ORIGINAL CONTRIBUTION Autosomal Dominant Familial Dyskinesia and Facial Myokymia Single Exome Sequencing Identifies a Mutation in Adenylyl Cyclase 5 Ying-Zhang Chen, MD, PhD; Mark M. Matsushita, BS; Peggy Robertson, PhD; Mark Rieder, PhD; Santhosh Girirajan, MBBS, PhD; Francesca Antonacci, PhD; Hillary Lipe, MN, ARNP; Evan E. Eichler, PhD; Deborah A. Nickerson, PhD; Thomas D. Bird, MD; Wendy H. Raskind, MD, PhD Background: Familial dyskinesia with facial myoky- Results: The exome contained 23 428 single-nucleotide mia (FDFM) is an autosomal dominant disorder that is ex- variants, of which 9391 were missense, nonsense, or splice acerbated by anxiety. In a 5-generation family of German site alterations. The critical region contained 323 variants, ancestry, we previously mapped FDFM to chromosome 5 of which were not present in 1 of the sequence databases. band 3p21-3q21. The 72.5-Mb linkage region was too large Adenylyl cyclase 5 (ADCY5) was the only gene in which for traditional positional mutation identification. the variant (c.2176GϾA) was co-transmitted perfectly with disease status and was not present in 3510 control white Objective: To identify the gene responsible for FDFM exomes. This residue is highly conserved, and the change by exome resequencing of a single affected individual. is nonconservative and predicted to be damaging. Participants: We performed whole exome sequencing Conclusions: ADCY5 is highly expressed in striatum. in 1 affected individual and used a series of bioinformatic Mice deficient in Adcy5 develop a movement disorder that filters, including functional significance and presence in is worsened by stress. We conclude that FDFM likely re- dbSNP or the 1000 Genomes Project, to reduce the num- sults from a missense mutation in ADCY5. This study dem- ber of candidate variants. Co-segregation analysis was per- onstrates the power of a single exome sequence com- formed in 15 additional individuals in 3 generations. bined with linkage information to identify causative genes for rare autosomal dominant mendelian diseases. Main Outcome Measures: Unique DNA variants in the linkage region that co-segregate with FDFM. Arch Neurol. 2012;69(5):630-635 N 2001, FERNANDEZ ET AL1 DE- tion with myokymia or chorea and 2 re- scribed a white family of German gions that contain single or clustered ion ancestry manifesting a disorder channel genes.1 We mapped the disorder characterized by predominantly to a 71.73-cM region, corresponding to perioral and periorbital myoky- 72.5 Mb, on chromosome band 3p21- miaI and face, neck, and upper limb dystonic/ 3q21.3 Interrogation of such a large re- choreic movements (familial dyskinesia gion, deemed by National Center for Bio- Author Affiliations: with facial myokymia [FDFM]; OMIM technology Information Entrez Gene Build Author Affil Departments of Medicine 606703). Initially, the disorder had been de- 37 to contain 253 annotated genes, be- Departments (Medical Genetics) (Drs Chen, (Medical Ge Bird, and Raskind and scribed as familial essential (benign) cho- came feasible by the recent development Bird, and Ra 2 Mr Matsushita), Genome rea. The onset of symptoms ranged from of “next-generation” massively parallel se- Mr Matsushi Sciences (Drs Robertson, 2 1⁄2 to 19 years. Initially paroxysmal and quencing technologies. We provide evi- Sciences (Dr Rieder, Girirajan, Antonacci, precipitated or worsened by stress, the dys- dence that adenylyl cyclase 5 (ADCY5)is Rieder, Girir Eichler, and Nickerson), kinetic episodes become nearly constant by a strong candidate gene for FDFM and Eichler, and Neurology (Ms Lipe and the end of the third decade of life, but in demonstrate the power of exome sequenc- Neurology (M Dr Bird), and Psychiatry and Dr Bird), and Behavioral Sciences some individuals, the dyskinetic episodes ing to identify causative genes for auto- Behavioral S (Dr Raskind), University of may diminish in frequency and severity at somal dominant disorders. (Dr Raskind Washington, Seattle; Howard older ages. Autosomal dominant transmis- Washington, Hughes Medical Institute, sion of FDFM is supported by the pres- METHODS Hughes Med Chevy Chase, Maryland ence of disease in all 5 generations, appar- Chevy Chase (Dr Eichler); and Veterans ently complete penetrance in males and (Dr Eichler) Integrated Service Network 20 HUMAN SUBJECTS Integrated Se Mental Illness Research, females, and instances of male-to-male Mental Illnes Education, and Clinical Center, transmission. An updated pedigree is shown in Figure 1. Education, a Department of Veterans Affairs, By targeted genotyping, we excluded 10 Patients were evaluated and blood samples were Department Seattle (Drs Bird and Raskind). candidate genes chosen for their associa- collected from 16 individuals in 3 generations Seattle (Drs ARCH NEUROL / VOL 69 (NO. 5), MAY 2012 WWW.ARCHNEUROL.COM 630 ©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 I II 2 2 6 7 III 1 2 3 4 5 6 7 8 9 10 