Volume 5, Number 3, June 2019 Neurology.org/NG

A peer-reviewed clinical and translational neurology open access journal

ARTICLE HACE1 defi ciency leads to structural and functional neurodevelopmental defects e330

ARTICLE Novel pathogenic XK mutations in McLeod syndrome and interaction between XK protein and chorein e328

ARTICLE HTT haplogroups in Finnish patients with Huntington disease e334

ARTICLE Brain-derived neurotrophic factor, epigenetics in stroke skeletal muscle, and exercise training e331 TABLE OF CONTENTS Volume 5, Number 3, June 2019 Neurology.org/NG

e331 Brain-derived neurotrophic factor, epigenetics in stroke skeletal muscle, and exercise training A.S.Ryan,H.Xu,F.M.Ivey,R.F.Macko,and C.E. Hafer-Macko Open Access

e332 Novel pathogenic VPS13A gene mutations in Japanese patients with chorea-acanthocytosis Y. Nishida, M. Nakamura, Y. Urata, K. Kasamo, H. Hiwatashi, I. Yokoyama, M. Mizobuchi, K. Sakurai, Y. Osaki, Y. Morita, M. Watanabe, K. Yoshida, K. Yamane, N. Miyakoshi, R. Okiyama, T. Ueda, N. Wakasugi, Y. Saitoh, T. Sakamoto, Y. Takahashi, K. Shibano, H. Tokuoka, A. Hara, K. Monma, K. Ogata, K. Kakuda, H. Mochizuki, T. Arai, M. Araki, T. Fujii, K. Tsukita, H. Sakamaki-Tsukita, and A. Sano Open Access

e334 HTT haplogroups in Finnish patients with Huntington disease S. Ylonen,¨ J.O.T. Sipil¨a, M. Hietala, and K. Majamaa Open Access

e335 Oligogenic basis of sporadic ALS: The example of SOD1 p.Ala90Val mutation L. Kuuluvainen, K. Kaivola, S. Monk¨ ¨are, H. Laaksovirta, M. Jokela, B. Udd, M. Valori, P. Pasanen, A. Paetau, B.J. Traynor, D.J. Stone, J. Schleutker, M. Poyh¨ onen,P.J.Tienari,and¨ L. Myllykangas Open Access

e336 Hybrid gel electrophoresis using skin fibroblasts to aid in diagnosing mitochondrial disease C. Newell, A. Khan, D. Sinasac, J. Shoffner, M.W. Friederich, Editorial J.L.K. Van Hove, S. Hume, J. Shearer, and I. Sosova Open Access e326 HACE1, RAC1, and what else in the pathogenesis of SPPRS? e337 Novel mutation in TNPO3 causes congenital H.-X. Deng limb-girdle myopathy with slow progression Open Access Companion article, e330 A. Vihola, J. Palmio, O. Danielsson, S. Penttil¨a, D. Louiselle, S. Pittman, C. Weihl, and B. Udd Articles Open Access fi e330 HACE1 de ciency leads to structural and functional e338 DMPK gene DNA methylation levels are neurodevelopmental defects associated with muscular and respiratory profiles V. Nagy, R. Hollstein, T.-P. Pai, M.K. Herde, P. Buphamalai, in DM1 P. Moeseneder, E. Lenartowicz, A. Kavirayani, G.C. Korenke, ´ ´ I. Kozieradzki, R. Nitsch, A. Cicvaric, F.J. Monje Quiroga, C. Legare, G. Overend, S.-P. Guay, D.G. Monckton, J. Mathieu, M.A. Deardorff, E.C. Bedoukian, Y. Li, G. Yigit, J. Menche, E.F. Perçin, C. Gagnon, and L. Bouchard B. Wollnik, C. Henneberger, F.J. Kaiser, and J.M. Penninger Open Access Open Access Editorial, e326 e328 Novel pathogenic XK mutations in McLeod syndrome Clinical/Scientific Notes and interaction between XK protein and chorein e302 A novel cathepsin D mutation in 2 siblings with late Y. Urata, M. Nakamura, N. Sasaki, N. Shiokawa, Y. Nishida, K. Arai, H. Hiwatashi, I. Yokoyama, S. Narumi, Y. Terayama, T. Murakami, infantile neuronal ceroid lipofuscinosis Y. Ugawa, H. Sakamoto, S. Kaneko, Y. Nakazawa, R. Yamasaki, J. Thottath, S.K. Vellarikkal, R. Jayarajan, A. Verma, M. Manamel, S. Sadashima, T. Sakai, H. Arai, and A. Sano A. Singh, V.R. Rajendran, S. Sivasubbu, and V. Scaria Open Access Open Access TABLE OF CONTENTS Volume 5, Number 3, June 2019 Neurology.org/NG

e327 X-linked myotubular myopathy and recurrent spontaneous pneumothorax: A new phenotype? P.-O. Carstens, E.M.C. Schwaibold, K. Schregel, C.D. Obermaier, A. Wrede, S. Zechel, S. Pauli, and J. Schmidt Open Access e329 First TMEM126A missense mutation in an Italian proband with optic atrophy and deafness C. La Morgia, L. Caporali, F. Tagliavini, F. Palombo, M. Carbonelli, R. Liguori, P. Barboni, and V. Carelli Open Access Cover image Neuropathologic findings of the autopsied patient. Plastic-embedded e333 Double somatic mosaicism in a child with Dravet sections from the dorsal spinal root show normal density of axons. syndrome See e335 A.M. Muir, C. King, A.L. Schneider, A.S. Buttar, I.E. Scheffer, L.G. Sadleir, and H.C. Mefford Open Access Academy Officers Neurology® is a registered trademark of the American Academy of Neurology (registration valid in the United States). James C. Stevens, MD, FAAN, President Neurology® Genetics (eISSN 2376-7839) is an open access journal published Orly Avitzur, MD, MBA, FAAN, President Elect online for the American Academy of Neurology, 201 Chicago Avenue, Ann H. Tilton, MD, FAAN, Vice President Minneapolis, MN 55415, by Wolters Kluwer Health, Inc. at 14700 Citicorp Drive, Bldg. 3, Hagerstown, MD 21742. Business offices are located at Two Carlayne E. Jackson, MD, FAAN, Secretary Commerce Square, 2001 Market Street, Philadelphia, PA 19103. Production offices are located at 351 West Camden Street, Baltimore, MD 21201-2436. Janis M. Miyasaki, MD, MEd, FRCPC, FAAN, Treasurer © 2019 American Academy of Neurology. Ralph L. Sacco, MD, MS, FAAN, Past President Neurology® Genetics is an official journal of the American Academy of Neurology. Journal website: Neurology.org/ng, AAN website: AAN.com Executive Office, American Academy of Neurology Copyright and Permission Information: Please go to the journal website (www.neurology.org/ng) and click the Permissions tab for the relevant Catherine M. Rydell, CAE article. Alternatively, send an email to [email protected]. Chief Executive Officer General information about permissions can be found here: https://shop.lww.com/ journal-permission. 20l Chicago Ave Disclaimer: Opinions expressed by the authors and advertisers are not Minneapolis, MN 55415 necessarily those of the American Academy of Neurology, its affiliates, or of the Publisher. The American Academy of Neurology, its affiliates, and the Tel: 612-928-6000 Publisher disclaim any liability to any party for the accuracy, completeness, efficacy, or availability of the material contained in this publication (including drug dosages) or for any damages arising out of the use Editorial Office or non-use of any of the material contained in this publication. Patricia K. Baskin, MS, Executive Editor Advertising Sales Representatives: Wolters Kluwer, 333 Seventh Avenue, Kathleen M. Pieper, Senior Managing Editor, Neurology New York, NY 10001. Contacts: Eileen Henry, tel: 732-778-2261, fax: 973-215- 2485, [email protected] and in Europe: Craig Silver, tel: +44 Lee Ann Kleffman, Managing Editor, Neurology® Genetics 7855 062 550 or e-mail: [email protected]. Sharon L. Quimby, Managing Editor, Neurology® Clinical Practice Careers & Events: Monique McLaughlin, Wolters Kluwer, Two Commerce fl Square, 2001 Market Street, Philadelphia, PA 19103, tel: 215-521-8468, fax: 215- Morgan S. Sorenson, Managing Editor, Neurology® Neuroimmunology & Neuroin ammation 521-8801; [email protected]. Andrea Rahkola, Production Editor, Neurology Reprints: Meredith Edelman, Commercial Reprint Sales, Wolters Kluwer, Two Robert J. Witherow, Senior Editorial Associate Commerce Square, 2001 Market Street, Philadelphia, PA 19103, tel: 215-356-2721; Karen Skaja, Senior Editorial Associate [email protected]; [email protected]. Special projects: US & Canada: Alan Moore, Wolters Kluwer, Two Kaitlyn Aman Ramm, Editorial Assistant Commerce Square, 2001 Market Street, Philadelphia, PA 19103, tel: Kristen Swendsrud, Editorial Assistant 215-521-8638, [email protected]. International: Andrew Wible, Senior Manager, Rights, Licensing, and Partnerships, Wolters Kluwer; Justin Daugherty, Editorial Assistant [email protected]. Madeleine Sendek, MPH, Editorial Assistant

Publisher Wolters Kluwer Baltimore, MD

Publishing Staff Kim Jansen, Executive Publisher Jessica Heise, Production Team Leader, Neurology Journals Megen Miller, Production Editor Steve Rose, Editorial Assistant Stacy Drossner, Production Associate A peer-reviewed clinical and translational neurology open access journal Neurology.org/NG

Neurology® Genetics

Editor Stefan M. Pulst, MD, Dr med, FAAN Vision Neurology®: Genetics will be the premier peer- reviewed journal in the field of neurogenetics. Deputy Editor Massimo Pandolfo, MD, FAAN Mission Neurology: Genetics will provide neurologists Associate Editors and clinical research scientists with Alexandra Durr, MD, PhD outstanding peer-reviewed articles, Margherita Milone, MD, PhD editorials, and reviews to elucidate the role Raymond P. Roos, MD, FAAN of genetic and epigenetic variations in Jeffery M. Vance, MD, PhD diseases and biological traits of the central Editorial Board and peripheral nervous systems. Hilary Coon, PhD Giovanni Coppola, MD ChantalDepondt, MD, PhD Editorial Tel: 612-928-6400 Brent L. Fogel, MD, PhD, FAAN Inquiries Toll-free: 800-957-3182 (US) AnthonyJ. Griswold, PhD Fax: 612-454-2748 Orhun H. Kantarci, MD [email protected] Julie R. Korenberg, PhD, MD Davide Pareyson, MD Shoji Tsuji, MD,PhD DinekeS. Verbeek,PhD Stay facebook.com/NeurologyGenetics Connected David Viskochil, MD,PhD twitter.com/greenjournal JulianeWinkelmann, MD Juan I. Young, PhD youtube.com/user/NeurologyJournal

Neurology® Journals

Editor-in-Chief Classification of Evidence Robert A. Gross, MD, PhD, FAAN Review Team Melissa J. Armstrong, MD Deputy Editor Richard L. Barbano, MD,PhD, FAAN Bradford B. Worrall, MD, MSc, FAAN RichardM.Dubinsky,MD,MPH,FAAN Jeffrey J. Fletcher, MD, MSc Section Editors Gary M. Franklin, MD, MPH, FAAN David S. Gloss II, MD,MPH&TM Biostatistics John J. Halperin, MD,FAAN Richard J. Kryscio, PhD Jason Lazarou, MSc, MD Sue Leurgans, PhD Steven R. Mess´e, MD, FAAN V. Shane Pankratz, PhD Pushpa Narayanaswami, MBBS, DM, FAAN fi Classi cation of Evidence Evaluations Alex Rae-Grant, MD Gary S. Gronseth, MD, FAAN

Podcasts Stacey L. Clardy, MD, PhD Jeffrey B. Ratliff, MD, Deputy Podcast Editor

Ombudsman David S. Knopman, MD, FAAN

Scientific Integrity Advisor Robert B. Daroff, MD, FAAN EDITORIAL OPEN ACCESS HACE1, RAC1, and what else in the pathogenesis of SPPRS?

Han-Xiang Deng, MD, PhD Correspondence Dr. Deng Neurol Genet 2019;5:e326. doi:10.1212/NXG.0000000000000326 [email protected]

Spastic paraplegia and psychomotor retardation with or without seizures (SPPRS) is a complex RELATED ARTICLE neurodevelopmental disorder with an autosomal recessive inheritance. SPPRS typically shows an infantile onset, starting with hypotonia either at birth or by age 3–4 months, followed by HACE1 deficiency leads to severely impaired global development and delayed early motor milestones.1 All patients with structural and functional SPPRS develop slowly progressive bilateral lower limb spasticity, leaving them wheelchair neurodevelopmental bound and bed bound by their 20s. In some cases, patients may never walk. Most patients defects develop seizures in childhood and have a speech delay. Other variable features include ocular Page e330 abnormalities, sensorineural hearing loss, skeletal abnormalities, obesity, and double in- continence. Some male patients have hypoplastic genitalia. Brain imaging may show generalized cerebral atrophy, ventricular dilatation, hypoplasia of the corpus callosum, and decreased white matter.1

By using family-based and unbiased genotype-driven whole-exome sequencing approaches, Hollstein et al. and Akawi et al. identified mutations of HACE1 in several patients with SPPRS.1,2 HACE1 encodes a HECT domain and ankyrin repeat-containing ubiquitin ligase (HACE1), which is involved in specific tagging of target proteins, leading to their subcellular localization or proteasomal degradation. Most HACE1 mutations in patients with SPPRS lead to a premature stop codon, suggesting that loss of HACE1 function causes SPPRS. However, the pathogenic mechanism remains largely unknown.

In this issue, Nagy et al. provide important information for understanding the pathogenic mechanism underlying SPPRS.3 They identified 2 novel homozygous truncating mutations in HACE1 in 3 patients from 2 families. More importantly, they performed detailed molecular and phenotypic analyses of Hace1 knockout mice and SPPRS patient fibroblasts. They showed several clinical features in the Hace1 knockout mice, which are similar to those observed in patients with SPPRS, including deficiencies in locomotion and learning/memory, enlarged ventricles, and hypoplastic corpus callosum. Pathologic and neurophysiologic studies dem- onstrated a reduced number of synaptic puncta and altered hippocampal synaptic transmission. The authors observed increased levels of active Rac1 in the Hace1 knockout mouse brain and SPPRS patient–derived fibroblasts. RAC1 is a small GTPase with diverse roles in signaling, and HACE1 targets RAC1 to the ubiquitin/proteasome system for degradation.4 Therefore, the authors hypothesize that upregulation of the RAC1 pathway may underlie the pathogenesis of SPPRS because of defective degradation of RAC1 by HACE1 deficiency. This is the first in vivo study to show a molecular pathway underlying SPPRS.

A total of 11 mutations in 17 SPPRS cases have been reported to date.1,2,5 Except for a single amino acid deletion (p.Leu832del), all the others are truncation mutations. Although these truncation mutations presumably have almost identical functional consequences, great varia- tions of clinical symptoms and disease severity were observed in these patients with SPPRS, suggesting that other genetic and environmental modifiers influence phenotype expression. It is known that the ankyrin repeats of HACE1 are responsible for substrate recognition, whereas the HECT domain is essential for ubiquitinylation. The p.Leu832del mutation is located in the

From the Ken and Ruth Davee Department of Neurology, Northwestern University Feinberg School of Medicine, Chicago, IL.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 HECT domain, suggesting that the loss of ubiquitinylation This suggest that HACE1 may have other substrates. In- activity, rather than the loss of the entire HACE1 protein, is deed, HACE1 also regulates other small GTPases, in- critical for development of SPPRS. cluding RAB11a, RAB6a, and RAB8a.13,14 It has also been reported that Hace1 promotes the stability of Nrf2 and Loss of HACE1 was initially noted in human malignancies, plays an important role in antioxidant response, and loss of and Hace1 knockout mice were shown to develop spontane- hace1 in a mouse model of Huntington disease accelerates ous, late-onset multiple tumors after age 1 year.6 The tumor motor deficits and exacerbates cognitive and psychiatric incidence was almost tenfold higher in the Hace1 knockout phenotypes.15 homozygotes than the heterozygotes in 2-year-old mice (12% vs 1.3%). Loss of Hace1 also rendered mice susceptible to The molecular mechanism by which increased RAC1 leads to second environmental and genetic hits for the development of the abnormal structure and function of synapses and the multiple cancers. This led to the hypothesis that HACE1 is pathogenic roles of other HACE1 regulated proteins in the a tumor suppressor gene, which prevents tumorigenesis by pathogenesis of SPPRS are still not understood. These issues suppressing cyclin D levels and reactive oxygen species remain to be addressed in future studies. generation.6,7 However, the neurodevelopmental phenotype and pathology in the Hace1 knockout mice have not been Author contributions comprehensively investigated until the present study. H.-X. Deng: drafting/revising the manusrcript.

The hypothesis that upregulation of the RAC1 pathway under- Study funding lies the pathogenesis of SPPRS is compatible with the previous No targeted funding reported. data. It is well known that RAC1 plays an essential role in development and structural plasticity of dendrites and dendritic Disclosure spines.8,9 Transgenic mice overexpressing constitutively active The author reports no disclosures. Full disclosures available: RAC1 in Purkinje neurons lead to ataxia and reduced Purkinje Neurology.org/NG. neuron axon terminals and smaller but increased number of dendritic spines.8 Recently, heterozygous missense mutations in References fi 1. Hollstein R, Parry DA, Nalbach L, et al. HACE1 deficiency causes an autosomal RAC1 were identi ed in developmental disorders with diverse – 10 recessive neurodevelopmental syndrome. J Med Genet 2015;52:797 803. phenotypes. Among 7 RAC1 mutations, p.Tyr64Asp appears 2. Akawi N, McRae J, Ansari M, et al. Discovery of four recessive developmental dis- to be constitutively active. The patient with this mutation orders using probabilistic genotype and phenotype matching among 4,125 families. Nat Genet 2015;47:1363–1369. showed some clinical features overlapping with those in SPPRS, 3. Nagy V, Hollstein R, Pai T-P, et al. HACE1 deficiency leads to structural and func- including severely impaired global development and delayed tional neurodevelopmental defects. Neurol Genet 2019;5:e330. doi: 10.1212/ NXG.0000000000000330. early motor milestones, hypoplastic corpus callosum and geni- 4. Torrino S, Visvikis O, Doye A, et al. The E3 ubiquitin-ligase HACE1 catalyzes the talia, ocular abnormalities, and sensorineural hearing loss. ubiquitylation of active Rac1. Dev Cell 2011;21:959–965. ff 5. Hariharan N, Ravi S, Pradeep BE, et al. A novel loss-of-function mutation in HACE1 is However, marked di erences were also observed. Notably, the linked to a genetic disorder in a patient from India. Hum Genome 2018;5:17061. patient with p.Tyr64Asp showed hypotonia soon after birth, but 6. Zhang L, Anglesio MS, O’Sullivan M, et al. The E3 ligase HACE1 is a critical chro- fi mosome 6q21 tumor suppressor involved in multiple cancers. Nat Med 2007;13: he did not seem to develop progressive spasticity, a speci c 1060–1069. 10 feature in SPPRS, even by age 12 years. This may suggest that 7. Daugaard M, Nitsch R, Razaghi B, et al. Hace1 controls ROS generation of vertebrate ff Rac1-dependent NADPH oxidase complexes. Nat Commun 2013;4:2180. upregulation of RAC1 is one of the multiple pathways a ected 8. Luo L, Hensch TK, Ackerman L, Barbel S, Jan LY, Jan YN. Differential effects of the by the HACE1 deficiency in SPPRS. Rac GTPase on Purkinje cell axons and dendritic trunks and spines. Nature 1996;379: 837–840. 9. Bongmba OY, Martinez LA, Elhardt ME, Butler K, Tejada-Simon MV. Modulation of Upregulation of RAC1 in SPPRS suggests a potential thera- dendritic spines and synaptic function by Rac1: a possible link to Fragile X syndrome peutic approach by using specific pharmacologic inhibition of pathology. Brain Res 2011;1399:79–95. 11 10. Reijnders MRF, Ansors NM, Kousi M, et al. RAC1 missense mutations in de- RAC1. However, caution should be taken because the de- velopmental disorders with diverse phenotypes. Am J Hum Genet 2017;101:466–477. velopment and function of the brain requires RAC1 to be 11. Martinez LA, Tejada-Simon MV. Pharmacological inactivation of the small GTPase fi Rac1 impairs long-term plasticity in the mouse hippocampus. Neuropharmacology nely tuned, as shown by the observations that either loss (or 2011;61:305–312. dominant-negative effect) or gain of RAC1 function led to 12. Iimura A, Yamazaki F, Suzuki T, Endo T, Nishida E, Kusakabe M. The E3 ubiquitin 10 ligase Hace1 is required for early embryonic development in Xenopus laevis. BMC developmental disorders in humans, and both depletion and Dev Biol 2016;16:31. overexpression of Rac1 resulted in abnormal phenotypes in 13. Lachance V, Degrandmaison J, Marois S, et al. Ubiquitylation and activation of a Rab 12 GTPase is promoted by a beta(2)AR-HACE1 complex. J Cell Sci 2014;127:111–123. Xenopus laevis. 14. Tang D, Xiang Y, De Renzis S, et al. The ubiquitin ligase HACE1 regulates Golgi membrane dynamics during the cell cycle. Nat Commun 2011;2:501. 15. Ehrnhoefer DE, Southwell AL, Sivasubramanian M, et al. HACE1 is essential for Upregulation of RAC1 may explain a part of the clinical astrocyte mitochondrial function and influences Huntington disease phenotypes in symptoms in SPPRS, but it does not cover the full spectrum. vivo. Hum Mol Genet 2018;27:239–253.

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG ARTICLE OPEN ACCESS HACE1 deficiency leads to structural and functional neurodevelopmental defects

Vanja Nagy, PhD, Ronja Hollstein, PhD, Tsung-Pin Pai, PhD, Michel K. Herde, PhD, Pisanu Buphamalai, MSc, Correspondence Paul Moeseneder, MSc, Ewelina Lenartowicz, MSc, Anoop Kavirayani, DVM, Dip. ACVP, Dr. Penninger [email protected] Georg Christoph Korenke, MD, Ivona Kozieradzki, MSc, Roberto Nitsch, PhD, Ana Cicvaric, PhD, or Dr. Nagy Francisco J. Monje Quiroga, PhD, Matthew A. Deardorff, MD, PhD, Emma C. Bedoukian, MS, Yun Li, PhD, [email protected] Gokhan¨ Yigit, PhD, Jorg¨ Menche, PhD, E. Ferda Perçin, MD, PhD, Bernd Wollnik, MD, Christian Henneberger, MD, Frank J. Kaiser, PhD, and Josef M. Penninger, MD

Neurol Genet 2019;5:e330. doi:10.1212/NXG.0000000000000330

Abstract RELATED ARTICLE Objective Editorial We aim to characterize the causality and molecular and functional underpinnings of HACE1 HACE1, RAC1, and what deficiency in a mouse model of a recessive neurodevelopmental syndrome called spastic else in the pathogenesis of paraplegia and psychomotor retardation with or without seizures (SPPRS). SPPRS? Page e326 Methods By exome sequencing, we identified 2 novel homozygous truncating mutations in HACE1 in 3 patients from 2 families, p.Q209* and p.R332*. Furthermore, we performed detailed molecular and phenotypic analyses of Hace1 knock-out (KO) mice and SPPRS patient fibroblasts.

Results We show that Hace1 KO mice display many clinical features of SPPRS including enlarged ventricles, hypoplastic corpus callosum, as well as locomotion and learning deficiencies. Mechanistically, loss of HACE1 results in altered levels and activity of the small guanosine triphosphate (GTP)ase, RAC1. In addition, HACE1 deficiency results in reduction in synaptic puncta number and long-term potentiation in the hippocampus. Similarly, in SPPRS patient–derived fibroblasts, carrying a disruptive HACE1 mutation resembling loss of HACE1 in KO mice, we observed marked upregulation of the total and active, GTP-bound, form of RAC1, along with an induction of RAC1-regulated downstream pathways.

Conclusions Our results provide a first animal model to dissect this complex human disease syndrome, establishing the first causal proof that a HACE1 deficiency results in decreased synapse number and structural and behavioral neuropathologic features that resemble SPPRS patients.

From the IMBA (V.N., T.-P.P., P.M., A.K., I.K., R.N., J.M.P.), Institute of Molecular Biotechnology of the Austrian Academy of Sciences, VBC—Vienna BioCenter Campus, Austria; Department of Medical Genetics (J.M.P.), Life Science Institute, University of British Columbia, Vancouver, Canada; Ludwig Boltzmann Institute for Rare and Undiagnosed Diseases (V.N., E.L.), Vienna, Austria; Section for Functional Genetics at the Institute of Human Genetics (R.H., F.J.K.), University of Lubeck;¨ German Center for Cardiovascular Research (DZHK e.V.) (F.J.K.), Partner Site Hamburg/Kiel/Lubeck,¨ Lubeck;¨ Institute of Cellular Neurosciences (M.K.H., C.H.), University of Bonn Medical School, Germany; Centre for Neuroendocrinology (M.K.H.), Department of Physiology, School of Biomedical Sciences, University of Otago, Dunedin, New Zealand; Department of Neurophysiology and Neuropharmacology (A.C., F.J.M.Q.), Center for Physiology and Pharmacology, Medical University of Vienna, Austria; Drug Safety and Metabolism (R.N.), IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden; Division of Genetics and the Roberts Individualized Medical Genetics Center (M.A.D., E.C.B.), Children’s Hospital of Philadelphia, PA; Departments of Pediatrics (M.A.D.), University of Pennsylvania Perelman School of Medicine, Philadelphia, PA; Institute of Human Genetics (Y.L., G.Y., B.W.), University Medical Center Gottingen,¨ Germany; Institute of Neurology (C.H.), University College London, UK; German Center for Neurodegenerative Diseases (DZNE) (C.H.), Bonn, Germany; Zentrum fur¨ Kinder- und Jugendmedizin (G.C.K.), Neurop¨adiatrie, Klinikum Oldenburg, Germany; Department of Medical Genetics (E.F.P.), Faculty of Medicine, Gazi University, Ankara, Turkey; CeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences (P.B., J.M.), Vienna, Austria.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing Charge was funded by IMBA, Austrian Academy of Sciences, ERC. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (CC BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary ANOVA = analysis of variance; HACE1 = HECT Domain And Ankyrin Repeat Containing E3 Ubiquitin Protein Ligase 1; HD = healthy donor; KO = knock out; LTP = long-term potentiation; NADPH = nicotinamide adenine dinucleotide phosphate oxidase; OFT = open-field test; PPI = pre-pulse inhibition; RAC1 = Ras-related C3 botulinum toxin substrate 1; ROS = reactive oxygen species; SPPRS = spastic paraplegia and psychomotor retardation with or without seizures; WT = wild type.

Homologous to the E6-AP carboxyl terminus domain and patients 10 and 11 has been approved by the ethics committee ankyrin repeat containing E3 ubiquitin-protein ligase 1 (HACE1), at The University of G¨ottingen. Biological materials of was identified to be downregulated in a number of tumors, as well patients and healthy donors (HDs) and written informed – as to play a role in inflammatory responses.1 4 Analysis of Hace1 consents were obtained in accordance with the Declaration of knock-out (KO) mice showed HACE1 is ubiquitously expressed, Helsinki. Fibroblasts from 2 female patients and 1 male pa- with relatively high expression in the brain.2 tient only, 6, 7, and 10, respectively, are used in this study, as no consent was available for the others. A neurodevelopmental disorder named spastic paraplegia and psychomotor retardation with or without seizures (SPPRS) Mice has recently been described in 15 patients of 6 unrelated Animals were housed at the Institute of Molecular Bio- families, associated with mutations throughout the HACE1 technology, Vienna, Austria, maintained under a 12-hour – gene.5 7 Mutations are predicted to be deleterious to HACE1 light/dark cycle, and provided with food and water ad libitum. protein function, either disrupting folding or causing a frame Experiments described in this study were approved by the shift and early termination of translation, resulting in no de- Bundesministerium fur Wissenschaft, Forschung und Wirt- tectable protein product in patient-derived fibroblasts.6,7 schaft (BMWFW-66.015/0004-WF/V/3b/2015) and per- Clinical manifestations are variable and include early-onset formed according to EU-directive 2010/63/EU. developmental delays, severe intellectual disability, epilepsy, hypotonia, spasticity, ataxia, and difficulty with verbal com- Immunoblotting munication. MRI findings include enlarged ventricles, hypo- The following antibodies were used for standard Western plastic corpus callosum, and altered white and gray brain blotting protocols: anti-HACE1 at 1:1,000 dilution (AbCam), matter ratios. There are several known targets for HACE1 anti-Cyclin D1 at 1:1,000 (AbCam), anti-Glyceraldehyde-3- ubiquitination, including Ras-related C3 botulinum toxin Phosphate Dehydrogenase at 1:1,000 dilution (Cell signal- substrate 1 (RAC1), small GTPase, important for different ling), anti-RAC1 at 1:1,000 dilution (Millipore), anti-ß-actin – aspects of brain development and neuronal function.4,8 11 In at 1:5,000 dilution (Sigma, clone AC-74), and appropriate – addition to cytoskeletal remodeling, RAC1 can regulate re- secondary horseradish peroxidase linked whole antibodies active oxygen species (ROS) levels as part of the nicotinamide (GE Healthcare). adenine dinucleotide phosphate oxidase (NADPH) com- Histology plex.12 Correspondingly, Hace1 KO mice were reported to Selected brains were isolated, processed, embedded, sec- have elevated ROS levels.13 While the role for hace1 was tioned, and stained by the Histopathology Facility at the recently described in embryonic development in Xenopus Vienna Biocenter Core Facilities (VBCF), member of the laevis14 and murine HACE1 has been implicated in Hun- Vienna Biocenter (VBC), Austria. Briefly, 2-μm-thick coronal tington disease,13 nothing is known about the role of HACE1 or sagittal paraffin-embedded sections were prepared by during mammalian nervous system development or the mo- routine microtomy and stained with hematoxylin and eosin lecular underpinnings of SPPRS. To elucidate the neuro- (H&E, Shandon* Harris Hematoxylin Acidified; Thermo developmental pathophysiology of SPPRS, we therefore Scientific Shandon Eosin Y; Fisher Scientific), Luxol Fast performed detailed analysis of Hace1 KO mice and confirmed Blue–Cresyl Violet (LFBCV, Sigma), or with anti–Myelin Basic our findings in SPPRS patient–derived fibroblasts. Protein antibody (Abcam, 1:100). Slides were imaged using a Zeiss Axioskop 2 MOT microscope (Carl Zeiss Microscopy) Methods and subsequently digitized with the Pannoramic FLASH 250 II automated slide scanner (3D Histech). Images were acquired See supplemental information (e-methods, links.lww.com/ with the Pannoramic Viewer software (3D Histech) and with NXG/A151) for more details. SPOT Insight camera (Diagnostic Instruments, Inc.). Standard protocol approvals, registrations, and patient consents MRI Genetic and clinical data of patients 6–8 were published.6 Our Anesthetized male C57Bl6/J mice 8–10 weeks of age and newly identified patients are labeled as patients 9–11. Genetic their Hace1 KO littermates were imaged in the Preclinical data regarding patient 9 were produced as part of a clinical Imaging Facility at VBCF (pcPHENO, VBCF), member of diagnostic service (GeneDx), and the research study including the VBC, Austria, using a 15.2 T MRI (Bruker BioSpec,

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Ettlingen, Germany). Values were normalized to brain size, Migration assay averaged, and presented as percentage of total brain volume, Primary dermal fibroblasts from patients 6, 7, and 10 and HDs and unpaired Student’s t test was used to determine statistical were seeded at high density in 24-well plates for 24 hours and significance. scratched using a pipette tip. The area of the scratch at 0 and 24 hours following the scratch was measured digitally in Behavioral testing pixels, and the relative gap closure was calculated and nor- fi Open- eld test (OFT) and elevated plus maze were performed malized to wild-type cells. Unpaired Student’s t test was used as described previously using an automated activity system to determine significance. (TSE-Systems, Germany).15 Unpaired Student’s t test was used to determine significance. Accelerating rotarod was performed ROS detection in the mouse brain and human on a 5–40 rpm accelerating apparatus (Ugo Basile), as described primary dermal fibroblasts previously.16 Significance was determined by 2-way analysis of Twenty-micrometer coronal mouse brain sections were washed variance (ANOVA) with Sidak’s multiple comparisons. In the 2X in phosphate buffered saline and incubated in 10-mM dihy- ladder rung walking task (pcPHENO, VBCF), mice were trained droethidium for 15 minutes at 37°C. Sections were washed twice to walk across a ladder (length: 80 cm; spacing of rungs: 1 cm) in phosphate buffered saline and imaged with a Zeiss Axioplan2 toward an escape ladder in the TSE MotoRater system (tse- microscope. Levels of ROS in fibroblasts were investigated using systems.com). Mice were recorded and manually scored using CellROX Deep Red staining (Thermo Fischer) followed by flow SimiMotion software 8.5.0.327.17 Unpaired Student’s t test was cytometric analysis according to the manufacturer’s instructions. used to determine significance. T-maze was performed as de- Statistical analysis was determined using unpaired Student’s t test. scribed previously.18 Significance of deviation of the number of successes in mutant vs wild-type (WT) mice was evaluated using a binomial test. For acoustic fear conditioning (pcPHENO, VBCF), mice were trained to associate the conditioned sound Results stimulus (CS = 85 dB, 10 kHz) with the unconditioned foot Identification of novel HACE1 mutations in shock stimulus delivered by the floor grid (US = 1.5 mA), as SPPRS patients described previously, using Coulbourn Habitest operant cages Family 1 originating from Saudi Arabia had 1 affected female (Coulbourn Instruments, MA and FreezeFrame, Actimetrics, individual, patient 9, born to healthy, consanguineous parents 19 IL). Unpaired Student’s t testwasusedtodeterminesignifi- (figure e-1A, links.lww.com/NXG/A148). The patient pre- cance. Mice were trained in the Morris water maze (pcPHENO, sented with similar clinical symptoms as previously reported 20 VBCF) as described previously. Unpaired Student’s t test was cases of SPPRS, including severe intellectual disability, de- used to analyze the short- and long-term memory data and 2-way velopmental delay, and inability to sit or speak by the age of 5 ANOVA with Sidak’s multiple comparison test for the latency to years.6 Cranial MRI revealed microcephaly and brachyceph- reach the platform. Mice were tested for startle responses and aly, as well as hypoplastic corpus callosum and likely brain- 21 prepulse inhibition (PPI) as described previously using the SR- stem abnormality, small sella with ectopic neurohypophysis, LAB Startle Response System (San Diego Instruments) cham- and mild ventriculomegaly (figure e-1B). Additional clinical ber. Two-way ANOVA with Sidak’s multiple comparison test findings include mild facial dysmorphia, skeletal abnormalities, was used to determine significance. and ulnar deviation of the wrists and small feet (figure e-1C). In an unrelated consanguineous family 2 originating from Turkey, Hippocampal slices preparation and born to a healthy father and mother with Hashimoto disease are electrophysiologic recordings one female (patient 10) and one male (patient 11), 4 and 6 Electrophysiologic field recordings in the stratum radiatum of months of age at initial admission (Figure e-1D). During preg- acute hippocampal slices were performed as described pre- nancy, the mother was euthyroid and both children were carried viously.22 Two-population Student’s t test were used to de- to term. Patients had variable clinical symptoms in comparison termine significance. with patient 9 and those previously reported, including hypo- tonia, small feet, enlarged head circumference, inverted and wide Synaptic number analysis spaced nipples, facial dysmorphic features, and strabismus Hace1 KO; Thy1-green fluorescent protein (GFP) line M and (figures e-1, E and F). Considering the pronounced hypotonia hace1 WT; and Thy1-GFP line M16 were analyzed using the in all 3 patients, ataxia was difficult to determine. For detailed LSM700 Axio Imager. Synapse number counts and mea- clinical symptoms of all patients, please refer to table. surement of neurite lengths were performed using ImageJ.23 Statistical analysis was performed using Student’s t test. In these 3 patients, mutations in HACE1 were identified by exome sequencing and verified by Sanger sequencing: patient Active RAC1 pulldown 9 carries a homozygous c.625C>T; p.Q209* HACE1 muta- Cellular levels of active RAC1 in fibroblasts were analyzed tion in exon 8, NM_020771.3:c.625C>T, p.Q209* (chr6:104 using the Active Rac1 Pull-Down and Detection Kit 797 018 on the hg38 build); patients 10 and 11 carry a ho- (Thermo Fisher Scientific) according to the manufacturer’s mozygous c.994C>T; p.R332* HACE1 mutation in exon 11, instructions. NM_020771.3:c.994C>T, p.R332* (chr6:104 791 584)

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 4 erlg:Gntc oue5 ubr3|Jn 09Neurology.org/NG 2019 June | 3 Number 5, Volume | Genetics Neurology:

Table Comparison of clinical characteristics and genetic findings in reported patients

Patient 6 Patient 7 Patient 9 Patient 10 Patient 11

Background HACE1 mutation Nonsense; c.2242C > T; p.R748* Nonsense; c.2242C > T; Nonsense; c.625C > T; p.Q209X Nonsense; c.994C > T; p.R332X Nonsense; c.994C > T; p.R332X and Frameshift; p.R748* and Frameshift; c.2019_ c.2019_ Saudi Arabia Turkey Turkey 2020insTTTAGGTATTTTTAGGT; 2020insTTTAGGTATTTTTAG ATT; p.P674Ffs*5 (Hollstein GTATT; p.P674Ffs*5 Family 1 Family 2 Family 2 et al.6) (Hollstein et al.6)

Sex F M F F M

Birth weight (kg) 3.37 3.4 3.0 4.140 (97th percentile) 3.480

Birth OFC (cm) 33 34 UK 37 cm (97th percentile) UK

Gestational age 39 40 40 40 40

Development Recent growth parameters

Age 10 y 11 mo 7 y 8 mo 5 y 4 mo 4 mo 6 mo

Height (cm) BMI 24.8 BMI 21.17 86.5, Z = −5.67 (scoliosis) 67 (>97th percentile) 66 (75th percentile)

Weight (kg) 15.5, Z = −1.6 9.5 (>97th percentile) 9.7 (>97th percentile)

OFC (cm) 54 (+0.5 SD) 53 (+0.2 SD) 4y, 44, Z = −3.72 44 (>97th percentile) 43.5 (75th percentile)

Neurology Sitting and walking 4 y 3–4 y Not at 5 y Not at 4 mo Not at 6 mo

Language Inarticulate speech Inarticulate speech None at 5 y Too young to be evaluated Too young to be evaluated

General Developmental delay Developmental delay Developmental delay Developmental delay Developmental Delay

Epilepsy None Myoclonic seizures and None None None focal epilepsy

Stereotypies None None Bruxism None None

Mobility Hypotonia and unstable ataxic Hypotonia and unstable Hypotonia Hypotonia Hypotonia gait ataxic gait

Ophthalmic findings None None Retinal dystrophy Strabismus of the right eye Bilateral strabismus

Hearing No abnormality No abnormality Bilateral sensorineural loss noted No abnormality No abnormality at 4 y

Intellectual disability Yes Yes Yes Too young to be evaluated Too young to be evaluated

Neuroradiology MRI Hypoplastic corpus callosum Hypoplastic corpus Microcephaly and brachycephaly; Normal Normal callosum hypoplastic corpus callosum and likely brainstem abnormality; small sella with ectopic neurohypophysis; mild ventriculomegaly

Continued fi

he ( gure 1A). Both are nonsense mutations predicted to result in a truncated HACE1 protein product with no catalytic ac- tivity. Parents in both families were determined to be het- erozygous carriers. Figure 1A also indicates all previously reported HACE1 mutations in SPPRS patients.5,6 Thus, we have identified three new cases with biallelic HACE1 muta- tions associated with marked and diverse neurologic as well as ureteral reflux; inguinal hernia anterior fontanelle Frontal bossing, prominent glabella, depressed nasal bridge, anteverted nares, and everted lower lip non-neurologic abnormalities, diagnosed to be SPPRS.

Hace1-mutant mice exhibit brain morphologies similar to SPPRS patients To determine whether Hace1 KO mice develop any clinical symptoms of SPPRS patients, we first analyzed structural and cellular features of mutant mouse brains using 15.2 T MRI and/or histologic staining at different developmental stages. Histologic examination of mouse brains at embryonic day (E) Normal Accessory spleen; bilateral Normal Delayed closure of the Normalglabella, depressed and short nasal bridge, anteverted nares, everted lower lip, high palate, and dysplastic ears Normal 18.5, postnatal day (P) 0.5, and P14.5 revealed no evident defects cortical lamination, cerebellar foliation, hippocampal architecture, or cellular aberration and density in any de- velopmental period examined (figure 1, B–D). However, enlarged ventricles and hypoplastic corpus callosum (figure 1, B and C) were noted in Hace1 KO animals as compared with

s visit. WT littermates, as early as E18.5 and P0.5, respectively. ’

Western blot analyses of 8-week-old mouse brain lysates revealed HACE1 to be expressed throughout the adult mouse neurogenic bladder and bladder stroma thoracolumbar kyphoscoliosis; ulnar deviation of the wristssmall and feet insufficiency brain, which was absent in the brains of Hace1 KO mice (figure 2A). Total volume and morphology of the Hace1 KO (continued) adult mouse brain as compared with WT littermates mea- sured by MRI and/or histologic examination revealed no whole-brain volumetric or cellularity differences as compared with WT littermates (figure 2, B and C). Review of histologic sections also revealed no abnormality in cortical laminar or- fi Normal Intestinal pseudo-obstruction; ganization and cellular density in the adult brain ( gure 2D). Detailed analysis of different brain regions via MRI also if the features were not noted at time of doctor ’ showed no significant differences in the volumes of the cortex, colliculi, thalamus, and putamen, comparing adult WT and unknown

‘ Hace1 KO brains, with a small decrease in the volume of the

that have donated fibroblasts for the study, as well as newly identified three affected individuals from families 1 and 2. Features were left blank if t olfactory bulb (data not shown). The adult cerebellar volume 13 and foliation as well as cortical laminar organization were comparable between Hace1 KOs and their control WT lit- termates (figure 2, E and F). Hippocampal volume was sig- Neurogenic bladder and esophageal reflux Lumbar lordosis Pes planus Bilateral hip dislocation; Normal Normal Hypothyroidism; adrenal Normal Normalnificantly Downturneddecreased mouth as Frontal bossing,measured prominent by MRI (figure 2G); however, there were no evident aberrations in cellularity and architecture on microscopic examination of hippocampi in adult animals (figure 2H).

