz, K. Devriendt, K. Dierickx, Disclosing incidental findings in genet-ics contexts: a review of the empirical ethical research, Eur. J. Med. Genet. 56 (10)(2013) 529–540.

A2 variants in Stickler syndrome detected by targetedNGS and exomee.2014.09.001