Genes Genom DOI 10.1007/s13258-016-0508-1

Online ISSN 2092-9293 Print ISSN 1976-9571

RESEARCH ARTICLE

Whole-exome sequencing in Tricho-rhino-phalangeal syndrome (TRPS) type I in a Korean family Byulee Yoon1,2,3 · Yun-Ji Kim1,2,3 · Seung-Yeol Son4 · Kyudong Han1,2,3 · Byung Cheol Park5 

Received: 14 November 2016 / Accepted: 15 December 2016 © The Genetics Society of Korea and Springer-Science and Media 2016

Abstract Tricho-rhino-phalangeal syndrome (TRPS) is a rare autosomal dominant and monogenic disease. Among three types of TRPS, it is known that TRPS type I and type III are caused by deletions or substitutions in the TRPS1 gene, located on chromosome 8 (8q23.3). Although the mutations in TRPS1 gene are responsible for human TRPS, some cases are not detected by the mutations of TRPS1 gene and several cases are presented with different genetic variations. The present case was a sporadic and without TRPS1 mutation. Therefore, we performed whole-exome sequencing (WES) with one patient and his family (father, mother, and brother) and validated novel mutations using PCR and Sanger sequencing. Through family-based WES, we found the two de novo mutations such as ZNF 134 and EXD 3 genes. Through functional effect prediction using disease association Ensembl database, we propose that the Electronic supplementary material The online version of this article (doi:10.1007/s13258-016-0508-1) contains supplementary material, which is available to authorized users. * Kyudong Han [email protected] * Byung Cheol Park [email protected] 1

Department of Nanobiomedical Science, Dankook University, Cheonan 31116, Republic of Korea

2

BK21 PLUS NBM Global Research Center for Regenerative, Dankook University, Cheonan 31116, Republic of Korea

3

DKU-Theragen Institute for NGS Analysis (DTiNa), Cheonan, Republic of Korea

4

Department of Microbiology, College of Natural Science, Dankook University, Cheonan 31116, Republic of Korea

5

Department of Dermatology, College of Medicine, Dankook University, Cheonan 31116, Republic of Korea

de novo mutation of ZNF134 (p.Ser207Arg) could be one of potential candidate genes for causing TRPS and develope the TRPS phenotype in the present case. Keywords TRPS1 · Next generation sequencing · Whole-exome sequencing · ZNF134 · EXD3 · Tricho-rhinophalangeal syndrome Abbreviations TRPS Tricho-rhino-phalangeal syndrome NGS Next-generation sequencing WES Whole-exome sequencing EXD3 Exonuclease 3′–5′ domain containing 3 ZNF134 Zinc finger protein 134 SNVs Single-nucleotide variants InDels Insertions/deletions UTR Untranslated region

Introduction Tricho-rhino-phalangeal syndrome (TRPS) is a rare genetic syndrome caused by mutations in the TRPS1 gene that is affiliated with the GATA DNA-binding zinc (Momeni et al. 2000). TRPS is characterized by clinical features such as sparse, slow-growing scalp hair, and craniofacial dysmorphism, and it can be divided into three types: TRPS type I (OMIM #190,350), TRPS type II (OMIM #150,230), and TRPS type III (OMIM #190,351) (Momeni et  al. 2000). TRPS types I and III are known to be monogenic disorders that are caused by mutations in the TRPS1 gene and to have autosomal dominant inheritance (Kobayashi et  al. 2002). On the other hand, it was reported that TRPS was caused without a genetic variation in the TRPS1 gene and no family history was found. Unfortunately, genetic mutations

13

Vol.:(0123456789)

Genes Genom

have not been reported or found so far in cases which had no genetic mutation in the TRPS 1 gene (Chen et al. 2010; Ludecke et al. 2001). Recently, next-generation sequencing (NGS) technologies have been used in the analysis of human genetic diseases and they are making a rapid progress (Cheon et  al. 2014; Jin et al. 2014; Ku et al. 2011). Among them, wholeexome sequencing (WES) is a cost-effective, convenient and sensitive method for mutation detection and it captures most of the phenotype-altering mutations. Through WES, we captured all of the protein coding genes and analyzed them through computational analysis by comparing with the human reference genome. In this study, we suggest that other candidate genes besides those in the TRPS 1 gene can be postulated to cause TRPS through WES analysis in a Korean TRPS patient and his family who did not have genetic variation in TRPS 1.