Of interest, volumetric MRI reconstructions revealed a sig- nificant reduction in whole-brain white matter (figure e-2, A–D, links.lww.com/NXG/A149) in Hace1 KO adult mice as compared with their WT littermates, substantiated by myelin basic protein staining and volumetric MRI analysis (figure e-2, – B–D), similar to findings in reported SPPRS patients5 7 and patient 9 described here (figure 1C and table). Also, noted

Comparison of clinical characteristics and genetic findings in reported patients was a reduction of the anterior commissure in Hace1 KO mice as compared with WT littermates (figure e-2C). In addition, fi Gastrointestinal patients were too young for diagnosis at time of admission or labeled as Table Skeletal Endocrine Facial features Abbreviations: BMI = body massSummary index; of OFC clinical = features occipital of frontal the circumference. previously published Patient 6, 7 in line with ndings in SPPRS patients, as well as adult and

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 Figure 1 Histologic analysis of brain regions in neonatal and young Hace1 knock-out (KO) mice

(A) Schematic representation of HACE1 protein and locations of mutations of previously pub- lished (in gray) and novel mutations reported here (red arrows, black letters). (B) H&E-stained sections of wild-type (WT) and Hace1 KO mice at E18.5 reveal no major structural abnormality across the regions studied, with slight ven- triculomegaly in the Hace1 KO brain as compared with controls, left panel (yellow arrows). (C) H&E- stained coronal sections of the newborn brain of WT and Hace1 KO mice at P0.5 reveal ventricular dilation in the Hace1 KO brain (yellow arrows) and attenuation of the corpus callosum (black arrows). (D) H&E-stained sections of WT and Hace1 KO mice at P14.5. Cortical lamination, hippocampal formation, and the cerebellar ver- mis are comparable between WT and Hace1 KO in all 3 age groups studied. All brain regions are la- beled in the figure and magnifications are as in- dicated. A = ankyrin domain; H&E = hematoxylin and eosin; HECT = homologous to the E6-AP carboxyl terminus; MID = middle region.

6 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Figure 2 Morphology of the Hace1 KO mouse brain

(A) Western blot analysis depicting HACE1 abundance in the indicated 8–12-week-old male mouse brain regions of wild-type (WT) and Hace1 knock-out (KO) mice. (B) Pseu- docolored 3D reconstruction of lateral MRI images of representative whole mouse brains, and quantification (mean values ± SEM) showing no difference in total brain volumes in 8–12-week-old male WT and Hace1 KO littermates. Orange— cerebellum, yellow—hippocampus, red— corpus callosum, light blue—thalamus, purple—colliculi, turquoise—olfactory bulbs, and dark blue—putamen; WT n = 12 and KO n = 9; p value as indicated in the bar graph; unpaired Student’s t test. (C) LFB- CV–stained coronal sections of whole WT and Hace1 KO adult mouse brains. Ven- triculomegaly (red arrows) is evident in the KO brain as compared with the WT litter- mate. (D) LFB-CV–stained coronal sections of the adult cortex show no defects in cel- lularity or lamination (laminae I to VI) in Hace1 KO mice. (E) Pseudocolored (orange) 3D reconstructed MRI images of represen- tative lateral views of WT and Hace1 KO lit- termate cerebelli showing no significant volumetric differences in Hace1 KO vs WT control littermates. Right, quantification of volumetric differences; mean values ± SEM; WT n = 12; KO n = 9; p value as indicated in the figure; unpaired Student’s t test. (F) Se- lected LFB-CV–stained sagittal sections of the cerebellum showing no apparent dif- ferences between WT and Hace1 KO brains. (G) Pseudocolored (yellow) 3D re- construction of MRI images of representa- tive lateral view of hippocampi of WT and Hace1 KO mice shows a significant re- duction in hippocampal volume in Hace1 KO brains; quantification is shown in the right panel; mean values ± SEM; WT n = 12 and KO n = 9; *p ≤ 0.05; unpaired Student’s t test. (H) LFB-CV–stained coronal sections of the hippocampus show no notable mor- phological difference in cellularity between WT and Hace1 KO brains. Scale bars as la- beled in the images. CA = cornu ammonis; CC = corpus callosum; cereb = cerebellum; Ctx = cortex; DG = dentate gyrus; hipp = hippocampus; hypo = hypothalamus; LFB- CV = luxol fast blue - cresyl violet; LV = lat- eral ventricle; stria = striatum; WM = white matter.

developing mouse brains, we observed enlarged ventricles in of behavioral assays. The most notable clinical features of – Hace1 KO mice as compared with control littermates (figure SPPRS are intellectual disability and ataxia5 7; we therefore 2C). These data show that Hace1 KO mice phenocopy key focused on hippocampal- and cerebellar-dependent tasks, structural features associated with SPPRS patients. principal brain regions regulating those behaviors in mice. Hace1 KO mice generated in our group previously are gen- HACE1-deficient mice exhibit locomotion and erally healthy, viable, and fertile.2 First, we found no signifi- associative learning disabilities cant differences in anxiety levels in Hace1 WT and KO To assess what, if any, behavioral defects can be detected in littermates analyzed by the OFT and the elevated plus maze Hace1 KO mice, we subjected cohorts to an extensive battery (figure e-3, A and B, links.lww.com/NXG/A150); however,

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 7 we noted a reduction in the distance traveled during the rod in the first trial, Hace1 KO mice exhibited shorter latencies testing (figure 3A). These data indicated that while our mice to fall on all subsequent trials (figure 3B). We further analyzed exhibited normal anxiety levels, there may be a locomotion the mice on a ladder rung walking task, which enables the defect. Therefore, we tested Hace1 KO mice for any loco- observer to analyze coordinated front and hind paw place- motion and gait defects. WT and Hace1 KO mice were tested ment on each individual rung of the ladder.17 While Hace1 in the accelerating rotarod task, a standard assay to determine KO animals were capable of performing the task without locomotion and gait defects.16 While Hace1 KO mice and falling off the apparatus and showed similar percentages of control littermates had similar latencies to fall off the rotating correct placement over all limbs analyzed, percent of major

Figure 3 Behavioral analysis of Hace1 KO mice reveals deficits in specific locomotion, learning, and memory tasks

(A) Open-field maze (OFM) and elevated plus maze (EPM) showed slight and significant reductions in the distances Hace1 KO mice had traveled as compared with their WT littermates, respectively; data are shown as mean values ± SEM; WT n = 10 and KO n = 10; *p ≤ 0.05 or as indicated in the bar graphs; unpaired Student’s t test. (B) Hace1 KO mice show significantly shorter latencies to fall off the accelerating rotarod; mean values ± SEM; WT n = 19 and KO n = 21; *p ≤ 0.05; 2-way ANOVA with Sidak’s multiple comparison test. (C) Ladder rung walking task illustrates a significant increase in the percentage of total misses, deep slips, and/or slight slips (0–2 score) in Hace1 KO mice compared with WT littermate controls; mean values ± SEM; WT n = 12 and KO n = 9; *p ≤ 0.05; unpaired Student’s t test. No significant differences were noted in the 3–4 score (replacement and correction), 5 score (partial placement), or 6 score (correct placement). Bar graphs depict mean values of all paws ± SEM; p values are indicated in figures; WT n = 12 and KO n = 9; unpaired Student’s t test. (D) Significant deficiency in long-term contextual memory as measured by auditory fear conditioning was detected in Hace1 KO mice as compared with WT littermates (left panel); mean values ± SEM; WT n = 16 and KO n = 13; *p ≤ 0.05; unpaired Student’s t test. Cued fear conditioning was not significantly different in Hace1 KO mice as compared with WT littermates; mean values ± SEM; WT n = 16; KO n = 13; p value = 0.21; unpaired Student’s t test. (E) Hace1 KOs show significant reduction in acoustic startle inhibition at 110 and 120 dB as compared with WT littermates. Mean values ± SEM; WT n = 14; KO n = 11; *p ≤ 0.05; ***p ≤ 0.001; 2-way ANOVA with Sidak’s multiple comparison test. ANOVA = analysis of variance; KO = knock out; WT = wild type.

8 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG missteps made were significantly higher as compared with this synaptic pathway and found that LTP is significantly reduced WT littermates. Both groups of animals had similar placement in acute hippocampal slices from Hace1 KO mice compared with and correction scores, while WT animals had a higher per- WT littermates (figure 4, D and E). Of note, nerve conduction centage of partial placement on rungs (figure 3C). Thus, velocity was not affected in Hace1 KO recordings as compared Hace1 KO mice exhibit locomotion and gait deficits. with controls (data not shown), in line with the report that nerve conduction velocity is unaffected in SPPRS patients.6 The second notable clinical feature of SPPRS patients is in- – tellectual disability (table).5 7 To determine whether learning Because we observed hippocampal-dependent contextual fear and memory processes are altered in Hace1 KO animals, we learning deficiency and dysfunctions in electrophysiologic analyzed mutant mice in various hippocampal-dependent properties of pyramidal cells within the CA1 region of the behavioral tasks. In the Morris water maze, KO mice were able hippocampus, we analyzed morphological properties of py- to swim normally and there were no apparent differences in ramidal neurons in this brain region. To this end, we crossed latencies between WT and Hace1 KO littermates to find the Hace1 KO mice with Thy1-GFP-M mice to drive neuronal visible platform (not shown), in escape latencies during any of GFP expression, which allowed us to visualize neuronal the training days (figure e-3C, links.lww.com/NXG/A150), morphology, including spines along dendrites of CA1 pyra- nor in short- or long-term memory (figure e-3C, right panels). midal cells.27 Quantification of selected dendritic segments Likewise, there were no differences in T-maze between WT indeed revealed a significant reduction in spine numbers in and Hace1 KO animals (figure e-3D). However, subjecting Hace1 KO neurons as compared with hippocampal neurons the animals to cued and contextual fear conditioning revealed from WT littermates (figure 4F). Whether such reduction in significantly reduced freezing responses in the contextual as- the synapse number is also present throughout the brain and pect of the task in Hace1 KO mice compared with WT lit- might explain defective locomotion in Hace1 KO mice needs termates (figure 3D). to be determined. These data indicate that, as a correlate to hippocampus-dependent impaired associate learning, In addition, Hace1 KO mice exhibited significantly enhanced HACE1 deficiency leads to a deficit of synaptic plasticity and startle amplitude responses or reduced startle inhibition, in a marked reduction of synapses in the hippocampus. the acoustic startle task at presentation of 110- and 120-dB startle sounds as compared with their WT littermates (figure Elevated levels of active Rac1 in Hace1 KO 3E). These results also suggest Hace1 KO animals do not have mouse brains and SPPRS any hearing deficiencies. PPI was reduced for all prepulse patient–derived fibroblasts intensities presented (80, 85, 90, and 95 dB); however, this To determine molecular alterations downstream of HACE1 reduction was not significant when comparing Hace1 KO deficiency, we focused on Ras-related C3 botulinum toxin animals with WT controls (figure e-3E, links.lww.com/NXG/ substrate 1, RAC1, a well-characterized substrate for HACE1- A150). These data indicate that, besides locomotion defects, mediated proteasomal degradation.8,9 Indeed, Western blot associative learning is altered in Hace1 KO mice, paralleling analysis showed that RAC1 protein abundance was increased behavioral findings in human SPPRS patients. in Hace1 KO brains when compared with WT littermates (figure 5A). In addition, Cyclin D1 abundance and ROS Hace1 deficiency results in altered levels, as downstream read-outs for RAC1 activity,12 were hippocampal synaptic transmission and likewise increased in Hace1 KO brain regions as compared reduced synapse numbers with controls (figure 5, A and B). We and others have previously shown that altered synaptic transmission and plasticity at CA3-CA1 Schaffer collaterals The role of RAC1 has been extensively studied in the context can underlie deficits of contextual fear memory.22,24 We tested of mouse brain development.10,11 Of interest, perturbation of whether synaptic transmission and plasticity of this pathway RAC1 levels, either overexpression or downregulation, results are affected in Hace1 KO mice. Recordings of field excitatory in decrease in dendritic spines in the mouse brain, similar to post synaptic potentials in response to CA3-CA1 synapse our Hace1 KO mice.10,28 To determine whether RAC1 is also stimulation (schematic of the set-up in figure 4A) revealed an changed in HACE1-mutant SPPRS patients, we performed upregulation of excitatory synaptic transmission in hippo- Western blotting analysis of SPPRS patient–derived fibro- campal slices from Hace1 KO mice (figure 4, A and B). Fur- blasts, isolated from patients 6, 7, and 10 (table). Of note, in ther analysis showed that the increase in synaptic transmission these patients, the mutations result in undetectable HACE1 at the CA3-CA1 synaptic pathway was most likely due to protein expression (figure 5C), resembling our Hace1 KO a postsynaptic effect of the Hace1 deletion because neither the mice. In all patients’ fibroblasts, total abundance of RAC1 and axonal fiber volley (figure 4C) nor the paired-pulse ratio, an Cyclin D1 was increased as compared with fibroblasts from indirect measure of the presynaptic release probability,25 was HD controls (figure 5C). It is important that increased affected (interstimulus interval 50 ms, paired-pulse ratio: WT abundance of active GTP-bound RAC1 was detected in all 3 1.70 ± 0.066, n = 19, Hace1 KO 1.66 ± 0.036, n = 20, p = 0.64, patient fibroblasts (figure 5D). As a functional consequence of 2-population t test, not shown). We then probed long-term po- increased RAC1 activity, ROS upregulation was also evident tentiation (LTP), a cellular correlate of memory formation,26 of in SPPRS patient–derived fibroblasts (figure 5E). In addition,

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 9 Figure 4 Altered long-term potentiation and reduced hippocampal synapse numbers in Hace1 KO mice as compared with WT controls

(A) Upper left panel, schematic fEPSP recording configuration from acute hippocampal slices: stimulation (red bolt) of CA3-CA1 Schaffer collaterals (green), the fEPSP extracellular recording electrode in the CA1 region (white cone, ep = extracellular pipette). Synaptic transmission at CA3-CA1 hippocampal synapses is increased in Hace1 KO mice compared with WT littermates. fEPSP slopes were normalized to axonal fiber volleys to account for variations in axonal stimulation efficacy between slice preparations and animals. Analysis was performed over a range of stimulus intensities. Average fEPSP slopes were significantly higher in Hace1 KO slices; mean values ± SEM; WT n = 19 and KO n = 20; *p ≤ 0.05; 2-population Student’s t test. (B) Representative traces of fEPSPs in response to CA3-CA1 synapse stimulation obtained from WT (black trace) and Hace1 KO (red trace) mice. Field EPSP (black arrow) recordings were normalized to fiber volleys (gray arrow) to account for varying stimulation efficiencies across experiments and preparations. Time scale of recording is indicated. (C) No significant difference was found between fiber volley amplitudes (mean values ± SEM) between the indicated genotypes; WT n = 19 and KOn = 20. (D) Long-term potentiation (LTP) of synaptic transmission was induced by HFS after a 10-minute baseline recording. fEPSP slopes were normalizedto their baseline levels and plotted over time. LTP was reduced in Hace1 KO animals; mean values ± SEM; WT n = 19 and KO n = 18; **p ≤ 0.01; 2-population Student’s t test. (E) Representative fEPSP traces illustrating LTP experiments displayed in panel (D). Solid lines represent baseline fEPSPs, and dashed lines represent fEPSPs 30 minutes after LTP induction (Hace1 WT: black, upper panel; Hace1 KO: red, lower panel). (F) Representative 3D composite of z-stacked ×63 −/− magnified (zoom factor of 2) confocal images of hippocampal CA1 pyramidal cell distal dendrites of Hace1+/+;Thy1-GFP and Hace1 ;Thy1-GFP mice. Uncolored dendritic segments are on the left, synaptic spines are pseudocolored in magenta on the right panels; scale bar is 2.5 μm. Right bar graphs show quantifi- cations of spine numbers; mean values ± SEM; WT n = 3 and KO n = 3; **p ≤ 0.01; unpaired Student’s t test. fEPSPs = field excitatory post synaptic potentials; KO = knock out; WT = wild type.

10 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Figure 5 Hace1 KO mouse brains and SPPRS patient–derived fibroblasts have elevated Rac1 and ROS levels

(A) Western blot analysis of the cortex (ctx.), hippocampus (hipp.) and cerebellum (cereb.) from WT and Hace1 KO mice revealed a strong upregulation of total RAC1 abundance (lower band, arrow) and Cyclin D1 protein in the absence of HACE1, in all indicated brain regions tested. Actin was used as a loading control. (B) ROS (detected by DHE staining) is elevated in the hippocampus of Hace1 KO mice as compared with WT littermates; ×20 magnification. (C) Fibroblasts harvested from SPPRS patients (P6, P7, and P10) have no detectable HACE1 protein product as compared with normal donor fibroblasts (ND) as revealed by Western blotting using an antibody recognizing the HACE1 C-terminus; however, patient fibroblasts express elevated total abundance of RAC1 and Cyclin D1 protein product. GAPDH was used as a loading control. (D) SPPRS patient–derived fibroblasts exhibit elevated abundance of active RAC1 in comparison with normal donor fibroblasts, as measured by the Active RAC1 Pull-Down and Detection Kit. (E) Left panel, FACS analysis of SPPRS patient–derived fibroblasts incubated with CellROX deep red reagent revealed significantly more ROS in patient cells as compared with normal donor (ND) fibroblasts. Right panel, quantification of ROS production in the indicated patient and control fibroblasts; mean values ± SD; n = 7 for ND, P6, and P7 fibroblasts, and n = 3 biological replicates for patient 10 fibroblasts; *p ≤ 0.05; unpaired Student’s t test. (F and G) Confluent SPPRS patient P6, P7, P10 and ND-derived fibroblasts were subjected to scratch assays, and migration was monitored. Panel (F) shows a quantification of fibroblast migration rates; mean values ± SD; n = 3; **p ≤ 0.01 and ***p ≤ 0.001 comparing patient fibroblasts with the ND controls; unpaired Student’s t test. Representative images are shown in (G). KO = knock out; ROS = reactive oxygen species; SPPRS = spastic paraplegia and psychomotor retardation with or without seizures; WT = wild type. in a scratch assay, patient fibroblasts migrated significantly show that loss of HACE1 expression results in increased levels faster, as compared with control fibroblasts (figure 5, F and of active RAC1 in both Hace1 KO mouse brains and HACE1- – G), confirming enhanced RAC1 activity.29 32 These data mutant SPPRS patient–derived fibroblasts.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 11 Discussion 3 newly reported SPPRS patients here, who either had micro- cephaly (1/3) or normal head circumference (2/3). Hypoplastic Here, we report 2 novel mutations in HACE1, p.Q209* and corpus callosum, hypotonia, intellectual disability, and hearing p.R332*, discovered in 3 patients from 2 unrelated consan- impairment were noted in some but not all patients carrying guineous families with a complex neurodevelopmental dis- gain-of-function RAC1 mutations, requiring further examination order. These new patients have variable clinical symptoms, of the specific or common molecular pathology of RAC1- 5–7 overlapping with previous descriptions of SPPRS. The associated neurodevelopmental disorder and SPPRS. genotype–phenotype correlations of the different mutations need to be addressed in future studies, as more SPPRS Importantly, it has been previously established that dysregu- patients are diagnosed. We find that HACE1 is expressed lation of RAC1 abundance in mouse models results in a de- throughout the adult mouse brain. Hace1 KO mice have crease in dendritic spine number as well as reduction in defects in basic sensorimotor processing, deficiencies in spe- hippocampal plasticity, as we have uncovered in our Hace1 cific learning and memory tasks, reduced LTP of hippocampal KO mice.10,28,38 We therefore speculate that RAC1 is a key CA3-CA1 synapses, and fewer synaptic spines at CA1 pyra- factor underlining neuronal pathology in Hace1 KO mice and midal neurons. In addition, they exhibit locomotion defects HACE1-deficient patients. Whether RAC1-dependent cyto- as compared with littermate controls. Furthermore, we skeletal modulation and/or ROS homeostasis is the key de- show a marked upregulation of RAC1 levels throughout terminant feature of HACE1-dependent neurodevelopmental the mutant mouse brain and elevated ROS levels. Verifying deficiencies remains to be elucidated. The present study adds that our findings are relevant in humans, active RAC1 HACE1 to the large RAC1 regulatory complex, which when abundance, downstream signaling components, ROS perturbed has deleterious consequence to brain development production, and cellular migration in SPPRS patient– and neuronal function. derived fibroblasts are likewise dysregulated, all indicative of a hyperactive RAC1 pathway. Thus, HACE1 deficiency Acknowledgment leads to neuroradiologic and behavioral manifestations in The authors thank patients and their families for their mice reminiscent of the clinical features seen in SPPRS participation; members of the IMP Biooptics facility for their patients. technical advice and assistance; Dr. David Keays, IMP, Vienna, for the use of behavioral equipment, and members of A well-characterized target for HACE1 ubiquitination and his team for helpful discussions. In addition, the authors thank subsequent proteasomal degradation is RAC1, a small mem- Dr. Daniel Lin from Sunjin Lab for the technical support with ber of the Rho family of GTPases, critical for neurogenesis, 3D Images of dendritic spines. The work was supported by migration, axonal elongation, synaptogenesis, and activity- the Human Frontiers Science Program (HFSP, to C.H.), the driven plasticity.11 HACE1 is therefore poised to regulate many German Research Foundation (DFG, SPP1757, SFB1089 to functions of RAC1, by modulating its active GTP-bound levels, C.H.), FWF Lise Meitner Postdoctoral Fellowship to V.N., including NADPH-dependent ROS production.12,13 and EMBO Long-Term Fellowship to T.-P.P. F.J.K. is supported by the German Research Foundation (DFG) Mutations in several genes encoding guanine exchange factors Research Unit 2488, 'ProtectMove’.J.M.P.issupportedby and GTPase-activating proteins that regulate the on/off state The Austrian Academy of Sciences, the European Commun- of RAC1 have been identified to cause rare neuronal ity’s Seventh Framework Programme/Advanced ERC grant, disorders.33,34 Indeed, analysis of a protein–protein interaction and Era of Hope/DOD Excellence grant. subnetwork extracted from the human-integrated protein– protein interaction reference database35 reveals RAC1 to have 25 Study funding first-order interactors reported to be causative of intellectual No targeted funding reported. disabilities. Neuronally expressed RAC1 first-order interactors areenrichedinGTPasesignaltransductionbyGeneOntology Disclosure analysis. Furthermore, RAC1 was also identified as a candidate V. Nagy has received government research funding from the gene for intellectual disability in a meta-analysis of several Austrian Science Fund and the Lise Meitner Postdoctoral thousand trios,36 and RAC1 missense mutations were recently Fellowship; and has received foundation/society research reported to cause a neurodevelopmental disorder with variable support from the Ludwig Boltzmann Society. R. Hollstein, T.- symptoms in 7 individuals, some of which are overlapping with P. Pai, M.K. Herde, P. Buphamalai, and P. Moeseneder report SPPRS patients.37 Of interest, mutations in RAC1 caused re- no disclosures. E. Lenartowicz has received foundation/ markably different neurologic phenotypes that ranged from society research support from the Ludwig Boltzmann Society. microcephaly to normal head circumference to macrocephaly, as Anoop Kavirayani is employed with VBCFs, which receives well as variable intellectual disability (4/7), hypotonia (4/7), government funding from the Austrian Federal Ministry of epilepsy (3/7), behavioral problems (3/7), and stereotypic Education, Science, and Research and the City of Vienna. G. movements (2/7).37 The individuals who were determined to Christoph Korenke, I. Kozieradzki, R. Nitsch, A. Cicvaric, F.J. carry gain-of-function RAC1 mutations (3/7) had normal head Monje Quiroga, M.A. Deardorff, E.C. Bedoukian, Y. Li, and G. circumference (1/3) or megalocephaly (2/3), in contrast to the Yigit report no disclosures. J¨org Menche has received

12 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG government research support from the Vienna Science and Technology Fund. E. Ferda Perçin and B. Wollnik report no Appendix (continued) disclosures. Christian Henneberger has served on the editorial Name Location Role Contribution board for Frontiers in Molecular Neuroscience and Brain Research Ewelina LBI-RUD, Vienna, Author Assisted with Bulletin; has received government research support from NRW- Lenartowicz, Austria behavioral R¨uckkehrerprogramm; and has received foundation/society re- MSc experiments. Assisted with search support from the Human Frontiers Science Program. F.J. manuscript Kaiser has received government research support from the editing. German Research Foundation. J.M. Penninger reports no dis- Anoop IMBA, Institute of Author Coordinated and Kavirayani, Molecular analyzed mouse closures. Go to Neurology.org/NG for full disclosures. DVM, Dip. Biotechnology of the histologic findings. ACVP Austrian Academy of Assisted with Sciences, VBC—Vienna manuscript editing. Publication history BioCenter Campus, Received by Neurology: Genetics July 19, 2018. Accepted in final form Austria

February 5, 2019. Georg C. Klinikum Oldenburg, Author Diagnosed and Korenke, MD Oldenburg, Germany treated the patients, obtained consents and/or biomaterials, and Appendix Authors analyzed patient clinical data. Name Location Role Contribution Assisted with manuscript editing. Vanja Nagy, IMBA, Institute of First and co- Conceptualized PhD Molecular corresponding and coordinated Ivona IMBA, Institute of Author Assisted with Biotechnology of the author the project. Kozieradzki, Molecular Western blotting. Austrian Academy of Performed all MSc Biotechnology of the Assisted with Sciences, VBC—Vienna mouse behavioral Austrian Academy of manuscript BioCenter Campus, experiments, Sciences, VBC—Vienna editing. Austria and LBI-RUD, analyzed the data, BioCenter Campus, Vienna, Austria and wrote the Austria manuscript. Assisted with Roberto IMBA, Institute of Author Assisted with early manuscript editing. Nitsch, PhD Molecular conceptualization Biotechnology of the of the project and Ronja University of Luebeck, Author Performed all Austrian Academy of provided molecular Hollstein, PhD Germany experiments Sciences, biology tools. pertaining to VBC—Vienna Assisted with human fibroblasts. BioCenter Campus, manuscript editing. She assisted in Austria, and coordinating AstraZeneca, clinical input and Gothenburg, Sweden assisted with data analysis. Assisted Ana Cicvaric, Medical University of Author Designed, with manuscript PhD Vienna, Austria performed, and editing. analyzed electrophysiologic Tsung-Pin Pai, IMBA, Institute of Author Performed image experiments. PhD Molecular acquisition and Assisted with Biotechnology of the analysis of dendritic manuscript editing. Austrian Academy of spines. Assisted Sciences, VBC—Vienna with manuscript Francisco J. Medical University of Author Designed, BioCenter Campus, editing. Monje Vienna, Austria performed, and Austria Quiroga, PhD analyzed electrophysiologic Michel K. University of Bonn Author Designed, experiments. Herde, PhD Medical School, performed, and Assisted with Germany, and Centre analyzed manuscript for electrophysiologic editing. Neuroendocrinology, experiments. Department of Assisted with Matthew A. Univ. of Penn Author Diagnosed and Physiology, School of manuscript editing. Deardorff, MD Perelman School of treated the Biomedical Sciences, Med., and Children’s patients, University of Otago, Hospital of obtained Dunedin, New Philadelphia consents and/or Zealand biomaterials, and analyzed patient Pisanu CeMM, Vienna, Author Performed and clinical data. Buphamalai, Austria analyzed the RAC1 Assisted with MSc network analysis. manuscript Assisted with editing. manuscript editing. Emma C. Children’s Hospital of Author Diagnosed and Bedoukian, Philadelphia, PA treated the Paul IMBA, Institute of Author Assisted with MS patients, obtained Moeseneder, Molecular behavioral consents and/or MSc Biotechnology of the experiments. biomaterials, and Austrian Academy of Assisted with analyzed patient Sciences, manuscript clinical data. VBC—Vienna editing. Assisted with BioCenter Campus, manuscript Austria editing.

Continued

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 13 3. Diskin SJ, Capasso M, Schnepp RW, et al. Common variation at 6q16 within HACE1 fl Appendix (continued) and LIN28B in uences susceptibility to neuroblastoma. Nat Genet 2012;44: 1126–1130. Name Location Role Contribution 4. Tortola L, Nitsch R, Bertrand MJ, et al. The tumor suppressor Hace1 is a critical regulator of TNFR1-mediated cell fate. Cell Rep 2016;15:1481–1492. 5. Akawi N, McRae J, Ansari M, et al. Discovery of four recessive developmental dis- Yun Li, PhD University Medical Author Analyzed exome ¨ orders using probabilistic genotype and phenotype matching among 4,125 families. Center Gottingen, data of family 2, – Germany identified the Nat Genet 2015;47:1363 1369. fi causative mutation, 6. Hollstein R, Parry DA, Nalbach L, et al. HACE1 de ciency causes an autosomal and confirmed recessive neurodevelopmental syndrome. J Med Genet 2015;52:797–803. cosegregation in the 7. Hariharan N, Ravi S, Pradeep BE, et al. A novel loss-of-function mutation in HACE1 is family. Assisted with linked to a genetic disorder in a patient from India. Hum Genome 2018;5:17061. manuscript editing. 8. Torrino S, Visvikis O, Doye A, et al. The E3 ubiquitin-ligase HACE1 catalyzes the ubiquitylation of active Rac1. Dev Cell 2011;21:959–965. Goekhan University Medical Author Analyzed exome 9. Castillo-Lluva S, Tan CT, Daugaard M, Sorensen PH, Malliri A. The tumour sup- Yigit, PhD Center Goettingen, data of family 2, pressor HACE1 controls cell migration by regulating Rac1 degradation. Oncogene Goettingen, Germany identified the 2013;32:1735–1742. causative mutation, ff ff and confirmed 10. Luo L, Hensch TK, Ackerman L, Barbel S, Jan LY, Jan YN. Di erential e ects of the Rac – cosegregation in the GTPase on Purkinje cell axons and dendritic trunks and spines. Nature 1996;379:837 840. family. Assisted with 11. Stankiewicz TR, Linseman DA. Rho family GTPases: key players in neuronal de- manuscript editing. velopment, neuronal survival, and neurodegeneration. Front Cel Neurosci 2014;8:314. 12. Daugaard M, Nitsch R, Razaghi B, et al. Hace1 controls ROS generation of vertebrate Jorg¨ Menche, CeMM, Vienna, Author Performed and Rac1-dependent NADPH oxidase complexes. Nat Commun 2013;4:2180. PhD Austria analyzed the RAC1 13. Rotblat B, Southwell AL, Ehrnhoefer DE, et al. HACE1 reduces oxidative stress and network analysis. mutant Huntington toxicity by promoting the NRF2 response. Proc Natl Acad Sci U S Assisted with A 2014;111:3032–3037. manuscript 14. Iimura A, Yamazaki F, Suzuki T, Endo T, Nishida E, Kusakabe M. The E3 ubiquitin editing. ligase Hace1 is required for early embryonic development in Xenopus laevis. BMC Dev Biol 2016;16:31. E. Ferda Gazi University, Author Diagnosed and 15. Carola V, D’Olimpio F, Brunamonti E, Mangia F, Renzi P. Evaluation of the elevated Percin, MD, Ankara, Turkey treated the fi PhD patients, obtained plus-maze and open- eld tests for the assessment of anxiety-related behaviour in consents and/or inbred mice. Behav Brain Res 2002;134:49–57. biomaterials, and 16. Deacon RM. Measuring motor coordination in mice. J Vis Exp 2013:e2609. analyzed patient 17. Metz GA, Whishaw IQ. The ladder rung walking task: a scoring system and its clinical data. practical application. J Vis Exp 2009:1204. Assisted with 18. Deacon RM, Rawlins JN. T-maze alternation in the rodent. Nat Protoc 2006;1:7–12. manuscript editing. 19. Wehner JM, Radcliffe RA. Cued and contextual fear conditioning in mice. Curr Protoc Neurosci 2004, chapter 8:Unit 8.5C. Bernd University Medical Author Analyzed exome 20. Vorhees CV, Williams MT. Assessing spatial learning and memory in rodents. ILAR J Wollnik, MD Center Goettingen, data of family 2, 2014;55:310–332. Goettingen, Germany identified the fi causative mutation, 21. Breuss MW, Hansen AH, Landler L, Keays DA. Brain-speci c knockin of the path- and confirmed ogenic Tubb5 E401K allele causes defects in motor coordination and prepulse in- cosegregation in the hibition. Behav Brain Res 2017;323:47–55. family. Assisted with 22. Minge D, Senkov O, Kaushik R, et al. Heparan sulfates support pyramidal cell ex- manuscript editing. citability, synaptic plasticity, and context discrimination. Cereb Cortex 2017;27: 903–918. Christian University of Bonn Author Designed, 23. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image Henneberger, Medical School, Bonn, performed, and analysis. Nat Methods 2012;9:671–675. MD Germany, and analyzed 24. Kochlamazashvili G, Henneberger C, Bukalo O, et al. The extracellular matrix mol- University College electrophysiologic ecule hyaluronic acid regulates hippocampal synaptic plasticity by modulating post- London, London, UK, experiments. synaptic L-type Ca(2+) channels. Neuron 2010;67:116–128. and German Center Assisted with for manuscript editing. 25. Regehr WG. Short-term presynaptic plasticity. Cold Spring Harb Perspect Biol 2012; Neurodegenerative 4:a005702. Diseases (DZNE), 26. Bliss TV, Gardner-Medwin AR. Long-lasting potentiation of synaptic transmission in Bonn, Germany the dentate area of the unanaesthetized rabbit following stimulation of the perforant path. J Physiol (lond) 1973;232:357–374. Frank J. University of Luebeck, Author Assisted with 27. Feng G, Mellor RH, Bernstein M, et al. Imaging neuronal subsets in transgenic mice Kasier, PhD Luebeck, Germany project expressing multiple spectral variants of GFP. Neuron 2000;28:41–51. DZHK e.V. (German coordination, 28. Bongmba OY, Martinez LA, Elhardt ME, Butler K, Tejada-Simon MV. Modulation of Center for clinical dendritic spines and synaptic function by Rac1: a possible link to Fragile X syndrome Cardiovascular collaborations, data pathology. Brain Res 2011;1399:79–95. Research), Partner analysis, and 29. Nobes CD, Hall A. Rho GTPases control polarity, protrusion, and adhesion during Site Hamburg/Kiel/ writing the – Lubeck,¨ Lubeck,¨ manuscript. cell movement. J Cel Biol 1999;144:1235 1244. Germany Assisted with 30. Kraynov VS, Chamberlain C, Bokoch GM, Schwartz MA, Slabaugh S, Hahn KM. Lo- manuscript editing. calized Rac activation dynamics visualized in living cells. Science 2000;290:333–337. 31. Liu S, Kapoor M, Leask A. Rac1 expression by fibroblasts is required for tissue repair in Josef M. IMBA, Vienna, Austria Co- Conceived the vivo. Am J Pathol 2009;174:1847–1856. Penninger, Department of corresponding study and wrote 32. Kaverina I, Krylyshkina O, Small JV. Regulation of substrate adhesion dynamics MD Medical Genetics, Life author the manuscript during cell motility. Int J Biochem Cel Biol 2002;34:746–761. Science Institute, together with VN. 33. van Bokhoven H. Genetic and epigenetic networks in intellectual disabilities. Annu University of British Assisted with Rev Genet 2011;45:81–104. Columbia, Vancouver, manuscript editing. 34. Pengelly RJ, Greville-Heygate S, Schmidt S, et al. Mutations specific to the Rac-GEF domain Canada of TRIO cause intellectual disability and microcephaly. J Med Genet 2016;53;735–742. 35. Alanis-Lobato G, Andrade-Navarro MA, Schaefer MH. HIPPIE v2.0: enhancing meaningfulness and reliability of protein-protein interaction networks. Nucleic Acids Res 2017;45:D408–D414. References 36. Lelieveld SH, Reijnders MR, Pfundt R, et al. Meta-analysis of 2,104 trios provides 1. Anglesio MS, Evdokimova V, Melnyk N, et al. Differential expression of a novel support for 10 new genes for intellectual disability. Nat Neurosci 2016;19:1194–1196. ankyrin containing E3 ubiquitin-protein ligase, Hace1, in sporadic Wilms’ tumor 37. Reijnders MRF, Ansor NM, Kousi M, et al. RAC1 missense mutations in de- versus normal kidney. Hum Mol Genet 2004;13:2061–2074. velopmental disorders with diverse phenotypes. Am J Hum Genet 2017;101:466–477. 2. Zhang L, Anglesio MS, O’Sullivan M, et al. The E3 ligase HACE1 is a critical chro- 38. Zamboni V, Armentano M, Berto G, et al. Hyperactivity of Rac1-GTPase pathway mosome 6q21 tumor suppressor involved in multiple cancers. Nat Med 2007;13: impairs neuritogenesis of cortical neurons by altering actin dynamics. Sci Rep 2018;8: 1060–1069. 7254.

14 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG ARTICLE OPEN ACCESS Novel pathogenic XK mutations in McLeod syndrome and interaction between XK protein and chorein

Yuka Urata, MD, Masayuki Nakamura, MD, PhD, Natsuki Sasaki, MD, PhD, Nari Shiokawa, MD, PhD, Correspondence Yoshiaki Nishida, MD, Kaoru Arai, MD, Hanae Hiwatashi, MS, Izumi Yokoyama, BS, Shinsuke Narumi, MD, Dr. Nakamura nakamu36@ Yasuo Terayama, MD, PhD, Takenobu Murakami, MD, PhD, Yoshikazu Ugawa, MD, PhD, Hiroki Sakamoto, MD, m.kufm.kagoshima-u.ac.jp Satoshi Kaneko, MD, PhD, Yusuke Nakazawa, MD, Ryo Yamasaki, MD, PhD, Shoko Sadashima, MD, Toshiaki Sakai, MD, Hiroaki Arai, MD, and Akira Sano, MD, PhD

Neurol Genet 2019;5:e328. doi:10.1212/NXG.0000000000000328 Abstract Objective To identify XK pathologic mutations in 6 patients with suspected McLeod syndrome (MLS) and a possible interaction between the chorea-acanthocytosis (ChAc)- and MLS-responsible proteins: chorein and XK protein.

Methods Erythrocyte membrane proteins from patients with suspected MLS and patients with ChAc, ChAc mutant carriers, and normal controls were analyzed by XK and chorein immunoblotting. We performed mutation analysis and XK immunoblotting to molecularly diagnose the patients with suspected MLS. Lysates of cultured cells were co-immunoprecipitated with anti-XK and anti-chorein antibodies.

Results All suspected MLS cases were molecularly diagnosed with MLS, and novel mutations were identified. The average onset age was 46.8 ± 8 years, which was older than that of the patients with ChAc. The immunoblot analysis revealed remarkably reduced chorein immunoreactivity in all patients with MLS. The immunoprecipitation analysis indicated a direct or indirect chorein-XK interaction.

Conclusions In this study, XK pathogenic mutations were identified in all 6 MLS cases, including novel mutations. Chorein immunoreactions were significantly reduced in MLS erythrocyte mem- branes. In addition, we demonstrated a possible interaction between the chorein and XK protein via molecular analysis. The reduction in chorein expression is similar to that between Kell antigens and XK protein, although the chorein-XK interaction is a possibly noncovalent binding unlike the covalent Kell-XK complex. Our results suggest that reduced chorein levels following lack of XK protein are possibly associated with molecular pathogenesis in MLS.

From the Department of Psychiatry (Y. Urata, M.N., N. Sasaki, N. Shiokawa, Y. Nishida, K.A., H.H., I.Y., A.S.), Kagoshima University Graduate School of Medical and Dental Sciences; Department of Neurology and Gerontology (S.N., Y.T.), Iwate Medical University, Morioka; Department of Neurology (T.M.), School of Medicine, Fukushima Medical University; Department of Neuro-regeneration, Department of Neurology (Y. Ugawa), School of Medicine, Fukushima Medical University; Department of Neurology (H.S., S.K.), Kansai Medical University, Hirakata; Department of Neurology (Y. Nakazawa, R.Y., S.S.), Neurological Institute, Graduate School of Medical Sciences, Kyushu University, Fukuoka; Department of Neurology (T.S.), Nagano Matsushiro General Hospital; and Department of General Medicine (H.A.), Nagano Matsushiro General Hospital, Japan.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing Charge was paid for by the Authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the 2019 American Academy of Neurology. 1 Glossary MLS = McLeod syndrome; ChAc = chorea-acanthocytosis; NA = neuroacanthocytosis.

Neuroacanthocytosis (NA) syndromes are rare neurodegen- Sweden) and rabbit anti-XK protein (HPA019036; Atlas erative disorders exhibiting neurologic abnormalities and Antibodies) primary antibodies, which show no cross- erythrocyte acanthocytosis. The core NA syndromes are reactivity with spectrin. Donkey anti-rabbit IgG, HRP- characterized by degeneration of the striatum and hunting- linked whole Ab (GE Health care, Little Chalfont, England) tonism. They comprise 2 main diseases: chorea-acanthocytosis and VeriBlot for IP Detection Reagent (HRP) (ab131366; (ChAc) and McLeod syndrome (MLS). ChAc is caused by Abcam, Cambridge, UK) were used as secondary antibodies. loss-of-function mutations in VPS13A,1,2 leading to an absent Proteins were visualized using ECL Prime Western Blotting or markedly reduced level of the encoding protein, chorein.3,4 Detection Reagent (GE Health care), and images were MLS is caused by loss-of-function mutations in the XK,leading recorded with a digital analyzer (FUSION-SOLO.7S.WL; to absent XK protein.5 Although later onset and cardiomyop- Vilber Lourmat, Marne-la-Vall´ee, France). athy may occur predominantly in MLS,6 the 2 diseases share almost their entire symptomology in the CNS and erythrocyte Standard protocol approvals, registrations, membrane. Although molecular interactions are assumed to and patient consents exist between these diseases, no studies have as yet established Genomic DNAs and/or proteins from peripheral blood a direct association. samples were taken from all participants who provided written informed consent. The research protocol and consent form were approved by the Institutional Review Boards of Methods Kagoshima University.

