Hum Genet (2015) 134:181–190 DOI 10.1007/s00439-014-1509-2 ORIGINAL INVESTIGATION Characterization of ANKRD11 mutations in humans and mice related to KBG syndrome Katherina Walz · Devon Cohen · Paul M. Neilsen · Joseph Foster II. · Francesco Brancati · Korcan Demir · Richard Fisher · Michelle Moffat · Nienke E. Verbeek · Kathrine Bjørgo · Adriana Lo Castro · Paolo Curatolo · Giuseppe Novelli · Clemer Abad · Cao Lei · Lily Zhang · Oscar Diaz-Horta · Juan I. Young · David F. Callen · Mustafa Tekin Received: 10 October 2014 / Accepted: 9 November 2014 / Published online: 21 November 2014 © Springer-Verlag Berlin Heidelberg 2014 Abstract Mutations in ANKRD11 have recently been the degradation of the protein. Analysis of 11 pathogenic reported to cause KBG syndrome, an autosomal domi- ANKRD11 variants in humans, including six reported in this Yod/ nant condition characterized by intellectual disability (ID), study, and one reported in the Ankrd11 + mouse, shows behavioral problems, and macrodontia. To understand the that all mutations affect the C-terminal regions and that the pathogenic mechanism that relates ANKRD11 mutations mutant proteins accumulate aberrantly. In silico analysis with the phenotype of KBG syndrome, we studied the cellu- shows the presence of D-box sequences that are signals for lar characteristics of wild-type ANKRD11 and the effects of proteasome degradation. We suggest that ANKRD11 C-ter- mutations in humans and mice. We show that the abundance minus plays an important role in regulating the abundance of wild-type ANKRD11 is tightly regulated during the cell of the protein, and a disturbance of the protein abundance cycle, and that the ANKRD11 C-terminus is required for due to the mutations leads to KBG syndrome. K. Walz and D. Cohen contributed equally to this work. Introduction Electronic supplementary material The online version of this article (doi:10.1007/s00439-014-1509-2) contains supplementary Heterozygous mutations in ANKRD11 (MIM611192), material, which is available to authorized users. encoding ankyrin repeat domain 11, have recently been K. Walz (*) · D. Cohen · J. Foster II. · C. Abad · C. Lei · F. Brancati · G. Novelli L. Zhang · O. Diaz-Horta · J. I. Young · M. Tekin (*) Medical Genetics Unit, Policlinico Tor Vergata University Dr. John T. Macdonald Foundation Department of Human Hospital, Viale Oxford 81, 00133 Rome, Italy Genetics and John P. Hussman Institute for Human Genomics, Miller School of Medicine, University of Miami, 1501 NW 10 K. Demir Ave, BRB 610, M-860, Miami, FL 33136, USA Division of Pediatric Endocrinology, Dokuz Eylül University e-mail: [email protected] Faculty of Medicine, ˙Izmir 35340, Turkey M. Tekin R. Fisher e-mail: [email protected] Northern Genetics Service Teesside Genetics Unit, The James Cook University Hospital, Marton Road, Middlesbrough TS4 K. Walz 3BW, UK Department of Medicine, Miller School of Medicine, University of Miami, 1501 NW 10 Ave, BRB 418, M-860, M. Moffat Miami, FL 33136, USA Department of Paediatric Dentistry, Newcastle Dental Hospital and School, Richardson Road, Newcastle upon Tyne NE2 4AZ, P. M. Neilsen · D. F. Callen UK Swinburne University of Technology Sarawak Campus, Kuching, Sarawak, Malaysia N. E. Verbeek Department of Medical Genetics, University Medical Center F. Brancati Utrecht, Lundlaan 6, 3584 EA Utrecht, The Netherlands Department of Medical, Oral and Biotechnological Sciences, Gabriele D’Annunzio University, 66100 Chieti, Italy 1 3 182 Hum Genet (2015) 134:181–190 shown to cause KBG syndrome (MIM148050), charac- fragile X, and microarray studies showed normal results in terized by intellectual disability (ID), behavioral abnor- the affected individuals. The institutional review board at malities, macrodontia of upper central incisors, and skel- the University of Miami approved this study. All participat- etal anomalies (Brancati et al. 2006; Herrmann et al. 1975; ing individuals provided written informed consent. Skjei et al. 2007; Sirmaci et al. 2011). Heterozygous dele- tions at 16q24.3 that contain the ANKRD11 gene were Animals also described in patients with ID, autistic spectrum dis- Yod/ order, and in some cases with brain anomalies, a clinical Ankrd11 + or C3H.Cg-Ankrd11 <Yod>/H (EM:00380) phenotype overlapping with KBG syndrome (Marshall congenic on a C3H/HeH genetic background was obtained et al. 2008; Willemsen et al. 2010; Handrigan et al. 2013; from the European Mouse Mutant Archive EMMA. The Isrie et al. 2012; Sacharow et al. 2012; Khalifa et al. 2013; colony was expanded at the University of Miami. At Yod/ Miyatake et al. 2013). weaning age, Ankrd11 + and wild-type littermates were ANKRD11 contains two transcriptional repression housed 2–4 per cage in a room with a 12-h light/dark cycle domains at