Fig. 2 Pedigree of TRPS patient’s family. I-1 father, I-2 mother, II-1 and II-3 siblings, II-2 TRPS patient. Father and TRPS patient show sparse hairs (asterisk). The patient’s brothers and parents has no definite clinical features of TRPS

Ion proton exome capture and sequencing

Materials and methods Patient and pedigree analysis A 15-years-old Korean boy visited the Department of Dermatology, Dankook University Hospital and he had sparse and slow growing scalp hair since his childhood (BASP classification M2V1). He continued to lose his hair over 6  months (BASP Classification U1 or N–H Classification Va). He had bulbous nose, long philtrum, thin upper lip, and abnormally short fingers and toes (Fig. 1). In terms of family history (Fig.  2), his father (I:1) had male pattern baldness (BASP classification V2,) but his mother (I:2), elder (II:1) and younger (II:3) brothers were normal healthy persons. We obtained peripheral blood samples from the patient and his family. The institutional review board of Dankook hospital approved this study (DKUH 2014-08-005).

Genomic DNAs (gDNAs) from the patient (II:2) and his family (I:1, I:2, and II:3) were isolated from blood samples using gDNA purification kit (GeneAll, Seoul, Korea). The gDNA (50  ng) with 5X Ion Ampliseq™ HiFi Mix and premixed primer pool were used for target amplification according to the instructions of the manufacturer of the Ion AmpliSeq Exome Kit (Life Technologies, Victoria, Australia). The PCR amplicons were performed to partially digest primer sequences and were ligated to IonP1 adapter and Ion Xpress™ Barcode adapter. Adapter-ligated libraries were purified using Agencourt® AMPure® XP reagent. The quantity and the quality of the libraries were assessed using a Library Quantification Kit (Life Technologies, Victoria, Australia) and an Agilent High Sensitivity DNA Kit (Agilent Technologies), respectively. In addition, emulsion PCR was performed using the OneTouch 2 instrument (Life Technologies, Victoria, Australia) with an Ion PI Template OT2 200 Kit v2 according to the manufacturer’s

Fig. 1 Photograph of the TRPS patient. He has a sparse and slowly growing scalp hairs since his childhood (BASP classification M2V1). He has bulbous nose, long philtrum, thin upper lip (a), and abnormally short fingers (b) and toes (c)

13

Genes Genom

instructions. The enrichment of template-positive Ion Sphere Particles (ISP) in the Ion PI chip was achieved using the Ion OneTouch ES enrichment system (Life Technologies, Victoria, Australia). After being purified from the ISP, the Ion PI chip was prepared and loaded according to the manufacturer’s recommendations. Raw sequence data were processed using the Ion Torrent platform-specific pipeline software (Torrent Suite v.4.2). These include read alignment to the human genome reference (hg19), a targetregion coverage analysis, variant calling and poor signal reads, and PCR duplicate removal. Illumina HiSeq exome capture and sequencing The gDNA from the patient (II:2) was randomly fragmented using Covaris S2 (Covaris, Woburn, MA, USA) and analyzed for a peak between 200 and 300  bp by the Agilent Bioanalyzer 2100 (Agilent Technologies, Palo Alto, USA). DNA fragments were used for adaptor ligation and were purified by AMPure XP beads (Beckman Coulter, Inc., Brea, CA, USA). DNA was selected and it was extracted to the size 300–400 bp on a 2% agarose gel. DNA was denatured, hybridized with RNA probes overnight and captured via streptavidin beads. Next, the DNA was eluted from the probe. DNA was purified again using the AMPure XP Beads and analyzed by the Agilent Bioanalyzer 2100 within the size of 300–400 bp. The exome-enriched libraries were transferred to Illumina HiSeq2500 (Illumina, San Diego, CA, USA) and sequenced 100  bp paired-end. Raw reads in FASTQ format from exome sequencing were aligned to the hg19 human reference genome from UCSC with BWA (v. 0.7.12). PCR duplicates were removed using Picard (v. 1.98), and variant calling and detection of single-nucleotide variants and detection of insertions/deletions were performed using the Genome Analysis Toolkit (GKTK). Concordant variants from Ion proton and Illumina Hiseq The single-nucleotide variants (SNVs) and insertions/deletions (InDels) were annotated by the genomic region, such as the 5′ untranslated region (UTR), the 3′ UTR, intron, exon, and splicing site from ion proton data analysis (I:1, I:2,II:2, and II:3) and Illumina Hiseq data analysis (II:2) using SnpEff (v.4.1). SNVs within the coding sequence were annotated as synonymous, missense, nonsense, and InDels within coding sequences were annotated as inframe or frameshift using dbNSFP (v. 2.9). Multi-calling variants were processed from each annotated variant from ion proton data (I:1, I:2, II:2, and II:3) and later de novo variants (heterozygous variants) in the affected subject (II:2) were collected. Among the de novo variants, we applied