Human samples and mutation analysis Data availability statement All 6 patients with suspected MLS were Japanese males with The data sets generated during and/or analyzed during the clinically suspected NA (table). Six healthy male controls and current study are available from the corresponding author on 6 male patients with ChAc were matched to suspected MLS reasonable request. cases by age. A further 6 heterozygous ChAc mutant carriers were used for the analyses. Lymphoblastoid cell lines from MLS_17 and a healthy control were established by SRL Results (Tokyo, Japan). Molecular diagnosis and clinical features of Coding and flanking regions of XK (NC_000023.10) and MLS cases VPS13A (NC_000009.11) were analyzed by Sanger se- For all suspected MLS cases, XK protein immunoreactivity quencing on an ABI PRISM 3100 Avant Genetic Analyzer was lacking in the immunoblot analysis of the erythrocyte (Thermo Fisher Scientific, Waltham, MA).3 In the case of membrane (figure 1A). Clinical symptoms and pathologic XK MLS_6, we performed a whole-genome sequence, long-range mutations are presented in the table In MLS_6, compre- PCR covering the deletion region, and Sanger sequencing. hensive mutation analysis revealed a mutation, which was a combination of a gross deletion and an insertion Immunoprecipitation and (figure 1, B–D). immunoblot analysis Co-immunoprecipitation (co-IP) and reverse co-IP assays Chorein immunoreactivity reductions in all were performed using Dynabeads Protein G (Thermo Fisher MLS cases Scientific). K562 and HEK293 cells that stably overexpressed We found a marked reduction in chorein immunoreactivity in chorein8 were lysed with Mammalian Protein Extraction Re- all patients with MLS (figure 2A). The mean density level of agent (Thermo Fisher Scientific). K562 cells that were sub- patients with MLS was significantly lower (p = 0.00127, d = cultured at 1 × 106 cells/mL and incubated for 24 hours were 2.6) at 0.55, relative to controls (figure 2D). The average used. The cell lysates (input) were used for the Dynabeads- reductions in the levels of chorein immunoreactivity in the antibody complex and Dynabeads-IgG complex. The cell ly- erythrocyte membranes of MLS patients were equivalent to sate was diluted 5 times with 1× Tris-buffered saline because those found in heterozygous ChAc mutation carriers (figure 2, delicate surfactant conditions were required to maintain the C and D), although no pathogenic mutations were identified IP interaction. The cell lysate and each bead were incubated in VPS13A in any patients with MLS. On the other hand, the for 2 hours at room temperature. average density levels of the XK immunoreactions did not significantly differ between ChAc and ChAc mutant carriers Protein samples were analyzed by immunoblotting using and healthy controls in either the immunoblot or densito- rabbit anti-chorein (HPA021662; Atlas Antibodies, Bromma, metric analyses (figure 2, B–E). Chorein immunoreactions of

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG erlg.r/GNuooy eeis|Vlm ,Nme ue2019 June | 3 Number 5, Volume | Genetics Neurology: Neurology.org/NG

Table Profile of patients with MLS in this study

Clinical symptoms XK pathogenic mutations

Age Main Atrophy of the Age at at Initial psychiatric EEG CK corpus striatum Protein Case sampling onset symptom symptom Acanthocytes Chorea abnormality (IU/L) DTRs Cardiomyopathy on MRI Exon DNA change change

MLS_1 64 47 Muscle Insomnia + + Intermittent 5465 Absent − + Exon c.del669_ p.V225Lfs*12 weakness theta wave 3 673TGTAGinsGGTCCTCTTTACC

MLS_2 61 53 Involuntary ??+??aHyperreflexia ? + Exon c.1013delT p.F338Sfs*70 movement 3

MLS_3 69 43 Muscle Persecutory + ? ? 920 Absent + ? Exon c.451dupC p.Q151Pfs*47 weakness delusion 2

MLS_4 56 33 Involuntary Depression + + 3–4 Hz slow 821 Normal − + Exon c.370C>T p.Q124* movement wave 2

MLS_5 65 50 Involuntary Cognitive + + ? 2,422 ? ? ? Exon c.397C>T p.R133* movement decline 2

MLS_6 70 55 Involuntary Obsessiveness + + ? 1,052 Absent −− Exon Gross deletion Unknown movement 3

Abbreviations: CK = creatine kinase; DTR = deep tendon reflex. Novel mutations were identified in MLS_4 and MLS_6. Age at onset: age when first signs or symptoms appeared (yr). a When DTR tests were performed, MLS_2 was affected with bacterial meningitis. 3 Figure 1 Molecular diagnosis of 6 MLS cases

(A) The results of XK immunoblotting revealed a lack of XK immunoreac- tivity in all patients with MLS. Equal loading was shown by staining with MemCode reversible protein stain (Pierce), shown in the lower panel. (B) In the case of MLS_6, the XK gene mutation was predicted to be a gross deletion including exon 3 based on the results of gDNA amplification. To identify the breakpoints of this mu- tation, we performed a whole-ge- nome analysis. Based on the results, we performed a long-range PCR cov- ering the deletion region. The results of long-range PCR for MLS_6 showed a gross deletion mutation that was approximately 5500 bp in size. (C) Sanger sequencing results revealed a combination of a gross deletion, from intron 2 (c.509–636) to an XK- CYBB intergenic region (c.*3667 + 670), and an insertion of a comple- mentary sequence of 380 bp in intron 2. (D) The schematic shows the structure of the gross deletion region in MLS_6.

the lymphoblastoid cell lysates from MLS_1 and the control overexpressing HEK293 cells. Signals positive for chorein and were equivalent. In addition, there was no immunoreaction XK protein were detected in the XK and chorein immuno- corresponding to the XK protein band in both control and precipitants, respectively (figure 2H). MLS_1 lymphoblastoid cells (figure 2F). Chorein-XK protein interaction in Discussion cultured cells Cell lysates extracted from K562 cells were immunoprecipi- In the present study, we analyzed 6 cases with MLS and tated with anti-XK antibody. In the subsequent immunoblot confirmed the molecular diagnosis, as well as identifying 2 analysis, positive chorein bands were detected in the XK additional novel pathogenic mutations (table). The profile of immunoprecipitants (figure 2G). Because the endogenous 6 cases of MLS in this study was similar to those reported chorein level was low, XK protein immunoreactivity was not previously.6 In our MLS cases, the average onset age was 46.8 visually observed in chorein immunoprecipitants. Therefore, ± 8 years, which is approximately 13 years older than that of co-IP and reverse co-IP assays were conducted in a similar patients with ChAc.3 The disease duration for MLS may be manner using the lysate extracted from chorein stably longer than 30 years, which is typically longer than for ChAc.6

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Figure 2 Immunoblotting of erythrocyte membranes, lymphoblastoid cells, and co-immunoprecipitants

(A and C) Chorein immunoblotting results revealed reduced chorein immunoreactivity in the erythrocyte membranes of patients with MLS, which are equivalent to the heterozygous mutant carriers of ChAc (ChAc_hetero). (B and C) XK immunoblotting results revealed normal XK immunoreactivity in the erythrocyte membranes of ChAc and ChAc_hetero. (A–C) Each imaging was performed by underexposed condition for semiquantitative analysis. (D) The histograms show the chorein relative density ratio in patients with MLS and ChAc and ChAc_hetero. Each band density was normalized by protein density staining with MemCode reversible protein stain. Error bars represent the SD (each n = 6); 2-sample Student t tests were performed accordingly on different sets of data. **p < 0.01; ***p < 0.001. d shows the effect size (Cohen’s d). (E) The histograms show the XK protein relative density ratio in patients with MLS and ChAc and ChAc_hetero. Each band density was normalized by protein density staining with memcode reversible protein stain. (F) Chorein immunoblotting results revealed equivalent chorein immunoreactivity in the lymphoblastoid cells of MLS_1 and the control (MLS_1_lympho and Control_lympho, re- spectively). The XK immunoblotting of lymphoblastoid cells shown in the lower panel of figure 2F reveals a lack of XK immunoreactivity. The Control_EM and MLS_EM lanes show the erythrocyte membrane for control and MLS, respectively. (G) Co-immunoprecipitation (IP) assay using K562 cells was performed with anti-XK antibody. Immunoblot analyses used anti-chorein (upper panel) and anti-XK antibodies (lower panel). (H) Co-IP and reverse co-IP assays using human embryonic kidney 293 (HEK293) cells stably overexpressing Myc-DDK–tagged chorein confirmed an interaction between XK protein and chorein. Immunoblot analyses used anti-XK protein and anti-chorein antibodies.

In the present study, semiquantitative chorein immunoblot- of immunoblotting analysis might be unavailable for ting using erythrocyte membranes from all patients with MLS semiquantification. revealed significantly reduced chorein immunoreactions compared with age- and sex-matched healthy controls (figure In the present study, XK immunoblotting of lymphoblastoid 2A). Chorein immunoreactivities in heterozygous ChAc cell lysate from healthy controls showed no XK protein band, mutation carriers are also reduced to the same level as in suggesting no expression of XK protein in lymphoblastoid patients with MLS (figure 2, C and D). These findings were cells. This may account for the normal chorein immuno- demonstrated in at least triplicate independent experiments. reaction found by chorein immunoblotting of lymphoblastoid Some ChAc mutation carriers exhibit partial symptoms of NA cell lysate from MLS_1, although further investigation is such as acanthocytosis.9 Taken together, the later onset and required. slower progression found for MLS compared with ChAc suggest that the chorein level reductions found in MLS may XK protein covalently interacts with Kell antigens, which are be directly associated with MLS molecular pathology. The remarkably reduced in erythrocyte membranes of MLS erythrocyte membrane from 1 patient with MLS and lym- patients.6 In this study, based on the finding of reduced phoblastoid cells from another patient with MLS showed chorein in the erythrocyte membranes of patients with MLS, normal chorein levels in previous study.4 In that study, we hypothesized that the XK protein directly or indirectly chorein immunoblotting of heterozygous ChAc mutant car- interacts with chorein. In the present study, we performed IP riers showed normal chorein levels, suggesting that the results assays, which revealed the possible interaction. In erythrocyte

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 membranes, the absence of XK led to reduced chorein levels, although the absence of chorein was unrelated to XK levels. Appendix (continued)

Computational analysis revealed a number of the corre- Name Location Role Contribution sponding impaired phosphorylation pathways in MLS and ChAc, suggesting a common molecular background bridging Nari Kagoshima Author Advised on 10 Shiokawa, University laboratory work the generation of acanthocytes. In the present study, protein MD, PhD Graduate School of staining on blotting membranes revealed upshift in band 3 from Medical and Dental Sciences, both MLS and ChAc (data not shown), suggesting results of Kagoshima phosphorylation. Taken together, these results suggests that Yoshiaki Kagoshima Author Advised on data reduced chorein is associated with MLS phosphorylation- Nishida, MD University analysis related molecular pathology in the erythrocyte membranes. Graduate School of Medical and Dental However, the direct mechanisms of reduced chorein in eryth- Sciences, rocytes of MLS are unknown. In addition, our study did not Kagoshima include molecular investigations of the CNS. Further studies Kaoru Arai, Kagoshima Author Advised on data are needed to elucidate the molecular mechanism of NA. MD University analysis Graduate School of Medical and Dental Acknowledgment Sciences, The authors thank the patients with MLS, patients with ChAc, Kagoshima

ChAc mutant carriers, and healthy controls for their participa- Hanae Kagoshima Author Performed tion, and Ms. Kyoko Meguro for her technical assistance. Hiwatashi, University laboratory work MS Graduate School of Medical and Dental Study funding Sciences, This study was supported by Grants-in Aid from the Research Kagoshima Committee of CNS Degenerative Diseases, Research on Izumi Kagoshima Author Performed Policy Planning and Evaluation for Rare and Intractable Yokoyama, University laboratory work BS Graduate School of Diseases, Health, Labour and Welfare Sciences Research Medical and Dental Grants, and the Ministry of Health, Labour and Welfare, Ja- Sciences, Kagoshima pan, and in part by the Ministry of Education, Culture, Sports, Science and Technology KAKENHI (Grant No. 17H04250 Shinsuke Iwate Medical Author Collected clinical Narumi, MD University, Morioka data and blood to A.S. and No. 18K07606 to M.N.). samples of study patients

Disclosure Yasuo Iwate Medical Author Collected clinical Disclosures available: Neurology.org/NG. Terayama, University, Morioka data and blood MD, PhD samples of study patients Publication history Received by Neurology: Genetics November 18, 2018. Accepted in final Takenobu Fukushima Medical Author Collected clinical Murakami, University, data and blood form March 7, 2019. MD, PhD Fukushima samples of study patients

Yoshikazu Fukushima Medical Author Collected clinical data Ugawa, MD, University, and blood samples of Appendix Authors PhD Fukushima study patients

Name Location Role Contribution Hiroki Kansai Medical Author Collected clinical Sakamoto, University, Hirakata data and blood Yuka Urata, Kagoshima Author Performed MD samples of study MD University laboratory work and patients Graduate School of data analysis and Medical and Dental prepared the Satoshi Kansai Medical Author Collected clinical Sciences, manuscript Kaneko, University, Hirakata data and blood Kagoshima MD, PhD samples of study patients Masayuki Kagoshima Corresponding Supervised the Nakamura, University author project, advised on Yusuke Kyushu University, Author Collected clinical MD, PhD Graduate School of laboratory work and Nakazawa, Fukuoka data and blood Medical and Dental data analysis, and MD samples of study Sciences, prepared the patients Kagoshima manuscript Ryo Kyushu University, Author Collected clinical data Natsuki Kagoshima Author Advised on Yamasaki, Fukuoka and blood samples of Sasaki, MD, University laboratory work MD, PhD study patients PhD Graduate School of Medical and Dental Shoko Kyushu University, Author Collected clinical data Sciences, Sadashima, Fukuoka and blood samples of Kagoshima MD study patients

6 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG References Appendix (continued) 1. Ueno S, Maruki Y, Nakamura M, et al. The gene encoding a newly discovered protein, chorein, is mutated in chorea-acanthocytosis. Nat Genet 2001;28:121–122. Name Location Role Contribution 2. Rampoldi L, Dobson-Stone C, Rubio JP, et al. A conserved sorting-associated protein is mutant in chorea-acanthocytosis. Nat Genet 2001;28:119–120. 3. Tomiyasu A, Nakamura M, Ichiba M, et al. Novel pathogenic mutations and copy Toshiaki Nagano Matsushiro Author Collected clinical data number variations in the VPS13A Gene in patients with chorea-acanthocytosis. Am J Sakai, MD General Hospital, and blood samples of Med Genet B 2011;156:620–631. Nagano study patients 4. Dobson-Stone C, Velayos-Baeza A, Filippone LA, et al. Chorein detection for the diagnosis of Chorea-acanthocytosis. Ann Neurol 2004;56:299–302. Hiroaki Nagano Matsushiro Author Collected clinical data 5. Ho M, Chelly J, Carter N, Danek A, Crocker P, Monaco AP. Isolation of the gene for McLeod Arai, MD General Hospital, and blood samples of syndrome that encodes a novel membrane transport protein. Cell 1994;77:869–880. Nagano study patients 6. Jung HH, Danek A, Walker RH. Neuroacanthocytosis syndromes. Orphanet J Rare Dis 2011;6:68. Akira Sano, Kagoshima Author Advised on data 7. Narumi S, Natori T, Miyazawa H, et al. Case of McLeod syndrome with a novel MD, PhD University Graduate analysis, prepared genetic mutation. Neurol Clin Neurosci 2016;4:115–117. School of Medical the manuscript, and 8. Shiokawa N, Nakamura M, Sameshima M, et al. Chorein, the protein responsible for and Dental Sciences, served as a scientific chorea-acanthocytosis, interacts with β-adducin and β-actin. Biochem Biophys Res Kagoshima advisor Commun 2013;441:96–101. 9. Ichiba M, Nakamura M, Kusumoto A, et al. Clinical and molecular genetic assessment of a chorea-acanthocytosis pedigree. J Neurol Sci 2007;263:124–132. 10. De Franceschi L, Scardoni G, Tomelleri C, et al. Computational identification of phospho-tyrosine sub-networks related to acanthocyte generation in neuro- acanthocytosis. PLoS One 2012;7:e31015.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 7 ARTICLE OPEN ACCESS Brain-derived neurotrophic factor, epigenetics in stroke skeletal muscle, and exercise training

Alice S. Ryan, PhD, Huichun Xu, PhD, Frederick M. Ivey, PhD, Richard F. Macko, MD, and Correspondence Charlene E. Hafer-Macko, MD Dr. Ryan [email protected] Neurol Genet 2019;5:e331. doi:10.1212/NXG.0000000000000331 Abstract Objective (1) To compare paretic (P) vs nonparetic (NP) skeletal muscle brain-derived neurotrophic factor (BDNF) and the effects of resistive training (RT) on systemic and skeletal muscle BDNF mRNA expression in stroke; and (2) to compare the DNA methylation profile for BDNF and BDNFAS (BDNF antisense RNA) between P and NP muscle and the effects of aerobic exercise training (AEX) on DNA methylation in stroke.

Methods In this longitudinal investigation, participants (50–76 years) with chronic stroke underwent a fasting blood draw, a 12-week (3×/week) RT intervention (n = 16), and repeated bilateral vastus lateralis muscle tissue biopsies (n = 10) with BDNF expression determined by RT-PCR. Five stroke survivors completed 6 months of AEX (3×/week) and had bilateral muscle biopsies. DNA methylation status in gene BDNF and BDNFAS was assessed by Illumina 450k meth- ylation array.

Results P muscle had ;45% lower BDNF mRNA expression than NP muscle (6.79 ± 1.30 vs 10.52 ± 2.06 arbitrary units [AU], p < 0.05), and P muscle exhibited differential methylation status in the DNA sequences of BDNF (3 CpG [59-C-phosphate-G-39] sites, p = 0.016–0.044) and BDNFAS (1 CpG site, p = 0.016) compared to NP. Plasma BDNF and muscle BDNF mes- senger RNA (mRNA) expression did not significantly change after RT. BDNFAS DNA methylation increased after AEX in P relative to NP muscle (p = 0.017).

Conclusions This is the first evidence that stroke hemiparesis reduces BDNF skeletal muscle expression, with our findings identifying methylation alterations on the DNA sequence of BDNF and BDNFAS gene. Preliminary results further indicate that AEX increases methylation in BDNFAS gene, which presumably could regulate the expression of BDNF.

From the VA Maryland Health Care System, Research Service, Department of Medicine, Division of Gerontology and Geriatric Medicine, Department of Neurology, at the University of Maryland School of Medicine, and the Baltimore Geriatric Research, Education and Clinical Center (GRECC), MD.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing Charge was funded by the Veterans Affairs. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary AEX = aerobic exercise training; BDNF = brain-derived neurotrophic factor; BDNFAS = BDNF Antisense RNA; NP = non- paretic; P = paretic; RT = resistive training.

It is well established that stroke leads to long-term disability and leg curl machines (Keiser; pneumatic resistance, Fresno, CA) cognitive dysfunction. Physical activity improves motor and at every session.3 cognitive ability perhaps, in part, due to enhanced neurotrophic factor signaling. Brain-derived neurotrophic factor (BDNF) is Aerobic exercise training a protein that promotes neuroplasticity, including the growth The progressive treadmill protocol consisted of three 40- and survival of neurons, regulation of axonal and dendritic minute sessions per week at a target aerobic intensity of branching, and synaptic transmission.1 BDNF may mediate the 60%–70% heart rate reserve performed over a 6-month improvement in cognition after acute exercise.2 Yet, it is un- training period. Training started at low intensity (40%–50% known whether BDNF changes with exercise training in neu- heart rate reserve) for 10–20 minutes and gradually pro- 4 rologic conditions, such as chronic stroke. We hypothesize that gressed to target levels. BDNF is involved in the skeletal muscle changes poststroke and that exercise (resistive training [RT] or aerobic exercise training Plasma and muscle analyses [AEX]) exerts beneficial effects through epigenetic regulation of Plasma BDNF was measured in duplicate using Human BDNF expression. Therefore, in this 2-part preliminary in- BDNF Emax ImmunoAssay Kit. Vastus lateralis biopsies of vestigation, we sought to compare paretic (P) vs nonparetic the P and NP muscle were performed under local anesthesia (NP) skeletal muscle BDNF and the effects of RT on systemic after a 12-hour fast at baseline and after RT (n = 10) and after – and skeletal muscle BDNF mRNA expression in stroke and to AEX (n = 5), 24 36 hours after the last bout of exercise. − compare the DNA methylation profile for BDNF and BDNF Muscle was immediately freeze-clamped and stored at 80°C. RNA extraction, reverse transcription, and quantitative real- antisense RNA (BDNFAS) between P and NP muscle and 3 assess the effectsofAEXonDNAmethylationinstroke. time PCR for gene expression of BDNF was conducted. The Illumina 450k methylation array was used for DNA methyl- ation according to manufacturer’s protocol. Briefly, DNA Methods purified from muscle was bisulfite-converted using EZ-96 DNA methylation kit (D5004, Zymo Research), amplified, Twenty-one participants who had a prior stroke aged 43–81 and then enzymatically fragmented followed by hybridization 2 years with body mass index between 21 and 39 kg/m who with the 450k array and fluorescent staining. had mild-to-moderate hemiparetic gait deficits and completed conventional rehabilitation therapy provided written informed Statistical analyses consent to participate in this longitudinal investigation at the Baseline BDNF levels between P and NP thigh muscle was Baltimore Veterans Affairs Medical Center and the University determined using paired Student t tests. Changes were of Maryland School of Medicine. Evaluations included medical assessed using repeated measures analysis of variance using history, physical examination, and fasting blood profile. Partic- SPSS 22.0. The 450k array image file (IDAT files) processed ipants with stroke were excluded for unstable angina, congestive using the minfi package to calculate intensity and perform heart failure (NYHA II), severe peripheral arterial disease, Subset-quantile within array normalization method was used major poststroke depression, dementia, severe receptive apha- to reduce the technical variability between type I and type II sia, and orthopedic or chronic pain conditions. All methods and assay designs. Methylation scores are calculated as β values, procedures were approved by the institutional review board of ranging from 0 (totally unmethylated) to 1 (totally methyl- the University of Maryland and the VA R&D Committee. ated). M values were derived as logit (β), which have prop- erties suitable for statistical tests. Differential methylation Tests between P and NP thigh muscle and between post- and pre- Height, weight, fat mass, lean tissue mass, and %body fat were AEX thigh muscle was evaluated by paired Student t test on determined by dual-energy x-ray absorptiometry (Prodigy the M values. Data are presented as means ± SEM. p values < LUNAR GE version 7.53.002). Exercise testing with open- 0.05 are statistically significant. circuit spirometry was conducted to measure peak aerobic capacity (VO2 peak) using a graded treadmill test. One- Data availability statement repetition maximum strength tests assessed leg press and leg De-identified data and the study protocol may be available. extension strength of each leg (Keiser, Fresno, CA). More information regarding the data is available from the corresponding author on reasonable request. Resistive training Participants trained 3×/week for 12 weeks, performing 2 sets Clinical trials identifiers of 20 unilateral repetitions on the leg press, leg extension, and NCT02347995, NCT00387712, NCT00891514.

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Figure 1 P and NP skeletal muscle brain-derived neuro- Table 2 Paretic skeletal muscle BDNF methylation after trophic factor mRNA expression AEX

Probeset ID Gene β (pre-AEX) β (post-AEX) p Value

cg22830701 BDNFAS 0.042 0.068 0.017

cg22973087 BDNFAS/ 0.029 0.039 0.019 LIN7C

Abbreviations: AEX = aerobic exercise training; BDNFAS = BDNF antisense RNA.

relative to NP muscle (p = 0.017, figure 2). DNA methylation in BDNFAS did not change significantly in NP muscle.

Discussion * p < 0.05. P = paretic; NP = nonparetic. Our results provide the first preliminary evidence of reduced BDNF expression in P skeletal muscle as well as differential Results methylation status for BDNF and BDNFAS between P and NP muscle and suggests that AEX modulates methylation for Resistive training BDNF in P muscle. We are unaware of any studies examining Leg press strength increased by 26% for the P leg (304 ± 40 vs BDNF expression in skeletal muscle of stroke patients, or 363 ± 41 lbs derived from pneumatic resistance equipment) and changes with exercise training in this population. The differ- 25% for NP leg (449 ± 36 vs 560 ± 33 lbs) (both p < 0.0001) ence in BDNF expression between the P and NP limb in both with RT. Leg extension strength increased by 47% and 25% (P: gene expression and methylation provide important evidence 57 ± 10 vs 84 ± 11 lbs, NP: 113 ± 10 vs 141 ± 9 lbs respectively, for epigenetic modifications after stroke. both p < 0.0001). Plasma BDNF protein increased by 25% with RT but this was not significant (0.64 ± 0.08 vs 0.87 ± 0.20 ng/ There is limited work in human skeletal muscle for BDNF. mL). P muscle has ;45% lower BDNF mRNA expression (p < Acute exercise increases BDNF mRNA and protein expression 0.05) than NP muscle (figure 1). Skeletal muscle BDNF mRNA 5 in young healthy men. Evidence remains inconclusive re- did not change with RT (P: 7.21 ± 1.38 vs 7.06 ± 1.85 AU and garding the effectiveness of exercise training on serum BDNF. NP: 11.39 ± 2.12 vs 7.84 ± 1.32 AU). Our results indicate that circulating BDNF did not change after – Aerobic exercise training RT. Other studies report that 8 12 weeks of RT either in- crease6 or do not change7 serum BDNF in older adults. Pos- VO2 peak increased by 18% with AEX (22.5 ± 2.5 vs 26.6 ± ff tulated mechanisms for improved cognition with exercise 3.5, p < 0.05). P and NP muscle have di erential methylation fl status for DNA sequences of BDNF (3 CpG sites, p = 0.016 to includes increased blood ow or oxidative stress leading to increased endothelial function and release of BDNF from the 0.044) and BDNFAS (1 CpG site, p = 0.016) (table 1) at 2 baseline (prior to exercise training). Table 1 also provides the brain microvascular endothelial cells. Indeed, we have shown β coefficients and the probe set ID. After AEX, there were significant changes in methylation of P muscle at 2 CpG sites Figure 2 DNA %methylation for cg22830701 BDNFAS after (table 2). DNA methylation in BDNFAS increased in P aerobic exercise training, n = 5

Table 1 Baseline P and NP skeletal muscle BDNF methylation

Probeset ID Gene β (NP muscle) β (P muscle) p Value

Cg22830701 BDNFAS 0.069 0.042 0.016

Cg23426002 BDNF 0.934 0.913 0.016

Cg20108357 BDNF 0.869 0.900 0.028

Cg06260077 BDNF 0.123 0.144 0.044

Abbreviations: BDNF = brain-derived neurotrophic factor; BDNFAS = BDNF p Antisense RNA; P = paretic; NP = nonparetic. * < 0.05. P = paretic; NP = nonparetic.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 increased cerebral vasomotor reactivity after aerobic training in chronic stroke4 providing evidence of changes in cerebral blood Appendix Authors fl fl ow, which could potentially in uence BDNF levels. Name Location Role Contribution

Epigenetic regulation of numerous genes, specifically methyla- Alice S. University of Author Design and conceptualized Ryan, PhD Maryland, study; performed and tion at the BDNF gene, was reported in a recent systematic Baltimore collected the data; performed review related to neurodegenerative diseases.8 Peripheral BDNF statistical analysis; drafted the manuscript for methylation is promoted as a surrogate for central BDNF intellectual content methylation, given the direct relationships observed between Huichun University of Author Performed statistical analysis BDNF promoter methylations in the hippocampus and quad- Xu, MD, Maryland, and revised the manuscript riceps tissue of patients with bipolar disorder.9 Patients with mild PhD Baltimore

depressive disorder not only have higher BDNF promoter Frederick University of Author Performed and collected the methylation than healthy controls but also thinning of several Ivey, PhD Maryland, data; revised the manuscript cortical regions of the brain. The latter finding is inversely cor- Baltimore 10 related with BDNF promoter methylation. We are unaware of Richard University of Author Revised the manuscript fi Macko, MD Maryland, studies in chronic stroke. Our nding that methylation decreased Baltimore the gene expression of BDNFAS supports an increase in BDNF. Charlene University of Author Performed and collected the Hafer- Maryland, data; revised the manuscript Limitations of the current trial include the relatively small Macko, MD Baltimore sample size and limited human muscle sample available for analysis. In addition, a broader analysis on genome-wide DNA fi methylation and transcription pro le changes in P vs NP References muscles would be informative in future studies. However, the 1. Calabrese F, Rossetti AC, Racagni G, Gass P, Riva MA, Molteni R. Brain-derived neurotrophic factor: a bridge between inflammation and neuroplasticity. Front Cell current studies were highly controlled and included super- Neurosci 2014;8:430. vised exercise training programs and standardization of 2. Borror A. Brain-derived neurotrophic factor mediates cognitive improvements fol- muscle biopsy timing and conditions and novel with respect lowing acute exercise. Med Hypotheses 2017;106:1–5. 3. Ryan AS, Ivey FM, Prior S, Li G, Hafer-Macko C. Skeletal muscle hypertrophy and to skeletal muscle methylation in chronic stroke. Our earlier muscle myostatin reduction after resistive training in stroke survivors. Stroke 2011;42: work has established that P muscle after chronic stroke has 416–420. 4. Ivey FM, Ryan AS, Hafer-Macko CE, Macko RF. Improved cerebral vasomotor metabolic and structural differences relative to the NP reactivity after exercise training in hemiparetic stroke survivors. Stroke 2011;42: – muscle.11 15 The clinical relevance of the difference in DNA 1994–2000. 5. Matthews VB, Astr¨om MB, Chan MH, et al. Brain-derived neurotrophic factor is methylation in P vs NP muscle and following exercise training produced by skeletal muscle cells in response to contraction and enhances fat oxi- cannot be determined in this preliminary investigation. dation via activation of AMP-activated protein kinase. Diabetologia 2009;52: 1409–1418. 6. FortiLN,VanRoieE,NjeminiR,etal.Dose-andgender-specificeffects A larger randomized controlled trial is necessary to establish of resistance training on circulating levels of brain derived neurotrophic ff factor (BDNF) in community-dwelling older adults. Exp Gerontol 2015;70: the e ects of exercise on BDNF methylation and other epi- 144–149. genetic modifications in stroke skeletal muscle. Future re- 7. Walsh JJ, Scribbans TD, Bentley RF, Kellawan JM, Gurd B, Tschakovsky ME. Neu- rotrophic growth factor responses to lower body resistance training in older adults. search could also examine changes in epigenetics with exercise Appl Physiol Nutr Metab 2016;41:315–323. training and whether training-induced changes associate with 8. Wen KX, Miliç J, El-Khodor B, et al. The role of DNA methylation and histone modifications in neurodegenerative diseases: a systematic review. PLoS One 2016;11: improved cognitive ability after stroke. e0167201 9. Stenz L, Zewdie S, Laforge-Escarra T, et al. BDNF promoter I methylation correlates Study funding between post-mortem human peripheral and brain tissues. Neurosci Res 2015;91: 1–7. This study was supported by funds from VA RR&D Senior 10. Na KS, Won E, Kang J, et al. Brain-derived neurotrophic factor promoter meth- Research Career Scientist Award (ASR), NIH grants R01- ylation and cortical thickness in recurrent major depressive disorder. Sci Rep 2016; 6:21089 AG030075, VA Merit Awards, Claude D. Pepper Older Amer- 11. Ryan AS, Dobrovolny CL, Smith GV, Silver KH, Macko RF. Hemiparetic muscle icans Independence Center (P30AG028747), the Baltimore VA atrophy and increased intramuscular fat in stroke patients. Arch Phys Med Rehabil 2002;83:1703–1707. Geriatric Research, Education, and Clinical Center (GRECC). 12. Ryan AS, Ivey FM, Serra MC, Hartstein J, Hafer-Macko CE. Sarcopenia and physical function in middle-aged and older stroke survivors. Arch Phys Med Rehabil 2017;98: 495–499. Disclosure 13. Ivey FM, Hafer-Macko CE, Ryan AS, Macko RF. Impaired leg vasodilatory function Disclosures available: Neurology.org/NG. after stroke: adaptations with treadmill exercise training. Stroke 2010;41: 2913–2917. 14. Hafer-Macko CE, Yu S, Ryan AS, Ivey FM, Macko RF. Elevated tumor necrosis factor- Publication history alpha in skeletal muscle after stroke. Stroke 2005;36:2021–2023. fi 15. De Deyne PG, Hafer-Macko CE, Ivey FM, Ryan AS, Macko RF. Muscle mo- Received by Neurology: Genetics January 24, 2019. Accepted in nal form lecular phenotype after stroke is associated with gait speed. Muscle Nerve 2004; March 25, 2019. 30:209–215.

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG ARTICLE OPEN ACCESS Novel pathogenic VPS13A gene mutations in Japanese patients with chorea-acanthocytosis

Yoshiaki Nishida, MD, Masayuki Nakamura, MD, PhD, Yuka Urata, MD, Kei Kasamo, MD, Hanae Hiwatashi, MS, Correspondence Izumi Yokoyama, BS, Masahiro Mizobuchi, MD, PhD, Kotaro Sakurai, MD, PhD, Yasushi Osaki, MD, Dr. Nakamura nakamu36@ Yukari Morita, MD, Masako Watanabe, MD, PhD, Kenji Yoshida, MD, Kiyomi Yamane, MD, PhD, m.kufm.kagoshima-u.ac.jp Natsuki Miyakoshi, MD, Ryouichi Okiyama, MD, Takehiro Ueda, MD, PhD, Noritaka Wakasugi, MD, Yuji Saitoh, MD, PhD, Takashi Sakamoto, MD, PhD, Yuji Takahashi, MD, PhD, Ken Shibano, MD, PhD, Hideki Tokuoka, MD, Atsushi Hara, MD, Kazunari Monma, MD, PhD, Katsuhisa Ogata, MD, PhD, Keita Kakuda, MD, Hideki Mochizuki, MD, PhD, Takeo Arai, MD, PhD, Manabu Araki, MD, PhD, Takeshi Fujii, MD, PhD, Kazuto Tsukita, MD, Haruhi Sakamaki-Tsukita, MD, and Akira Sano, MD, PhD

Neurol Genet 2019;5:e332. doi:10.1212/NXG.0000000000000332 Abstract Objective To identify mutations in vacuolar protein sorting 13A (VPS13A) for Japanese patients with suspected chorea-acanthocytosis (ChAc).

Methods We performed a comprehensive mutation screen, including sequencing and copy number variation (CNV) analysis of the VPS13A gene, and chorein Western blotting of erythrocyte ghosts. As the results of the analysis, 17 patients were molecularly diagnosed with ChAc. In addition, we investigated the distribution of VPS13A gene mutations and clinical symptoms in a total of 39 molecularly diagnosed Japanese patients with ChAc, including 22 previously reported cases.

Results We identified 11 novel pathogenic mutations, including 1 novel CNV. Excluding 5 patients with the unknown symptoms, 97.1% of patients displayed various neuropsychiatric symptoms or forms of cognitive dysfunction during the course of disease. The patients carrying the 2 major mutations representing over half of the mutations, exon 60–61 deletion and exon 37 c.4411C>T (R1471X), were localized in western Japan.

Conclusions We identified 13 different mutations in VPS13A, including 11 novel mutations, and verified the clinical manifestations in 39 Japanese patients with ChAc.

From the Kagoshima University Graduate School of Medical and Dental Sciences (Y.N., M.N., Y.U., K. Kasamo, H.H., I.Y., A.S.), Department of Psychiatry, Kagoshima, Japan; Epilepsy Center (M.M.), Department of Neurology, Nakamura Memorial Hospital, Hokkaido, Japan; Department of Psychiatry and Neurology (K. Sakurai.), Hokkaido University Graduate School of Medicine, Hokkaido, Japan; Department of Neurology (Y.O., Y.M.), Kochi Medical School, Kochi, Japan; Shinjyuku Neuro Clinic (M.W.), Tokyo, Japan; Department of Neurology (K. Yoshida and K. Yamane), Neurological Institute, Ohta-Atami Hospital, Fukushima, Japan; Department of Neurology (N.M., R.O.), Tokyo Metropolitan Neurological Hospital, Tokyo, Japan; Division of Neurology (T.U., H.T.), Kobe University Graduate School of Medicine, Hyogo, Japan; Department of Neurology (N.W., Y.S., T.S., Y.T., M.A.), National Center of Neurology and Psychiatry Hospital, Tokyo, Japan; Department of Neurology (K. Shibano), Akita Red Cross Hospital, Japan; Amagasaki General Medical Center (A.H.), Hyogo, Japan; Department of Neurology (K.M., K.O.), National Hospital Organization Higashisaitama National Hospital, Saitama, Japan; Department of Neurology (K. Kakuda, H.M.), Graduate School of Medicine, Osaka University, Japan; Ikebe Clinic (T.A.), Shizuoka, Japan; Department of Psychiatry (T.F.), National Center Hospital, National Center of Neurology and Psychiatry, Tokyo, Japan; Department of Neurology (K.T., H.S.-T.), Tenri Hospital, Nara, Japan; and Department of Neurology (K.T., H.S.-T.), Graduate School of Medicine, Kyoto University, Japan.

Go to Neurology.org/NG for full disclosures. Funding information are provided at the end of the article.

The Article Processing Charge was funded by the authors. This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary cDNA = complementary DNA; ChAc = chorea-acanthocytosis; CNV = copy number variation; gDNA = genomic DNA; qPCR = quantitative PCR; VPS13A = vacuolar protein sorting 13A.

Chorea-acanthocytosis (ChAc) is a rare, autosomal recessive Methods neurodegenerative disease characterized by adult-onset chorea, involuntary orofacial movement, peripheral Mutation analysis acanthocytes, and various neuropsychiatric symptoms Coding and flanking regions of VPS13A (NC_000009.11) with loss-of-function mutations in vacuolar protein sorting were analyzed by Sanger sequencing on an ABI PRISM 3130 13A (VPS13A), which consists of 73 exons spanning ap- Avant Genetic Analyzer (Thermo Fisher Scientific, Waltham, 4,5 proximately 250 kb of chromosome 9q21. VPS13A encodes MA). For patients 16 and 17, we performed copy number a protein with a molecular weight of approximately 360 kDa, variation (CNV) analysis that has been previously described 4,6 named chorein.1,2 It is estimated that there are likely around in detail. 1000 ChAc cases in the world.3 Although more than 100 patients withChAchavesofarbeenreportedinJapan,thedistributionof Chorein analysis VPS13A mutations in Japan has not been conclusively de- We performed chorein Western blotting analysis that has been termined.Inthisstudy,wereport novel mutations in Japanese previously described in detail4,7 with minor modifications. We patients with ChAc. In addition, we investigate their clinical used polyvinylidene difluoride membranes from GE Health- symptoms. care (Little Chalfont, United Kingdom) or Merck Millipore

Table 1 Profiles of the patients with ChAc in this study

Pt no. AO S C Ac Ch O Ep NPS FS CK AS Res or Ori

1 35 M ND ? ? + − DI, Pica ? + + Kochi

2 26 F − + ? ? + OCS, AOP, CDc Sei + + Tokyo

3 25 F − +++− Ins IMTL + − Fukushima

4 18 F − + + + + Del, OCS, DI, CDc Sei + + Hokkaido

5 18 F − + + + + EI, Hal, FLD, CDc Sei + + Hokkaido

6 35 M − +++− DI, EFD Cho + + Tokyo

7 34 M − +++− ? GD + + Hyogo

8 39 M − +++− CDc OIM + + Nagano

9 33 F − + ? + + LOM, Vio, CDc Sei − + Nagano

10 42 F − +? − + Dem Sei + + Akita

11 36 F + ? + −− OCS, CDc OCS + + Saitama

12 25 F − +++− Cop OIM + + Kagawa

13 25 F − + + + + Mon, CDc Sei + + Ibaraki

14 20s F − +++− CDc Cho + + Shizuoka

15 37 M − + + + ? Mon OIM + + Tokyo

16 23 M − + + + + Irr, CDc OIM + + Nara

17 26 F − + + + + CDc Sei + + Nara

Abbreviations: ? = unknown; Ac = acanthocyte; AO = age at onset of first signs or symptoms (y), S = sex; AOP = alteration of personality; AS = atrophy of the corpus striatum on MRI or CT; C = chorein; CDc = cognitive decline; Ch = chorea; CK = elevated creatine kinase; Cop = coprolalia; Del = delusion; Dem =dementia;DI=disinhibition;EFD=executivefunction disorder; EI = emotional instability; Ep = epileptic episode; F = female; FLD = frontal lobe dysfunction; FS = first signs or symptoms; GD = gait disturbance; Hal = hallucination; IMTL = involuntary movement of the tongue and limbs; Ins = insomnia; Irr = irritability; LOM = lack of motivation; M = male; Mon = monologue; ND = not determined; NPS = neuropsychiatric symptom; OCS = obsessive-compulsive syndrome; OIM = orofacial involuntary movement; Pt = patient; Res or Ori = place of residence or origin (Japanese prefecture); Sei = seizure; Vio = violence.