the N and C termini, respectively, plus a tran- with access to food and water ad libitum. All procedures scriptional activation domain (Zhang et al. 2007). It were approved by the University of Miami Institutional interacts with the p160 coactivator and nuclear receptor Animal Care and followed the NIH Guidelines, “Using complex and recruits histone deacetylases to modify tran- Animals in Intramural Research”. scriptional activation (Zhang et al. 2004). ANKRD11 local- izes mainly to the nuclei of neurons and accumulates in Sanger sequencing discrete inclusions when neurons are depolarized (Sirmaci et al. 2011). Thus, it is not surprising that ANKRD11 muta- Mutations in ANKRD11 in patients were screened via tions lead to a neurobehavioral phenotype (Willemsen et al. conventional capillary sequencing. Once a mutation 2010; Handrigan et al. 2013; Sacharow et al. 2012; Spen- was identified in the proband, available parents were gler et al. 2013). However, how ANKRD11 mutations act at tested for the same variant. Primers were designed to the molecular level remains unknown. amplify the coding exons and respective flanking intronic The mouse ortholog Ankrd11 has a 79 % identity at regions, along with the 5′ UTR using Primer3 v. 0.4.0 the amino acid level with human ANKRD11. A chemi- (http://frodo.wi.mit.edu/) and are available upon request. cally induced (ENU) mutagenesis screen generated the Mouse DNA fragments were amplified using primers Yod/ Ankrd11 + allele, a missense mutation at a highly con- F5′-CTGAAGGAGCTGTTCAAGCAA-3′, and R5′- served residue (E2502K) (Barbaric et al. 2008). The CCTGGCTGCCCGGCAATGAA-3′. 20 ng of genomic Yod/ Ankrd11 + or Yoda mouse was reported before the identi- DNA was used in conjunction with Betaine (Sigma) fication of ANKRD11 mutations in KBG syndrome. and Platinum Taq (Life Technologies) in each PCR reac- In this study, to understand the pathogenic mechanisms tion. A touchdown protocol was used to amplify target underlying the KBG syndrome, we screened ANKRD11 in DNA regions. Amplicons were visualized on 2 % agarose additional patients for the documentation of mutation spec- gels, and cleanup was conducted using Sephadex. Patient trum, studied the cellular characteristics of the wild-type DNA sequencing was obtained using Big Dye Termina- ANKRD11 protein, and analyzed pathogenic variants in tor Cycle Sequencing V3.1 Ready Reaction Kit, and ABI humans and Yoda mice. PRISM 3130/3730 DNA analyzers (Applied Biosys- tems). Sequence traces were analyzed with Sequencher 4.7 Software (Gene Codes Corporation). NM_013275.5 Materials and methods was used as reference for nucleotide and codon number- ing in humans. In silico analysis to assess disruption of the Patients protein function was performed using Polyphen2 (http:// genetics.bwh.harvard.edu/pph2/index.shtml) and Mutation Each patient was evaluated by an experienced clinical genet- Taster (http://www.mutationtaster.org/). icist to exclude other syndromic forms of ID. Chromosome, RT-PCR K. Bjørgo Department of Medical Genetics, Oslo University Hospital, Human whole blood samples were collected from the Kirkeveien 166, 0450 Oslo, Norway child heterozygous for an ANKRD11 mutation, unaf- fected parents, and a control. Total RNA was extracted A. Lo Castro · P. Curatolo Department of Neuroscience, Pediatric Neurology and Psychiatry using PAXgene Blood RNA Kit (Qiagen), and cDNA was Unit, Tor Vergata University of Rome, 00165 Rome, Italy synthesized using Superscript III reverse transcriptase 1 3 Hum Genet (2015) 134:181–190 183 (Invitrogen). Nested PCR was performed to amplify released from the nocodazole block through three media a 295-bp region followed by a 218-bp region. Sanger washes and cultured in fresh media for the indicated time sequencing was used to confirm the presence of the muta- points. Cells were subsequently fixed and permeabilized tion in the RNA. Primers were designed to quantify total for immunofluorescence as previously described (Neilsen ANKRD11 mRNA (F5′-CAGCCCAGTGGACACAAC-3′ et al. 2008). Cells were incubated with a mouse anti-myc and R5′-CTCTCCACGCTCGTTTCTCT-3′) in the human antibody (Kumar et al. 2005) in 5 % donkey serum (over- family heterozygous for a truncating mutation. Addi- night, 4 °C) with subsequent incubation with Alexa Fluor tional primer pairs were designed to hybridize directly on 488-conjugated anti-mouse antibody (Molecular Probes, the mutation site, where the last two nucleotides on the Eugene, OR) in 5 % donkey serum (1 h, RT). Chomosomes/ forward primer matched either the mutant or wild-type DNA-rich regions were visualized through initial treatment allele. Primers designed to anneal to the wild-type DNA with RNase A (1 µg/mL, 37 °C, 30 min) and subsequent were F5′-AAGCCAGTGAGGAAGAGGC-3′ and R5′-