the following criteria. (1) Variants were detected simultaneously with Illumina Hiseq and Ion proton. (2) Among them, non-synonymous variants were selected. PCR validation and Sanger sequencing We conducted PCR amplification and Sanger sequencing to validate a total of 7 candidate loci of 5 genes and to analyze the entire exon sequence of the TRPS1 gene. We obtained the TRPS1 gene sequence from the UCSC Genome Browser Gateway (http://genome.ucsc.edu/cgi-bin/hgGateway). A primer pair (Supplementary Tables 2, 3) for PCR amplification was designed in target sequences using OligoCalc (http://www.basic.northwestern.edu/biotools/oligocalc.html) and Oligo Analysis Tools (http://www.operon. com/tools/oligo-analysis-tool.aspx). The PCR was performed in a total volume of 20  ul of the reaction mixture including 10 ul of 2X EF-Taq Premix4 (Biofact; Daejeon, Korea), 10 uM of oligo nucleotide primers, 1 ul of template and nuclease-free water. The PCR amplification conditions were as follows: first denaturation step of 3  min at 95 °C, followed by 35 cycles of 30 s at 95 °C, annealing for 30 s at the optimal temperature, and extension for 1 min at 72 °C, and final extension for 2  min at 72 °C. The PCR product was confirmed by 1% agarose gel electrophoresis and purified by PCR purification kit (Favorgen, Pingtung country, Taiwan) and then, we performed DNA sequencing using ABI3500 Genetic analyzer (Thermo Fisher Scientific, Foster City, USA). The DNA sequencing data of candidate loci was compared with exome sequencing data, and the TRPS1 gene sequencing data was compared with human reference sequence (RefSeq #NM_014112.4) through multiple alignment by Bioedit (v. 7.2.5) program.

Results Sanger sequencing analysis of the TRPS1 gene We screened mutations of the TRPS1 gene in the affected individual (II:2) and the unaffected individuals (I:1, I:2, II:1, and II:3) using Sanger sequencing. We excluded exon1 and part of exon7 with the UTRs in the seventh coding exon, and designed 10 primer pair sets (Supplementary Table 3). After Sanger sequencing analysis, we did not find mutations in the TRPS1 gene in our patient and his family. Whole-exome sequencing (WES) in a Korean TRPS family To detect de novo mutations, we performed the exome sequencing in the affected individual (II:2) and three unaffected individuals (I:1, I:2, and II:3) because we did not

13

Genes Genom

find any mutations in the TRPS1 gene by Sanger sequencing. First, we conducted WES of four individuals (I:1, I:2, II:2, and II:3) by Ion proton as single-end of ~174 bp. We obtained an average 6.3  Gbp exome sequencing data that was mapped ~92% on the target region (Table 1). After filtering, we compared the exome sequencing data from the affected individual to those from one unaffected sibling and unaffected parents according to extraction variants of exon, splicing site, and promoter region that could explain the relation of the disease in the patient. In addition, we selected de novo variants from only the patient. We also performed Illumina Hiseq in the patient (II:2) and generated a total of 8.9 Gbp exome sequencing data as pair-end 100 bp reads. In addition, the variants were detected simultaneously with Ion proton and Illumina Hiseq, and nonsynonymous variants were collected (see “Materials and Table 1 Summary statistics for whole-exome sequencing