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Table 2 VPS13A mutations identified in this study

Mutation Protein Type of no. Position DNA changeb changeb mutation Patient ID (genotype)

1a Exon 3 c.145-2A>T Unknown Splice Pt 2 (ht)

2a Exon 11 c.799C>T p.R267X Nonsense Pt 10 (hm)

3a Exon 25 c.2532dupT p.D845X Nonsense Pt 12 (ht)

4a Exon 25 c.2593C>T p.R825X Nonsense Pt 6 (ht)

5a Exon 26 c.2824+1 G>T Unknown Splice Pt 4 (ht), Pt 5 (ht)

6a Exon 33 c.3562C>T p.Q1182X Nonsense Pt 16 (ht), Pt 17 (ht)

7a Exon 36- c.4115-459_5991+6444dupc Unknown Large Pt 16 (ht), Pt 17 (ht) 45 duplication

8 Exon 37 c.4411C>T p.R1471X Nonsense Pt 3 (hm), Pt 6 (hm), Pt 7 (hm), Pt 8 (hm), Pt 9 (hm), Pt 13 (hm)

9a Exon 45 c.5881C>T p.R1961X Nonsense Pt 14 (hm)

10a Exon 60 c.8325 G>A p.K2775K Splice Pt 4 (ht), Pt 5 (ht)

11 Exon 60- c.8211+1232_8472- p.V2738AfsX5 Large deletion Pt 1 (hm), Pt 2 (ht), Pt 6 (ht), Pt 11 (ht), Pt 15 (hm) 61 245delinsTC

12a Exon 63 c.8653dupT p.Y2885LfsX2 Small insertion Pt 12 (ht)

13a Exon 63 c.8667+3A>T Unknown Splice Pt 11 (ht)

Abbreviations: hm = homozygous; ht = heterozygous; Pt = patient. a Novel mutation. b Mutations are described according to the nomenclature recommended by the Human Genome Variation Society (hgvs.org). c The DNA change was predicted by the sequencing of the duplication breakpoints.

(Carrigtwohill, County Cork, Ireland). We used 2 primary Results antibodies, a commercially available rabbit polyclonal antibody against chorein (NBP1-85641; Novus Biologicals, Littleton, CO) Mutations identified by Sanger and a generated rabbit polyclonal antibody against a synthetic sequencing analysis oligopeptide antigen corresponding to amino acid residues Using Sanger sequencing, we identified 10 novel mutations 1816–1830 (ESDPEEENYKVPEYK) encoded by exon 43 of the and 2 previously reported mutations in 15 patients (table 2). VPS13A gene (Asahi Techno Glass, Chiba, Japan). Images were These comprised homozygous or compound heterozygous recorded by digital analyzers (Fujifilm LAS-1000; Fujifilm, Tokyo, mutations. Five novel nonsense mutations (799C>T, 2532dupT, Japan, or Fusion-Solo.7S; Vilber Lourmat, Coll´egien, France). 2593C>T, 3562C>T, and 5881C>T) were found in 6 patients. In addition, 4 splice site mutations were found among 4 patients. Patients These splice site mutations (145-2A>T, 2824+1 G>T, 8325G>A, As the results of mutation analysis and chorein analysis, 17 Jap- and 8667+3A>T) were predicted to lead to exon skipping be- anese patients were molecularly diagnosed with ChAc (table 1). cause of the loss of a functional splice acceptor or donor site. Exon We extracted the patient’s symptoms based on the clinical records. skipping events in exons 3, 26, and 60 caused by 145-2A>T, 2824+1 G>T, and 8325G>A, respectively, were predicted to Standard protocol approvals, registrations, cause a frameshift resulting in a premature stop codon. On the and patient consents other hand, exon 63 skipping caused by 8667+3A>T does not Total DNA, RNA, and erythrocyte membrane protein from result in a frameshift because exon 63 consists of 114 bp multiples peripheral blood samples were taken from participants who of codon length. Nonsense mutation of 4411C>T in exon 37 and had given written informed consent. Total DNA and RNA gross deletion of exons 60–61, which have been previously from postmortem brains were collected after written in- reported,1,4,8 were found in 6 and 5 patients, respectively. A single formed consent was obtained from a family member. The nucleotide insertion mutation, which would cause a frameshift research protocol and consent form were approved by the and premature stop codon, was found in patient 12. Institutional Review Board of Kagoshima University. Mutations identified by CNV analysis Data availability statement CNV analysis was performed in samples from patients 16 and The data sets pertaining to the current study are available 17, for whom only a single heterozygous mutation was found from the corresponding author upon reasonable request. by Sanger sequencing analysis. Quantitative PCR (qPCR)

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 Figure Results of duplication analysis and patient 11’s chorein analysis and geographical distribution of VPS13A mutations

Results of qPCR for each exon of the VPS13A gene (A), results of genomic PCR performed with a for- ward primer located in intron 45 and reverse primer located in intron 35 (B), breakpoint in ge- nomic sequence (C), Western blotting for patient 11 (D), and geographical distribution of VPS13A mutations (E). (A) The RQ value of normal controls is approximately 1.0. The extent of a predicted duplication is indicated by arrows. The figure shows the results of qPCR for each exon of the VPS13A gene in patient 16. Comparable results were observed in patient 17. These results suggest heterozygous duplication of exon 36–45. (B) A di- rect connection between introns 45 and 35 was observed in the genomic DNA of patients. In this connection, a repeated AAAA sequence, which was common between the 59 end of intron 35 and the 39 end of intron 45, was observed. (C) Forward primer was located in intron 45. Reverse primer was located in intron 35. Genomic PCR using this combination of primers led to an approximately 7,300 bp PCR product (theoretically 7,319 bp long; arrow) including the predicted junction in 2 patients, but no PCR product in the control. (D) Equal loading was shown by staining with Mem- code Reversible Protein Stain (Thermo Fisher Sci- entific, Waltham, MA), shown in the right panel. Chorein immunoreactivity at 360 kDa was ob- served in the normal control, but not in patients with ChAc other than patient 11. A considerable reduction in chorein levels was observed for pa- tient 11. (E) Solid-colored circles represent patients who have homozygous mutations. Gra- dient-colored circles represent patients who have heterozygous mutations. Single red circles in- dicate exon 60–61 deletion, and single blue circles indicate exon 37 4411C>T mutations. The patients carrying these 2 mutations are localized in west- ern Japan. Some patients who could not identify their ancestral origin provided their address. The map was obtained from aoki2.si.gunma-u.ac.jp/ map/map.cgi. ChAc = chorea-acanthocytosis; qPCR = quantitative PCR; VPS13A = vacuolar protein sorting 13A.

and long-range PCR suggested a single gross duplication of Chorein analysis exons 36–45 because the relative quantification values for We performed chorein Western blotting of erythrocyte these exons were approximately 1.5 fold (figure, A). Conse- membranes of 16 patients. Western blotting revealed the quently, we performed individually designed PCR assays for complete absence of chorein in 15 patients. However, in pa- both patients to enable sequencing of the duplication break- tient 11, chorein immunoreactivity was markedly reduced, points. Sanger sequencing analysis, in which the PCR tem- although the chorein band remained faintly present plate included the junction of the duplication, revealed an (figure, D). abnormal sequence connecting exons 45 and 36 (figure, B). Long-range PCR of gDNA covering the junction between Summary of 39 patients with ChAc exons 45 and 36 in both patients revealed bands corre- A summary of the distribution of VPS13A gene mutations and sponding to approximately 7,300 bp (figure, C). Exons 36–45 clinical symptoms in a total of 39 Japanese patients with were tandemly duplicated, according to cDNA sequencing. ChAc, including 22 previously reported cases,4 can be given as The cDNA length of the duplication was 3754 bp, which follows: (1) average onset age was 29.9 ± 7.0 years; (2) the would cause a frameshift and premature stop codon. main symptoms at onset were involuntary movements,

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG epilepsy, neuropsychiatric symptoms, and/or cognitive In the present study, we summarized the distribution of dysfunction; (3) excluding 4 patients with the unknown VPS13A mutations and manifestations in Japanese patients data, all patients showed peripheral acanthocytosis; (4) with molecularly diagnosed ChAc. To understand the natural excluding 2 patients with the unknown imaging results, disease history and for accurate prediction of ChAc prognosis, 97.3% of patients showed atrophy of bilateral caudate heads much longer monitoring periods of the disease course are in brain MRI or CT; (5) excluding 5 patients with the required. unknown symptoms, 97.1% of patients showed various Acknowledgment psychiatric symptoms or forms of cognitive dysfunction; The authors thank all patients and their families for their (6) excluding 5 patients with the unknown symptoms, participation. They also thank Ms. Meguro, Ms. Nishimura, 94.3% of patients showed involuntary orofacial movement; and Ms. Shimomura for their technical assistance. (7) excluding 2 patients with the unknown data, 91.9% of patients showed elevated creatine kinase; (8) excluding 6 Study funding patients with the unknown symptoms, 90.9% of patients This study was funded by Grants-in Aid from the Research showed chorea affecting all 4 limbs and trunk; (9) 55.1% of Committee of CNS Degenerative Diseases, Research on the mutations in Japanese patients with ChAc carried the 2 Policy Planning and Evaluation for Rare and Intractable major mutations, exon 37 4411C>T (R1471X) and de- Diseases, Health, Labour and Welfare Sciences Research letion of exons 60–61; and (10) there were individually Grants, the Ministry of Health, Labour and Welfare, Japan, different mutations in the remaining 44.9% of Japanese and in part by the Ministry of Education, Culture, Sports, patients with ChAc. Science and Technology KAKENHI (Grant No. 17H04250 to A.S. and No. 18K07606 to M.N.). Discussion Disclosure Disclosures available: Neurology.org/NG. In the present study, we identified 11 novel pathogenic mutations and 2 previously reported mutations1,4,8 in 17 Publication history patients with ChAc and verified the clinical manifestations Received by Neurology: Genetics December 19, 2018. Accepted in final in 39 Japanese patients with ChAc. These mutations were form March 25, 2019. distributed throughout the VPS13A gene, as were those in previous reports.4,8 Although we could not identify genotype- phenotype correlations, over a half of the Japanese patients with ChAc carried exon 37 4411C>T (R1471X) or deletion of Appendix Authors – exons 60 61. The patients carrying these mutations were Name Location Role Contribution mainly localized in Tokyo and western Japan, suggesting ff fi Yoshiaki Kagoshima Author Performed partial founder e ects ( gure, E). Nishida, University laboratory work MD Graduate School and data analysis of Medical and and prepared the In the CNV analysis, we found c.4115-459_5991+6444dup. Dental Sciences, manuscript At the break point junction, a repeated AAAA sequence, Kagoshima 9 which was common between the 5 end of intron 35 and the Masayuki Kagoshima Corresponding Supervised the 39 end of intron 45, was observed. This is presumed to be Nakamura, University author project, advised a microhomology-mediated break-induced replication.9 MD, PhD Graduate School on laboratory of Medical and work and data Dental Sciences, analysis, and Patient 11 carried an exon-intron junction mutation resulting Kagoshima prepared the manuscript in the removal of exon 63 during splicing. Although exon 63 consists of 114 bp with multiple codon lengths, chorein Yuka Urata, Kagoshima Author Advised on MD University laboratory work Western blotting revealed a considerable reduction of chorein Graduate School in patient 11 (figure, D). Because the region of chorein cor- of Medical and Dental Sciences, responding to exon 63 contains a tetratricopeptide repeat Kagoshima motif, which has been reported to be involved in protein- Kei Kagoshima Author Advised on data protein interaction domains, we suggest that exon 63 is es- Kasamo, University analysis sential in the critically important protein interaction function MD Graduate School of Medical and of chorein. Dental Sciences, Kagoshima

In addition to the motor symptoms, patients with ChAc dis- Hanae Kagoshima Author Performed played high frequency of psychiatric symptoms, which may Hiwatashi, University laboratory work explain the previous report that VPS13A mutations pre- MS Graduate School of Medical and dispose individuals to psychiatric disorders.6 Continued Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 Appendix (continued) Appendix (continued)

Dental Sciences, Yuji National Center Author Collected clinical Kagoshima Takahashi, of Neurology and data and blood MD, PhD Psychiatry samples of study Izumi Kagoshima Author Performed Hospital, Tokyo patients Yokoyama, University laboratory work BS Graduate School Ken Akita Red Cross Author Collected clinical of Medical and Shibano, Hospital, Akita data and blood Dental Sciences, MD, PhD samples of study Kagoshima patients

Masahiro Nakamura Author Collected clinical Hideki Kobe University Author Collected clinical Mizobuchi, Memorial data and blood Tokuoka, Graduate School data and blood MD, PhD Hospital, samples of study MD of Medicine, samples of study Hokkaido patients Hyogo patients

Kotaro Hokkaido Author Collected clinical Atsushi Amagasaki Author Collected clinical Sakurai, University data and blood Hara, MD General Medical data and blood MD, PhD Graduate School samples of study Center, Hyogo samples of study of Medicine, patients patients Hokkaido Kazunari National Hospital Author Collected clinical Yasushi Kochi Medical Author Collected clinical Monma, Organization data and blood Osaki, MD School, Kochi data and blood MD, PhD Higashisaitama samples of study samples of study National patients patients Hospital, Saitama

Yukari Kochi Medical Author Collected clinical Katsuhisa National Hospital Author Collected clinical Morita, MD School, Kochi data and blood Ogata, MD, Organization data and blood samples of study PhD Higashisaitama samples of study patients National patients Hospital, Saitama Masako Shinjyuku Neuro Author Collected clinical Watanabe, Clinic, Tokyo data and blood Keita Osaka University, Author Collected clinical MD, PhD samples of study Kakuda, Osaka data and blood patients MD samples of study patients Kenji Ohta-Atami Author Collected clinical Yoshida, Hospital, data and blood Hideki Osaka University, Author Collected clinical MD Fukushima samples of study Mochizuki, Osaka data and blood patients MD, PhD samples of study patients Kiyomi Ohta-Atami Author Collected clinical Yamane, Hospital, data and blood Takeo Arai, Ikebe Clinic, Author Collected clinical MD, PhD Fukushima samples of study MD, PhD Shizuoka data and blood patients samples of study patients Natsuki Tokyo Author Collected clinical Miyakoshi, Metropolitan data and blood Manabu National Center Author Collected clinical MD Neurological samples of study Araki, MD, of Neurology and data and blood Hospital, Tokyo patients PhD Psychiatry samples of study Hospital, Tokyo patients Ryouichi Tokyo Author Collected clinical Okiyama, Metropolitan data and blood Takeshi National Center Author Collected clinical MD Neurological samples of study Fujii, MD, Hospital, data and blood Hospital, Tokyo patients PhD National Center samples of study of Neurology and patients Takehiro Kobe University Author Collected clinical Psychiatry, Tokyo Ueda, MD, Graduate School data and blood PhD of Medicine, samples of study Kazuto Kyoto University, Author Collected clinical Hyogo patients Tsukita, MD Kyoto data and blood samples of study Noritaka National Center Author Collected clinical patients Wakasugi, of Neurology and data and blood MD Psychiatry samples of study Haruhi Kyoto University, Author Collected clinical Hospital, Tokyo patients Sakamaki- Kyoto data and blood Tsukita, MD samples of study Yuji Saitoh, National Center Author Collected clinical patients MD, PhD of Neurology and data and blood Psychiatry samples of study Akira Sano, Kagoshima Author Advised on data Hospital, Tokyo patients MD, PhD University analysis, Graduate School prepared the Takashi National Center Author Collected clinical of Medical and manuscript, and Sakamoto, of Neurology and data and blood Dental Sciences, served as MD, PhD Psychiatry samples of study Kagoshima a scientific Hospital, Tokyo patients advisor

6 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG References 5. Ichiba M, Nakamura M, Kusumoto A, et al. Clinical and molecular genetic assessment – 1. Ueno S, Maruki Y, Nakamura M, et al. The gene encoding a newly discovered of a chorea-acanthocytosis pedigree. J Neurol Sci 2007;263:124 132. protein, chorein, is mutated in chorea-acanthocytosis. Nat Genet 2001;28: 6. Shimo H, Nakamura M, Tomiyasu A, Ichiba M, Ueno S, Sano A. Comprehensive 121–122. analysis of the genes responsible for neuroacanthocytosis in mood disorder and 2. Rampoldi L, Dobson-Stone C, Rubio JP, et al. A conserved sorting-associated protein schizophrenia. Neurosci Res 2011;69:196–202. is mutant in chorea-acanthocytosis. Nat Genet 2001;28:119–120. 7. Dobson-Stone C, Velayos-Baeza A, Filippone LA, et al. Chorein detection for the 3. Jung HH, Danek A, Walker RH. Neuroacanthocytosis syndromes. Orphanet J Rare diagnosis of Chorea-acanthocytosis. Ann Neurol 2004;56:299–302. Dis 2011;6:68. 8. Dobson-Stone C, Danek A, Rampoldi L, et al. Mutational spectrum of the CHAC 4. Tomiyasu A, Nakamura M, Ichiba M, et al. Novel pathogenic mutations and copy gene in patients with chorea-acanthocytosis. Eur J Hum Genet 2002;10:773–781. number variations in the VPS13A Gene in patients with chorea-acanthocytosis. Am J 9. Ottaviani D, LeCain M, Sheer D. The role of microhomology in genomic structural Med Genet B 2011;156:620–631. variation. Trends Genet 2014;30:85–94.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 7 ARTICLE OPEN ACCESS HTT haplogroups in Finnish patients with Huntington disease

Susanna Ylonen,¨ MSc, Jussi O.T. Sipil¨a, MD, DMSc, Marja Hietala, MD, DMSc, and Kari Majamaa, MD, DMSc Correspondence Dr. Sipil¨a Neurol Genet 2019;5:e334. doi:10.1212/NXG.0000000000000334 [email protected] Abstract Objective To study genetic causes of the low frequency of Huntington disease (HD) in the Finnish population, we determined HTT haplogroups in the population and patients with HD and analyzed intergenerational Cytosine-Adenosine-Guanosine (CAG) stability.

Methods A national cohort of patients with HD was used to identify families with mutant HTT (mHTT). HTT haplogroups were determined in 225 archival samples from patients and from 292 population samples. CAG repeats were phased with HTT haplotypes using data from parent- offspring pairs and other mHTT carriers in the family.

Results The allele frequencies of HTT haplotypes in the Finnish population differed from those in 411 non-Finnish European subjects (p < 0.00001). The frequency of haplogroup A was lower than that in Europeans and haplogroup C was higher. Haplogroup A alleles were significantly more common in patients than in controls. Among patients with HD haplotypes A1 and A2 were more frequent than among the controls (p = 0.003). The mean size of the CAG repeat change was +1.38 units in paternal transmissions being larger than that (−0.17) in maternal trans- missions (p = 0.008). CAG repeats on haplogroup A increased by 3.18 CAG units in paternal transmissions, but only by 0.11 units in maternal transmissions (p = 0.008), whereas hap- logroup C repeat lengths decreased in both paternal and maternal transmissions.

Conclusions The low frequency of HD in Finland is partly explained by the low frequency of the HD- associated haplogroup A in the Finnish population. There were remarkable differences in intergenerational CAG repeat dynamics that depended on HTT haplotype and parent gender.

From the Division of Clinical Neuroscience (S.Y., K.M.), Neurology, University of Oulu; Department of Neurology and Medical Research Center (S.Y., K.M.), Oulu University Hospital; Department of Neurology (J.O.T.S.), North Karelia Central Hospital, Siun Sote, Joensuu; Division of Clinical Neurosciences (J.O.T.S.), Turku University Hospital; Neurology (J.O.T.S.), University of Turku; Department of Clinical Genetics (M.H.), Turku University Hospital; and Institute of Biomedicine (M.H.), University of Turku, Finland.

Go to Neurology.org/NG for full disclosures. Funding information is provided at the end of the article.

The Article Processing Charge was funded by the University of Oulu. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary CAG = cytosine-adenosine-guanosine; DNA = deoxyribonucleic acid; HD = Huntington’s disease; LNA = locked nucleic acid; mHTT = mutant HTT; SNP = single-nucleotide polymorphism; THL = National Institute for Health and Welfare.

Huntington disease (HD) is caused by an expansion of a The study has been approved by the ethics committee of Cytosine-Adenosine-Guanosine (CAG) repeat in the HTT Hospital District of Southwestern Finland (Dnro ETMK 19/ gene.1 The expansion is driven by paternal transmissions, 180/2010) and received the national study permits from the – where the repeat instability correlates with repeat length.2 5 National Institute for Health and Welfare (THL, Dnro THL/ The prevalence of HD varies markedly between populations. 1456/5.05.00/2010) and the National Supervisory Authority The disease is most common among white populations, for Welfare and Health, Valvira (Dnro 1195/06.01.03.01/ whereas the prevalence is only a 10th or less in Asian pop- 2012). The study involved no contact with patients. Hence, ulations.6 Among whites, the Finns and Icelanders have low no informed consent was stipulated. prevalence, whereas the prevalence rates in other Scandina- vian countries are similar to those in other whites.7,8 Molecular methods HTT haplogroups A, B, and C were defined using 2 intragenic Shortly after the identification of the HTT gene, it was SNPs, rs762855 and rs4690073, and haplotype A variants established that the risk of CAG expansion is associated with were further defined using 4 additional intragenic SNPs, 10 certain chromosome 4 haplotypes.9 Indeed, differences in the rs2857936, rs363096, rs2276881, and rs362307. The SNPs frequency of HD between populations correlate with dif- were determined with restriction fragment analysis (FastDi- ferences in the frequency of HTT haplogroups defined by gest; Thermo Fisher Scientific, Waltham, MA), allele-specific single-nucleotide polymorphisms (SNPs). Haplogroup A is amplification using locked nucleic acid (LNA) primers (Exi- the most frequent haplogroup in white patients with HD,10 qon, Vedbaek, Denmark), or by sequencing (ABI3500xL while haplogroup C is the most common one in white general Genetic Analyzer, Applied Biosystems, Foster City, CA) as populations. On the other hand, East Asian populations lack appropriate (table e-1, links.lww.com/NXG/A153). The haplotypes A1 and A2 and HD cases harbor haplogroup C.11 PCRs were performed using Phire Hot Start II DNA Poly- HTT haplogroups have been suggested to differ in the rate of merase (Thermo Fisher Scientific) according to the standard new mutations,11 implying differences in the intergenerational procedures. The LNA amplifications were conducted in stability of the CAG repeats. duplicates or confirmed by sequencing. Additional SNPs (table e-1, links.lww.com/NXG/A153) were analyzed in We have previously estimated that the frequency of hap- samples, where haplogroup A variant remained undeter- 10,14 logroup A is lower in the Finnish general population than mined. Haplotypes were manually annotated and then that in most European populations.12 Here we report the fre- phased to CAG repeat length by familial relationship or by quencies of HTT haplogroups in Finnish patients with HD homozygosity of haplotype. and in the general population and assess intergenerational stability of CAG repeat expansion across HTT haplogroups. Statistical analysis Exact test of population differentiation as implemented in Arlequin 3.515 was used to compare frequencies of hap- logroups between cases and controls. Chi-square test or t test Methods was used to compare 2 groups as appropriate. Patients and controls A previously reported national cohort of patients with HD12 Data availability statement was used to identify families with mutant HTT (mHTT). Access to data was regulated by the Finnish law, THL, and Archival deoxyribonucleic acid (DNA) samples (N = 225) Valvira. Permission to conduct this study prohibits disclosing remaining from diagnostic or predictive HTT analyses were data to third parties without explicit permission from THL and fi fi obtained from the 2 national laboratories that perform HD Valvira. Those ful lling the requirements for viewing con - diagnostics. Family relationships were determined using data dential data as required by the Finnish law and who receive the from the Population Register Centre, a governmental au- permission from THL and Valvira are able to access the data. thority keeping records of citizens, the archives of Family Federation of Finland, a nongovernmental nonprofit welfare Results organization that has provided genetic counselling to Finnish families with HD, and hospital records. Phasing of CAG re- Frequencies peat and HTT haplotype was conducted using data from TheallelefrequenciesofHTT haplotypes in the Finnish pop- parent-offspring pairs and other mHTT carriers in the family. ulation differed from those in 411 non-Finnish European sub- As controls, we had 292 healthy Finnish blood donors with jects11 (p < 0.00001, exact test of population differentiation). a mean age of 40.5 years and range 19–64 years.13 The frequency of haplogroup A was lower (p = 0.0003, χ2 test)

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG than that in Europeans, and haplogroup C was higher (p = 0.046, χ2 test), but there was no difference in the proportion of Table 2 The change in the size of mHTT CAG repeat in 65 haplotypes A1 and A2 out of all haplogroup A alleles. transmissions Total Mother Father The allele frequencies of HTT haplotypes among Finnish patients with HD and unsymptomatic mHTT carriers differed Unchanged alleles, N (%) 18 (28) 14 (34) 4 (17) from those in the controls (p < 0.00001, exact test of pop- Contractions, N (%) 21 (32) 15 (37) 6 (25) ff fi ulation di erentiation). Haplogroup A alleles were signi - Expansions, N (%) 26 (40) 12 (29) 14 (58) cantly more common in patients than in controls, while haplogroup B and C alleles were more common in population Mean change, N (%) +0.4 (+1.0) −0.2 (−0.3) +1.4 (+3.1) controls (table 1). Frequencies of haplotypes A1–A5 differed Range of change, N −7 to +12 −3to+4 −7 to +12 between controls and patients with HD (p < 0.00001, exact Range of change, % −14 to +27 −7 to +10 −14 to +27 test of population differentiation). Among patients with HD (including unsymptomatic mHTT carriers), haplotypes A1 Abbreviation: SNP = single-nucleotide polymorphism. and A2 were more frequent and haplotypes A4 and A5 less The median size of the parent CAG repeat was 43 triplets (percentiles 41, 46) in each group. The 65 transmissions include 50 transmissions, where mHTT frequent than among the controls (p = 0.003, exact test of haplotype SNPs and CAG repeat could be phased. population differentiation).

Transmissions gender. HTT haplotype A1 was not only associated with HD, We were able to analyze the change in the size of mHTT CAG but also showed a predilection to CAG repeat expansion in repeat in 65 transmissions (table 2). The mean size of the paternal transmissions. Repeat tracts on haplotypes A2 and change was +1.38 CAG units in paternal transmissions being A3 showed propensity to expand, whereas those on hap- larger than that in maternal transmissions (−0.17 CAG units) logroup C showed a preference for contraction irrespective of (p = 0.008, t test). mHTT haplotype SNPs and CAG repeat parental gender. could be phased in 50 transmissions (haplogroup A, 38; haplogroup C, 10; other haplogroup, 2; figure 1). mHTT Previous studies have shown that paternal transmission drives – CAG repeats in haplogroup A (42.6 CAG units) were shorter intergenerational CAG expansion in HTT.2 5 We identified than those in haplogroup C (48.4 CAG units). In haplogroup 65 Finnish parent-child pairs, where paternal transmission C, the mHTT CAG repeats decreased in paternal as well as was associated with repeat expansion (+1.38 CAG units), maternal transmissions (p = 0.74 for gender difference), whereas maternal transmission mainly resulted in repeat whereas the mean increase of the repeats in haplogroup A was contraction (−0.17 CAG units). Strikingly similar figures considerable in paternal transmissions (3.18 CAG units) but have been obtained in 337 parent-offspring transmissions in meager in maternal transmissions (0.11 CAG units) (p = Dutch patients, where paternal transmission was associated 0.008 for difference) (figure 2). The gender difference in with CAG repeat expansion (+1.76 CAG units) and maternal mHTT CAG change was significant in haplotype A1 inher- transmissions resulted in contraction (−0.07 CAG units).3 itances (p = 0.022), but not in haplotype A2 or A3 inheritance. These findings suggest that intergenerational stability of mHTT in Finnish HD families does not differ from that in Discussion other white populations. Interestingly, the intergenerational stability of the CAG repeat differed between mHTT hap- We found remarkable differences in intergenerational CAG lotypes. On average, repeat tracts on haplotype A1 gained repeat dynamics that depended on HTT haplotype and parent length in paternal transmissions, while maternally inherited tracts decreased in length. Haplotype A1 seems to be a major contributor to paternal anticipation in HD, as 48% of all the Table 1 Frequencies of HTT haplotypes in Finnish patients with HD and controls pathogenic expansions were on haplotype A1. Patients with Controls, Non-Finnish We found that repeat tracts on haplogroup C were larger than Haplogroup HD, N (%) N(%) Europeans, N (%) those on haplogroup A and that repeat tracts on haplogroup C A 291 (64.7) 209 (35.8) 197 (47.9) showed a tendency to contract. Indeed, the ratio between the

B 27 (6.0) 63 (10.8) 17 (4.1) number of contraction events and the number of expansion events was 2.0 in haplogroup C transmissions, while the C 128 (28.4) 310 (53.1) 190 (46.2) corresponding ratio was 0.4 in haplogroup A transmissions O 4 (0.9) 2 (0.3) 7 (1.7) (figure 1A). The skew toward contraction suggests that HTT haplogroup C harbors DNA sequence elements that favor Total 450 584 411 contraction in previously elongated repeat tracts. These Abbreviation: HD = Huntington disease. findings should be interpreted with caution, however, as HTT Patients with HD includes unsymptomatic carriers. Data on non-Finnish Europeans are from Warby et al.11 haplotype and CAG repeat could be phased in only 10 hap- logroup C samples. The haplogroup C samples showed

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 Figure 1 Proportions of changes in mHTT CAG repeat tracts Figure 2 Mean change in mHTT CAG repeat length in ma- in transmission events in 48 Finnish families with ternal and paternal transmission events in 48 Huntington disease Finnish families with Huntington disease

Two families belonging to haplogroups other than A or C are not included. Changes are shown for haplogroup A haplotypes and for haplogroup C.

Finnish general population differ from those among non-Finnish Europeans. In particular, haplogroup A was low, and thus the low frequency of the susceptibility haplogroups partly explains the low HD prevalence in the Finnish population. Similar to other white populations, the majority of HD cases in Finland harbored haplotype A1 or A2 and the 2 haplotypes constituted similar proportions of haplogroup A chromosomes.

In addition to HTT haplotype, other possible contributors to variation in prevalence rates include differences in the rate of new mutations and founder effect. A recent analysis has suggested that a higher frequency of haplotype A1 in the 18 Two families belonging to haplogroups other than A or C are not included. general population yield a higher rate of new HD mutations. (A) Changes in haplogroups A and C. (B) Changes in haplogroup A haplotypes. Hence, the relatively low rate of new mutations would be the mechanism producing low prevalence of HD in the Finnish population by virtue of the low proportion of HTT haplotypes A1 and A2. In Finland, the haplotype distribution may be a tendency for contraction in both paternal and maternal because of small founder population and geographic, cultural, transmissions, which resembles the findings in some Cretan and linguistic isolation have shaped the gene pool of the HD families.16 The repeat tracts are stable or contract in Finnish population.19 These events have yielded Finnish transmission in these families and, clinically, the families do disease heritage, where certain Mendelian diseases are not show anticipation. enriched in the population, while some others are lacking or occur at a low prevalence, such as HD.12 The population Remarkable ethnic differences in the prevalence of HD have genomics also shows some ancient features that are un- suggested a genetic contribution. The frequency is highest detectable in most other regions of Europe.20 among European populations and a recent systematic review compiled 14 studies and found that the median prevalence is Taking into account the low frequency of HD in Finland, our 5.7/100,000 (range 0.96–11.8).6 Thelowestprevalencerates sample set of 225 is rather comprehensive. The archival nature included in the review were 0.96/100,000 in Iceland17 and 2.14/ of the samples posed some limitations, as lack of samples 100,000 in Finland.12 In the European population, CAG precluded phasing in many families. However, homozygosity expansions occur predominantly on haplotypes A1 and A2, of the HTT haplotype or results from 79 out of the 225 which are present in 20% of the individuals with <27 CAG samples enabled us to phase HTT haplotype and CAG repeat repeats.10,11 We found that haplogroup frequencies in the in 50 transmissions.

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Here we found that the intergenerational stability of the CAG repeat differed between mHTT haplotypes. Repeat tracts on Appendix (continued) haplotype A1 gained length only in paternal transmissions, Name Affiliation Role Contribution while paternally as well as maternally inherited tracts de- creased in length on haplogroup C. In addition, we showed Marja Turku University Author Helped in data Hietala, Hospital, Turku acquisition; that the frequency of the HD-associated HTT haplogroup A MD, DMSc interpreted the data; was lower in the Finnish population than that in the non- revised the manuscript for intellectual Finnish European population and that the frequency of hap- content; supervised logroup A was higher in patients with HD than that in the the study fi population. These ndings at least partly explain the low Kari University of Oulu, Author Design and prevalence of HD in Finland. Majamaa, Oulu and Oulu conceptualized study; MD, DMSc University Hospital interpreted the data; performed the statistical analyses; analyzed the data; Disclosure revised the manuscript for intellectual S. Yl¨onen has nothing to declare. J.O.T. Sipil¨a has received content; supervised honoraria (Merck, Pfizer, Sanofi Genzyme), has received the study a consultancy fee (Rinnekoti Foundation), has received travel grants and congress sponsorship (Orion Corporation, Merck Serono, Sanquin, Lundbeck, Novartis), and holds shares (Orion Corporation). M. Hietala has nothing to declare. K. Majamaa References has received travel support from MSD Finland and honoraria 1. MacDonald ME, Ambrose CM, Duyao MA, et al. A novel gene containing a tri- fi nucleotide repeat that is expanded and unstable on Huntington’s disease chromo- from Sano Genzyme and UCB. This study was not industry- somes. The Huntington’s Disease Collaborative Research Group. Cell 1993;72: sponsored. Go to Neurology.org/NG for full disclosure. 971–983. 2. Nørremølle A, Sørensen SA, Fenger K, Hasholt L. Correlation between magnitude of CAG repeat length alterations and length of the paternal repeat in paternally inherited Acknowledgment Huntington’s disease. Clin Genet 1995;47:113–117. The expert technical assistance of Ms. Anja Heikkinen is 3. Aziz NA, van Belzen MJ, Coops ID, Belfroid RDM, Roos RAC. Parent-of-origin differences of mutant HTT CAG repeat instability in Huntington’s disease. Eur J Med gratefully acknowledged. The authors would like to thank Dr. Genet 2011;54:e413–e418. Annukka M. Tuiskula, PhD, and Mr. Antti-Jussi Kortevaara 4. Ramos EM, Cerqueira J, Lemos C, Pinto-Basto J, Alonso I, Sequeiros J. In- tergenerational instability in Huntington disease: extreme repeat changes among 134 for valuable assistance in obtaining the DNA samples archived transmissions. Mov Disord 2012;27:583–585. by HUSLAB and the staff of the genetic laboratory of the 5. Semaka A, Kay C, Doty C, et al. CAG size-specific risk estimates for intermediate allele repeat instability in Huntington disease. J Med Genet 2013;50:696–703. Hospital District of Southwest Finland for gathering and 6. Baig SS, Strong M, Quarrell OW. The global prevalence of Huntington’s disease: dispatching the samples stored there. a systematic review and discussion. Neurodegener Dis Manag 2016;6:331–343. 7. Gilling M, Budtz-Jørgensen E, Boonen SE, et al. The Danish HD Registry- a nationwide family registry of HD families in Denmark. Clin Genet 2017;92: Study funding 338–341. 8. Roos AK, Wiklund L, Laurell K. Discrepancy in prevalence of Huntington’s disease in This study was supported in part by grants from the Sigrid two Swedish regions. Acta Neurol Scand 2017;136:511–515. Juselius Foundation, Turku University Foundation, and the 9. Andrew SE, Hayden MR. Origins and evolution of Huntington disease chromosomes. Neurodegeneration 1995;4:239–244. Finnish Parkinson Foundation. 10. Warby SC, Montpetit A, Hayden AR, et al. CAG expansion in the Huntington disease gene is associated with a specific and targetable predisposing haplogroup. Am J Hum Publication history Genet 2009;84:351–366. 11. Warby SC, Visscher H, Collins JA, et al. HTT haplotypes contribute to differences in Received by Neurology: Genetics February 8, 2019. Accepted in final form Huntington disease prevalence between Europe and East Asia. Eur J Hum Genet March 20, 2019. 2011;19:561–566. 12. Sipil¨a JO, Hietala M, Siitonen A, P¨aiv¨arinta M, Majamaa K. Epidemiology of Hun- tington’s disease in Finland. Parkinsonism Relat Disord 2015;21:46–49. 13. Meinil¨a M, Finnil¨a S, Majamaa K. Evidence for mtDNA admixture between the Finns – Appendix Authors and the Saami. Hum Hered 2001;52:160 170. 14. Kay C, Collins JA, Skotte NH, et al. Huntingtin haplotypes provide prioritized target panels for allele-specific silencing in Huntington disease patients of European an- Name Affiliation Role Contribution cestry. Mol Ther 2015;23:1759–1771. 15. Excoffier L, Laval G, Schneider S. Arlequin (version 3.0): an integrated software Susanna University of Oulu, Author Performed all package for population genetics data analysis. Evol Bioinform Online 2007;1:47–50. Ylonen,¨ Oulu laboratory work; 16. Tsagournissakis M, Fesdjian OC, Shashidharan P, Plaitakis A. Stability of the Hun- MSc analyzed the data; tington disease (CAG)n repeat in a late onset form occurring on the island of Crete. drafted the Hum Mol Genet 1996;4:2239–2243. manuscript for 17. Sveinsson O, Halld´orsson S, Olafsson E. An unusually low prevalence of Huntington’s intellectual content disease in Iceland. Eur Neurol 2012;68:48–51. 18. Kay C, Collins JA, Wright GEB, et al. The molecular epidemiology of Huntington Jussi North Karelia Central Author Design and disease is related to intermediate allele frequency and haplotype in the general pop- Sipila,¨ MD, Hospital, Joensuu and conceptualized study; ulation. Am J Med Genet B Neuropsychiatr Genet 2018;177:346–357. DMSc University of Turku major role in the 19. Norio R. Finnish Disease Heritage I: characteristics, causes, background. Hum Genet and Turku University acquisition of data; 2003;112:441–456. Hospital, Turku analyzed the data; 20. Neuvonen AM, Putkonen M, Oversti¨ S, et al. Vestiges of an ancient border in the revised the manuscript contemporary genetic diversity of north-eastern Europe. PLoS One 2015;10: for intellectual content e0130331.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 ARTICLE OPEN ACCESS Oligogenic basis of sporadic ALS The example of SOD1 p.Ala90Val mutation

Liina Kuuluvainen, MD, Karri Kaivola, MD, Saana Monk¨ ¨are, MSc, Hannu Laaksovirta, MD, Correspondence Manu Jokela, MD, PhD, Bjarne Udd, MD, PhD, Miko Valori, MSc, Petra Pasanen, PhD, Anders Paetau, MD, PhD, Dr. Myllykangas [email protected] Bryan J. Traynor, MD, PhD, David J. Stone, PhD, Johanna Schleutker, PhD, Minna Poyh¨ onen,¨ MD, PhD, Pentti J. Tienari, MD, PhD,* and Liisa Myllykangas, MD, PhD*

Neurol Genet 2019;5:e335. doi:10.1212/NXG.0000000000000335 Abstract Objective To characterize the clinical and neuropathologic features of patients with amyotrophic lateral sclerosis (ALS) with the superoxide dismutase 1 (SOD1) p.Ala90Val mutation, as well as the mutation frequency and the role of oligogenic mechanisms in disease penetrance.

Methods An index patient with autopsy-proven ALS was discovered to have the SOD1 p.Ala90Val mutation, which was screened in 2 Finnish ALS cohorts (n = 453). Additional contributing variants were analyzed from whole-genome or whole-exome sequencing data.

Results Seven screened patients (1.5%) were found to carry the SOD1 heterozygous mutation. Allele- sharing analysis suggested a common founder haplotype. Common clinical features included limb-onset, long disease course, and sensory symptoms. No TDP43 pathology was observed. All cases were apparently sporadic, and pedigree analysis demonstrated that the mutation has reduced penetrance. Analysis of other contributing genes revealed a unique set of additional variants in each patient. These included previously described rare ANG and SPG11 mutations. One patient was compound heterozygous for SOD1 p.Ala90Val and p.Asp91Ala.

Conclusions Our data suggest that the penetrance of SOD1 p.Ala90Val is modulated by other genes and indicates highly individual oligogenic basis of apparently sporadic ALS. Additional genetic variants likely contributing to disease penetrance were very heterogeneous, even among Finnish patients carrying the SOD1 founder mutation.

*These authors contributed equally to this work.

From the Department of Clinical Genetics (L.K.), Helsinki University Hospital; Department of Medical Genetics (L.K.), University of Helsinki, Helsinki, Finland; Molecular Neurology (K.K., M.V., P.J.T.), Research Programs Unit, Biomedicum, University of Helsinki, Helsinki, Finland; Department of Medical Genetics (S.M.), University of Helsinki, Helsinki, Finland and Turku; University Hospital (S.M.), Laboratory Division, Genetics and Saske, Department of Medical Genetics, Turku, Finland; Department of Neurology (H.L.), Helsinki University Hospital, and Molecular Neurology, Research Programs Unit, Biomedicum, University of Helsinki, Helsinki, Finland; Neuromuscular Research Center (M.J., B.U.), Tampere University Hospital and University of Tampere, Tampere, Finland; Division of Clinical Neurosciences (M.J.), Turku University Hospital and University of Turku, Turku, Finland; Folkh¨alsan Research Center (B.U.), Biomedicum, University of Helsinki, Helsinki, Finland; Institute of Biomedicine (P.P., J.S.), University of Turku; Turku University Hospital (P.P., J.S.), Laboratory Division, Genetics and Saske, Department of Medical Genetics, Turku, Finland; Department of Pathology (A.P.), University of Helsinki and Helsinki University Hospital, Helsinki, Finland; Laboratory of Neurogenetics (B.J.T.), National Institute on Aging, National Institutes of Health, Bethesda, MD; Merck & Co. (D.J.S.), Inc., West Point, PA; Department of Clinical Genetics (M.P.), Helsinki University Hospital; Department of Medical Genetics (M.P.), University of Helsinki, Helsinki, Finland; Department of Neurology (P.J.T.), Helsinki University Hospital; and Department of Pathology (L.M.), University of Helsinki and Helsinki University Hospital, Helsinki, Finland.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing Charge was funded by the University of Helsinki. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary ALS = amyotrophic lateral sclerosis; IHC = immunohistochemistry; SOD1 = superoxide dismutase 1; WES = whole-exome sequencing; WGS = whole-genome sequencing.