IV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 V 1 2 3 4 5 6 7 Male Affected Deceased Female Unaffected Individual whose exome Individuals who Monozygotic twins was sequenced provided DNA Figure 1. Family with autosomal dominant familial dyskinesia with facial myokymia. under a protocol approved by the Institutional Review Board were selected based on nonoverlap with variants present in of the University of Washington. dbSNP131 (http://www.ncbi.nlm.nih.gov/SNP), the 1000 Ge- nomes Project (http://www.1000genomes.org/page.php), and TARGETED CAPTURE AND EXOME SEQUENCING 3510 control exomes of European ancestry from the UW Ge- nome Sciences Genomic Resource Center Exome Sequencing Genomic DNA was extracted from peripheral blood using stan- Projects. The UW Exome Variant Server (http://evs.gs.washington dard procedures. Five micrograms of DNA from an affected .edu/EVS/) catalogs variants identified through exome sequenc- member of the FDFM family (V-3 in Figure 1) was used to con- ing primarily of individuals with heart, lung, or blood disor- struct a shotgun sequencing library as described previously.4 ders. (3) Location in the linkage region. Variants in the 72.5-Mb The shotgun library was hybridized to NimbleGen microar- region on chromosome band 3p21-3q21 bounded by D3S1582 rays (Nimblegen_solution_V2refseq2010.HG19) for target en- and D3S3606 were further considered. (4) Conservation scores richment, followed by washing, elution, and additional ampli- as estimated by University of California Santa Cruz phastCons fication. We targeted all protein-coding regions as defined by and Genomic Evolutionary Rate Profiling. phastCons scores are RefSeq 36.3. Exome sequencing was performed using the Illu- based on sequence conservation in 17 vertebrate species.5 The mina GAIIx platform (Illumina, Inc) with paired-end 76–base score from 0 to 1 denotes the probability that each nucleotide pair (bp) reads. Sequence alignment and variant calling were belongs to a conserved element. Genomic Evolutionary Rate performed against the reference human genome (National Cen- Profiling scores range from −11.6 to 5.86 and are based on se- ter for Biotechnology Information 37/hg19). For single- quence conservation in 46 mammals.6 (5) Potential effect of nucleotide polymorphisms (SNPs) and short insertions or de- an amino acid substitution on protein structure and function letions (indels), calling and genotyping were performed using as predicted by SIFT (Sorting Intolerant From Tolerant; the GATK Unified Genotyper. Annotations of variants were http://sift.jcvi.org/www/SIFT_intersect_coding_submit.html)7 based on the National Center for Biotechnology Information and PolyPhen-2 (Polymorphism Phenotyping v2; (http:// and University of California Santa Cruz (http://genome.cse genetics.bwh.harvard.edu/pph2/).8,9 .ucsc.edu/cgi-bin/hgGateway) databases using an in-house server (SeattleSeq Annotation 131; http://gvs.gs.washington.edu VARIANT VALIDATION BY SANGER SEQUENCING /SeattleSeqAnnotation/). Capillary sequencing using customized primers was per- RANKING OF CANDIDATE GENES/VARIANTS formed to confirm 5 unique coding variants in the linkage re- gion and to investigate co-transmission with the FDFM phe- Candidate genes/variants were ranked on the basis of the fol- notype in all family members. Primer sequences are given in lowing: (1) Coding region sequence effect. Missense, nonsense, eTable 1 (http://www.archneurol.com). Genomic DNA was poly- coding indels, and splice acceptor and donor sites were selected merase chain reaction amplified in a thermal cycler (DNA En- for further evaluation. (2) Variant frequency. Unique variants gine Tetrad 2; MJ Research). Sequencing was performed using ARCH NEUROL / VOL 69 (NO. 5), MAY 2012 WWW.ARCHNEUROL.COM 631 ©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 able inferior wall defect. The participant provided infor- Table 1. Variants Detected in the Exome of Individual V-3 mation that 4 of his neurologically affected relatives had also experienced CHF. The age at onset of CHF and the 72.5-Mb age at and cause of death of his grandfather (II-6) were Whole Chromosome 3 Variant Exome Linkage Region unknown. His father (III-7) had undergone coronary ar- tery bypass grafting and died of CHF at age 83 years, but Single-nucleotide variants, No. 23 428 323 2 uncles developed CHF at much earlier ages, 1 dying at Synonymous 10 500 130 Coding-notMod3a 240 0 age 63 years (III-3) and the other in his 50s (III-5). In- Intergenic 837 13 dividual IV-11 died of cancer at age 62 years. Records on Intron 1963 40 these individuals were unavailable for review. Near gene (5Ј and 3Ј)983 5Јand 3Ј UTR 399 4 ARRAY COMPARATIVE GENOMIC Nonsense 88 2 HYBRIDIZATION Splice site (5Ј and 3Ј)291 Missense 9274 130 Indels, No.b 414 9 To evaluate the possibility of a large CNV, array com- parative genomic hybridization was performed on DNA Abbreviations: indels, short insertions or deletions; UTR, untranslated from individuals IV-8 and V-2.