Ion proton

Total bases (Gbp) Total number of reads Total number of mapped reads Percentage of mapped reads (%) Total number target region read Percentage target region reads (%) Coverage 5× rate (%) Coverage 10× rate (%) Coverage 20× rate (%)

Fig. 3 Sanger sequencing validation of candidate variants identified through WES in TRPS patient’s family. Two variants in EXD3 (Exonuclease 3′–5′ Domain Containing 3) and ZNF134 (Zinc Finger Protein 134) genes were validated by Sanger sequencing. Heterozygous mutation denotes overlapping peaks

13

methods” section). After applying the data, we obtained 7 loci of 5 genes (ZNF717, SLC39A4, EXD3, SRRM5, and ZNF134) as candidate variants. All candidates appeared to be heterozygous variants that showed missense mutations in the affected individual (II:2) (Supplementary Table  1). A total of seven candidate loci were validated in the patient (II:2) and his family (I:1, I:2, II:1, and II:3) by Sanger sequencing. We could identify two candidate genes of 7 loci as disease causing variants not only in the patient (II:2) but also in his father (I:1): EXD3 p.Pro626Leu (NM_017820.4, c.1877C>T) and ZNF134 p.Ser207Arg (NM_003435.3, c619A>C). Sanger sequencing results were differently detected than those of WES in his father that were validated as homozygous with thymine in EXD3 and heterozygous in ZNF134 that showed the same peak as in the patient (Fig. 3). In addition, we determined the allele

Illumina Hiseq

Sample I:1

Sample I:2

Sample II:3 Sample II:2 Sample II:2

5.6 32,077,897 31,768,242 99.03 30,145,263 93.98 96.25 94.04 89.28

6.6 38,705,203 38,094,778 98.42 35,527,851 91.79 96.94 95.31 91.99

7 40,383,843 39,673,600 98.24 37,254,852 92.25 97.1 95.6 92.58

6 33,509,052 33,091,019 98.75 31,195,563 93.1 96.44 94.62 90.77

8.9 88,703,466 86,327,173 99.51 64,819,431 74.72 99.41 98.53 95.76

Genes Genom

frequency of two candidate variants on the basis of NCBI dbSNP database (http://www.ncbi.nlm.nih.gov/projects/ SNP/), and both variants showed minor allele frequencies (Table 2). Functional effect prediction Variations in EXD3 and ZNF134 were non-synonymous mutations causing the amino acid change. We conducted functional effect prediction of these variants on disease association using Ensembl database (http://asia.ensembl. org/index.html; hg19). Among the variants, p.Ser207Arg in the ZNF134 gene showed more than 5 % allele frequency as well as Polymorphism Phenotyping v2 (Polyphen2) and Sorting Intolerant from Tolerant (SIFT) results indicated that the variant were possibly damaging and deleterious (Table 2). These results indicate that ZNF134 p.Ser207Arg could be more associated with the development of diseases than EXD3 p.Pro626Leu.

Discussion TRPS1 (OMIM #604,386) is located on 8q23.3 with seven exons and encodes a protein that is affiliated with nine zincfinger motifs of four different types including C2H2-type zinc finger, GATA-type zinc finger, and two C-terminal Ikaros-like zinc finger motifs (Piccione et al. 2009). Previous studies reported that most of the TRPS type I patients have nonsense, missense or in-frame splice site mutations in the TRPS1 gene (Chen et al. 2010; Hatamura et al. 2001; Ludecke et al. 2001; Momeni et al. 2000; Nan et al. 2013; Zhou et al. 2013). In the present study, we tried to confirm the genetic mutations in the TRPS1 syndrome through familial phenotypic and genotypic analyses. We analyzed the entire coding exons of TRPS1, which was more frequently associated with the syndrome (Kobayashi et  al. 2002; Momeni et  al. 2000) by Sanger sequencing in the patient and his family. However, all seven exons of the TRPS1 gene in all family members did not contain any variants on Sanger sequencing and WES. In the previous investigation, some patients