Superoxide dismutase 1 (SOD1) mutations are the second approval for the use of patient tissue samples was obtained most common cause of familial amyotrophic lateral sclerosis from the National Supervisory Authority for Welfare and (ALS) explaining approximately 12%–20% of familial and Health (Valvira). 1%–2% of sporadic ALS.1 Usually, SOD1 mutations have an autosomal dominant pattern of inheritance.1 Data availability statement The data set is available upon reasonable request from the SOD1 mutation c.269C>T, p.Ala90Val (previously called corresponding author. A89V) has been described in 3 family members with ALS with variable age at onset, incomplete penetrance, and sensory – neuropathy2 and in 4 additional individuals with ALS.3 5 The Results ethnicity of the patients was not reported. Genetic analyses fi The SOD1 mutation NM_000454.4 c.269C>T, p.Ala90Val We identi ed the SOD1 p.Ala90Val mutation through whole- found in the index patient was analyzed in the Helsinki and exome sequencing (WES) in our neuropathologically exam- Turku cohorts (n = 453). Seven additional heterozygous cases ined index patient with ALS and investigated its frequency were found (1.5%). This mutation is in the gnomAD data- and additional genetic burden in 2 Finnish ALS cohorts. base7 in 1/8,367 Finnish samples (heterozygote) but absent in all other populations (95,693 samples) after removing neurologic patients. There is a statistically significant differ- Methods ence in the carrier frequency of the p.Ala90Val between the The index patient was autopsied because of a clinically atyp- Finnish patients with ALS (7/453, excluding index) and the −9 ical motor neuron disease. DNA was extracted from his liver Finnish gnomAD population (1/8,367) (p = 6.9 × 10 , tissue, and a heterozygous SOD1 p.Ala90Val mutation was Fisher exact test). found in WES performed at the Institute for Molecular Medicine Finland (FIMM, Helsinki, Finland). This mutation Although the patients were not known to be related, allele- was screened in 2 ALS cohorts. The Helsinki cohort (n = sharing analysis of the samples indicates a common haplotype 300), collected 1995–2014, was subjected to whole-genome of at least 379,7 kb (Chr21:32723906-33103636) with 8 rare sequencing (WGS) at Broad Institute, Boston, MA. The single nucleotide polymorphism markers, implying a common Turku cohort (n = 153) consisted of samples sent to the ancestor (table e-2, links.lww.com/NXG/A152). TYKS Laboratory of Medical Genetics between 2007 and 2016 for SOD1 sequencing with the diagnosis of definitive or None of the patients had a family history of ALS, and alto- probable ALS or phenotype consistent with motor neuron gether 6 unaffected carriers (aged 50–87 years) of p.Ala90Val disease in the referral. WES was performed at FIMM to the were identified in the families of P6 and P8 (figure e-1, links. p.Ala90Val mutation–positive samples of the Turku cohort. lww.com/NXG/A152). Analysis of other neurodegeneration Sequencing details are shown in e-Methods. All p.Ala90Val- implicated genes (n = 1,115) revealed that all patients had positive samples were screened for the C9orf72 repeat ex- additional potentially contributing variants (table and table pansion using the previously described method.6 e-3, links.lww.com/NXG/A152). Each patient had a unique profile of other variants, the number of possibly or probably To identify additional coding or splicing variants in the contributing variants varied between 4 and 14 per patient. p.Ala90Val-positive samples, we analyzed other neurodegener- Seven of the 8 patients had at least 1 variant that we consid- ative disease and SOD1 pathway genes from their WES/WGS ered “probably pathogenic” (table and table e-3, links.lww. data (e-methods and table e-1, links.lww.com/NXG/A152). com/NXG/A152). Three patients had mutations previously described in ALS: P6, a heterozygous ANG mutation; P7, Neuropathologic analysis was performed following the stan- a heterozygous SPG11 mutation; and P8 was compound dard protocol. Clinical information was examined from heterozygous for SOD1 p.Ala90Val and p.Asp91Ala con- medical records. firmed by family member testing (figure e-1, links.lww.com/ NXG/A152). Four other patients had probably pathogenic Standard protocol approvals, registrations, variants in genes previously associated with motor neuron and patient consents disease or peripheral neuropathy: P1 in ARHGEF28,P3in This study was approved by the local ethics committees. In- UNC13A,P4inARHGEF10, and P5 in ADGRB2/BAI2.P2 formed consent was given by the patients/relatives, or the was the only one who did not have any probably pathogenic

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Table Clinical features and selected genetic findings of patients 1–8

P1 P2 P3 P4 P5 P6 P7 P8

Patient Male Female Female Male Female Female Male Female

Age at onset 40 51 70 47 43 32 48 50 (y)

Disease 14 7 7 18 25a 6a 15a 7 duration (y)

Site of onset Lower limb Limbb Upper limb Lower limb Lower limb Lower limb Lower limb Lower limb

Initial Cramps, NA Weakness of Difficulty Muscle Distal lower Pain and later Distal lower symptoms difficulties with limbs, walking, twitches limb weakness in the limb balance, and predominantly stumbling, weakness lower limbs weakness diminished right upper and control of legs limb problems with balance

Sensory Yes NA No Yes Yes No Yes No symptoms

Initial EMG Sensorimotor NA Motor axon Consistent Consistent Compatible Compatible with Compatible polyneuropathy damage, with motor with motor with motor motor neuron with motor suggestive of neuron neuron neuron disease neuron motor neuron disease disease disease disease disease

Sensory Yesc NA No Yesc No No No No neuropathy in EMG

Cognitive No NA Yesd No No No No No symptoms

Cerebral No NA Yesd Yes No NA NA NA infarct in MRI

Creatine Elevated NA Normal Elevated Slightly Normal Normal kinase elevated

Cause of ALS Suspected ALS Respiratory aa a ALS death myocardial failure infarction

Family No NA No No No No No No history of ALS

C9orf72 Normal Normal Normal Normal Normal Intermediate Normal Normal allele (23 repeats)

Probably WES: ARHGEF28: WGSe WGS: UNC13A: WGS: WGS: WES: ANG: WES: SPG11: WES: pathogenic p.T248R p.R298W ARHGEF10: ADGRB2/ p.K78Ef p.Q1875X 1.SOD1: variants in p.P234T BAI2p.S63L CACNA1H: p.D91Af WES/WGS/ p.R1231C other tests homozygous SMN2 deletiong

Abbreviations: ALS = amyotrophic lateral sclerosis; NA: information not available; WES = whole-exome sequencing; WGS = whole-genome sequencing. a The patient is alive. b More detailed information about the site of onset is not available. c The amplitude of antidromic sensory potentials of patient P1 at age 40 years: median nerve 4.8 mV (normal value ³20 mV), ulnar nerve 3.6 mv (normal value ³17 mV), and sural nerve 5.8 mV (normal value ³6 mV), and of patient P4 at age 53 years: ulnar nerve 4.9 mV and sural nerve 8.7 mV; the median nerve had no response in the study. The EMG studies were performed using the standard protocol. d In addition to small old infarcts in the left occipital lobe and right posterior frontal area, there was a mild expansion in the cortical liquor spaces and mild atrophy in the hippocampi, changes in the pons area and in the periventricular white matter that were interpreted as degenerative. This patient also had cognitive symptoms, and a neuropsychological assessment at age 76 years revealed predominantly frontal lobe problems that were not at the level of dementia. e The patient had variants in 3 SOD1 pathway genes: FBXW8, NOB1, and ALOX15. f Mutation has been previously reported in patients with ALS; the references are in the supplemental material (e-references, links.lww.com/NXG/A152). g Deletions in exons 7 and 8 of the SMN1 and SMN2 genes were investigated by the PCR-restriction fragment length polymorphism method. Comprehensive list and information of genetic variants are in table e-3, links.lww.com/NXG/A152.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 variants according to our interpretation; she had nevertheless Neuropathologic features variants in 3 SOD1 pathway genes: FBXW8, NOB1, and The index patient’s brain weighed 1527 g and appeared ALOX15 (table e-3, links.lww.com/NXG/A152). None of the macroscopically normal. The anterior roots of the spinal cord patients had a C9orf72 repeat expansion, but P6 had 23 hex- were atrophic. Microscopically, the anterior horns showed anucleotide repeats in C9orf72 (the significance of which is significant loss of neurons (figure, C). presently unclear). The axon density was markedly lowered in the anterior roots Clinical features compared with the dorsal roots (figure, A–B). The patients’ clinical features are summarized in the table. The age at onset was variable (32–70 years). All had a limb- There was mild neurodegeneration in the hypoglossal nucleus onset disease, with typical presenting symptoms including at the level of the medulla oblongata (figure, D). Immunohis- fasciculations, weakness, and difficulties with walking and tochemistry (IHC) showed no TDP43-positive inclusions in balance. The initial EMG and nerve conduction study of P1 the anterior horns, cortical areas, or in the hypoglossal nucleus. (index) revealed sensorimotor polyneuropathy; later, he had No hyaline conglomerate inclusions, reported to be specificfor stocking-like sensory abnormalities in both feet, and both some SOD1 mutations,8 were detected on neurofilament soles showed hyperesthesia in addition to the motor symp- (SMI32) IHC. Tau, and beta amyloid stainings were negative. toms. The initial EMG of P4 was consistent with motor neuron disease, and a later EMG revealed additional distal P62 staining showed only a few positive neurites, but no sensory polyneuropathy. P7 had reduced vibration sense in intraneuronal inclusions. The muscle samples showed very his feet, and P5 had paresthesia in her hands. All patients had strong group atrophy and fairly abundant reinnervation a long disease course, 7–25+ years; 3 of the patients were still (figure, E–F). The cause of death was concluded to be motor alive at the time of this study. neuron disease.

Figure Neuropathologic findings of the autopsied patient (index)

(A) Plastic-embedded sections from the dorsal spinal root show normal density of axons, whereas (B) se- vere loss of both myelinated and unmyelinated axons is seen in the anterior spinal roots (toluidine blue ×600 magnification). (C) There is severe neuronal loss in the anterior spinal columns, and the remaining neurons appear chromatolytic (hematoxylin and eo- sin [HE]-stained section from the lumbar spinal cord, ×400 magnification). (D) The hypoglossal nucleus was mildly degenerated (HE-stained section from the medulla oblongata, ×400 magnification. (E) Muscle biopsy taken from the vastus lateralis showed atro- phic small groups (arrow) and overrepresentation of type 2 fibers, suggesting abundant reinnervation (double immunohistochemistry for myosin, ×200 magnification). (F) Higher magnification shows that both type 1 (brown) and type 2 (red) fibers (arrows) are atrophic (×400 magnification).

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Discussion Institute on Aging, NIH (Z01-AG000949-02). The whole- genome sequencing was funded by Merck Sharp & Dohme In this study, 1.5% of the patients with ALS carried the SOD1 Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, and mutation p.Ala90Val, making it a major mutation in Finnish Intramural Research Program of the NIH. patients with ALS based on its frequency, although it had 2–5 previously been described in only 7 patients. In the Hel- Disclosure sinki cohort, it is the third most common currently known Disclosures available: Neurology.org/NG. ALS mutation after C9orf72 repeat expansion and SOD1 p.Asp91Ala (unpublished data). There is a clear enrichment Publication history of p.Ala90Val in the Finnish population. Received by Neurology: Genetics January 7, 2019. Accepted in final form April 1, 2019. There were 6 unaffected family members who were confirmed to carry the p.Ala90Val mutation illustrating the proposed Appendix Authors reduced penetrance and oligogenic mechanisms in ALS.4 The SOD1 p.Ala90Val probably plays a dominating role in our Name Location Role Contribution patients despite the additional rare variant burden because (1) Liina University of Author Design, analysis, and the clinical features were similar in all patients thus far Kuuluvainen, Helsinki, Helsinki, interpretation of 2 MD Finland data, WES and WGS reported and (2) the neuropathology of the index patient analyses, and 9 was consistent with SOD1-related ALS. The p.Ala90Val drafted and revised the manuscript mutation has been shown to cause a conformational change critically for 10 on the SOD1 protein, and SOD1 enzymatic activity has important been shown to be reduced in the CSF of a patient with the intellectual content 5 mutation. In silico analysis with MutationTaster (muta- Karri Kaivola, University of Author Design, analysis, and tiontaster.org/), PolyPhen-2 (genetics.bwh.harvard.edu/ MD Helsinki, Helsinki, interpretation of Finland data, WES and WGS pph2/), and SIFT (provean.jcvi.org/index.php) predicts p. analyses, and Ala90Val to be deleterious. drafted and revised the manuscript critically for We cannot exclude the role of environmental factors in disease important intellectual content penetrance with total confidence. However, 3 of the 8 patients had mutations previously described in ALS, and 4 additional Saana University of Author C9orf72 screening, Monk¨ ¨are, Helsinki, Helsinki, Sanger sequencing patients had probably pathogenic rare variants in genes pre- MSc Finland and data collection, viously implicated in motor neuron disease or peripheral and drafted and revised the neuropathy. Our data represent an illustrative example of manuscript critically a mutation whose penetrance appears to require additional for important genetic factors. It also demonstrates the genetic heterogeneity intellectual content of sporadic ALS: despite sharing a founder mutation, the Hannu University of Author Clinical data and Laaksovirta, Helsinki, Helsinki, sample collection spectrum of other variants was very heterogeneous; each pa- MD Finland and drafted and tient had a unique set of variants. The small sample size and revised the varying sequencing methodology preclude powerful analyses of manuscript critically for important the discovered variants on clinical features. At present, it is not intellectual content possible to make firm conclusions on the pathogenic role of the Manu Jokela, University of Author Design, analysis, and potentially contributing variants in individual patients, although MD, PhD Tampere, Tampere, interpretation of in the p.Asp91Ala compound heterozygous P8, the disease- Finland, and data, clinical data ff University of Turku, and sample causing e ect is clear. The allele frequency of many variants Turku, Finland collection, family (table e-3, links.lww.com/NXG/A152) suggests predisposing member testing, and ff drafted and revised or disease-modifying rather than disease-causing e ects. the manuscript critically for important Acknowledgment intellectual content The authors thank Lilja Jansson, Leena Saikko ja Kristiina Bjarne Udd, University of Author Design, analysis, and Nokelainen for technical assistance in this study. MD, PhD Tampere, Tampere, interpretation of Finland data, clinical data and sample Study funding collection, family This study was supported by Helsinki University Hospital, member testing, and drafted and revised Sigrid Juselius Foundation, Finnish Cultural Foundation, the the manuscript Academy of Finland (294817), Liv och H¨alsa Foundation and critically for Finska L¨akares¨allskapet, The Finnish Medical Foundation, important intellectual content and the Intramural Research Program of the National Continued Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 References Appendix (continued) 1. Marangi G, Traynor BJ. Genetic causes of amyotrophic lateral sclerosis: new genetic analysis methodologies entailing new opportunities and challenges. Brain Res 2015; – Name Location Role Contribution 1607:75 93. 2. Rezania K, Yan J, Dellefave L, et al. A rare Cu/Zn superoxide dismutase mutation causing familial amyotrophic lateral sclerosis with variable age of onset, incomplete Miko Valori, University of Author WES and WGS penetrance and a sensory neuropathy. Amyotroph Lateral Scler Other Motor Neuron MSc Helsinki, Helsinki, analyses, selection of Disord 2003;4:162–166. Finland the SNP markers 3. Andersen PM, Sims KB, Xin WW, et al. Sixteen novel mutations in the Cu/Zn superoxide used in the allele- dismutase gene in amyotrophic lateral sclerosis: a decade of discoveries, defects and dis- sharing analysis, and putes. Amyotroph Lateral Scler Other Motor Neuron Disord 2003;4:62–73. drafted and revised 4. Cady J, Allred P, Bali T, et al. Amyotrophic lateral sclerosis onset is influenced by the the manuscript burden of rare variants in known amyotrophic lateral sclerosis genes. Ann Neurol critically for 2015;77:100–113. important 5. Jacobsson J, Jonsson PA, Andersen PM, Forsgren L, Marklund SL. Superoxide dis- intellectual content mutase in CSF from amyotrophic lateral sclerosis patients with and without CuZn- superoxide dismutase mutations. Brain 2001;124:1461–1466. Petra University of Turku, Author Sanger sequencing 6. Renton AE, Majounie E, Waite A, et al. A hexanucleotide repeat expansion in Pasanen, PhD Turku, Finland and data collection C9ORF72 is the cause of chromosome 9p21-linked ALS-FTD. Neuron 2011;72: and drafted and 257–268. revised the 7. Lek M, Karczewski KJ, Minikel EV, et al. Analysis of protein-coding genetic variation manuscript critically in 60,706 humans. Nature 2016;536:285–291. for important 8. Ince PG, Tomkins J, Slade JY, Thatcher NM, Shaw PJ. Amyotrophic lateral intellectual content sclerosis associated with genetic abnormalities in the gene encoding Cu/Zn superoxide dismutase: molecular pathology of five new cases, and comparison Anders University of Author Neuropathologic with previous reports and 73 sporadic cases of ALS. J Neuropathol Exp Neurol Paetau, MD, Helsinki, Helsinki, analysis and drafted 1998;57:895–904. PhD Finland and revised the 9. Mackenzie IR, Bigio EH, Ince PG, et al. Pathological TDP-43 distinguishes sporadic manuscript critically amyotrophic lateral sclerosis from amyotrophic lateral sclerosis with SOD1 muta- for important tions. Ann Neurol 2007;61:427–434. intellectual content 10. Fujisawa T, Homma K, Yamaguchi N, et al. A novel monoclonal antibody reveals a conformational alteration shared by amyotrophic lateral sclerosis-linked SOD1 Bryan J Laboratory of Author Drafted and revised mutants. Ann Neurol 2012;72:739–749. Traynor, MD, Neurogenetics, the manuscript PhD National Institute on critically for Aging, NIH, important Bethesda, MD intellectual content

David J Stone, Merck & co., Inc., Author Drafted and revised PhD West Point, PA the manuscript critically for important intellectual content

Johanna University of Turku, Author Clinical data and Schleutker, Turku, Finland sample collection, PhD drafted and revised the manuscript critically for important intellectual content

Minna University of Author Clinical data and Poyh¨ onen,¨ Helsinki, Helsinki, sample collection MD, PhD Finland and drafted and revised the manuscript critically for important intellectual content

Pentti J University of Author Design, analysis, and Tienari, MD, Helsinki, Helsinki, interpretation of PhD Finland data, statistical analysis, and drafted and revised the manuscript critically for important intellectual content

Liisa University of Author Design, analysis, and Myllykangas, Helsinki, Helsinki, interpretation of MD, PhD Finland data, neuropathologic analysis, and drafted and revised the manuscript critically for important intellectual content

6 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG ARTICLE OPEN ACCESS Hybrid gel electrophoresis using skin fibroblasts to aid in diagnosing mitochondrial disease

Christopher Newell, PhD, Aneal Khan, MD, MSc, David Sinasac, PhD, John Shoffner, MD, Correspondence Marisa W. Friederich, PhD, Johan L.K. Van Hove, MD, PhD, Stacey Hume, PhD, Jane Shearer, PhD, and Dr. Khan [email protected] Iveta Sosova, PhD

Neurol Genet 2019;5:e336. doi:10.1212/NXG.0000000000000336 Abstract Objective We developed a novel, hybrid method combining both blue-native (BN-PAGE) and clear- native (CN-PAGE) polyacrylamide gel electrophoresis, termed BCN-PAGE, to perform in-gel activity stains on the mitochondrial electron transport chain (ETC) complexes in skin fibroblasts.

Methods Four patients aged 46–65 years were seen in the Metabolic Clinic at Alberta Children’s Hospital and investigated for mitochondrial disease and had BN-PAGE or CN-PAGE on skeletal muscle that showed incomplete assembly of complex V (CV) in each patient. Long-range PCR per- formed on muscle-extracted DNA identified 4 unique mitochondrial DNA (mtDNA) deletions spanning the ATP6 gene of CV. We developed a BCN-PAGE method in skin fibroblasts taken from the patients at the same time and compared the findings with those in skeletal muscle.

Results In all 4 cases, BCN-PAGE in skin fibroblasts confirmed the abnormal CV activity found from muscle biopsy, suggesting that the mtDNA deletions involving ATP6 were most likely germline mutations that are associated with a clinical phenotype of mitochondrial disease.

Conclusions The BCN-PAGE method in skin fibroblasts has a potential to be a less-invasive tool compared with muscle biopsy to screen patients for abnormalities in CV and other mitochondrial ETC complexes.

From the Department of Medical Genetics (C.N., A.K., D.S.) and Department of Pediatrics (A.K.), Alberta Children’s Hospital Research Institute, Cumming School of Medicine, University of Calgary, Canada; Atlanta (J. Shoffner), GA; Departments of Pediatrics (M.W.F., J.L.K.V.H.), Section of Clinical Genetics and Metabolism, University of Colorado; Department of Medical Genetics (S.H.), University of Alberta, Canada; Faculty of Kinesiology (J. Shearer), University of Calgary, Alberta, Canada; and Departments of Laboratory Medicine and Pathology (I.S.), University of Alberta, Edmonton, Canada.

Go to Neurology.org/NG for full disclosures. Funding information is provided at the end of the article.

Competing interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Ethics approval statement: All experimental procedures of this study were performed in accordance with the recommendations of the University of Calgary’s Conjoint Health Research Ethics Board, REB# 13-0753, MITO-FIND (Mitochondrial Functional and Integrative Next Generation Diagnostics) Study.

Patient consent statement: All patients gave written informed consent in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki).

The Article Processing Charge was funded by authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary BBG = Brilliant Blue G; ETC = electron transport chain; BN-PAGE = blue-native polyacrylamide gel electrophoresis; BCN- PAGE = blue- and clear-native polyacrylamide gel electrophoresis; CN-PAGE = clear-native polyacrylamide gel electrophoresis; CV = complex VC; DDM = n-Dodecyl β-D-maltoside; MNG = Medical Neurogenetics; mtDNA = mitochondrial DNA; NBT = nitro blue tetrazolium; nDNA = nuclear DNA; NGS = next-generation sequencing; OXPHOS = oxidative phosphorylation.

Mitochondrial diseases can have abnormal electron transport PCR for private deletions and point mutations, muscle for chain (ETC) dysfunction.1 Electrons are transferred through BN-PAGE or CN-PAGE, and skin biopsy for fibroblast cul- 5 protein complexes (I, II, III, IV, and V) that interact and ture. Four patients aged 46–65 years were seen in the Meta- form supercomplexes (respirasomes) in the inner mito- bolic Clinic at Alberta Children’s Hospital (Calgary, AB) and chondrial membrane. Tissues with high energy requirements investigated for mitochondrial disease (table 1). Four controls may be more vulnerable to disruption of ETC function caused aged 46–62 years were also selected from a bank of skin by either nuclear DNA (nDNA) or mitochondrial DNA fibroblasts that had previously been investigated for and found (mtDNA) mutations.2 Skeletal muscle biopsy is the preferred not to have a diagnosis of an inborn error of metabolism or source to measure ETC protein integrity and function be- a mitochondrial disease through the Metabolic Clinic at cause of the higher mitochondrial density, but can be Alberta Children’s Hospital. invasive.3 Muscle and skin biopsies were performed as part of standard- While a skin biopsy is less invasive,4,5 there have been some of-care diagnostic procedures using a needle muscle biopsy.13 concerns whether existing procedures can represent function Approximately 150 mg total muscle sample was biopsied from in skeletal muscle due to a lower mitochondrial density, ac- the vastus lateralis before being snap frozen without preser- tivity and some metabolic defects are not expressed in skin vatives and briefly stored in liquid nitrogen. A portion of fibroblasts.6 Most techniques involve measuring enzyme ac- muscle was sent for respiratory chain enzyme analysis and tivity and using ETC protein immunoblotting to detect pro- either BN-PAGE or CN-PAGE14,15 to either University of tein abundance7,8 and their utility in the clinical setting in Colorado Denver Biochemical Genetics Laboratory (Aurora, patients with disease is not clear.9,10 CO) or Medical Neurogenetics Laboratories (Atlanta, GA) in accordance with the provincial health plan. The remaining Our aim was to determine whether low-level deletions found muscle was sent for mtDNA sequencing and assessment in a muscle samples also existed in cultured skin fibroblasts (Sanger or next-generation sequencing (NGS) and Southern using either one or a combination of blue-native or clear- blot) at the University of Alberta—Molecular Diagnostic native polyacrylamide gel electrophoresis (BN-PAGE/CN- Laboratory (Edmonton, AB). BN-PAGE or CN-PAGE PAGE) of ETC proteins. We found that a hybrid method, identified incomplete assembly of CV in each patient, char- blue-native and clear-native polyacrylamide gel electropho- acterized by the CV doublet (figure 1). The combined se- resis (BCN-PAGE), was able to resolve the difficulty in ab- quencing and Southern blot analyses performed on muscle normal detecting complex V (CV) patterns in patients with tissue successfully identified 4 unique mtDNA deletions the disease.4,5 spanning the ATP6 gene of CV (table 1).16 Skin samples were collected using either a 4-mm circular punch biopsy or a 4 × 2-mm linear piece removed from the incision site of a muscle Methods biopsy and transferred to the Biochemical Genetics Labora- tory at Alberta Children’s Hospital (Calgary, AB) for sub- Standard protocol approvals, registrations, fi and patient consents sequent broblast culturing. All experimental procedures of this study were performed in accordance with the regulations of the University of Calgary’s Skin fibroblast culture Conjoint Health Research Ethics Board (REB13-0753) and Patient and control skin biopsy tissues were each passaged to the Declaration of Helsinki, and written informed patient P5andexpandedintoT175flasks as per the protocol used by consent was obtained. the Biochemical Genetics Laboratory at Alberta Children’s Hospital. Cellular media, composed of minimum essential Patients and tissues medium with 2 mM glutamine (Life Technologies, Bur- The mitochondrial clinic at Alberta Children’s Hospital uses lington, ON), 10% fetal bovine serum (Life Technologies), a standard protocol for mitochondrial disease testing, which 1 mM sodium pyruvate (Life Technologies), 20 mM uridine includes a muscle needle biopsy for light and electron (Life Technologies), and 100 U/mL penicillin- microscopy11,12, mtDNA extraction for Kearns-Sayre syn- streptomycin, were removed and replaced after 3 succes- drome Southern blot, targeted mutation analysis, long-range sive days of cell incubation.17 Previously shown to reduce

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Table Patient characteristics at the time of diagnosis confirmation

Age Sex Diagnosis Heteroplasmy

Controls 60 F NA NA

62 F NA NA

46 M NA NA

50 M NA NA

Patients 54 F m.8753_16,566 (CV)a <10%

63 F ATPase _CytB (CV)a <10%

48 M m.9090_m.16070 (CV) <10%

64 M m.9928, ATPase6 (CV, junction point unknown)b 25%

Abbreviations: CV = complex V; NA = not applicable; NGS = next-generation sequencing. Data were provided by the Molecular Diagnostics Laboratory at the University of Alberta (Edmonton, AB). a Denotes patient diagnoses and disease heteroplasmy levels confirmed by Sanger sequencing—the remaining patients were confirmed using the MiSeq NGS platform. b Southern blot analysis corroborated sequencing results in only 1 patient. false-negative mtDNA results in skin fibroblasts, uridine and 7.4) (Sigma-Aldrich, Oakville, ON) before being resuspended sodium pyruvate were added to the media to maintain survival in 5 mL of cold sucrose buffer (4°C). The cell suspension was of abnormal mitochondria.18 On reaching 80% confluence, then transferred into a glass and Teflon homogenizer and skin fibroblasts were detached from their respective flasks underwent 20 passes on ice. Homogenized cell suspension using 3 mL 0.25% trypsin/1 mM EDTA (Life Technologies) was then centrifuged at 3,000 rpm for 7 minutes at 4°C and monitored under light microscopy to establish successful (Sorvall Legend Micro 21R Microcentrifuge; Thermo Fisher separation. Detached skin fibroblasts from 8 T175 flasks per Scientific, Burlington, ON). The supernatant was then re- patient/control cell line were aspirated with a 25-mL pipette moved (pellet discarded) and centrifuged at 10,000 rpm for and pooled in a 50-mL conical tube. Pooled cells were 10 minutes at 4°C (Sorvall Legend Micro 21R Micro- centrifuged at 200g for 5 minutes at room temperature centrifuge). The supernatant was then discarded, and the (Beckman Spinchron R Centrifuge; Beckman Coulter, pellet was resuspended in 2 mL of cold (4°C) sucrose buffer Ramsey, MN). The supernatant was discarded, and the pellet before another centrifugation step at 10,000 rpm for 10 was carefully washed 3× with 2 mL sucrose buffer—stock minutes at 4°C (Sorvall Legend Micro 21R Microcentrifuge). solution (250 mM sucrose, 20 mM Tris, 0.1 mM EDTA, pH The supernatant was then discarded, and the pellet was resuspended in 250 μLofbuffer C—stock solution (1.5 M aminocaproic acid, 50 mM Bis-Tris, pH 7.0) (Sigma-Aldrich). Figure 1 Clinical identification of incompletely assembled At this stage, a 10-μL aliquot was removed for protein quanti- mitochondrial CV from skeletal muscle. A repre- fi fi sentative image identifying incomplete assembly cation of skin broblast mitochondria, with bovine serum al- of mitochondrial CV from skeletal muscle bumin as the standard, using the Bradford method (Bio-Rad, Hercules, CA). Solubilization of the mitochondrial membranes was then achieved by incubating 1.6 mg of the nonionic de- tergent n-Dodecyl β-D-maltoside (DDM—from a stock solu- tion) per 1 mg mitochondrial protein for 20 minutes on ice. After solubilization, samples were centrifuged at 14,800 rpm for 30 minutes at 4°C (Sorvall Legend Micro 21R Microcentrifuge). The supernatant was discarded, and the pellet was resuspended in Brilliant Blue G (BBG) (Sigma-Aldrich)—300 μgBBG(from a stock solution) per 100 μg of mitochondrial protein. The mitochondrial protein product suspended in BBG was stored at −80°C until BCN-PAGE analysis.

Clinical diagnostics of patient mitochondrial enzyme assemblies were per- formed using in-gel activity staining of BN-PAGE or CN-PAGE. Diagnostic Blue-native and clear-native polyacrylamide assessments were conducted on patient skeletal muscle tissues with gel electrophoresis BN-PAGE14,15 performed at University of Colorado Denver Biochemical Ge- netics Laboratory (n = 3) (Aurora, CO) and CN-PAGE at MNG Laboratories (n = BCN-PAGE was performed using an XCell SureLock Mini- 1) (Atlanta, GA). Representative image pertains to the BN-PAGE cohort (patient fi 1). BN-PAGE or CN-PAGE = blue-native or clear-native polyacrylamide gel Cell electrophoresis system at 4°C (Thermo Fisher Scienti c). electrophoresis; CV = complex V; MNG = Medical Neurogenetics. NativePAGE 4%–16% Bis-Tris gels (Thermo Fisher Scien- tific) were directly loaded with 40 μg mitochondrial protein

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 per well using the stored solutions of mitochondrial protein phosphate buffer pH 7.4 (Sigma-Aldrich) containing 10 mg product and BBG. Gels were run at 160 V for the first hour, diaminobenzidine tetrahydrochloride (Sigma-Aldrich), 20 mg and the electric field strength was then modified to 100 V for cytochrome c (Sigma-Aldrich), and 1.5 g sucrose (Sigma- the remaining 4.5 hours, as this was found to optimize protein Aldrich) at room temperature with mild agitation. Banding separation. Each electrophoresis experiment used a single developed within 4–12 hours. batch of anode (50 mM Tricine, 15 mM Bis-Tris, pH 7.0) and cathode (50 mM Tricine, 15 mM Bis-Tris, 0.01% wt/vol CV activity DDM, pH 7.0) (Sigma-Aldrich) buffers without replacement. The gel was rinsed thoroughly with water and mild agitation (3 × 10 minutes) before being preincubated in 50 mM Tris In-gel enzyme activity staining pH 8.6 (Sigma-Aldrich) for 1 hour at room temperature with Following BCN-PAGE, adapted from previously published pro- mild agitation. During this time, a solution containing the fol- tocols, each individual oxidative phosphorylation (OXPHOS) lowing chemicals added in the following order was prepared: protein complex was investigated for its respective enzymatic 35 mM Tris, 270 mM glycine, 14 mM MgSO4,and8mMATP 19–21 activity and protein complex assembly. Individual gels were (Sigma-Aldrich). The solution was then adjusted to a pH of 7.8 run for each OXPHOS enzyme complex, with each gel con- before the addition of 0.2% Pb (NO3)2 (Sigma-Aldrich). Finally, taining all 4 control and all 4 patient samples. Starting with the solution was adjusted to a pH of 8.6, and the gel was in- complex I, gels were completed in ascending order of associated cubated at 37°C with mild agitation. Banding developed within OXPHOS enzyme complex number up to CV. Each activity stain 1–2 hours, with optimal band development after 18 hours. was prepared fresh to an approximate volume of 20 mL (see below), and before scanning or imaging, each gel was rinsed Protein immunoblotting thoroughly with water and mild agitation (3 × 10 minutes). Following BCN-PAGE, protein immunoblotting adapted from For complex III, a protein abundance stain was used because previously published work21 was performed using separate gels the published activity stain is not deemed effective in skin for patient and control samples. After being transferred to fibroblasts.22 polyvinylidene difluoride membranes (Millipore, Billerica, MA), primary antibodies were used to probe the membranes Complex I activity overnight at 4°C on a shaking platform. Secondary antibodies The gel was preincubated in 2 mM Tris-HCl pH 7.0 (Sigma- were then incubated on the membrane for 1.5 hours at room Aldrich) for 15 minutes at room temperature with mild agi- temperature. Primary antibodies were as follows: Total tation. It was then transferred to a solution of 2 mM Tris-HCl OXPHOS (catalog no. ab110413; Abcam, Cambridge, MA) pH 7.4 (Sigma-Aldrich) containing 0.1 mg/mL nicotinamide and β-actin (catalog no. ab8226; Abcam). Membranes were adenine dinucleotide (Sigma-Aldrich) and 0.25 mg/mL nitro exposed to the chemiluminescent SuperSignal West Femto blue tetrazolium (NBT) (Sigma-Aldrich) at room tempera- Maximum Sensitivity Substrate (Life Technologies) and im- ture with mild agitation. Banding began to develop within 2 aged using the ChemiGenius imaging system (Syngene, hours, with optimal band visualization >24 hours. Frederick, MD). Densitometry was performed using Gene- β Complex II activity Tools (Syngene), with -actin as a loading control. The gel was preincubated in 200 mM Tris-HCl pH 7.4 (Sigma- Whole-exome sequencing Aldrich) for 15 minutes at room temperature with mild agitation. Two of the patients (1 and 4) also participated in our NGS The gel was then incubated in a solution of 200 mM Tris-HCl study and had whole-exome sequencing performed. Exomes pH 7.4 (Sigma-Aldrich) containing 30 mM succinic acid (Sigma- were sequenced using the 5500XL SOLiD System (Life Aldrich), 0.2 mM phenazine methosulfate (Sigma-Aldrich), 2 mM Technologies), and exome enrichment was performed using EDTA (Sigma-Aldrich), 2 mM potassium cyanide (Sigma- Agilent’s SureSelect XT Human All Exon V5 (Agilent Tech- Aldrich), and 1.0 mg/mL NBT (Sigma-Aldrich) at room tem- nologies). For secondary analysis, the sequencing data were perature with mild agitation. Banding developed within 12 hours. uploaded to the Galaxy instance of University of Calgary (vpn. Complex III protein abundance chgi.ucalgary.ca/), which used the Genome Analysis Tool Kit The gel was preincubated in 5 mM Tris-HCl pH 7.4 (Sigma- and sequence alignment map tools to generate a variant call Aldrich) for 15 minutes at room temperature with mild agi- file. Filtering and interpretative analysis of the resultant an- tation. The gel was then incubated in approximately 20 mL of notated variants were conducted in .xlsx format. The filtering 1-Step tetramethybenzidine-Blotting Substrate Solution strategy consisted of sequencing quality parameters (variant (Thermo Fisher Scientific) at room temperature with mild reads), frequency of the variant (≤MAF 0.01), zygosity, var- agitation. Banding developed within 6–12 hours, with optimal iant context, and computational evidence such as PolyPhen, band visualization after 12 hours. scale-invariant feature transform, and genomic evolutionary rate profiling. The assembly used was GRCh37/hg19. Complex IV activity The gel was preincubated in 50 mM phosphate buffer pH 7.4 Statistical analyses (Sigma-Aldrich) for 15 minutes at room temperature with Statistical analysis was performed using GraphPad Prism for mild agitation. It was then incubated in a solution of 50 mM Windows, Version 7.02 (GraphPad Software Inc., La Jolla,

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG CA). Differences between groups were determined by Stu- Complex II and complex IV activities dent t tests where p < 0.05 was significant. Data are expressed Clinical in-gel activity staining identified normal activity of as mean ± SEM. both complex I and complex IV for each patient (n = 4, per complex) compared with an internal control sample. BCN- Data availability PAGE analysis corroborated these findings by discerning no fi Anonymized data will be shared by request from any quali ed visual differences between patient and control samples (n = 4, investigators. per group) (figure 2).

Complex III protein abundance Results Clinical in-gel activity staining of complex III is commonly not reported because of reproducibility in fibroblasts. However, In-gel enzyme activity staining BCN-PAGE analysis can resolve a protein abundance of A representative image comparing BCN-PAGE resolved mi- complex III using the 1-Step tetramethybenzidine-Blotting tochondrial OXPHOS protein complexes between patient 1 Substrate Solution (figure 2). No visual differences between and control 1 is reported in figure 2. The results of our protein abundance or assembly were identified between pa- modified BCN-PAGE assay demonstrate the ability to both tient and control samples (n = 4, per group). successfully identify each of the 5 mitochondrial OXPHOS protein complexes and to recapitulate clinical diagnostic CV activity fi ndings of incompletely assembled CV exclusively found in Clinical in-gel activity staining identified normal activity of patient samples. Individualized details of each OXPHOS CV, although each patient sample also identified a stronger- protein complex are described in the following sections. than-normal single band of incompletely assembled CV (n = 4) compared with an internal control. BCN-PAGE analysis Complex I activity confirmed these findings by identifying a similarly mis- Clinical in-gel activity staining identified normal activity for assembled CV doublet in patient samples, whereas each complex I in each patient (n = 4) compared with an internal control sample presented as a single band (n = 4, per group) control sample. BCN-PAGE analysis identified a pronounced (figure 2). difference in protein complex assembly between patient and fi control samples ( gure 2). A combination of lower and higher Protein immunoblotting protein assembly existed for control samples, whereas only Protein levels of individual mitochondrial OXPHOS protein a lower assembly of complex I was present in the patients (n = complexes were assessed to provide a quantitative measure of 4, per group). The higher assembly is most likely super- 23 protein abundance. Examination of the blots revealed no complexes, as documented previously. significant differences in protein abundance for mitochondrial OXPHOS complexes II, III, IV, or V (n = 4, per group) (figure 3). Interestingly, even in skin fibroblasts, we found abundant Figure 2 Clinical identification of incompletely assembled complex I in control samples compared with patients, po- mitochondrial CV from skin fibroblasts. A repre- tentially affirming visual contrast between patient and control sentative image identifying incomplete assembly samples from BCN-PAGE experiments. of mitochondrial CV from skin fibroblasts Whole-exome sequencing Exome sequencing in patients 1 and 4 did not reveal any nuclear gene candidates to explain the phenotype or mito- chondrial disease. However, long-range PCR of the region commonly deleted in the patients with Kearns-Sayre syndrome revealed a low level of mitochondrial genome deletions. Sanger sequencing of the deletion breakpoint revealed that all dele- tions involved generating a novel ATP6 protein that had the 39 terminal portion of the gene fused to sequence in the hyper- variable region. This fusion, if stable, would be predicted to cause the ATP6 protein to have a longer carboxy-terminal end.

Discussion Assessment of patient mitochondrial enzyme assemblies was performed We present 4 cases of patients with a low level of mtDNA using in-gel activity staining of hybrid BCN-PAGE. Affirmation of previous diagnostic assessments was conducted on isolated mitochondria from skin deletions in skeletal muscle and abnormal assembly of CV fibroblasts of patients and controls (n = 4 per group). Representative image pertains to data from patient 1 and control 1 samples. BCN-PAGE = blue- proteins. CV abnormalities can be associated with many 24,25 native and clear-native polyacrylamide gel electrophoresis; CV = complex V. pathogenic variants affecting both nDNA and mtDNA. In 3 of these patients, the level of heteroplasmy in skeletal muscle

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 was below 10%, which raised the concern that the mtDNA immunoblotting.32 On the other hand, CN-PAGE lacks BBG, deletions may be due to aging of the skeletal muscle and not and therefore, only acidic proteins migrate toward the an- due to a germline mitochondrial disease. We then modified ode.31 A modified CN-PAGE assay (high-resolution CN- established gel electrophoresis techniques to create a hybrid PAGE) has also been developed where both anionic and BCN-PAGE method that successfully detected an abnormal nonionic detergents are added to the cathode buffer. This CV in the skin fibroblasts from all 4 patients. These results creates micelles, which alter the charge of the native proteins were consistent with the mtDNA mutation, which showed the and facilitate migration toward the anode.33 However, the use presence of the same deletion in both muscle and skin and of anionic detergents may compromise the assembly of mi- thereby reducing the likelihood of somatically acquired tochondrial OXPHOS protein complexes. deletions in the skeletal muscle. Existing as a subset of inherited metabolic disorders, many mitochondrial diseases The development of a BCN-PAGE assay aimed to combine are characterized by a deficiency in OXPHOS function— the beneficial aspects of the various BN- and CN-PAGE which can result from either nDNA or mtDNA mutations.26 conditions while preserving both protein activity and protein Diagnosis of individual mitochondrial diseases is complicated complex assembly. Specifically, mitochondrial protein as- by the variability of clinical phenotypes and tissue-specific sembly was maintained during extraction using the mild, heteroplasmy of the mitochondrial genome, and thus, patients nonionic detergent (DDM) during membrane solubilization. commonly require a multifaceted diagnostic approach ex- Next, the cathode buffer of the BCN-PAGE assay was pre- amining tissues from multiple organ systems.27 Further pared without the addition of the anionic dye BBG, typically complicating mitochondrial genome analysis, skeletal muscle used in BN-PAGE. Instead, BBG was replaced with the can accumulate spontaneous mtDNA mutations, which per- nonionic detergent (DDM) to prevent BBG interference with sist because of the postmitotic nature of the tissue.28,29 For in-gel activity staining and protein immunoblotting.32 The this reason, we developed a BCN-PAGE assay using mito- modified BCN-PAGE cathode buffer therefore acts to retain genic skin fibroblasts from patients who had potentially OXPHOS protein activity and protein complex assembly. The pathogenic mtDNA mutations in skeletal muscle. BCN-PAGE sample buffer also contains an optimized con- centration of BBG. Acting as a charge-shift molecule, the Commonly performed during the clinical investigation of anionic BBG dye permits protein complex migration toward mitochondrial disease; BN-PAGE and CN-PAGE use tradi- the anode.31 In contrast to CN-PAGE protocols, BCN-PAGE tional nondenaturing (native) electrophoresis to separate does not use anionic detergents in the cathode or sample proteins based on their electrophoretic motility.30,31 The buffers, thus allowing native protein migration and complex primary difference between the aforementioned techniques is formation.34 Finally, the duration of electrophoresis and the presence of anionic dyes or detergents to enhance protein electric field strength in BCN-PAGE were selected for opti- migration by increasing the negative charge of the migrating mal protein separation. Collectively, the combination of BN- proteins.32 This is commonly achieved through the addition and CN-PAGE techniques provides a more comprehensive of BBG, an anionic dye that is added to both cathode and examination of the mitochondrial OXPHOS protein activity sample buffers during the beginning of running a BN-PAGE and protein complex assembly. gel, before being removed for the remainder of the assay. However, the addition of BBG has been shown to interfere The present study used a BCN-PAGE assay using skin with in-gel activity staining and downstream protein fibroblasts as a supplement to corroborate clinical findings

Figure 3 Examination of OXPHOS proteins isolated from patient skin fibroblasts with incompletely assembled mito- chondrial CV

Assessment of mitochondrial OXPHOS proteins in mito- chondria isolated from patients with mitochondrial CV deletions and controls, using skin fibroblasts. Results are accompanied by representative immunoblot images with n = 4 per group. Data are presented as mean ± SEM, with * indicating a significant difference between patient and con- trol samples at p < 0.05. CV = complex V; OXPHOS = oxidative phosphorylation.

6 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG from skeletal muscle biopsies. Following a common mito- improper storage of tissue42 or spontaneous accumulation of chondrial disease workup, each patient had a skeletal muscle mtDNA mutations,3 whereas false-negative results can be biopsy performed for biochemical analysis (figure 1), histol- caused by sample selection bias.43 Although they exhibit a low ogy (data not shown), and molecular analysis (table 1), and bioenergetic capacity, skin fibroblasts can have varying levels a skin biopsy, which was banked for future use. Of interest, the of mtDNA heteroplasmy in relation to the age of the patient. hallmark doublet of incompletely assembled CV, as identified Recent research has demonstrated that both DNA mutations in clinical BN- or CN-PAGE results, was visualized in patient and mtDNA heteroplasmy levels in skin fibroblasts remained skin fibroblasts using BCN-PAGE (figure 2). These findings unaltered following prolonged culture (passages 2–15), in- suggest that the mtDNA mutations of CV identified in muscle dicating that mtDNA mutations in skin fibroblasts are likely are likely of germline origin as opposed to being acquired inherited.44 Contrarily, some research has identified limited somatically or as a result of clonal expansion.35 Furthermore, utility for mtDNA depletion detection using skin fibroblasts.45 the systematic identification of complex I supercomplexes This demonstrates the importance of a comprehensive mi- may be another advantage of our BCN-PAGE technique. tochondrial diagnostic approach in which multiple techniques These visualized differences in complex I protein assembly should be applied. Combining this approach with other identified using BCN-PAGE (figure 2) were also elevated noninvasive approaches46 may allow a less invasive approach following protein immunoblotting (figure 3). This is consis- in an increasing number of cases, especially in diseases that tent with the visual contrast between patient and control may escape detection using classic methods of leukocyte samples from BCN-PAGE experiments. BCN-PAGE using DNA whole-exome or whole-genome sequencing. skin fibroblasts provides support for a germline mitochondrial genome mutation, and visualization of BCN-PAGE protein Author Contributions bands, after in-gel activity staining, may provide valuable in- C. Newell, A. Khan, D. Sinasac, S. Hume, J. Shearer, and I. formation regarding mitochondrial protein abundance. Sosova designed and developed the research. C. Newell, A. Khan, M.W. Friederich, and I. Sosova conducted experiments CV comprises an F0 and an F1 region, which can be further and collected and analyzed data. C. Newell, A. Khan, J. divided into 16 subunits, 2 of which are encoded by mtDNA. Shoffner, J.L. Van Hove, and J. Shearer wrote the manuscript. As part of the F0 region embedded in the inner mitochondrial All authors read and approved the final manuscript. membrane, subunits A and A6L are encoded by mtDNA genes ATP6 and ATP8, respectively. The functioning ATP6 Acknowledgment protein forms a proton pore connecting the noncatalytic F0 The authors thank Elizabeth Newell for editing an earlier draft region to the catalytically active F136. Apart from facilitating of this manuscript. the movement of protons between each region, ATP6 protein also physically connects the F0 and F1 regions via the pe- Study funding ripheral stalk. The existing hypothesis is that damage to This study was supported by Alberta Children’s Hospital subunit A may lead to impaired intramitochondrial protein Research Foundation (ACHRF) and MitoCanada (A. Khan). translation, instability of CV, and finally dissociation of the F0 This research was supported by PhD funding to C. Newell and an F1 regions—resulting in the visualization of CV as from MitoCanada and Alberta Innovates—Health Solutions a doublet.36 However, identification of these subcomplexes in MD/PhD Studentship. IS was a recipient of Alberta Child- cultured skin fibroblasts from patients has been variable, with ren’s Hospital Research Institute (ACHRI) Clinical Research heteroplasmy levels >95%.36 Considering that 3 of our Fellowship. patients were identified to have low heteroplasmy (<10%), with the fourth having a heteroplasmy of 25%, we hypothesize Disclosure that expression of the ATP6 protein may act as a dominant Disclosure available: Neurology.org/NG. negative mutation.37 In other words, the mutant ATP6 pro- tein binds with surrounding CV proteins, resulting in the Publication history effective sequestration of properly functioning CV. Received by Neurology: Genetics November 28, 2018. Accepted in final form March 1, 2019. Frequently used to complement existing diagnostic proce- dures for various inherited metabolic disorders, the skin References 1. Parikh S, Goldstein A, Koenig MK, et al. Diagnosis and management of mitochondrial punch biopsy is a simple procedure and can be used to study disease: a consensus statement from the Mitochondrial Medicine Society. Genet Med biochemical defects in patient skin fibroblasts.38 The gold 2015;17:689–701. 2. Gorman GS, Chinnery PF, DiMauro S, et al. Mitochondrial diseases. Nat Rev Dis standard for assessment of mitochondrial enzyme activity for Primers 2016;2:16080. suspected mitochondrial disease is, however, the skeletal 3. Haas RH, Parikh S, Falk MJ, et al. The in-depth evaluation of suspected mitochondrial 39 fi disease. Mol Genet Metab 2008;94:16–37. muscle biopsy. Speci cally, skeletal muscle is regarded as 4. McKenzie M, Lazarou M, Thorburn DR, Ryan MT. Analysis of mitochondrial subunit a superior tissue because it contains a high concentration of assembly into respiratory chain complexes using Blue Native polyacrylamide gel mitochondria,40 thereby lending to increased test sensitivity,3 electrophoresis. Anal Biochem 2007;364:128–137. ff 5. Wittig I, Carrozzo R, Santorelli FM, Sch¨agger H. Supercomplexes and subcomplexes and its function is commonly a ected by mitochondrial dis- of mitochondrial oxidative phosphorylation. Biochim Biophys Acta 2006;1757: 41 ease. However, false-positive results may result from 1066–1072.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 7 6. Emma F, Montini G, Parikh SM, Salviati L. Mitochondrial dysfunction in inherited 25. Tucci A, Liu Y-T, Preza E, et al. Novel C12orf65 mutations in patients with axonal renal disease and acute kidney injury. Nat Rev Nephrol 2016;12:267–280. neuropathy and optic atrophy. J Neurol Neurosurg Psychiatry 2014;85:486–492. 7. Williams SL, Scholte HR, Gray RG, Leonard JV, Schapira AH, Taanman JW. Im- 26. Dimmock DP, Lawlor MW. Presentation and diagnostic evaluation of mitochondrial munological phenotyping of fibroblast cultures from patients with a mitochondrial disease. Pediatr Clin North Am 2017;64:161–171. respiratory chain deficit. Lab Invest 2001;81:1069–1077 . 27. Liang C, Ahmad K, Sue CM. The broadening spectrum of mitochondrial disease: 8. Kramer KA, Oglesbee D, Hartman SJ, et al. Automated spectrophotometric analysis shifts in the diagnostic paradigm. Biochim Biophys Acta 2014;1840:1360–1367. of mitochondrial respiratory chain complex enzyme activities in cultured skin fibro- 28. Wachsmuth M, H¨ubner A, Li M, Madea B, Stoneking M. Age-related and blasts. Clin Chem 2005;51:2110–2116. heteroplasmy-related variation in human mtDNA copy number. PLoS Genet 2016; 9. Fiala GJ, Schamel WWA, Blumenthal B. Blue native polyacrylamide gel electropho- 12:e1005939. resis (BN-PAGE) for analysis of multiprotein complexes from cellular lysates. J Vis 29. Kazachkova N, Ramos A, Santos C, Lima M. Mitochondrial DNA damage patterns Exp 2011:24. Epub 2011 Feb 24. Available at: jove.com/index/Details.stp?ID=2164. and aging: revising the evidences for humans and mice. Aging Dis 2013;4: Accessed March 1, 2018. 337–350. 10. da Cunha ES, Domingues CC, de Paula E. Modified native electrophoresis protocol 30. Sch¨agger H, von Jagow G. Blue native electrophoresis for isolation of membrane for the solubilization and separation of mitochondrial protein complexes. Anal Bio- protein complexes in enzymatically active form. Anal Biochem 1991;199:223–231. chem 2011;418:158–160. 31. Sch¨agger H, Cramer WA, von Jagow G. Analysis of molecular masses and oligomeric 11. Bourgeois JM, Tarnopolsky MA. Pathology of skeletal muscle in mitochondrial dis- states of protein complexes by blue native electrophoresis and isolation of membrane orders. Mitochondrion 2004;4:441–452. protein complexes by two-dimensional native electrophoresis. Anal Biochem 1994; 12. Sarnat HB, Mar´ın-Garc´ıa J. Pathology of mitochondrial encephalomyopathies. Can J 217:220–230. Neurol Sci 2005;32:152–166. 32. Shukolyukov SA. Native electrophoresis in cell proteomic: BN-PAGE and CN-PAGE. 13. Tarnopolsky MA, Pearce E, Smith K, Lach B. Suction-modified Bergstr¨om muscle Cell Tissue biol 2011;5:311–318. biopsy technique: experience with 13,500 procedures. Muscle Nerve 2011;43: 33. Wittig I, Karas M, Schagger H. High resolution clear native electrophoresis for in-gel 716–725. functional assays and fluorescence studies of membrane protein complexes. Mol Cell 14. Coughlin CR, Scharer GH, Friederich MW, et al. Mutations in the mitochondrial Proteomics 2007;6:1215–1225. cysteinyl-tRNA synthase gene, CARS2, lead to a severe epileptic encephalopathy and 34. Wittig I, Braun H-P, Sch¨agger H. Blue native PAGE. Nat Protoc 2006;1:418–428. complex movement disorder. J Med Genet 2015;52:532–540. 35. Tuppen HAL, Blakely EL, Turnbull DM, Taylor RW. Mitochondrial DNA mutations 15. Chatfield KC, Coughlin CR, Friederich MW, et al. Mitochondrial energy failure in and human disease. Biochim Biophys Acta 2010;1797:113–128. HSD10 disease is due to defective mtDNA transcript processing. Mitochondrion 36. Smet J, Seneca S, De Paepe B, et al. Subcomplexes of mitochondrial complex V reveal 2015;21:1–10. mutations in mitochondrial DNA. Electrophoresis 2009;30:3565–3572. 16. Southern EM. Detection of specific sequences among DNA fragments separated by 37. Baquedano MS, Guercio G, Marino R, et al. Unique dominant negative mutation in gel electrophoresis. J Mol Biol 1975;98:503–517. the N-terminal mitochondrial targeting sequence of StAR, causing a variant form of 17. Vangipuram M, Ting D, Kim S, Diaz R, Sch¨ule B. Skin punch biopsy explant culture congenital lipoid adrenal hyperplasia. J Clin Endocrinol Metab 2013;98:E153–E161. for derivation of primary human fibroblasts. J Vis Exp e3779. MyJoVE Corporation; 38. Ceuterick C, Martin J-J. Diagnostic role of skin or conjunctival biopsies in neuro- Epub 2013 Jul 7. logical disorders: an update. J Neurol Sci 1984;65:179–191. 18. Chretien D, Rustin P. Mitochondrial oxidative phosphorylation: pitfalls and tips in 39. Desjardins CA, Gundersen-Rindal DE, Hostetler JB, et al. Comparative genomics of measuring and interpreting enzyme activities. J Inherit Metab Dis 2003;26:189–198. mutualistic viruses of Glyptapanteles parasitic wasps. Genome Biol 2008;9:R183. 19. Smet J, De Paepe B, Seneca S, et al. Complex III staining in blue native polyacrylamide 40. Stephenson EJ, Hawley JA. Mitochondrial function in metabolic health: a genetic and gels. J Inherit Metab Dis 2011;34:741–747. environmental tug of war. Biochim Biophys Acta 2014;1840:1285–1294. 20. Van Coster R, Smet J, George E, et al. Blue native polyacrylamide gel electrophoresis: 41. Taivassalo T, Haller RG. Exercise and training in mitochondrial myopathies. Med Sci a powerful tool in diagnosis of oxidative phosphorylation defects. Pediatr Res 2001; Sports Exerc 2005;37:2094–2101. 50:658–665. 42. Shoffner JM. Mitochondrial myopathy diagnosis. Neurol Clin 2000;18:105–123. 21. D´ıaz F, Barrientos A, Fontanesi F. Evaluation of the mitochondrial respiratory chain 43. Koenig MK. Presentation and diagnosis of mitochondrial disorders in children. and oxidative phosphorylation system using blue native gel electrophoresis. Curr Pediatr Neurol 2008;38:305–313. Protoc Hum Genet 2009;Chapter 19:Unit 19.4. Available at: ncbi.nlm.nih.gov/ 44. Kang E, Wang X, Tippner-Hedges R, et al. Age-related accumulation of somatic pubmed/19806591. Accessed March 23, 2017. mitochondrial DNA mutations in adult-derived human iPSCs. Cell Stem Cell 2016; 22. Smet J, De Paepe B, Seneca S, et al. Complex III staining in blue native polyacrylamide 18:625–636. gels. J Inherit Metab Dis 2011;34:741–747. 45. Dimmock D, Tang LY, Schmitt ES, Wong LJ. Quantitative evaluation of the mito- 23. Friederich MW, Erdogan AJ, Coughlin CR, et al. Mutations in the accessory subunit chondrial DNA depletion syndrome. Clin Chem 2010;56:1119–1127. NDUFB10 result in isolated complex I deficiency and illustrate the critical role of 46. Newell C, Hume S, Greenway SC, Podemski L, Shearer J, Khan A. Plasma-derived intermembrane space import for complex I holoenzyme assembly. Hum Mol Genet cell-free mitochondrial DNA: a novel non-invasive methodology to identify mito- 2016;26:ddw431. chondrial DNA haplogroups in humans. Mol Genet Metab 2018 Epub 2018 Oct 16. 24. Roels F, Verloo P, Eyskens F, et al. Mitochondrial mosaics in the liver of 3 infants with Available at: sciencedirect.com/science/article/pii/S1096719218303950. Accessed mtDNA defects. BMC Clin Pathol 2009;9:4. October 23, 2018.

8 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG ARTICLE OPEN ACCESS Novel mutation in TNPO3 causes congenital limb-girdle myopathy with slow progression

Anna Vihola, PhD, Johanna Palmio, MD, PhD, Olof Danielsson, MD, PhD, Sini Penttil¨a, PhD, Correspondence Daniel Louiselle, MS, Sara Pittman, BS, Conrad Weihl, MD, PhD, and Bjarne Udd, MD, PhD Dr. Vihola [email protected] Neurol Genet 2019;5:e337. doi:10.1212/NXG.0000000000000337 Abstract Objective We report a second family with autosomal dominant transportinopathy presenting with con- genital or early-onset myopathy and slow progression, causing proximal and less pronounced distal muscle weakness.

Methods Patients had clinical examinations, muscle MRI, EMG, and muscle biopsy studies. The MYOcap gene panel was used to identify the gene defect in the family. Muscle biopsies were used for histopathologic and protein expression studies, and TNPO3 constructs were used to study the effect of the mutations in transfected cells.

Results We identified a novel heterozygous mutation, c.2757delC, in the last part of the transportin-3 (TNPO3) gene in the affected family members. The mutation causes an almost identical frameshift affecting the stop codon and elongating the C-term protein product of the TNPO3 transcript, as was previously reported in the first large Spanish-Italian LGMD1F kindred. TNPO3 protein was increased in the patient muscle and accumulated in the subsarcolemmal and perinuclear areas. At least one of the cargo proteins, the splicing factor SRRM2 was normally located in the nucleus. Transiently transfected mutant TNPO3 constructs failed to localize to cytoplasmic annulate lamellae pore complexes in cells.

Conclusions We report the clinical, molecular genetic, and histopathologic features of the second trans- portinopathy family. The variability of the clinical phenotype together with histopathologic findings suggests that several molecular pathways may be involved in the disease patho- mechanism, such as nucleocytoplasmic shuttling, protein aggregation, and defective protein turnover.

From the Folkh¨alsan Institute of Genetics and Department of Medical Genetics (A.V.), Medicum, University of Helsinki; Neuromuscular Research Center (J.P.), Tampere University and University Hospital of Tampere, Finland; Neuromuscular Unit (O.D.), Division of Neurology, Department of Clinical and Experimental Medicine, Linkoping¨ University, Sweden; Neuromuscular Research Center (S. Penttil¨a), Tampere University and University Hospital of Tampere, Finland; Department of Neurology (D.L.), Department of Neurology (S. Pittman), Department of Neurology (C.W.), Washington University School of Medicine, Saint Louis, MO; Folkh¨alsan Institute of Genetics and Department of Medical Genetics (B.U.), Medicum, University of Helsinki; Neuromuscular Research Center (B.U.), Tampere University and University Hospital of Tampere; and Department of Neurology (B.U.), Vaasa Central Hospital, Vaasa, Finland.

Go to Neurology.org/NG for full disclosures. Funding information is provided at the end of the article.

The Article Processing Charge was waived at the discretion of the Editor. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary AL = annulate lamella; ALPC = annulate lamellae pore complex; BSA = Bovine serum albumin; CK = creatine kinase; DAB = diaminobenzinide; HA = hemagglutinin; IF = immunofluorescent; IHC = immunohistochemistry; PBS = Phosphate buffered saline; PC = pore complex; RBM4 = RNA binding motif protein 4; RT = room temperature; SG = stress granule; SR = serine/ arginine rich; SRRM2 = SR repetitive matrix 2; SQSTM1 = sequestosome 1; TA = tibialis anterior; TDP-43 = TAR DNA- binding protein 43; TIA1 = T-cell-restricted intracellular antigen-1; TNPO = transportin-3; WT = wild type.

The limb-girdle muscular dystrophies form a heterogeneous group of genetically transmitted myopathies with pre- Figure 1 Pedigree dominantly proximal, progressive muscle weakness.1 To date, 8 forms of dominant limb-girdle muscular dystrophy with known genetic cause have been identified; however, a new nomen- clature has recently been proposed,2 in which only 4 dominant forms fulfilled the required criteria: LGMD D1 DNAJB6 re- lated, D2 TNPO3 related, D3 heterogeneous nuclear ribonu- cleoprotein D like related, and D4 calpain3 related.

A dominant mutation in the TNPO3 gene was found to cause LGMD1F in a large Spanish-Italian family with proximal limb and axial muscle weakness.3,4 The causative mutation, c.2771delA p.*924Cext*15 in exon 22, results in the extension of the reading frame by 15 additional amino acids. There was wide variability in the age at onset and disease severity,3 and also, nonpenetrance was observed.5 Muscle weakness and atrophy of The affected family members were included in this study. The proband (II-1) is indicated with an arrow. the lower limbs were prominent. Additional features were dys- phagia, arachnodactyly, joint contractures, scapular winging, and hyperlordosis in some of the patients.3,6 Muscle histopathology Standard protocol approvals, registrations, was characterized by myopathic changes, including nuclear pa- and patient consents thology, myofibrillar protein accumulation in the cytoplasm, and All participants provided appropriate consent, and the study rimmed vacuolar pathology corresponding to accumulated was approved by the IRB of Tampere University Hospital. autophagosomal membranes at the ultrastuctural level.6,7 Transportin-3 (TNPO3) belongs to the importin beta su- Methods perfamily. It facilitates the nuclear import of Ser/Arg-rich (SR) proteins.8 SR motifs are commonly found on RNA- Molecular genetics binding proteins associated with splicing. TNPO3 has also Targeted massively parallel sequencing was performed for been identified as essential for HIV infection, and loss of DNA samples of patients II-1 and III-1, as previously de- TNPO3 function is protective against HIV.9 The role of scribed,10 and sample I-3 was Sanger sequenced. The se- TNPO3 in skeletal muscle and how mutations affect its quencing library was enriched using the probes of MYOcap function and lead to muscle disease have not been described. v3 gene panel that is targeted to the exons of 265 genes known or predicted to cause muscular dystrophy or myopathy. Patients Histologic techniques A Swedish family with 3 patients representing subsequent gen- Snap-frozen muscle biopsies were processed into sections for erations, the proband (II-1), his mother (I-3), and his son (III-1), histologic and immunohistochemical stainings. Conventional was included in this study (figure 1). The patients were followed hematoxylin and eosin (H&E), Herovici, modified Gomori up since early childhood because of walking difficulties or hy- trichrome, and nicotinamide adenine dinucleotide tetrazo- potonia at birth. All underwent neurologic examinations, muscle lium reductase staining techniques were applied.11 biopsy, and muscle MRI studies. EMG findings and creatine kinase (CK) levels were available in the proband and his mother. For immunohistochemistry (IHC), the Ventana GX auto- Muscle biopsies were performed at different time points: for I-3, mated immunostainer was used to get 3,39-diaminobenzidine at ages 31 and 48 years (both from the tibialis anterior muscle, immunolabeling, followed by hematoxylin and bluing coun- TA); for II-1, at ages 3, 24, and 35 years (all from the TA); and terstain (all by Roche Tissue Diagnostics/Ventana Medical for III-1, at age 16 months (vastus lateralis). Systems, Tucson, AZ).

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Immunofluorescent (IF) stainings were performed manually. extension constructs “+15” and “RG” were made by insertion The frozen muscle sections were fixed in 4% paraformaldehyde mutagenesis of the HA-TNPO3 plasmid using the Q5 Site- for 15 minutes, permeabilized with 0.05% Triton X-100 in Directed Mutagenesis Kit (NEB E0554S) and the following phosphate buffered saline (PBS) for 10 minutes, and blocked primers: with 2% bovine serum albumin (BSA) in PBS for 30 minutes. Primary antibody incubation was performed overnight at 8°C. Immunofluorescence After PBS washes, Alexa-488 and Alexa-546 Fluor–conjugated HeLa cells were grown on glass coverslips in a 12-well plate. The μ secondary antibodies were used for detection at room tem- next day, cells were transfected with 1.0 g of plasmid using ’ perature (RT) for 1 hour. Hoechst nuclear conterstain was Lipofectamine 2000 and following the manufacturer sprotocol. fi performed before mounting in Fluoromount (Sigma-Aldrich) Twenty-four hours later, cells were xedinice-cold4%para- medium. The following primary antibodies were used: poly- formaldehyde in PBS for 15 minutes at RT. Cells were then clonal anti-TNPO3 antibody (C-term) (Abcam ab109386), washed with PBS 3 times for 5 minutes each and permeabilized polyclonal anti-p62 (Millipore/Sigma P0067), monoclonal with0.2%TritonX-100inPBSfor10minutesatRT. anti-TAR DNA-binding protein 43 (TDP-43) (Sigma-Aldrich Blocking and detection of hemagglutinin only WH0023435M1-1), polyclonal anti-ubiquitin (Dako Z0458), Cells were incubated in blocking buffer made up of 1% BSA in polyclonal anti-myotilin (ProteinTech 10731-1-AP), mono- PBS for 1 hour at RT. Coverslips were then transferred to a hu- clonal anti-desmin (Abcam ab32362), monoclonal anti-alpha-B- midified chamber and incubated with Alexa Fluor 488-conjugated crystallin (CRYAB) (Leica Biosystems NCL-ABCrys-512), mouse anti-HA antibody diluted 1/200 in blocking buffer for monoclonal anti-tropomyosin (Abcam ab7786), polyclonal either 2 hours at RT or overnight at 4°C. Coverslips were washed anti-CHCHD10 (Novus Biologicals NBP1-91169), polyclonal with PBS 3 times for 5 minutes each at RT. They were then anti-RNA binding motif protein 4 (RBM4) (Atlas Antibodies carefully dipped in molecular-grade water for 10 seconds and HPA047849), and polyclonal anti-serine/arginine repetitive mounted on microscope slides using Mowiol mounting medium. matrix 2 (SRRM2) (Abcam ab122719). Blocking and double-labeling of both marker Western blotting and hemagglutinin Frozen muscle biopsies were homogenized in Laemmli sam- Cells were incubated in Blocking Buffer A (PBS + 1% BSA + 5% ple buffer and heated at 98°C for 5 minutes to prepare tissue serum of the secondary antibody host) for 1 hour at RT. Cov- lysates. Conventional sodium dodecyl sulphate erslips were then transferred to a humidified chamber and in- polyacrylamide gel electrophoresis and Western blotting cubated overnight with anti-HA, anti-T-cell-restricted intracellular protocols were used, with 4%–20% precast TGX (Bio-Rad antigen-1 (TIA1), RanGAP1 (Santa Cruz Biotechnology), or Laboratories, Hercules, CA) and Trans-Blot Turbo System Mab414 (BioLegend) in Blocking Buffer A. The next day, cov- (Bio-Rad) for protein transfer onto polyvinylidene difluoride erslips were washed 3 times for 5 minutes each with PBS and then membranes. For immunodetection, the membrane was in- incubated with Alexa Fluor 555–conjugated secondary antibody cubated overnight at 8°C with anti-TNPO3 antibody (Abcam diluted 1/500 in Blocking Buffer A for 1 hour at RT. Cells were ab109386) at 1/500 dilution in tris-buffered saline with washed 3 times for 5 minutes each with PBS and then incubated in tween-20/5% skimmed milk powder. The next day, after Blocking Buffer B (PBS + 1% BSA + 5% serum of the primary horseradish peroxidase-conjugated secondary antibody in- antibody host) for 1 hour at RT. Coverslips were then incubated cubation for 1 hour at RT, the bands were detected using for 2 hours at RT with Alexa Fluor 488–conjugated mouse anti- enhanced chemiluminescence (SuperSignal West Femto HA antibody diluted 1/200 in Blocking Buffer B. Coverslips were Maximum Sensitivity Substrate, Thermo Fisher Scientific). washed with PBS 3 times for 5 minutes each at RT. They were After blotting, the gels were recovered and stained with then carefully dipped in molecular-grade water for 10 seconds and Coomassie blue for myosin heavy chain, which was used as mounted on microscope slides using Mowiol mounting medium. a loading control. ChemiDoc reader and ImageLab software (Bio-Rad) were used to obtain images and for calculating the Arsenite assay relative quantities and molecular weights of the bands. Twenty-four hours after transfection, media were removed from the coverslips and replaced with warm media containing 0.5 mM Primer ID Sequence arsenite or an equivalent volume of PBS. Cells were incubated fi +15 forward tcacccaggaatgtcttttttaaAGCTCGAGTCTAGAGGGC for 45 minutes in a humidi ed 37°C, 5% CO2 incubator. After incubation, media were aspirated, and cells were washed once +15 reverse caggcacagtgcaggagtgtgagcATCGAAACAACCTGGTGAAG with PBS before fixing and staining as described above. RG forward gtgcctgtcacccaggaatgtcttttttaaAGGTTGTTTCGATAGCTCG Quantification of TNPO3 foci RG reverse agtgcaggagtgtgagctgtcgaagcatccGGTGAAGTCTCGCAAGGC Fixed and stained U2OS cells were viewed on low magnifi- cation, and transfected cells were scored as either TNPO3 foci Plasmids positive or TNPO3 foci negative. A minimum of 50 cells were Wild-type (WT) hemagglutinin (HA)-TNPO3 was provided by counted for each condition, and at least 3 independent Nathaniel Landau of New York University. The C-terminal experiments were conducted.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 Data availability statement adulthood, after which it slowly progressed. She started to use The data that support the findings of this study are available a wheelchair at age 64 years. from the corresponding author upon reasonable request. The son (III-1) was hypotonic at birth and had slightly high palate, no contractures. There was no facial weakness or at- Results rophy or hypertrophy of muscles. Tendon reflexes and rou- tine laboratory tests were normal. He started to walk with fi The proband, a 44-year-old man (II-1), was rst examined at waddling gait at 13 months. Gower sign was present, and age 13 months because of unsteady waddling gait and ab- slight girdle weakness and axillar hypotonia were observed. At normal head control. At age 3 years, he had mild myopathic age 7 years, he could not run and had difficulties walking on facies, narrow palate, and generalized muscle weakness and uneven ground. Trendelenburg was positive. He had no fl atrophy. There were also absent tendon re exes and Gower hyperlordosis or scoliosis. The Medical Research Council fl sign present when rising from the oor. Slight extension scale score was 3–4/5 in all proximal and distal limb muscles. contracture was observed in the right elbow, otherwise joint His vital capacity was 75% at age 8 years. contractures or hyperlordosis were not present. Cognitive functions were normal. During childhood, he was unable to Muscle imaging runorwalkontoesandheels.Themostpronouncedweakness Muscle MRI of the proband showed generalized and severe was present in ankle dorsiflexion and knee and hip flexion. EMG diffuse fatty degenerative changes in all pelvic and thigh showed no abnormalities, muscle enzymes were slightly ele- muscles and slightly less severe but still diffuse changes in all vated, whereas other routine laboratory tests yielded normal the distal lower limb muscles (figure 2). The changes in the results. His functional abilities were stable and weakness non- pelvic and thigh muscles were also severe in the mother, al- progressive until early adulthood. After that, weakness slowly though less severe in the hamstrings and anterior compart- progressed to encompass most upper and lower limb muscles, ment of the lower limbs. Early diffuse degenerative changes although more markedly proximal muscles. were seen in all muscles of the son of the proband.

His mother (I-3) was able to walk at 16 months and had Genetics proximal and distal lower limb and neck flexor weakness and The analysis of the MYOcap sequencing data revealed a het- Gower sign. She also had proximal upper limb weakness. erozygous variant c.2757delC p.(R920Gfs*20) in TNPO3 Hyperlaxity of most joints was noted. The CK level was (NM_012470) in II-1 and III-1, and the same mutation was later normal, and EMG was suggestive of myopathy. During confirmed with Sanger sequencing in I-3. The detected variant childhood, generalized muscle weakness was noted. Achilles was not listed in the Exome Variant Server, Exome Aggregation tendon reflex was normal, others were absent. She could walk Consortium, or 1000 Genomes databases. Other variants suit- on toes but not on heels. Her disease was also stable until able for autosomal dominant inheritance were not detected.

Figure 2 Muscle imaging

Severe generalized fatty degenerative changes in the proband (A). The ham- string muscles and anterior compart- ment of the lower limbs were relatively spared in the mother (B). Mild and diffuse early changes were seen in the son of the proband more pronounced in the sartorius, gracilis, adductor magnus, and peroneus lon- gus muscles (C).

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Histopathology sarcoplasmic basophilic material (Figure 3, D and F), and even The muscle biopsies obtained from the proband (II-1) and his some cytoplasmic bodies (figure 3D). Occasional ragged red mother (I-3) at adult age showed general myopathic changes fibers were encountered (figure 3F), and the number of including variation in fiber size and shape, fiber splitting, and Cytochrome C oxidase-negative fibers was slightly increased numerous internal nuclei (figure 3, A and B). Both hypertrophic (5%–10% fibers Cytochrome C oxidase-negative in I-3 at age 48 and atrophic fibers were present in large numbers, and prom- years). Sporadic targetoid, whorled, and moth-eaten fibers were inent type 1 fiber predominance was observed (>90% type 1 also observed (I-3 at age 48 years). fibers, whereas in the TA muscle, 60%–80% type 1 fibers is considered normal).12 Very few occasional necrotic and regen- In biopsies taken at young age (II-1 and III-1 at 3 years and 16 erating fibers were present. The nuclear pathology was re- months, respectively), there was fiber size variation but no internal markable, including swollen nuclei with central pallor (figure nuclei. Sarcoplasmic abnormalities were not observed in routine 3A), and accumulation of perinuclear/subsarcolemmal baso- histochemical stainings (H&E and Gomori), but the myofibrillar/ philic material was a frequent finding. Cytoplasmic abnormalities cytoskeletal pathology became visible when immunostained for were observed as well, including rimmed vacuoles (figure 3B), myotilin, desmin, and tropomyosin, showing several fibers with

Figure 3 Histopathology

Histopathology of patient II-1 TA bi- opsy (A–E, G–K, M–N, P) shows myo- pathic changes: fiber size variation and numerous internal nuclei, some with central pallor in H&E (A, arrow- heads) and rimmed vacuoles in Herovici (B, arrowheads). Sub- sarcolemmal TNPO3 accumulation is observed in TNPO3 IHC staining (C, arrowheads). A fiber with small cyto- plasmic bodies is seen in Gomori tri- chrome staining (D, black arrowhead) and a fiber with myofibrillar pathol- ogy (D, white arrowhead). In a serial section, mitochondrial NADH staining reveals an uneven staining pattern in central parts of the muscle fibers (E). Gomori trichrome of patient I-3 TA biopsy (F) shows a ragged red fiber (black arrowhead) and myofibrillar pathology (white arrowhead). Large (arrowhead) and small myotilin accu- mulations are observed in IHC stain- ing (G). In IF double staining, p62 (green) and TDP-43 (red) colocalize in inclusion bodies (H, arrowheads) in the rimmed vacuolar fiber. The mito- chondrial CHCHD10 shows sub- sarcolemmal accumulation in several muscle fibers (I). IF double staining of desmin (green) and alpha-B-crystallin (red) shows both cytoplasmic and subsarcolemmal overexpression and colocalization (J, arrowheads), as does desmin (green) and tropomyosin (red) in K. Notably, desmin (green) and tropomyosin (red) show over- expression in patient III-1 muscle at only age 16 months (L). In confocal microscopy, RBM4 (green) is often excluded from the nuclei (M, arrow- heads) in the patient biopsies, whereas SRRM2 (green) in N shows strictly nuclear localization in the pa- tient, a pattern similar to control muscles. Confocal analysis of TNPO3 shows normal nuclear localization in patient III-1 (O), but often perinuclear accumulation of TNPO3 in adult pa- tient biopsy (P, arrowheads), which shows as a nuclear rim. (A–L), scale bar = 100 μm; (M–P), scale bar = 50 μm. IF = immunofluorescent; IHC = immunohistochemistry; TA = tibialis anterior; TNPO3 = transportin-3.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 accumulation (figure 3L). However, the rimmed vacuolar pa- in II-1, variable nuclear and/or perinuclear localization was ob- thology, as well as perinuclear basophilic masses, appeared later in served (figure 3P). A similar staining pattern was observed with the disease course and was not present at childhood. Type 1 fiber RBM4 (figure 3M), one of the splicing factors TNPO3 binds predominance was present as early as at age 3 years (II-1). and translocates to the nucleus. Another Ser/Arg-rich splicing factor, SRRM2, showed normal nuclear (speckle) localization in Ultrastuctural examination of I-3 and II-1 showed subsarcolemmal patient II-1, with no perinuclear accumulation (figure 3N). deposits of amorphic material and degenerated mitochondria and sometimes membranous structures (data not shown). Areas of Western blotting myofibrillar disorganization, including Z-streaming, were en- In Western blotting, one band of approximately 96 kD was fi countered, as well as sporadic fibers with minicore-like pathology. observed in both controls and patients ( gure 4A); hence, we ff The nuclei were large and showed atypical morphology. could not di erentiate the mutant isoform from the WT by gel migration. However, when the protein bands were quanti- Immunohistochemical analysis showed that TNPO3 accumu- tated, we observed a marked (two- to threefold) increase in lated subsarcolemmally, surrounding the myonuclei, corre- the TNPO3 protein levels in patients II-1 and III-1 compared fi sponding to the perinuclear basophilic material observed in with controls ( gure 4B). fi H&E and Herovici ( gure 3C). Other markers present in these Functional studies subsarcolemmal masses included desmin, tropomyosin, Immunoblotting of Hela cells following transient transfection CRYAB, and mitochondrial protein CHCHD10 (in addition, with plasmids expressing HA-tagged TNPO3-WT, TNPO3 some lysosomal associated membrane protein 2-positive carrying the previously reported 15 amino acid C-terminal granules were observed in the perinuclear location.) The ac- fi extension (HA-TNPO3+15) or the currently reported cumulated myo brillar material in the cytoplasm stained for frameshift mutation (HA-TNPO3-RG) (figure 5A) revealed desmin, myotilin, CRYAB, ubiquitin, and p62/sequestosome 1 similar expression levels and a slight increase in the molecular (SQSTM1) and, to a lesser extent, for tropomyosin and lyso- fi weight of HA-TNPO3+15 and HA-TNPO3-RG as expected somal associated membrane protein 2. Rimmed vacuolar bers (figure 5B). Immunofluorescence of exogenously expressed were positive for p62/SQSTM1, microtubule-associated pro- fi TNPO3 using an HA antibody revealed similar nuclear lo- teins 1A/1B light chain 3, TDP-43, and ubiquitin ( gure 3H). calization in addition to cytosolic puncta that were signifi- cantly enriched in HA-TNPO3-WT–expressing cells No gross abnormality of nuclear markers matrin-3 and emerin compared with TNPO3 mutants (figure 5, C and D). was observed with the IF technique. Confocal analysis showed diffuse to granular nuclear localization of TNPO3 in the control fi To more clearly identify the cytosolic structure containing HA- muscle and patient III-1 (age 16 months) ( gure 3O), whereas TNPO3-WT, we performed dual immunofluorescence with an antibody to the stress granule (SG) marker TIA1 because it has recently been reported that nuclear import receptors associate Figure 4 Western blotting with SGs. Although HA-TNPO3-WT and HA-TNPO3-RG do localize to a subset of TIA1-positive SGs, the larger cytosolic HA-TNPO3-WT puncta do not colocalize even when cells are treated with arsenite to induce SGs in HeLa cells (figure 5E).

We reasoned that cytosolic TNPO3 puncta may colocalize with nuclear pore complexes (PCs) that assemble in the cytoplasm at structures termed annulate lamellae pore complexes (ALPCs). In addition to nucleoporins, ALPCs contain much of the machinery necessary for nuclear transport including Ran- GAP1 and both nuclear import and export receptors. Co- immunofluorescence with antibodies to nuclear PCs (Mab414) or RanGAP1 and HA in HeLa cells transfected with HA-TNPO3-WT or HA-TNPO3-RG demonstrated strong colocalization at ALPCs in only TNPO3-WT–expressing cells.