with TRPS phenotype including facial and various skeletal abnormalities did not have TRPS1 gene mutations (Chen et  al. 2010; Ludecke et  al. 2001). However, the TRPS is known to be a monogenic mutation disease. Ludecke and colleagues analyzed the mutations in the TRPS1 gene of patients with TRPS type I or TRPS type III. The study concluded that in 50 out of the 57 unrelated cases, mutations in the TRPS1 gene were identified, and in seven cases, genetic mutations were not identified (Ludecke et al. 2001). In addition, Chen et al. investigated the mutations in the TRPS1 gene of Asian patients with TRPS phenotype, and it was found that two out of the eight cases showed an autosomal dominant inheritance pattern but the other cases showed a sporadic pattern, and furthermore, three patients did not have mutations in the TRPS1 gene(Chen et al. 2010). Hence, there is a possibility that TRPS could develop sporadically or without genetic mutations in the TRPS1 gene although it is an autosomal dominant monogenic disease. The present case might be a sporadic case without TRPS1 gene mutation; hence, we tried to find other mutations which could cause the TRPS using WES technology In all reported cases, detection of genetic mutations in TRPS cases was only through the Sanger sequencing method. Therefore, we applied the WES in the patient (II:2) and family members (I:1, I:2, and II:3) using two kinds of WES platforms and tried to find genetic variations besides the use of Sanger sequencing in all family members. We obtained variants simultaneously detected by Ion proton and Illumina Hiseq and the candidate variants were confirmed by Sanger sequencing validation in the entire family. As a result, we could find two significant positions, which were EXD3 (p.Pro626Leu) and ZNF134 (p.Ser207Arg) (Table 2) from the other protein coding genes. As shown in Fig.  3, the Sanger sequencing peaks of EXD3 and ZNF134 genes were identified as variants in both the father and patient. The EXD3 gene encodes 3′–5′exonuclease domain with exonuclease Mut-7 homolog in Caenorhabditis elegans, which was transposon silencing (Ketting et al. 1999) that makes a protein with homology to RNaseD. However, the EXD3 was predicted as ‘Benign’ in Ensembl database analysis of the variant effect. The ZNF

Table 2 Genotypic features of mutation Location Gene

chr. 9 chr. 19

Coding DNA change

EXD3 c. 1877C>T ZNF134 c. 619A>C

Amino acid change

Minor Allele Prediction of variant effectb frequency(MAF)a PolyPhen 2 score PolyPhen 2 result SIFT score SIFT result

p.Pro626Leu p.Ser207Arg

A = 0.0118/59 A = 0.0605/303

0.227 0.758

Benign Possibly damaging

0.09 0.02

Tolerated Deleterious

ªFrequency of Single Nucleotide Polymorphisms: http://www.ncbi.nlm.nih.gov/projects/SNP/ b

Predicted variant effect from Ensembl database: http://asia.ensembl.org/index.html

13

Genes Genom

134 gene, which is a zinc finger protein, includes transcription factor activity and Cys2-His2 (C2H2) zinc finger domains.(Tommerup and Vissing 1995) The ZNF134 gene was predicted to affect protein function and it was ‘Possibly Damaging and Deleterious’ by Ensembl database (Table 2). Taken together, we suggest that ZNF134 mutation can have an adverse effect on the disease. We predicted the putative transcription factor binding site in promoter regions of EXD3 and ZNF134 using Transfac database (http://transfac.gbf.de/TRANSFAC/), and detected the binding sites of GATA family, GATA-1, -2, and -4 (data not shown). In addition, we tried to find the functional relationships between TRPS1 and the predicted related genes by the GeneMANIA prediction server (http://www.genemania. org). The Supplementary Fig.  1 (yellow line) shows sharing of protein domains of GATA zinc finger family with TRPS1 (Supplementary Fig. 1). GATA zinc finger regulates specific transcription activity in TRPS1 which includes GATA DNA-binding site, and missense mutation in the GATA motif leads to the TRPS phenotype.(Piccione et al. 2009) Therefore, we expected that potential disruption of the GATA zinc finger motifs of TRPS1 could be induced by the two variants. In summary, it can be postulated that the de novo mutation in ZNF134 (p.Ser207Arg) might be a potential candidate gene that causes TRPS and development of the TRPS phenotype in the present case based on our WES analysis, Sanger validation, and functional effect prediction. Acknowledgements This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF- 2015R1D1A1A02059462). Compliance with ethical standards Conflict of interest Byulee Yoon declares that she has no conflict of interest. Yun-Ji Kim declares that she has no conflict of interest. SeungYeol Son declares that he has no conflict of interest. Kyudong Han declares that he has no conflict of interest. Byung Cheol Park declares that he has no conflict of interest. Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