Discussion (A) Western blotting shows approximately 3-fold increase in TNPO3 protein Only 1 family with TNPO3 mutation causing LGMD1F has expression in the patient samples analyzed (III-1 and II-1) compared with the 3,4,13 pooled control sample (C100 and C50). (B) The bands were quantitated by been reported to date. In addition, a patient with sporadic calculating the relative quantities of TNPO3/MyHC in each sample and LGMD harboring a completely different heterozygous mis- normalizing the control samples average to 1. MyHC = myosin heavy chain; TNPO3 = transportin-3. sense mutation in the C-terminal part of TNPO3 has been described4,14 The follow-up study on the large family found

6 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Figure 5 Expression and localization of HA-tagged TNPO3 constructs

(A) Schematic of HA-tagged TNPO3- WT, the previously reported LGMD1F mutation TNPO3+15 or the currently described frameshift mu- tation TNPO3-RG. (B) Immunoblot with anti-HA, TNPO3, or GAPDH of HeLa cell lysates transfected with HA-TNPO3-WT, HA-TNPO3+15, or HA-TNPO3-RG. (C) HA immunofluo- rescence of HeLa cells transiently transfected with HA-TNPO3-WT, HA- TNPO3+15, or HA-TNPO3-RG. HA- positive cytosolic puncta (arrows). (D) Quantitation of the percent of cells in C with cytosolic HA-TNPO3 puncta (*** p = < 0.0001). (E) Im- munofluorescence using antibodies to HA (green) and TIA1 (red) of HeLa cells transfected with TNPO3-WT or TNPO3-RG pre- and post-stress granule induction with arsenite. HA- positive/TIA1-negative puncta (arrowheads). (F) Immunofluores- cence using antibodies to HA (red) and RanGAP1 (green) or nucleo- porin antibody Mab414 (green) of HeLa cells transfected with TNPO3- WT or TNPO3-RG. Cytosolic puncta (arrowheads). Scale bars (C, E, and F) =20μm. TNPO3 = transportin-3; WT = wild type.

that the mutation c.2771delA segregated in the total of 45 with main involvement of the vastus lateralis, rectus femoris, individuals of 115 studied family members with the estimated sartorius, and gracilis muscles and at later stages of the calf and penetrance rate of 84.7%.5 Age at onset ranged from less than peroneus longus muscles.3,6 1 year to 58 years. Based on the onset and the rate of pro- gression, a juvenile form (onset before age 15 years, severe We report the second transportinopathy family with a novel disability) and an adult form (onset in the third to fourth but almost identical frameshift mutation in the TNPO3 gene decades, slower progression) were presented.3,12 In the ju- compared with the one identified in the first family.3,12 The venile group, 6 patients had onset at age 1 year with delayed congenital very early-onset generalized limb weakness and early motor skills. Although our patients had very early-onset slow progression in our patients differed from the phenotype proximal weakness or congenital hypotonia, the disease was of most of the patients in the primary family, although some stable during childhood and slow progression started in patients with congenital disease also were encountered. Two adulthood. The youngest of our patients had a mild decrease of our patients had already very severe fatty degeneration in in respiratory functions; none had cardiomyopathy or most of their lower limb muscles making comparison difficult, hyperlordosis. Also, muscle MRI findings have been variable whereas earliest involvement can be observed in the sartorius,

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 7 gracilis, adductor magnus, and peroneus longus muscles develops later and may contribute to later events in the muscle (figure 2C). In contrast to the previous reports, we found that pathology. Rimmed vacuolar pathology with p62/SQSTM1, TNPO3 protein was quantitatively increased in the patient TDP-43, microtubule-associated proteins 1A/1B light chain muscle and accumulated in the subsarcolemmal and peri- 3, and valocin-containing protein-positive material and accu- nuclear areas in the muscle fibers. mulated autophagosomal membranes at the ultrastructural level indicate induction of the autophagosomal degradation The detected mutation c.2757delC is predicted to cause pathway. Again, the number of rimmed vacuolated fibers was a frameshift and transfer the stop codon (p.R920Gfs*20). The rather low. Defects in RNA metabolism are connected to predicted outcome on protein level is almost identical to the defective autophagy and rimmed vacuolar pathology, as seen original TNPO3 mutation c.2771delA p.*924Cext*15 (the in multisystem proteinopathy pathologies.16 amino acid sequence changed by the mutations is underlined): The muscle histopathologic findings and experimental data so far implicate that there may be several possible mechanisms involved in the transportinopathy molecular pathophysiology. WT …TRLFR* To elucidate the pathogenic effects of the described TNPO3 c.2771delA …TRLFRCSHSCTVPVTQECLF* mutations, studies clarifying how they affect the nuclear

c.2757delC …TGCFDSSHSCTVPVTQECLF* import/export dynamics will be required. In addition, the splicing of muscle genes, especially those regulated by the TNPO3 cargo proteins, needs to be analyzed. However, toxic Actually, there is just a 5 amino acid difference between the 2 proteins often shed little or no light on the functions of the protein products. Hence, it is expected that the patho- normal cognate. mechanisms are largely shared as well. Indeed, we found that expression of the previously reported TNPO3 variant be- Study funding haved similarly to our reported variant in cell culture. Spe- The study was funded by the Finnish Medical Foundation cifically, TNPO3-WT localizes to the nuclear envelope and (J.P.), the Academy of Finland (B.U.), the Sigrid Jus´elius a cytosolic organelle termed annulate lamellae (ALs). ALs are Foundation (B.U.), and the Folkh¨alsan Research Foundation cytosolic accumulations of nuclear PCs and the molecular (B.U.). This study is not industry sponsored. machinery necessary for nuclear import/export. The function of ALs and ALPCs is unclear but may serve as repository Disclosure necessary for replacement of nuclear pores and import/export Disclosures available: Neurology.org/NG. components. The effect of TNPO3 mutations on nuclear pore function remains to be established. TNPO3 is a nuclear Publication history transport receptor, regulating nucleocytoplasmic shuttling of Received by Neurology: Genetics February 11, 2019. Accepted in final several S/R-rich splicing factors. Perinuclear accumulation form March 27, 2019. and simultaneous nuclear depletion of TNPO3 and RBM4 observed in the patient muscle biopsy are in line with these mechanisms. Consequently, processing of specific mRNA species may be altered, leading to aberrant splicing and ex- Appendix Authors

pression of muscle genes. However, the nuclear localization of Name Location Role Contribution SRRM2 indicates that these transcription factors have also Anna Folkh¨alsan Institute of Author Study design and other alternatives to TNPO3 for nuclear entry. Physical ac- Vihola, PhD Genetics and conceptualization; cumulation of TNPO3 in the perinuclear/subsarcolemmal University of Helsinki, drafting of the regions, shown by IHC and Western blotting, together with Helsinki, Finland manuscript; acquisition of data; and other proteins and organelles, could disturb normal nuclear interpretation of data functions in several ways. Indeed, the patient biopsies show Johanna Tampere University Author Study design and nuclear anomalies. Palmio, MD, and University conceptualization; PhD Hospital of Tampere, drafting of the fi Finland manuscript; Myo brillar pathology was present in the patient biopsies in acquisition of data; and this study, although less extensively than in conventional interpretation of data fi myo brilliar myopathies such as myotilinopathy or zaspop- Olof Department of Clinical Author Acquisition of data 15 athy, shown by accumulation of myotilin and desmin, as Danielsson, and Experimental and interpretation of 6 fi MD, PhD Medicine, Linkoping¨ data reported before in the Spanish family. As a novel nding, we University, Sweden found that the myofibrillar changes were present as early as at Sini Tampere University Author Acquisition of data; age 16 months, which is in line with the congenital onset of Penttil¨a, and University interpretation of symptoms. Notably, we did not observe perinuclear TNPO3 PhD Hospital of Tampere, data; and drafting accumulation or depletion from nuclei (or other abnormal Finland and revision for intellectual content localization) at that early stage, suggesting that this pathology

8 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG 2. Straub V, Murphy A, Udd B. LGMD workshop study group. 229th ENMC in- — Appendix (continued) ternational workshop: limb girdle muscular dystrophies nomenclature and re- formed classification Naarden, the Netherlands, 17–19 March, 2017. Neuromuscul – Name Location Role Contribution Disord 2018;28:702 710. 3. Meli`a MJ, Kubota A, Ortolano S, et al. Limb-girdle muscular dystrophy 1F is caused by a microdeletion in the transportin 3 gene. Brain 2013;136:1508–1517. Daniel Washington Author Acquisition of data, fi Louiselle, University in St. Louis, editing of the 4. Torella A, Fanin M, Mutarelli M, et al. Next-generation sequencing identi es trans- MS MO manuscript; and portin 3 as the causative gene for LGMD1F. PLoS One 2013;8:e63536. interpretation of data 5. Fanin M, Peterle E, Fritegotto C, et al. Incomplete penetrance in limb-girdle muscular dystrophy type 1F. Muscle Nerve 2015;52:305–306. Sara Washington Author Acquisition of data 6. Peterle E, Fanin M, Semplicini C, Padilla JJ, Nigro V, Angelini C. Clinical phenotype, Pittman, University in St. Louis, and interpretation of muscle MRI and muscle pathology of LGMD1F. J Neurol 2013;260:2033–2041. B.S. MO data 7. Cenacchi G, Peterle E, Fanin M, Papa V, Salaroli R, Angelini C. Ultrastructural changes in LGMD1F. Neuropathology 2013;33:276–280. Conrad Washington Author Study design and 8. Maertens GN, Cook NJ, Wang W, et al. Structural basis for nuclear import of splicing Weihl, MD, University in St. Louis, conceptualization; factors by human Transportin 3. Proc Natl Acad Sci USA 2014;111:2728–2733. PhD MO editing of the 9. Bin Hamid F, Kim J, Shin CG. Cellular and viral determinants of retroviral nuclear manuscript; and entry. Can J Microbiol 2016;62:1–15. interpretation of data 10. Evil¨a A, Arumilli M, Udd B, Hackman P. Targeted next-generation sequencing assay for detection of mutations in primary myopathies. Neuromuscul Disord 2016;26: Bjarne Udd, Department of Author Study design and 7–15. MD, PhD Neurology, Vaasa conceptualization; 11. Dubowitz V, Sewry CA. Muscle Biopsy—A Practical Approach. 3 ed. London: Central Hospital, drafting and revision Saunders Elsevier; 2007. Vaasa, Finland of the manuscript; 12. Dahmane R, Djordjevic S, Simunic B, Valencic V. Spatial fiber type distribution in acquisition of data; normal human muscle Histochemical and tensiomyographical evaluation. J Biomech interpretation of 2005;38:2451–2459. data; and data 13. Gamez J, Navarro C, Andreu AL, et al. Autosomal dominant limb-girdle muscular collection and dystrophy: a large kindred with evidence for anticipation. Neurology 2001;56: analysis 450–454. 14. Gibertini S, Ruggieri A, Saredi S, et al. Long term follow-up and further molecular and histopathological studies in the LGMD1F sporadic TNPO3-mutated patient. Acta Neuropathol Commun 2018;6:141. References 15. Selcen D. Myofibrillar myopathies. Neuromuscul Disord 2011;21:161–171. 1. Nigro V, Savarese M. Genetic basis of limb-girdle muscular dystrophies: the 2014 16. Buchan JR, Kolaitis RM, Taylor JP, Parker R. Eukaryotic stress granules are cleared by update. Acta Myol 2014;33:1–12. autophagy and Cdc48/VCP function. Cell 2013;153:1461–1474.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 9 ARTICLE OPEN ACCESS DMPK gene DNA methylation levels are associated with muscular and respiratory profiles in DM1

C´ecilia L´egar´e, MSc, Gayle Overend, PhD, Simon-Pierre Guay, PhD, Darren G. Monckton, PhD, Correspondence Jean Mathieu, MD, MSc, Cynthia Gagnon, PhD, and Luigi Bouchard, PhD, MBA Dr. Bouchard [email protected] Neurol Genet 2019;5:e338. doi:10.1212/NXG.0000000000000338 Abstract Objective To assess the effects of dystrophia myotonica protein kinase (DMPK) DNA methylation (DNAme) epivariation on muscular and respiratory profiles in patients with myotonic dys- trophy type 1 (DM1).

Methods Phenotypes were assessed with standardized measures. Pyrosequencing of bisulfite-treated DNA was used to quantify DNAme levels in blood from 90 patients with DM1 (adult form). Modal CTG repeat length was assessed using small-pool PCR. The presence of Acil-sensitive variant repeats was also tested.

Results DNAme levels upstream of the CTG expansion (exon and intron 11) were correlated with − − − modal CTG repeat length (rs = 0.224, p = 0.040; rs = 0.317, p = 0.003; and rs = 0.241, p = 0.027), whereas correlations were observed with epivariations downstream of the CTG repeats (rs = 0.227; p = 0.037). The presence of a variant repeat was associated with higher DNAme levels at multiple CpG sites (up to 10% higher; p = 0.001). Stepwise multiple linear regression modeling showed that DNAme contributed significantly and independently to explain phe- notypic variability in ankle dorsiflexor (3 CpGs: p = 0.001, 0.013, and 0.001), grip (p = 0.089), and pinch (p = 0.028) strengths and in forced vital capacity (2 CpGs: p = 0.002 and 0.021) and maximal inspiratory pressure (p = 0.012).

Conclusions In addition to the CTG repeat length, DMPK epivariations independently explain phenotypic variability in DM1 and could thus improve prognostic accuracy for patients.

From the Department of Biochemistry (C.L., S.-P.G., L.B.), Universit´e de Sherbrooke, Sherbrooke; ECOGENE-21 Biocluster (C.L., S.-P.G., L.B.), Chicoutimi, Qu´ebec, Canada; Groupe de Recherche interdisciplinaire sur les maladies neuromusculaires (C.L., J.M., C.G., L.B.), Saguenay, Canada; Institute of Molecular (G.O., D.G.M.), Cell and Systems Biology, University of Glasgow, United Kingdom; and Centre de Recherche Charles-Le-Moyne-Saguenay-Lac-StJean sur les innovations en sant´e (J.M., C.G.), Facult´edem´edecine et des sciences de la sant´e, Universit´e de Sherbrooke, Canada.

Go to Neurology.org/NG for full disclosures. Funding information is provided at the end of the article.

The Article Processing Charge was funded by the authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary DM1 = dystrophy type 1; DMPK = dystrophia myotonica protein kinase; MIRS = Muscular Impairment Rating Scale.

Myotonic dystrophy type 1 (DM1) is an autosomal dominant (1) age ≥18 years, (2) molecular confirmation of diagnosis multisystemic1 disorder caused by a CTG repeat expansion in (late-onset vs adult phenotypes), and (3) capacity to provide the 3’untranslated region of the dystrophia myotonica protein informed consent. The adult phenotype was assigned to par- – kinase (DMPK) gene.2 4 DM1 affects all systems including ticipants who had at least 2 of the following characteristics: (1) muscular, cardiac, respiratory, endocrine, and CNS. The number of CTG repeats >200 in the standard diagnostic test, hallmark feature of DM1 is the involvement of the muscular (2) Muscular Impairment Rating Score (MIRS) ≥3, and (3) system including presence of myotonia, progressive skeletal age at symptom onset <40 years.17 Age at onset was defined as muscle impairment with a distal to proximal pattern, and the age at which the participant presented his first symptom. To respiratory muscle. The phenotypic variability observed in reduce phenotypic variability that is present between patients terms of age at onset, clinical presentation, and severity has led with DM1 presenting different phenotypes of the disease, only to a clinical classification that has evolved over time going 90 participants presenting the adult form of the disease were from 35 to 5 phenotypes recognized today.6 Although patients kept for the analysis. Patients with other forms of muscular born with a larger expansion usually present a more severe dystrophies including congenital and childhood DM1 were phenotype than those with a smaller one, the CTG repeat thus excluded. length only partially explains this large phenotypic variability. Muscle strength measurements DNA methylation (DNAme) regulates gene expression7 and An isometric make test using a MicroFET2 dynamometer alternative splicing8 without modifying the nucleic acid se- (Hoggan Health Industries, West Jordan, UT) was performed quence of the genes. DNAme occurs mainly on cytosines twice by a trained physiotherapist to measure the strength of upstream of a guanine (CpG dinucleotides). The DMPK the shoulder abductors, knee extensors, and ankle dorsi- CTG expansion locus harbors a long CpG island, a genomic flexors. The participant’s position was standardized for each structure enriched in CpGs prone to DNAme. This led to muscle group. If there was more than 10% discrepancy be- hypothesize that DNAme contributes to disease development tween 2 assessments for a muscle group, a third measure was and progression and explains part of the phenotypic variability performed, and the mean of the 2 closest values was kept as – reported in DM1.9 16 the final result. The maximal isometric torque was computed by multiplying strength units (N) by length of the corre- 18 Only a few studies have been conducted, and all have con- sponding lever arm (m). A Jamar dynamometer (Asimow siderable limitations such as small sample size and inclusion of Engineering Co, Sequim, WA) was used to measure grip 19 multiple DM1 phenotypes, limiting comparison between strength. Three measures were performed with a standard- studies and the conclusions, which can be drawn. None tested ized protocol, and the mean was used as the final result. A B & associations with DM1 clinical phenotypes, except for the L Pinch Gauge (Fabrication Enterprises Inc, New York, NY) Muscular Impairment Rating Scale (MIRS) in 1 study.12 was used to measure lateral pinch strength. Again, 3 measures Therefore, we have assessed DNAme levels at the DMPK were performed with a standardized protocol, and the mean 19 gene locus and tested for association with muscular and re- was used as the final result. spiratory impairments in the largest number of patients with DM1 with the adult phenotype. Respiratory profile assessment Forced vital capacity, peak expiratory flow, and maximal in- spiratory and expiratory pressure (cm H2O) were measured by Methods a respiratory therapist at the Jonquiere Hospital following a standardized protocol. Percentage of predicted values for Standard protocol approvals, registrations, forced vital capacity and peak expiratory flow (% predicted) and patient consents were calculated using the equation as in reference 20. Smoking The study was approved by the Ethics Review Board of the status was determined using a life habit questionnaire. Centre de sant´e et de services sociaux de Chicoutimi, Canada, and all participants provided written consent. DNA extraction and methylation analyses Blood was collected in EDTA tubes during the same visit Participants muscular and respiratory profiles were assessed. Buffy coats In 2002, 200 of the 416 patients with DM1 followed at the were kept at −80°C until DNA extraction using the Gentra Neuromuscular Clinic of the Centre de sant´e et des services Puregene Blood Kit (Qiagen). sociaux de Jonqui`ere (Qu´ebec, Canada) were randomly se- lected and recruited in a longitudinal study aimed at identifying DNA methylation analysis takes advantage of the chemical the determinants of social participation. Selection criteria were propriety of unmethylated cytosines, which are sensitive to

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG CTG repeat length analysis and variant Table 1 Characteristics of the sample repeat analysis

Mean ± SD Range Small-pool PCR and Southern blot hybridization were used to measure CTG repeat length. PCR amplification of the DMPK – Age 41.0 ± 7.7 20.0 59.0 CTG repeat was performed using the flanking primers DM-C Sex: women (n, %) n = 55 (61%) NA and DM-DR, as previously described in reference 21. The PCR buffer was Custom PCR Master Mix-No Taq (Thermo Weight (kg) 67.8 ± 16.3 34.9–115.8 Scientific#SM0005)supplementedwith69mM Body mass index (BMI) (kg/m2) 25.21 ± 5.61 14.34–43.05 2-mercaptoethanol. Taq polymerase (Sigma-Aldrich, United Smoking (pack-years) 7.3 ± 11.0 0–56.0 Kingdom) was used at 1 unit per 10 μL. Where required, reactions were supplemented with 10% dimethyl sulfoxide Muscular Impairment Rating Scale (MIRS) category (n, %) (Sigma-Aldrich), and the annealing temperature reduced from 68°C to 63.5°C. PCR products were digested with Acil (New 2 n = 7 (7.8%) NA England Biolabs UK Ltd) in accordance with the manu- 3 n=22 NA facturer’s instructions. PCR-amplified DM1 alleles were run on (24.4%) 1% agarose gels in 0.5X tris-borate-EDTA, blotted, and hy- 4 n=56 NA bridized as described in reference 21. α-32P-dCTP (Perkin (62.2%) Elmer UK) and the Invitrogen Random Primers DNA Labeling 5 n = 5 (5.6%) NA System (Thermo Fisher Scientific UK) were used to label the molecular weight marker (Invitrogen 1 kb+ ladder, Thermo Modal CTG repeat length 603.4 ± 68–1,321 267.9 Fisher Scientific UK) and DNA probes. Flanking primers DM- C and DM-DR and a patient with DM1 with 56 CTG repeats Variant repeats (present n, %) 8 (8.9%) NA were used to amplify the CTG probe. After autoradiography, Age at symptom onset 20.4 ± 7.7 10.0–43.0 blot images were scanned, and then, estimated progenitor allele

Shoulder abductors strength (Nm) 40.8 ± 15.7 14.3–94.2 length and baseline modal allele lengths were estimated from the lower boundary21 and the densest part of the expanded – Knee extensors strength (Nm) 80.9 ± 33.1 18.9 165.3 alleles,22 respectively. Allele length was estimated by compari- Ankle dorsiflexors (Nm) 14.2 ± 6.7 2.2–35.6 son to the 1 kb+ ladder, using CLIQS 1D gel analysis software (TotalLabs UK). Somatic instability was computed as modal Grip strength (kg) 8.6 ± 6.4 0.0–28.5 allele length minus progenitor allele length. Pinch strength (kg) 5.2 ± 1.5 2.2–9.2

Forced vital capacity (% of predicted 82.9 ± 18.7 35.8–130.9 Statistical analyses value) Normality of distribution was tested using the Kolmogorov- Smirnov test and then visually confirmed on histograms. The Peak expiratory flow (% of predicted 73.1 ± 16.2 33.4–113.8 value) following variables were rank transformed to respect the normality condition of the regression analysis: age, modal and Maximal inspiratory pressure (H Ocm) 66.4 ± 23.5 26.0–138.0 2 progenitor CTG repeat length, DNA methylation levels, – Maximal expiratory pressure (H2Ocm) 58.5 ± 22.4 23.0 123.0 smoking, weight, shoulder abductors, knee extensors, ankle dorsiflexors, grip and pinch strengths, percentage of predicted Abbreviation: NA = not applicable. forced vital capacity, percentage of predicted peak expiratory flow, maximal expiratory and inspiratory pressure, age at symptom onset, and instability. Spearman correlation coef- sodium bisulfite treatment (NaBis; EpiTect Bisulfite Kit, ficients were used to assess the association between DNAme Qiagen). Treatment of unmethylated cytosines thus generates and CTG repeat length. Mann-Whitney U tests were then aC→T (U) transition, whereas methylated cytosines are applied on residual scores to assess the differences in DNAme protected from deamination to uracils in the presence of between patients with and without repeat variants. For NaBis, thus remaining cytosines. Combined with NaBis Spearman and Mann-Whitney U analysis, linear regression treatment, pyrosequencing is a quantitative real-time se- and resultant unstandardized residuals were used to adjust quencing technology that measures the site-specific DNA DNAme levels for sex and age of the participants. Stepwise methylation percentage for each CpG of a given sequence. multiple linear regression analyses were applied to identify PCR and sequencing primers were designed using Pyromark variables independently contributing to explain muscular- and Assay Design 2.0 (Qiagen). Primer sequences are provided in respiratory-related phenotypes. Confounding factors were table e-1 (links.lww.com/NXG/A157). added to each model based on their association with the corresponding phenotype. The 4 CpGs correlated with CTG The PyroMark PCR Kit (Qiagen) was used for amplification repeat length without correction for age and sex were selected of bisulfite-treated DNA. A Pyromark model q24 (Qiagen) to be included into the stepwise regression model. We used was used to quantify DNA methylation levels. relaxed 0.1 and 0.15 cutoff p values, respectively, for entry and

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 Figure 1 Spearman correlation between DMPK DNA methylation and modal CTG repeat length in patients with myotonic dystrophy type 1 (DM1)

Each panel (A to E) represents a different CpG. Results corrected for age and sex.

removal of variables within the regression. Significance levels Results were set to p < 0.05. Bonferroni correction was applied to adjust for multiple testing. Statistical analyses were performed Description of study participants and overall on IBM SPSS statistics version 24 (IBM) software. DNAme profile within the DMPK locus The participants are, on average, middle aged and normal Data availability weight, with women being slightly overrepresented. Partic- The data sets generated and/or analyzed during the current ipants were aged 41.0 years (±7.7) and 603.4 repeats (±267.9) study are not publicly available for ethical reasons, but are on average (table 1). Eight of the 90 participants were carriers available from the corresponding author on reasonable request. of a CTG repeat variant (8.9%).

4 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Table 2 Multiple linear regression analysis of muscle impairments of patients with DM1

Shoulder Knee Ankle Grip Pinch

Age NR β = −0.223 NR NR β = −0.173 p = 0.022 p = 0.071

Sex β = 0.703 β = 0.482 p < 0.001 NR NR β = 0.339 p < 0.001 p = 0.001

Modal CTG repeat length β = −0.146 NR β = −0.442 β = −0.476 β = −0.387 p = 0.066 p < 0.001 p < 0.001 p < 0.001

DNA methylation L6b NR NR NR NR NR

DNA methylation L6c NR NR β = 0.340 β = 0.173 β = 0.219 p = 0.001 p = 0.089 p = 0.028

DNA methylation L1i NR NR β = -0.363 NR NR p = 0.013

DNA methylation L1k NR NR β = 0.467 NR NR p = 0.001

Variant repeat (presence/absence) NR β = −0.225 NR NR NR p = 0.021

R2 of the model (p value) 0.529 (<0.001) 0.336 (<0.001) 0.323 (<0.001) 0.222 (<0.001) 0.305 (<0.001) p Value, Bonferroni corrected <0.001 <0.001 0.001 0.002 <0.001

Abbreviations: ankle = ankle dorsiflexor strength; DM1 = myotonic dystrophy type 1; grip = grip strength; knee = knee extensors strength; NR = variables not retained in the model in the final model; pinch = pinch strength; shoulder = shoulder abductors strength.

DNAme levels were quantified at 8 different loci (L1 to L8) higher DNAme levels only with CpGs just downstream of the within the CpG island covering 35 and 17 CpG sites upstream repeat. On average, DNAme levels at any of the 5 CpGs tested and downstream of the CTG expansion, respectively (figure were 3–4 times higher (L6; figure e-2, links.lww.com/NXG/ e-1, links.lww.com/NXG/A154). Some of these CpGs/loci A155). For example, we observed that the mean DNAme level have previously been studied in patients with DM1 or in at L6c for patients with a variant repeat was 15.5%, whereas it – mouse models.9 16 Upstream of the CTG expansion, DNAme was 4.7% for patients without Acil-sensitive sequence (p < profiles are multidimensional with those proximal to the CTG 0.001), even after correction for modal CTG and progenitor expansion (L3, 4, and 5 in intron 14 and exon 15) being allele length. Although a nonparametric statistical test was unmethylated (<5%) and those more distally located showing applied, these associations seem primarily driven by 3 par- increasing levels of DNAme from ;50% methylation, at L2 ticipants. We have carefully reviewed their familial, metabolic, (intron 13) to close to fully methylated CpGs at site L1 (exon and genetic characteristics, and the 2 patients with the highest 11). Downstream of the CTG expansion, the CpG sites tested levels of DNAme are siblings. No associations were found showed low (L6; <10%) to very low (L7 and L8; unmethy- between the presence of a variant repeat and DNAme levels lated to <5%) levels of DNAme. upstream of the CTG expansion.

Associations between DNAme and CTG repeats Stepwise multiple linear regression analyses Although the coefficients are similar, we observed that DNAme We then assessed whether DNAme levels measured at the levels were correlated with the CTG repeat length (figure 1) in DMPK gene locus were associated with muscular strength and opposite directions depending on whether they were upstream or respiratory profile. Using multiple linear regression models and downstream of the CTG expansion. Briefly, DNAme levels in stepwise procedures, we found that DNAme contributed to exon and intron 11 were correlated with modal CTG repeat explain variability in muscular and respiratory profiles in- − length after correction for age and sex (L1g [rs = 0.224; p = dependently of the CTG repeat length. These results are shown − − 0.040], L1i [rs = 0.317; p = 0.003] and L1k [rs = 0.241; p = in detail in table 2 for tested muscular strength phenotypes and 0.027]), whereas those downstream showed a correlation (L6b table 3 for respiratory measures. Very briefly, DNAme levels at fi [rs = 0.227; p = 0.037] and L6c [rs = 0.201; p = 0.065]). The one or more loci contributed signi cantly to every DM1- results remain significant with slightly stronger correlation coef- phenotype tested, except for knee and shoulder strengths and ficients after excluding patients with variant repeat. peak expiratory flow. Of interest, robust and persistent asso- ciations were observed between DNAme levels at L6c and all 3 We then assessed the links between DNAme and presence or muscular-related phenotypes tested, whereas our results sug- absence of a variant repeat within the CTG expansion. We gest a similar pattern of associations between DNAme levels found that the presence of a variant repeat was associated with measured at L6b and all respiratory-related phenotypes.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 5 Table 3 Multiple linear regression analysis of respiratory impairments of patients with DM1

FVC PEF MIP MEP

Age NA NA β = −0.332 NR p = 0.008

Sex NA NA NR β = 0.417 p < 0.001

Modal CTG repeat length β = −0.283 NR NR NR p = 0.012

DNA methylation L6b β = 0.368 β = −0.236 β = 0.266 β = −0.211 p = 0.002 p = 0.039 p = 0.083 p = 0.067

DNA methylation L6c NR NR β = −0.399 NR p = 0.012

DNA methylation L1i NR NR NR NR

DNA methylation L1k β = 0.249 NR NR NR p = 0.021

Variant repeat (presence/absence) NR NR NR NR

Smoking (pack-years) NR β = −0.228 NR NR p = 0.046

Weight β = −0.266 NR NR NR p = 0.013

R2 of the model (p value) 0.283 (<0.001) 0.107 (0.018) 0.167 (0.012) 0.219 (0.001)

p Value, Bonferroni corrected <0.001 0.072 0.048 0.004

Abbreviations: DM1 = myotonic dystrophy type 1; FVC = percentage of predicted forced vital capacity; MEP = maximal expiratory pressure; MIP =maximal inspiratory pressure; NA = variables not considered in the model based on their lack of association with the phenotype; NR = variables not retained in the final model; PEF = percentage of predicted peak expiratory flow.

DNAme levels were also tested for associations with somatic the adult form of the disease, which we believe reduces the instability of the CTG repeat and age at symptom onset. We confounding effects of extreme phenotypic and age at sampling found that DNAme levels were associated with variability in variability that characterizes broader DM1 cohorts and allows somatic instability (differences between progenitor and av- us to reveal new insights into the role of DNAme in DM1. erage repeat length) of the CTG repeat, but not with age at symptom onset (table 4). All models except that for peak Figure e-3 (links.lww.com/NXG/A156) summarizes the expiratory flow remain significant after Bonferroni correction. results published so far. Our results for sites L1 and L2 are in In addition to DNAme levels at other loci, age, sex, and line with some studies15,16 showing constitutive DNAme in smoking status contributed to explain one or more of the intron 11, but not with 1 study,13 although we report differ- phenotypes tested. ences in DNAme levels between individuals at these CpG sites. Our results are also in agreement with those previously published13,15,16 showing that the L3 region is unmethylated, Discussion although they found some DNAme when patients with DM1 had more than 300 CTG repeats.13,15,16 For the L4 and L5 Although epigenetics (DNAme) has been a mechanism of regions, hypomethylation of the region has been observed in – interest in DM1 since 1993, only 7 studies have been pub- other studies,9,12 15 although 1 study12 found DNAme at lished so far. Of them, 5 focused on the DM1 congenital various degrees in patients with pure CTG repeats, as did an- – phenotype.9,11,13 15 Two other studies reported DNAme other.11 For L6, the region has been found to present some analyses related to DMPK and SIX5 gene expression,10,16 and variability in our study and in mouse,10,11 in patients with variant only one tested association between a clinical phenotype, repeats,12 and patients with DM1.9 This region has been asso- namely the MIRS.12 We are the first to study and demonstrate ciated with hypomethylation in patients with DM111,13,15 and that DNAme at the DMPK gene locus contributes to vari- patients with DM1 with a pure CTG tract.12 The L7 and L8 sites ability of both muscular strengths and respiratory profiles. have been found to be hypomethylated both by our group and in Our results also support that DNAme contributes to DM1 a previous study.13 The studies published so far are of small to phenotypic variability independently of the CTG repeat length. very small sample size. Also, inconsistencies between the studies Our study has been conducted exclusively in 90 patients with might be explained by the different DNAme quantification

6 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG Table 4 Multiple linear regression analysis of phenotypic variability of age at onset of patients with DM1 and somatic instability of their repeats

Age at onset Somatic instability

Modal CTG Progenitor CTG Modal CTG Progenitor CTG

Age NA NA NR NR

Sex NR NR β = −0.079 β = −0.137 p = 0.054 p = 0.078

Modal CTG repeat length β = −0.380 NA β = 0.978 NA p = 0.001 p < 0.001

Progenitor allele length NA β = −0.339 NA β = 0.811 p = 0.003 p = <0.001

DNA methylation L6b NR NR β = −0.202 β = −0.254 p = <0.001 p = 0.005

DNA methylation L6c NR NR NR NR

DNA methylation L1i NR NR NR NR

DNA methylation L1k NR NR NR NR

Variant repeat (presence/absence) NR NR NR NR

R2 of the model (p Value) 0.144 (0.001) 0.115 (0.003) 0.870 (<0.001) 0.533 (<0.001) p Value, Bonferroni corrected 0.002 0.006 <0.001 <0.001

Abbreviations: age at onset = age at symptom onset; DM1 = myotonic dystrophy type 1; NA = variables not considered in the model based on their lack of association with the phenotype; NR = variables not retained in the final model; somatic instability = somatic instability of the repeat.

methods, which are in some cases far from ideal. For example, Next, we tested whether DNAme was also associated with key methylation-sensitive restriction enzymes have been routinely impairments in DM1. DNAme at 2 different CpG sites, both used, but do not allow detection of small changes in DNAme within the same L6 region downstream of the CTG expan- unlike pyrosequencing. These studies have also analyzed sion, contributed independently of the CTG repeat length to a broader range of DM1 phenotypes than in our study. The variability of muscular- (L6c; ankle dorsiflexors, handgrip, and other DM1 phenotypes should thus be also investigated. pinch strength) and respiratory-related phenotypes (L6b; forced vital capacity, maximal inspiratory pressure, and max- Of interest, we also report that the CTG repeat length was cor- imal expiratory pressure). This suggests that DNAme at those related with DNAme, but in opposite directions depending on CpG sites might affect distinctive muscle groups by changing which side of the CTG expansion DNAme levels were assessed. affinity for different transcription factors. However, no spe- We report a correlation with DNAme measured in intron 11. cific transcription factor consensus sequences covering each Some groups reported similar associations for DNAme measured of the CpG sites were found in the ENCODE database. Also, in closer vicinity of the CTG expansion,9,10,12,15,16 although dis- we did not find that DNAme contributes to strength vari- crepancies exist.11,14 Overall, ours and previous results support ability of proximal muscle groups (shoulder abductors and the hypothesis that DNAme and modal CTG repeat length are knee extensors), which could be explained by several reasons correlated with each other. However, we cannot yet conclude including measurement errors, muscle impairment level, and which of the 2 influences the other. other factors.

Of interest, we also found that presence of a variant repeat within At least 3 hypotheses could explain how DNAme might drive the CTG expansion was associated with higher DNAme levels clinical variability in DM1. First, DNAme might block CTCFs downstream of the CTG repeat (L6). Similar results have been binding to one or both of the CTCF consensus sequences reported showing an association driven by participants who had located on each side of the CTG expansion, leading to CTCF more “CG” sequences in their repeat.12 The “CG” content has insulator disruption ability between DMPK and SIX5.26 Sec- not been assessed in our study, precluding us from drawing ond, DNAme might reduce the mRNA levels of DMPK and a clear conclusion. However, we found that 2 of the 3 patients SIX5 and thus reducing the number of DMPK and SIX5 with higher DNAme levels were siblings. Variant repeats were protein.10,16 Third, DNAme might alter CTG repeat in- first observed in a small number of families,24,25 suggesting that stability 27,28 through disruption of CTCF binding,28 although they might be inherited. As such, the pattern of each variant another study reported that it was DNA demethylation that repeat is also probably inherited and similar within families. was promoting repeat instability27 with another one showing

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 7 no link between DNAme and CTG repeat instability.14 Of interest, we found that increased DNAme was associated with Appendix Authors fi lower somatic instability. Over time, this might be bene cial Name Location Role Contribution for the patients, as longer repeats usually means a worse prognosis. C´ecilia Department of Author Performed the data L´egar´e, Biochemistry, collection, statistical MSc Universit´ede analyses, and wrote Sherbrooke, the manuscript. Sherbrooke, Qu´ebec, Strengths and limitations Canada; ECOGENE-21 Biocluster, Chicoutimi, Our study is the largest in terms of number of patients with Qu´ebec, Canada; Groupe de recherche DM1 with the same phenotype (adult form) and the second interdisciplinaire sur les largest overall. Our participants were extensively phenotyped maladies and genotyped, including measures of CTG repeat length and neuromusculaires occurrence of CTG variant repeats. DNAme was measured in Gayle Institute of Molecular, Author Contributed to the Overend, Cell and Systems data collection and blood only (this could also be seen as a limitation) and using PhD Biology, University of revised the pyrosequencing, the gold standard method to assess CpG- Glasgow manuscript. specific DNAme levels, thus reducing potential technical vari- Simon- Department of Author Revised the ability. One limitation is that we only covered a small part of the Pierre Biochemistry, manuscript. DM1 locus (<3%), leaving many other CpG sites to be tested. Guay, PhD Universit´ede Sherbrooke, Another limitation is that we did not have access to non-DM1 Sherbrooke, Qu´ebec, participants, and thus, comparisons with what could be con- Canada; ECOGENE-21 fi Biocluster, Chicoutimi, sidered a normal DNAme pro le were not possible. However, Qu´ebec, Canada the DMPK region has previously been found to be hypo- Darren G. Institute of Molecular, Author Conceived the study methylated near the CTG repeat and hypermethylated in the Monckton, Cell and Systems design and revised region of exons 11 and 12 in controls and the general population PhD Biology, University of the manuscript. in various studies.9,11,12,29 No new data on the methylation Glasgow status of the DMPK region are available on the epigenome Jean Centre de recherche Author Conceived the study 29 Mathieu, Charles-Le-Moyne, design and revised roadmap than the data already published. MD, MSc Facult´edem´edecine et the manuscript. des sciences de la DNAme at the DMPK gene locus contributes to specific sant´e, Universit´ede Sherbrooke. Groupe de clinical phenotypes (muscular strengths and respiratory recherche profiles) in DM1. These associations are independent of the interdisciplinaire sur les maladies CTG repeat length. Our results thus provide evidence that neuromusculaires. measuring DNAme might help to predict progression of the Cynthia Centre de recherche Author Conceived the study disease and to establish a more reliable prognosis for those Gagnon, Charles-Le-Moyne, design, supervised patients. We nevertheless agree that further studies, in- PhD Facult´edem´edecine et all steps of the study, des sciences de la and participated in cluding longitudinal studies and in other DM1 forms, are sant´e, Universit´ede manuscript writing needed before strong conclusions can be drawn and clini- Sherbrooke. Groupe de and revision. recherche cally applied. interdisciplinaire sur les maladies Acknowledgment neuromusculaires. Pascal-Denys Grenier performed the quantitative muscle Luigi Department of Author Conceived the study testing during a single visit. C´eline B´elanger revised the Bouchard, Biochemistry, design, supervised PhD, MBA Universit´ede all steps of the study, language of the manuscript. Sherbrooke, and participated in Sherbrooke, Qu´ebec, manuscript writing Canada; ECOGENE-21 and revision. Study funding Biocluster, Chicoutimi, This project has been supported by the Marigold Foundation, Qu´ebec, Canada; Groupe de recherche the Canadian Institutes of Health Research (CIHR) (#JNM- interdisciplinaire sur les 108412), the Fondation du grand d´efi Pierre Lavoie, and maladies Muscular Dystrophy UK. neuromusculaires

Disclosure Disclosures available: Neurology.org/NG. References 1. Thornton CA. Myotonic dystrophy. Neurol Clin 2014;32:705–719. 2. Brook JD, McCurrach ME, Harley HG, et al. Molecular basis of myotonic dystrophy: Publication history expansion of a trinucleotide (CTG) repeat at the 3’ end of a transcript encoding – fi a protein kinase family member. Cell 1992;68:799 808. Received by Neurology: Genetics January 31, 2019. Accepted in nal form 3. Fu YH, Pizzuti A, Fenwick RG Jr, et al. An unstable triplet repeat in a gene related to April 4, 2019. myotonic muscular dystrophy. Science 1992;255:1256–1258.

8 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG 4. Mahadevan M, Tsilfidis C, Sabourin L, et al. Myotonic dystrophy mutation: an 17. Gagnon C, Mathieu J, Noreau L. Life habits in myotonic dystrophy type 1. J Rehabil unstable CTG repeat in the 3’ untranslated region of the gene. Science 1992;255: Med 2007;39:560–566. 1253–1255. 18. Petitclerc E, Hebert LJ, Mathieu J, Desrosiers J, Gagnon C. Lower limb muscle 5. New nomenclature and DNA testing guidelines for myotonic dystrophy type 1 strength impairment in late-onset and adult myotonic dystrophy type 1 phenotypes. (DM1). The International Myotonic Dystrophy Consortium (IDMC). Neurology Muscle Nerve 2017;56:57–63. 2000;54:1218–1221. 19. American Society of Hand T. Clinical Assessment Recommendations.Chicago (401 6. De Antonio M, Dogan C, Hamroun D, et al. Unravelling the myotonic dystrophy type N. Michigan Ave, Chicago 60611-4267): The Society, 1992. 1 clinical spectrum: a systematic registry-based study with implications for disease 20. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung classification. Revue neurologique 2016;172:572–580. volumes and forced ventilatory flows. Eur Respir J 1993;6(suppl 16):5–40. 7. Bird A. DNA methylation patterns and epigenetic memory. Genes Dev 2002;16:6–21. 21. Monckton DG, Wong LJ, Ashizawa T, Caskey CT. Somatic mosaicism, germline 8. Shukla S, Kavak E, Gregory M, et al. CTCF-promoted RNA polymerase II pausing expansions, germline reversions and intergenerational reductions in myotonic dys- links DNA methylation to splicing. Nature 2011;479:74–79. trophy males: small pool PCR analyses. Hum Mol Genet 1995;4:1–8. 9. Barbe L, Lanni S, Lopez-Castel A, et al. CpG methylation, a parent-of-origin effect for 22. Morales F, Couto JM, Higham CF, et al. Somatic instability of the expanded CTG maternal-biased transmission of congenital myotonic dystrophy. Am J Hum Genet triplet repeat in myotonic dystrophy type 1 is a heritable quantitative trait and 2017;100:488–505. modifier of disease severity. Hum Mol Genet 2012;21:3558–3567. 10. Brouwer JR, Huguet A, Nicole A, Munnich A, Gourdon G. Transcriptionally re- 23. Raymond K, Levasseur M, Mathieu J, Desrosiers J, Gagnon C. A 9-year follow-up pressive chromatin remodelling and CpG methylation in the presence of expanded study of the natural progression of upper limb performance in myotonic dystrophy CTG-repeats at the DM1 locus. J Nucleic Acids 2013;2013:567435. type 1: a similar decline for phenotypes but not for gender. Neuromuscul Disord 11. Lopez Castel A, Nakamori M, Tome S, et al. Expanded CTG repeat demarcates 2017;27:673–682. a boundary for abnormal CpG methylation in myotonic dystrophy patient tissues. 24. Braida C, Stefanatos RK, Adam B, et al. Variant CCG and GGC repeats within the Hum Mol Genet 2011;20:1–15. CTG expansion dramatically modify mutational dynamics and likely contribute to- 12. Santoro M, Fontana L, Masciullo M, et al. Expansion size and presence of CCG/ ward unusual symptoms in some myotonic dystrophy type 1 patients. Hum Mol CTC/CGG sequence interruptions in the expanded CTG array are independently Genet 2010;19:1399–1412. associated to hypermethylation at the DMPK locus in myotonic dystrophy type 1 25. Musova Z, Mazanec R, Krepelova A, et al. Highly unstable sequence interruptions of the (DM1). Biochim Biophys Acta 2015;1852:2645–2652. CTG repeat in the myotonic dystrophy gene. Am J Med Genet A 2009;149A:1365–1374. 13. Shaw DJ, Chaudhary S, Rundle SA, et al. A study of DNA methylation in myotonic 26. Filippova GN, Thienes CP, Penn BH, et al. CTCF-binding sites flank CTG/CAG dystrophy. J Med Genet 1993;30:189–192. repeats and form a methylation-sensitive insulator at the DM1 locus. Nat Genet 2001; 14. Spits C, Seneca S, Hilven P, Liebaers I, Sermon K. Methylation of the CpG sites in the 28:335–343. myotonic dystrophy locus does not correlate with CTG expansion size or with the 27. Gorbunova V, Seluanov A, Mittelman D, Wilson JH. Genome-wide demethylation congenital form of the disease. J Med Genet 2010;47:700–703. destabilizes CTG.CAG trinucleotide repeats in mammalian cells. Hum Mol Genet 15. Steinbach P, Glaser D, Vogel W, Wolf M, Schwemmle S. The DMPK gene of severely 2004;13:29792989. affected myotonic dystrophy patients is hypermethylated proximal to the largely 28. Libby RT, Hagerman KA, Pineda VV, et al. CTCF cis-regulates trinucleotide repeat expanded CTG repeat. Am J Hum Genet 1998;62:278–285. instability in an epigenetic manner: a novel basis for mutational hot spot de- 16. Yanovsky-Dagan S, Avitzour M, Altarescu G, et al. Uncovering the role of hyper- termination. Plos Genet 2008;4:e1000257. methylation by CTG expansion in myotonic dystrophy type 1 using mutant human 29. Buckley L, Lacey M, Ehrlich M. Epigenetics of the myotonic dystrophy-associated embryonic stem cells. Stem Cel Rep 2015;5:221–231. DMPK gene neighborhood. Epigenomics 2016;8:13–31.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 9 CLINICAL/SCIENTIFIC NOTES OPEN ACCESS A novel cathepsin D mutation in 2 siblings with late infantile neuronal ceroid lipofuscinosis

Jineesh Thottath, MD,* Shamsudheen Karuthedath Vellarikkal, PhD,* Rijith Jayarajan, MSc,* Correspondence Ankit Verma, MSc, Manu Manamel, MD, Archana Singh, PhD, V. Raman Rajendran, MD, Dr. Sivasubbu [email protected] Sridhar Sivasubbu, PhD, and Vinod Scaria, PhD or Dr. Scaria [email protected] Neurol Genet 2019;5:e302. doi:10.1212/NXG.0000000000000302

Neuronal ceroid-lipofuscinosis (NCL) is a heterogeneous and rare lysosomal storage disorder characterized by the accumulation of autofluorescent materials—ceroid and lipofuscin—in the cytoplasm.1 It is manifested as a progressive destruction of neuronal cells resulting in brain atrophy, loss of vision, and other neurodegenerative phenotypes.1 Over 446 mutations in different genes have been cataloged in the NCL mutation database (ucl.ac.uk/ncl/mutation). The over- lapping phenotypes and involvement of multiple genes indicate the difficulty in the accurate diagnosis of NCLs.2 The genetic characterization using whole exome sequencing approach can accurately diagnose NCLs.3,4 We report a phenotypically inconclusive case of a rare occurrence of NCL in siblings and identified a mutation in CTSD using whole exome sequencing.

Case report This study has been approved by institutional ethical committee (IHECC proposal No. 8), and written informed consent was obtained from all participants. Two of 3 siblings, born to a third- degree consanguineous marriage (figure, A) were apparently normal at birth and during the immediate perinatal period. They attained normal developmental milestones until the age of 3 years. A regression in the milestone was noted after 3 years of age, which progressed to complete blindness in 3–4 years. The motor system involvement started as incoordination of the upper limb, later progressed with age to dysarthria, prosopagnosia, ataxia with tremor, and weakness of limbs. By the age of 7, both the siblings were incapacitated and bedridden. There was no history of seizures. The third sibling, the eldest among the 3, was apparently normal and had attained regular milestones of development.

Preliminary examination revealed spastic rigidity of all 4 limbs and bilateral extensor plantar responses. Ophthalmic examination showed pigmentary degeneration of retina with optic atrophy with a suggestive provisional diagnosis of retinitis pigmentosa. An MRI of the brain showed diffuse cerebral and cerebellar atrophy with optic nerve atrophy and white matter abnormalities. Gradient recalled echo-T2 weighted MRI images revealed symmetric hypo- intense signals in the region of substantia nigra. In addition, a large midline cyst was noted in the inferior part of posterior fossa communicating with the cisterna magna posteroinferiorly and the fourth ventricle anteriorly (figure, B). Both siblings had similar radiologic findings (figure, B). Echocardiogram showed concentric left ventricular hypertrophy, speckled appearance of myocardium, mild mitral regurgitation dynamic left ventricle outflow tract obstruction, high systolic anterior motion, and mild mitral regurgitation. The clinical and radiologic features

*Contributed equally and would like to be known as joint first authors.

From the Department of Radiodiagnosis (J.T., M.M., V.R.R.), Government Medical College, Kozhikode, Medical College PO, Kozhikode, Kerala 673008, India; Genomics and Molecular Medicine Unit (S.K.V., R.J., A.V., A.S., S.S.), CSIR Institute of Genomics and Integrative Biology; Academy of Scientific and Innovative Research (AcSIR) (S.K.V., S.S., V.S.), CSIR-IGIB South Campus; and GN Ramachandran Knowledge Center for Genome Informatics (V.S.), CSIR Institute of Genomics and Integrative Biology, Delhi, India.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing charge was funded by the authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Figure CTSD mutation in a family with neuronal ceroid-lipofuscinosis type 10

(A) Pedigree of the family. Affected indi- viduals are shaded in black. (B) T2 gra- dient recalled echo images showing a large midline cyst (highlighted with arrows) in the inferior part of posterior fossa communicating with the cisterna magna posteroinferiorly and the fourth ventricle anteriorly for both the children, respectively. (C) Chromatogram depict- ing the capillary sequencing result of CTSD c.A392G in the family, and the mutation position is highlighted with asterisks (red for affected individuals). (D) The domain structure of CTSD pro- tein marked with the identified mutation (p.Y131C). (E) Conservation of CTSD Y131 across other species. (F) TEM image of the skin biopsy showing vacuolating bodies of varying sizes largely localized around the nucleus. The vacuolating bodies are highlighted with red arrows.

suggested a provisional diagnosis of “neurodegeneration with from southern India. The mutation c.A392G introduces an brain iron accumulation syndrome” with the unusual clinical amino acid change of p.Y131C in CTSD. The Y131 lies in manifestation, possibly an autosomal recessive inheritance. the light chain of CTSD (figure, D). The CTSD chain (residues 64–412) contains crucial sites that are required for Whole exome sequencing analysis identified a homozygous the catalytic activity of the enzyme.5 Furthermore, the Y131 mutation c.A392G (p.Y131C) in CTSD to be the putative residue in CTSD was found to be conserved across species candidate, which was found to be cosegregated in parents in (figure, E), suggesting the importance of this residue in the heterozygous state using Sanger sequencing (figure, C). CTSD activity. Homozygous pathogenic variations in CTSD were previously reported to cause the rare neuronal ceroid lipofuscinosis type 10.6 The present variation was reported Discussion in the ExAC database as a single heterozygous case with an allele frequency of 0.0000083, whereas it was absent in the In the present study, we demonstrate the utility of whole 1000 Genome, NHLBI Exome Sequencing Project, or in- exome sequencing in the accurate diagnosis of a rare and ternal exome database of unrelated individuals from India, unusual neurodegenerative disorder in 2 siblings of a family suggesting the rarity of this variation. Although the majority

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG of the clinical features corroborated with a provisional di- Study funding agnosis of NCL, the absence of typical MRI findings and The authors acknowledge funding from the Council of Scien- epileptic seizures demanded further clinical evaluation. Be- tific and Industrial Research (CSIR), India, through Grant No. cause lipofuscin deposits can be observed elsewhere in the BSC0212. The funders did not have any role in study design; body in patients with NCLs, we attempted to evaluate skin data collection, analysis, and interpretation; writing the report; biopsies by transmission electron microscopy (TEM). TEM and the decision to submit the report for publication. analysis revealed that the skin layers were structurally nor- mal; however, vacuolated bodies of varying sizes similar to Disclosure typical lipofuscin deposits were observed in all the skin layers J.Thottath,S.K.Vellarikkal,R.Jayarajan,A.Verma,M.Manamel, especially localized around the nucleus (figure, F). Together, A. Singh, and V.R. Rajendran report no disclosures. S. Sivasubbu the genotype–phenotype correlation confirms the diagnosis has received research support from the Council of Scientificand of NCL type 10. Thus, genetic testing using whole exome Industrial Research (CSIR), India. V. Scaria has served on the sequencing has enabled the accurate and timely diagnosis in editorial boards of Journal of Translational Medicine, PLoS One, rare neurologic conditions such as NCL10. Frontiers in Systems Biology, Journal of Orthopaedics, International Journal of Rheumatic Diseases,andPeerJ and has received research Author contributions support from the Council of Scientific and Industrial Research Clinical and radiologic workup of the patient and collection (CSIR), India. Disclosures available: Neurology.org/NG. of samples for genetic testing: J. Thottath, M. Manamel, and V.R. Rajendran. DNA isolation, preparation of exome Publication history enrichment and sequencing, analysis, and validation and in- Received by Neurology: Genetics July 6, 2017. Accepted in final form terpretation: S.K. Vellarikkal, A. Verma, and R. Jayarajan. TEM April 5, 2018. experiments: A. Singh. Study concept and design and over- References seeing all the experiment and validation: S. Sivasubbu and V. 1. Haltia M. The neuronal ceroid-lipofuscinoses: from past to present. Biochim Biophys Scaria. Writing of the manuscript: S.K. Vellarikkal, R. Jayarajan, Acta 2006;1762:850–856. 2. Wisniewski KE, Zhong N, Kida E, et al. Atypical late infantile and juvenile forms of J. Thottath, V.R. Rajendran, V. Scaria, and S. Sivasubbu. neuronal ceroid lipofuscinosis and their diagnostic difficulties. Folia Neuropathol 1997;35:73–79. 3. Benitez BA, Alvarado D, Cai Y, et al. Exome-sequencing confirms DNAJC5 mutations Acknowledgment as cause of adult neuronal ceroid-lipofuscinosis. PLoS One 2011;6:e26741. The authors acknowledge Ms. Rowmika Ravi for helping with 4. Patiño LC, Battu R, Ortega-Recalde O, et al. Exome sequencing is an efficient tool for variant late-infantile neuronal ceroid lipofuscinosis molecular diagnosis. PLoS One manuscript preparation and the GUaRDIAN consortium for 2014;9:e109576. the scientific support. The authors also acknowledge Dr. 5. Fusek M, Vetvicka V. Dual role of cathepsin D: ligand and protease. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2005;149:43–50. Najeeba Riyaz for the skin biopsy and Dr. Riyaz Arakkal and 6. Hersheson J, Burke D, Clayton R, et al. Cathepsin D deficiency causes juvenile-onset Dr. Rajeevan for the clinical workup of the family. ataxia and distinctive muscle pathology. Neurology 2014;83:1873–1875.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 CLINICAL/SCIENTIFIC NOTES OPEN ACCESS X-linked myotubular myopathy and recurrent spontaneous pneumothorax A new phenotype?

Per-Ole Carstens, MD, Eva Maria Christina Schwaibold, MD, Katharina Schregel, MD, Correspondence Carolin D. Obermaier, PHD, Arne Wrede, MD, Sabrina Zechel, MD, Silke Pauli, MD, and Jens Schmidt, MD Dr. Schmidt [email protected] Neurol Genet 2019;5:e327. doi:10.1212/NXG.0000000000000327

X-linked myotubular myopathy (XLMTM) is a rare hereditary disorder of the skeletal muscle. Symptoms include impaired respiration and muscular hypotonia, usually present at birth and leading to death during infancy or early childhood.1 Pneumothorax, defined as trapping of air in the pleural cavity, can be caused by surgery or can occur spontaneously.2 Pneumothorax has been reported only in a small number of cases with hereditary myopathies, but usually not spontaneously and never in XLMTM.

Case report Our index patient presented at age 42 because of problems in raising his arms and climbing stairs. His one-year-older brother reported no symptoms. During childhood and adolescence, all motor milestones were reached normally. Both brothers were good at sports, but could not climb up a rope. Chronic bronchitis was diagnosed in both at around age 6, yet no treatment was required. Both started to smoke cigarettes at ages 15 and 16. Between age 18 and 26, the index patient had recurrent spontaneous pneumothorax (SP), which affected the right lung 9 times and the left lung once. The older brother had a first SP of the left lung at the age of 25. At the age of 29, 3 instances of SP of the left lung occurred within short intervals and a micro- surgical intervention was performed. Several days later, he developed SP of the right lung and received microsurgical intervention of that side. All 5 instances of SP were associated with physical activity. At age 44, the index patient developed respiratory insufficiency and required noninvasive ventilation during the nights and for some hours during the day. An inhalation therapy with tiotropium, formoterol, and budesonide was required, and he stopped smoking. At age 47, another SP of the right lung occurred during non-invasive ventilation (NIV) and a partial lung resection was performed. There was no family history for chronic bronchitis, SP, or neuromuscular disorders.

Chest x-ray revealed no evidence of lung emphysema (figure, A), and alpha-1-antitrypsin deficiency as the cause for SP was genetically excluded (SERPINA1 gene). A moderately combined respiratory disorder was found by body plethysmography.

Clinical examination of the index patient revealed a paresis of arm abduction (muscle research council [MRC] grade 2) and hip flexion (MRC 4). The older brother could not walk on his heels and had no other paresis. Myogenic changes on EMG and slightly elevated creatine kinase (CK) were noted in both patients. The muscle biopsy and MRI of the index patient revealed signs of myopathic damage and displayed fatty degeneration of the muscle (figure, B–C). The

From the Department of Neurology (P.-O.C., J.S.), University Medical Center Gottingen;¨ Institute of Human Genetics (E.M.C.S., S.P.), University Medical Center Gottingen;¨ Institute of Human Genetics (E.M.C.S.), Heidelberg University; Department of Neuroradiology (K.S.), University Medical Center Gottingen;¨ CeGaT GmbH and Praxis fur¨ Humangenetik Tubingen¨ (C.D.O.); Institute of Neuropathology (A.W., S.Z.), University Medical Center Gottingen;¨ and Institute of Neuropathology, Saarland University Medical Center (A.W.), Homburg; and Institute of Human Genetics (S.P.), University Medical Center Gottingen,¨ Germany.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing Charge was funded by University Medical Center Gottingen,¨ Germany. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Figure Diagnostic findings

(A) Routine chest x-ray of the older brother at the age of 43 displays dystelectasis of the middle lobe of the right side without evidence of lung emphysema. (B) Muscle biopsy of the index patient. Myopathic muscle tissue with internalized nuclei, fiber size variability, and increase of endomysial connective tissue (105 of 220 muscle fibers display an internalized nucleus [47.7%]). In the older brother, 861 of 1,429 muscle fibers depict an internalized nucleus [60.2%]). HE staining, scale bar = 200 μm. (C) Muscle MRI of the index patient. The arrows depict fatty degeneration of the quadriceps muscle. T1w sequence.

diagnosis of XLMTM was based on the detection of the a genetic predisposition can be assumed. The current frameshift variant c.98dupA (p.(Ala34Glyfs*13)) in exon 3 knowledge of myotubularin is not sufficient to explain why of the MTM1 gene in both patients by next-generation se- SP may occur. quencing. The patients have no other siblings and their mother died before genetic testing could be performed. Over This case report is of interest for clinicians and human the following 10 years, both brothers developed proximal and geneticists in 3 different ways: (1) A novel frameshift distal tetraparesis, ptosis, and dysphagia, which led to major variant in the MTM1 gene is described. (2) The new restrictions in daily life activities. variant is associated with a moderate, adult-onset course of XLMTM. (3) The variant might be associated with re- Discussion current SP, which has not been reported before for XLMTM. MTM1 codes for myotubularin, which is involved in mito- chondrial homeostasis and regulation of the cytoskeletal system Study funding 3 in muscles. To our knowledge, the here-described variant No targeted reported funding. c.98dupA (p.(Ala34Glyfs*13)) has not been described in data- bases (ExAC, gnomAD, Human Gene Mutation Database) or in Disclosure the literature before. Truncating variants of MTM1 are normally Disclosures available: Neurology.org/NG. associated with a more severe phenotype than nontruncating variants.1 Although the variant c.98dupA (p.(Ala34Glyfs*13)) Publication history fi leads to a frameshift with a premature stop codon and a trun- Received by Neurology: Genetics November 15, 2018. Accepted in nal cated protein or a nonsense-mediated mRNA-decay, interest- form March 11, 2019. ingly, it was associated with a much more moderate phenotype of adult onset. Appendix Authors In our 2 patients, muscular weakness became apparent after Name Location Role Contribution the age of 40 years and respiratory impairment occurred only in one brother after recurrent SP and partial lung resection. Per-Ole Department of Author Analyzed the whole Carstens, Neurology, University set of clinical data; Dyspnea on exercise was diagnosed as a chronic bronchitis MD Medical Center drafted the in childhood but might be interpreted as a mild respiratory Gottingen,¨ Germany manuscript for intellectual content failure because of the myopathy. Eva Maria Institute of Human Author Planned and Christina Genetics, Heidelberg supervised the SP in hereditary myopathy has been reported only in Schwaibold, University, Germany genetic testing; 4 a small number of patients with nemaline myopathy, limb- MD revised the 5 manuscript for girdle muscular dystrophy, collagen VI-related dystro- intellectual content phy,6 and Duchenne muscular dystrophy.7 However, most Katharina Institute of Author Analyzed the muscle pneumothoraces were nonrecurrent and associated with Schregel, Neuroradiology, MRI; revised the assisted ventilation. So far, no association between MD University Medical manuscript for XLMTM and SP has been reported. Because of the re- Center Gottingen,¨ intellectual content Germany current and bilateral appearance of the SP in both brothers,

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG References Appendix (continued) 1. McEntagart M, Parsons G, Buj-Bello A, et al. Genotype-phenotype correlations in X-linked myotubular myopathy. Neuromuscul Disord 2002;12:939–946. Name Location Role Contribution 2. Tschopp JM, Rami-Porta R, Noppen M, Astoul P. Management of spontaneous pneumothorax: state of the art. Eur Respir J 2006;28:637–650. 3. Hnia K, Tronch`ere H, Tomczak KK, et al. Myotubularin controls desmin intermediate Carolin D. CeGaT GmbH and Author Planned and filament architecture and mitochondrial dynamics in human and mouse skeletal Obermaier, Praxis fur¨ supervised the genetic muscle. J Clin Invest 2011;121:70–85. PHD Humangenetik testing; revised the 4. Sasaki M, Yoneyama H, Nonaka I. Respiratory muscle involvement in nemaline Tubingen,¨ Germany manuscript for myopathy. Pediatr Neurol 1990;6:425–427. intellectual content 5. Fayssoil A, Ogna A, Chaffaut C, et al. Natural history of cardiac and respiratory involvement, prognosis and predictive factors for long-term survival in adult Arne Institute of Author Analyzed the muscle patients with limb girdle muscular dystrophies type 2C and 2D. PLoS One2016; Wrede, MD Neuropathology, biopsy; revised the 11:e0153095. Saarland University manuscript for 6. Fraser KL, Wong S, Foley AR, et al. Pneumothoraces in collagen VI-related Medical Center, intellectual content dystrophy: a case series and recommendations for management. ERJ Open Res Homburg, Germany 2017;3. 7. Yamamoto T, Kawai M. Spontaneous pneumothorax in Duchenne muscular dys- Sabrina Institute of Author Analyzed the muscle trophy [in Japanese]. Rinsho Shinkeigaku 1994;34:552–556. Zechel, MD Neuropathology, biopsy; revised the University Medical manuscript for Center Gottingen,¨ intellectual content Germany

Silke Pauli, Institute of Human Author Planned and MD Genetics, University supervised the genetic Medical Center testing; revised the Gottingen,¨ Germany manuscript for intellectual content

Jens Department of Author Conceptualized and Schmidt, Neurology, University supervised the MD Medical Center manuscript; revised Gottingen,¨ Germany the manuscript for intellectual content

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 CLINICAL/SCIENTIFIC NOTES OPEN ACCESS First TMEM126A missense mutation in an Italian proband with optic atrophy and deafness

Chiara La Morgia, MD, PhD,* Leonardo Caporali, ScD, PhD,* Francesca Tagliavini, ScD, PhD, Correspondence Flavia Palombo, ScD, PhD, Michele Carbonelli, MD, Rocco Liguori, MD, Piero Barboni, MD, and Dr. La Morgia [email protected] Valerio Carelli, MD, PhD

Neurol Genet 2019;5:e329. doi:10.1212/NXG.0000000000000329

Recessively inherited optic neuropathy has been an elusive entity for a long time. Currently, a few – causative genes have been described,1 6 associated with a spectrum of isolated or syndromic optic atrophy. Among these genes, TMEM126A (OPA7) was the first to be reported, with a single causative mutation found in all pedigrees identified to date of North African ancestry (c.163C>T; p.Arg55X), thus possibly belonging to the same founder mutational event.1,7,8

Case report A 16-year-old girl was born by likely consanguineous parents (figure A). Delivery was uneventful, and psychomotor development was normal. Medical history was relevant for an isolated febrile seizure at 4 months and migraine since age 14 years. The only brother presents a mild language disorder improved by logopedic rehabilitation, and the grand-grandmother is affected by epilepsy.

Visual problems were recognized when the patient was aged 4 years with evidence of bilateral optic atrophy. We observed the patient at age 16 years. Neurologic examination was un- remarkable except for the presence of sporadic postural and rest myoclonic jerks at upper and lower limbs and brisk deep tendon reflexes. Ophthalmologic evaluation showed visual acuity OD 0.16, OS 0.125, bilateral temporal pallor at fundus examination, profound color deficit, and bilateral cecocentral scotoma at automated visual fields (figure B). Optical coherence tomog- raphy showed bilateral diffuse optic atrophy (figure B).

Laboratory examinations were relevant for the presence of increased lactic acid levels after stan- dardized exercise (35.5 mg/dL; normal values 5–22 mg/dL). Brain MRI was normal. Cardiologic examination showed only the presence of a mild mitral valve prolapse. Audiometry disclosed the presence of bilateral mild sensorineural deafness. EMG ruled out the presence of peripheral neuropathy. Pattern visual evoked potentials showed the absence of cortical responses in OD and increased latency in OS. Somatosensory evoked potentials and EEG with muscle recordings were normal and in particular did not reveal the presence of a cortical correlate of myoclonic jerks. Cognitive evaluation showed a profile within normal limits (Wechsler Intelligence Scale for Children-IV score = 88).

Genetic analysis, after informed consent and EC approval (CE AVEC 211/2018), by a cus- tom next-generation sequencing (NGS) panel of optic atrophy-related genes revealed the presence of a homozygous mutation affecting the TMEM126A gene (c.497A>G, p.Q166R), affecting one of the transmembrane helices. According to the public database GnomAD

*These authors contributed equally to the manuscript.

From the IRCCS Istituto delle Scienze Neurologiche di Bologna (C.L.M., L.C., F.T., F.P., M.C., R.L., V.C.), UOC Clinica Neurologica; Dipartimento di Scienze Biomediche e Neuromotorie (C.L.M., R.L., V.C.), Universit`a di Bologna; and Studio Oculistico d’Azeglio (P.B.), Bologna, Italy.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing Charge was funded by the authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Figure Pedigree and ophthalmologic findings

(A) The pedigree and the segregation of the pathogenic TMEM126A mutation c.497A>G, p.Q166R, confirmed by Sanger sequencing, are shown. As shown by the reconstructed pedigree, maternal grand-grandmother and paternal grand-grandfather carried the same family name, both originating from a small village in Sicily, thus strongly suggesting consanguinity and a common founder for the homozygous mutation found in the proband. (B) Ophthalmologic findings: In the upper panel, Humphrey visual fields showed bilateral central scotoma, and in the lower panel, fundus oculi revealed diffuse (more temporally) optic disc pallor, as reflected by the OCT measurements, displaying a generalized reduction of RNFL thickness, more pronounced on the temporal-inferior quadrants.

(gnomad.broadinstitute.org), this variant is reported in 3 compatible with the diagnosis of recessive optic neuropathy alleles, but never in homozygosity, and it is predicted to be associated with this TMEM126A missense variant. Careful pathogenic (CADD phred 26.2). This mutation is compatible clinical evaluation disclosed also a mild sensorineural deafness, with the diagnosis of recessive optic neuropathy (figure A). which has been previously reported in association with the North African TMEM126A mutation.1 Only 6 families of African ancestry (Algeria and Morocco) have been reported to Discussion date, all carrying the c.163C>T (p.Arg55X) mutation, sug- ff 1,7,8 The genetic landscape of inherited optic neuropathies, including gesting a founder e ect. The patients described in these the rare recessive forms, has been greatly expanding thanks to reports presented a variable phenotype, despite the association – the availability of NGS techniques.1 6 Our case is a non-African with the same mutation, ranging from isolated to syndromic patient carrying a recessive homozygous TMEM126A missense optic atrophy. Extraocular features included sensorineural mutation, born from likely consanguineous Italian parents and deafness, hypertrophic cardiomyopathy, and peripheral poly- presenting in early childhood with isolated bilateral optic atro- neuropathy. Moreover, a Leber’s hereditary optic neuropathy– phy. Both in silico predictions and segregation analysis were like presentation has been described in 1 patient.8

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG TMEM126A is a mitochondrial protein, located in the inner Hereditary Optic Neuropathy) and received travel reim- mitochondrial membrane with still unknown functions,9 bursements from Santhera Pharmaceuticals. R. Liguori served highly expressed in the brain, cerebellum, fetal brain, skeletal on the scientific advisory boards of Biogen, Sanofi Genzyme, muscle, testis, fetal retinal pigmentary epithelium, and fetal Argon Healthcare s.r.l., and Editree Eventi s.r.l.; received retina of humans.1 Polarographic tests and spectrophoto- lecture fees from Dynamicom Education, SIMG Service, metric assays on cultured skin fibroblasts showed normal re- Adnkronos Salute unipersonale s.r.l., DOC Congress s.r.l., and spiratory chain function, but partial deficiency of complex I in First Class s.r.l.; and is a consultant for Alfasigma and Amicus 1 patient from the original study.1 In our case, the abnormal Therapeutics s.r.l. P. Barboni is involved in clinical trials with lactic acid elevation after standardized effort confirms that Santhera Pharmaceuticals (Raxone in Leber’s Hereditary TMEM126A is somehow involved in oxidative phosphory- Optic Neuropathy) and GenSight Biologics (gene therapy lation, even if its precise role remains to be defined and further with GS10 in Leber’s Hereditary Optic Neuropathy) and functional studies are needed. received speaker honoraria for educational courses and travel reimbursements from Santhera Pharmaceuticals and Omi- Overall, we found a second recessive mutation in the cron Pharmaceuticals. V. Carelli is involved in clinical trials TMEM126A gene in an Italian proband, who, similarly to with Santhera Pharmaceuticals (Raxone in Leber’s Hereditary the previously reported cases with the same North African Optic Neuropathy), GenSight Biologics (gene therapy with founder mutation, is affected by optic atrophy and mild GS10 in Leber’s Hereditary Optic Neuropathy), and Stealth sensorineural deafness. The phenotype recurring with re- BioPharma (Elamipretide in Primary Mitochondrial Myopa- cessive TMEM126A mutations is quite consistent, and we thy) and received speaker honoraria for educational courses predict that more cases will be diagnosed, as NGS is now and travel reimbursements from Santhera Pharmaceuticals. largely available in diagnostic centers. He is also funded for research program by Stealth Pharma- ceuticals, and his research is supported by grants from the Author contributions Italian Ministry of Health, Telethon, the Emilia Romagna C. La Morgia, L. Caporali, and V. Carelli: conception, drafting, Region, the patient’s organization MITOCON, and by private and revision of the manuscript. F. Tagliavini and F. Palombo: donations. Disclosures available: Neurology.org/NG. genetic analysis and interpretation of results. M. Carbonelli and P. Barboni: ophthalmologic evaluation and revision of the Publication history manuscript. R. Liguori: interpretation of results and revision Received by Neurology: Genetics November 16, 2018. Accepted in final of the manuscript. form March 5, 2019.

Study funding References 1. Hanein S, Perrault I, Roche O, et al. TMEM126A, encoding a mitochondrial protein, This study was supported by the Grant GR-2016-02361449 to is mutated in autosomal-recessive nonsyndromic optic atrophy. Am J Hum Genet L.C. and by the “Ricerca Corrente” funding (L.C., F.T., F.P., 2009;84:493–498. 2. Angebault C, Guichet PO, Talmat-Amar Y, et al. Recessive mutations in RTN4IP1 M.C., and V.C.), both from the Italian Ministry of Health. cause isolated and syndromic optic neuropathies. Am J Hum Genet 2015;97: 754–760. 3. Metodiev MD, Gerber S, Hubert L, et al. Mutations in the tricarboxylic acid cycle Disclosure enzyme, aconitase 2, cause either isolated or syndromic optic neuropathy with en- C. La Morgia is involved in clinical trials with Santhera cephalopathy and cerebellar atrophy. J Med Genet 2014;51:834–838. 4. Hartmann B, Wai T, Hu H, et al. Homozygous YME1L1 mutation causes mito- Pharmaceuticals (Raxone in Leber’s Hereditary Optic Neu- chondriopathy with optic atrophy and mitochondrial network fragmentation. Elife ropathy) and GenSight Biologics (gene therapy with GS10 in 2016;5:e16078. ’ 5. Inoue H, Tanizawa Y, Wasson J, et al. A gene encoding a transmembrane protein is Leber s Hereditary Optic Neuropathy) and received speaker mutated in patients with diabetes mellitus and optic atrophy (Wolfram syndrome). honoraria for educational courses and travel reimbursements Nat Genet 1998;20:143–148. 6. Gerber S, Ding MG, G´erard X, et al. Compound heterozygosity for severe and from Santhera Pharmaceuticals. C. La Morgia is funded by the hypomorphic NDUFS2 mutations cause non-syndromic LHON-like optic neuropa- Ministry of Health grant GR-2013-02358026 and is co-PI for thy. J Med Genet 2017;54:346–356. 7. Meyer E, Michaelides M, Tee LJ, et al. Nonsense mutation in TMEM126A causing the Ministry of Health grant GR-2016-02361449. L. Caporali autosomal recessive optic atrophy and auditory neuropathy. Mol Vis 2010;16: is funded by the Ministry of Health grant GR-2016-02361449. 650–664. 8. D´esir J, Coppieters F, Van Regemorter N, De Baere E, Abramowicz M, Cordonnier F. Tagliavini and F. Palombo report no disclosures. M. Car- M. TMEM126A mutation in a Moroccan family with autosomal recessive optic at- bonelli is involved in clinical trials with Santhera Pharma- rophy. Mol Vis 2012;18:1849–1857. ’ 9. Hanein S, Garcia M, Fares-Taie L, et al. TMEM126A is a mitochondrial located ceuticals (Raxone in Leber s Hereditary Optic Neuropathy) mRNA (MLR) protein of the mitochondrial inner membrane. Biochim Biophys Acta and GenSight Biologics (gene therapy with GS10 in Leber’s 2013;1830:3719–3733.

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3 CLINICAL/SCIENTIFIC NOTES OPEN ACCESS Double somatic mosaicism in a child with Dravet syndrome

Alison M. Muir, PhD, Chontelle King, BSc, Amy L. Schneider, MGC, Aman S. Buttar, BSc, Correspondence Ingrid E. Scheffer, MB, BS, PhD, Lynette G. Sadleir, MB, ChB, MD, and Heather C. Mefford, MD, PhD Dr. Mefford [email protected] Neurol Genet 2019;5:e333. doi:10.1212/NXG.0000000000000333

Dravet syndrome, the prototypic infantile-onset developmental and epileptic encephalopathy, occurs secondary to de novo pathogenic variants in SCN1A in over 80% of cases1. One possible genetic etiology for patients without a heterozygous SCN1A mutation is a post-zygotic mosaic SCN1A variant below the level detected by diagnostic sequencing, which routinely identifies only variants with allele frequencies above ;20%. We used targeted deep resequencing to systematically investigate whether mosaicism could be the cause in individuals with molecularly unsolved Dravet syndrome.

Results Using single-molecule molecular inversion probes,2 we performed deep sequencing of SCN1A and 7 other Dravet-associated genes (SCN2A, SCN8A, HCN1, GABRA1, GABRG2, STXBP1, and PCDH19) using DNA derived from blood or saliva of 20 individuals with a clinical di- agnosis of Dravet or Dravet-like syndrome. Previous targeted sequencing had not identified a heterozygous pathogenic variant in SCN1A or 65 other epilepsy genes.

We identified an individual who was mosaic for 2 different pathogenic variants at the same nucleotide position in SCN1A: chr2:g.166848363A > G, p.(Phe1808Leu) and chr2: g.166848363A > C, p.(Phe1808Val). The SCN1A variants were present in the blood at allele frequencies of 8.3% and 6.9%, respectively, which is well below the level of detection by standard targeted and Sanger sequencing technologies. We were able to detect both variants at varying allele frequencies (ranging from 0.6% to 39.7%) in DNA derived from hair follicles and skin fibroblasts (figure, A and B). Through cloning and sequencing a fragment containing both a maternally inherited SNP (rs10497275) and the mosaic variant site, we determined that both variants are located on the paternal allele.

This 12-year-old girl presented at 6 months with a brief febrile generalized tonic-clonic seizure (GTCS). She continued to have occasional febrile GTCS until 1 year of age when she had 3 10- minute focal motor seizures; 1 left hemiclonic and 1 followed by a postictal right Todd’s paresis. At 2 years, she developed clusters (up to 15 seizures over 4 days) of focal impaired awareness facial clonic seizures every 6 weeks. Her first episode of status epilepticus occurred at 3 years. Development was normal until 2 years when language delay became apparent and attention deficit hyperactivity disorder and autism spectrum disorder were diagnosed. Despite trials of valproate, carbamazepine, lamotrigine, topiramate, clobazam, and cannabidiol, her seizures remain intractable. EEGs at 17 months and 3 years showed no epileptiform discharges. Sub- sequent EEGs have not been possible due to behavior. MRI and routine metabolic and CSF studies are normal.

From the Division of Genetic Medicine (A.M.M., A.S.B., H.C.M.), Department of Pediatrics, University of Washington, Seattle, WA; Department of Paediatrics and Child Health (C.K.), University of Otago, Wellington, New Zealand; Department of Medicine (A.L.S., I.E.S.), Epilepsy Research Centre, The University of Melbourne, Austin Health, Heidelberg, Australia; The Florey Institute and Murdoch Children’s Research Institute (I.E.S.), Parkville, Australia; Department of Neurology (I.E.S.), Royal Children’s Hospital, Parkville, Australia; and Department of Paediatrics and Child Health (L.G.S.), University of Otago, Wellington, New Zealand.

Funding information and disclosures are provided at the end of the article. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/NG.

The Article Processing charge was funded by the authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Figure Two SCN1A mosaic variants at the same nucleotide position in an individual with Dravet syndrome

(A) Sanger sequence of codons 1807–1809 of SCN1A from DNA derived from leukocytes, skin fibroblasts, and hair bulb cells. The mutated base is highlighted with a red box. (B) Quantifi- cation of mutant allele frequencies of mosaic mutations in varying tissues using single-molecule molecular inversion probes and next-generation sequencing technology. (C) Pro- posed model for the simultaneous generation of double mo- saic mutations. A post-zygotic mutation event occurred that affected both stands of 1 SCN1A allele (either consecutively or concurrently) creating a DNA mismatch at the site of the mu- tation. Before the cell’s DNA repair machinery could fix the mismatch, DNA replication occurred (green represents new DNA strands), resulting in the production of 2 cells each with a different pathogenic mutation.

Discussion variants with 17.6% and 30.6% mosaic allele frequencies in blood respectively, while their nonmosaic offspring had Dravet Double somatic mosaicism, where 2 mosaic variants occur at syndrome.2 In contrast to the milder phenotypes in the parents the same nucleotide, has not been reported in Dravet syn- with mosaic variants in these families, our patient reported here 3,4 drome and only rarely recognized in human diseases ; re- had the classical features of Dravet syndrome similar to 4 cently, a single case was reported in SCN2A encephalopathy. a reported individual who was heterozygous for the same We detected double mosaicism in SCN1A in a patient with p.(Phe1808Leu) pathogenic variant.8 Marked differences in Dravet syndrome. The variants are predicted to cause differ- phenotypic severity associated with mosaicism are likely ent amino acid changes at the same residue (p.Phe1808). explained by variation in mosaicism levels in disease-relevant Both variants are present in cells derived from ectodermal tissues. For the brain, perhaps this may even be determined by (hair follicles) and mesodermal (leukocytes and skin fibro- region-specific and cell-type-specificdifferences. Comprehen- blasts) tissues. That neither variant is limited to a single germ sive studies of disease-relevant tissue will be required to gain layer strongly suggests that the mutations occurred prior to a more accurate picture of somatic mosaicism levels and how gastrulation. Although brain tissue is unavailable for confir- this affects disease severity. mation, these findings suggest that both variants should be present in brain, another ectodermal tissue. Study funding Supported by grants from the NINDS (R01 NS069605, to Dr. It is estimated that every individual accumulates approxi- Mefford). mately 8 spontaneous mutations per somatic cell division.5 Since both mutations occurred after the first cell cleavage but Disclosure before gastrulation, it is improbable that spontaneous muta- Disclosures available: Neurology.org/NG. tions randomly occurred at the same nucleotide twice in such a narrow developmental window. It is also unlikely that this Publication history non-cytosine-phosphate-guanine nucleotide has an apprecia- Received by Neurology: Genetics January 23, 2019. Accepted in final form bly higher mutation rate than the genome average. The excess March 18, 2019. of triallelic sites present in the genome supports the occurrence of simultaneous mutations on opposite strands of the same DNA molecule.6 This model would explain the double mosaicism of Appendix Authors the same nucleotide in our patient and is supported by the Name Location Role Contribution finding that both mosaic variants occurred on the paternal allele fi Alison University of Author Contributed to conception ( gure, C). Muir, PhD Washington, and design of the study, Seattle acquisition and analysis of sequencing data, and Based on American College of Medical Genetics and Genomics drafting and revision of the 7 guidelines, both SCN1A:p.(Phe1808Leu) and SCN1A: manuscript fi p.(Phe1808Val) are classi ed as pathogenic variants. There- Chontelle University of Author Contributed to acquisition fore, approximately 15.2% of SCN1A alleles harbor pathogenic King, BSc Otago, and analysis of data of mutations in the blood of our patient. We recently reported 2 Wellington clinical data for the case report parents with febrile seizures who were mosaic for SCN1A

2 Neurology: Genetics | Volume 5, Number 3 | June 2019 Neurology.org/NG References Appendix (continued) 1. Claes L, Del-Favero J, Ceulemans B, Lagae L, Van Broeckhoven C, De Jonghe P, et al. De novo mutations in the sodium-channel gene SCN1A cause severe myoclonic – Name Location Role Contribution epilepsy of infancy. Am J Hum Genet 2001;68:1327 1332. 2. Myers CT, Hollingsworth G, Muir AM, et al. Parental mosaicism in “de novo” epileptic encephalopathies. N Engl J Med 2018;378:1646–1648. Amy The University Author Contributed to acquisition 3. Draaken M, Giesen CA, Kesselheim AL, et al. Maternal de novo triple mosaicism for Schneider, of Melbourne, and analysis of data clinical two single OCRL nucleotide substitutions (c.1736A>T, c.1736A>G) in a Lowe MGC Heidelberg data for the Dravet cohort syndrome family. Hum Genet 2011;129:513–519. 4. Stosser MB, Lindy AS, Butler E, et al. High frequency of mosaic pathogenic variants in Aman S. University of Author Determined the allele of genes causing epilepsy-related neurodevelopmental disorders. Genet Med 2018;20: Buttar, BSc Washington, origin for the mosaic 403–410. Seattle mutations 5. Milholland B, Dong X, Zhang L, Hao X, Suh Y, Vijg J , Differences between germline and somatic mutation rates in humans and mice. Nat Commun 2017;8: Ingrid The University Author Contributed to acquisition and 15183. Scheffer, of Melbourne, analysis of clinical data for the 6. Hodgkinson A, Eyre-Walker A. Human triallelic sites: evidence for a new mutational MB, BS, Heidelberg Dravet cohort, and drafting mechanism? Genetics 2010;184:233–241. PhD and revision of the manuscript 7. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Lynette University of Author Contributed to acquisition and Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Sadleir, Otago, analysis of clinical data for the Med 2015;17:405–424. MB, ChB, Wellington case report, and drafting and 8. Fujiwara T, Sugawara T, Mazaki-Miyazaki E, et al. Mutations of sodium channel alpha MD revision of the manuscript subunit type 1 (SCN1A) in intractable childhood epilepsies with frequent generalized tonic-clonic seizures. Brain 2003;126(pt 3):531–546. Heather University of Author Contributed to conception Mefford, Washington, and design of the study, and MD, PhD Seattle drafting and revision of the manuscript

Neurology.org/NG Neurology: Genetics | Volume 5, Number 3 | June 2019 3