13

References Chen LH, Ning CC, Chao SC (2010) Mutations in TRPS1 gene in trichorhinophalangeal syndrome type I in Asian patients. Br J Dermatol 163:416–419 Cheon CK, Sohn YB, Ko JM, Lee YJ, Song JS, Moon JW, Yang BK, Ha IS, Bae EJ, Jin HS et al (2014) Identification of KMT2D and KDM6A mutations by exome sequencing in Korean patients with Kabuki syndrome. J Hum Genet 59:321–325 Hatamura I, Kanauchi Y, Takahara M, Fujiwara M, Muragaki Y, Ooshima A, Ogino T (2001) A nonsense mutation in TRPS1 in a Japanese family with tricho-rhino-phalangeal syndrome type I. Clin Genet 59:366–367 Jin HS, Lee JB, Kim K, Lee KY, Choi VN, Kim JS, Jeong SY, Yim SY (2014) Identification of the rare compound heterozygous variants in the NEB gene in a Korean family with intellectual disability, epilepsy and early-childhood-onset generalized muscle weakness. J Hum Genet 59:643–647 Ketting RF, Haverkamp TH, van Luenen HG, Plasterk RH (1999) Mut-7 of C. elegans, required for transposon silencing and RNA interference, is a homolog of Werner syndrome helicase and RNaseD. Cell 99:133–141 Kobayashi H, Hino M, Shimodahira M, Iwakura T, Ishihara T, Ikekubo K, Ogawa Y, Nakao K, Kurahachi H (2002) Missense mutation of TRPS1 in a family of tricho-rhino-phalangeal syndrome type III. Am J Med Genet 107:26–29 Ku CS, Naidoo N, Pawitan Y (2011) Revisiting Mendelian disorders through exome sequencing. Hum Genet 129:351–370 Ludecke HJ, Schaper J, Meinecke P, Momeni P, Gross S, von Holtum D, Hirche H, Abramowicz MJ, Albrecht B, Apacik C et al (2001) Genotypic and phenotypic spectrum in tricho-rhino-phalangeal syndrome types I and III. Am J Hum Genet 68:81–91 Momeni P, Glockner G, Schmidt O, von Holtum D, Albrecht B, Gillessen-Kaesbach G, Hennekam R, Meinecke P, Zabel B, Rosenthal A et  al (2000) Mutations in a new gene, encoding a zinc-finger protein, cause tricho-rhino-phalangeal syndrome type I. Nat Genet 24:71–74 Nan X, Dai S, Li CT, Chen XR, Zhao HS, Zhang FS, Song QH (2013) Novel deletion mutation of TRPS1 gene in a Chinese patient of trichorhinophalangeal syndrome type I. Gene 523:88–91 Piccione M, Niceta M, Antona V, Di Fiore A, Cariola F, Gentile M, Corsello G (2009) Identification of two new mutations in TRPS 1 gene leading to the tricho-rhino-phalangeal syndrome type I and III. Am J Med Genet A 149A:1837–1841 Tommerup N, Vissing H (1995) Isolation and fine mapping of 16 novel human zinc finger-encoding cDNAs identify putative candidate genes for developmental and malignant disorders. Genomics 27:259–264 Zhou S, Jin D, Song Y (2013) Identification of a new mutation in TRPS1 gene leading to tricho-rhino-phalangeal syndrome I in a Chinese patient. J Dermatol 40:277–278