Wong MMK, Watson LM, Becker EBE. J Neurol Neuromedicine (2017) 2(7): 11-15 Neuromedicine www.jneurology.com www.jneurology.com Journal of Neurology & Neuromedicine

Mini Review Open Access

Recent advances in modelling of cerebellar ataxia using induced pluripotent stem cells Maggie M. K. Wong1, Lauren M. Watson1 and Esther B. E. Becker1* 1Department of Physiology, Anatomy and Genetics, University of Oxford, United Kingdom

Article Info ABSTRACT

Article Notes The cerebellar ataxias are a group of incurable brain disorders that are caused Received: May 15, 2017 primarily by the progressive dysfunction and degeneration of cerebellar Purkinje Accepted: July 10, 2017 cells. The lack of reliable disease models for the heterogeneous ataxias has hindered the understanding of the underlying pathogenic mechanisms as well *Correspondence: Dr. Esther B. E. Becker as the development of effective therapies for these devastating diseases. Recent University of Oxford, Department of Physiology, Anatomy and advances in the field of induced pluripotent stem cell (iPSC) technology offer Genetics, Sherrington Road, Oxford OX1 3PT, United Kingdom, new possibilities to better understand and potentially reverse disease pathology. Email: [email protected] Given the neurodevelopmental phenotypes observed in several types of ataxias, iPSC-based models have the potential to provide significant insights into disease © 2017 Becker EBE. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License progression, as well as opportunities for the development of early intervention therapies. To date, however, very few studies have successfully used iPSC-derived Keywords cells to model cerebellar ataxias. In this review, we focus on recent breakthroughs iPSC in generating human iPSC-derived Purkinje cells. We also highlight the future Cerebellum challenges that will need to be addressed in order to fully exploit these models for Ataxia the modelling of the molecular mechanisms underlying cerebellar ataxias and the Disease modelling development of effective therapeutics. Purkinje cells Development Neurodegeneration Cerebellar ataxias The cerebellar ataxias are a clinically and genetically diverse group of neurological disorders that primarily affect the cerebellum. Characterized by a progressive loss of motor coordination, other symptoms include dysarthria and disturbance of eye movements1. Cerebellar ataxias may be acquired or inherited, with hereditary ataxias

toclassified date . intoAmong autosomal these, the dominant, autosomal autosomal recessive recessive, disorders X-linked Friedreich or mitochondrial2-4 forms. Over 60 genetic subtypes have been identified dominant polyglutamine spinocerebellar ataxias (SCAs) are the ataxia (FRDA) and ataxia-telangiectasia (A-T), as well as the autosomal heterogeneity, emerging evidence points to the existence of common pathogenicmost studied mechanisms forms of ataxias. that may Despite be significantshared by clinicalseveral andgenetically genetic distinct forms of cerebellar ataxias (reviewed in ). However, it is still unclear how the proposed pathological pathways5-8 ultimately result in cerebellar dysfunction and degeneration, predominantly affecting

Purkinje cells. neurodegenerative disorders. The heterogeneous nature of the cerebellar Understanding disease mechanisms is key to treating of reliable disease models have, however, hampered our understanding ofataxias the molecular combined disease with the mechanisms unavailability underlying of human brain cerebellar tissue andataxias the lackand

Page 11 of 15 Wong MMK, Watson LM, Becker EBE. J Neurol Neuromedicine (2017) 2(7): 11-15 Journal of Neurology & Neuromedicine thus, the development of effective therapies. Although mouse models of several cerebellar ataxias, including FRDA and SCAs, have provided valuable insights into the morphologycells expressed with mature an extensively PC markers arborized including dendritic L7/PCP2 tree. pathophysiology of these disorders (reviewed in9), many and GRID2, and possessed a characteristic Purkinje cell questions remain about the observed species differences in disease phenotypes and the effectiveness of potential matureMoreover,22,23 these. cells displayed the distinctive firing pattern drugs in clinical trials. To help translate research from of Purkinje cells, suggesting that they were functionally Wang et al. animal models into novel treatments for ataxia patients, with human foetal used acerebellar similar self-inductiveslices . Although approach these to produce Purkinje cells from human 24 iPSCs co-cultured Theit is essentialcurrent obstacles to validate might findings be overcomein the relevant by exploiting affected human cell types, particularly in cerebellar Purkinje cells. and were electrically active, they did not possess the recently developed human induced pluripotent stem cell cells appeared to display Purkinje cell-specific markers (iPSC) technology and neuronal differentiation protocols. Generation of cerebellar neurons using human iPSCs duecharacteristic to associated morphology ethical issues. of mature Purkinje cells. Moreover, this protocol is likely to be limited in its uptake Most recently, the Mugumura group used a very similar allowed for the robust reprogramming of human somatic cellsMajor into advancesiPSCs in the field of stem cell technology have cells from iPSCs, but with the inclusion of two additional 10-12 approach to their original method to generate Purkinje neuronal subtypes and. Human other affected ataxia patient-specific cell types in cerebellar iPSCs, ataxias,which canwould subsequently be an ideal tool be differentiatedfor studying the into pathogenic specific mediumgrowth 25 factors (brain-derived neurotrophic factor (BDNF) and neurotrophin-3 (NT-3)) in their maturation neurons possessed. After co-culture extensively with branched mouse primary dendritic granule trees drugmechanisms screening in in human vitro and disease-relevant neurons, cell precursors for 75 days, the human iPSC-derived and might also be highly13,14 valuable for high-throughput to differentiate in vitro, owing. Cerebellar to their large Purkinje size, cellscomplex are throughoutwith a thick soma stem, and characteristic dendrites, andof mature the synaptic Purkinje receptor cells, particularly vulnerable in cerebellar ataxia but difficult GRID2and expressed on dendritic the mature spines. PC-specificThus, this markersprotocol L7/PCP2not only of maturation offers a shorter and more feasible protocol to generate beenmorphology, achieved unique8,13 recently firing and properties are reviewed and below. extensive period . However, significant breakthroughs have that more closely resemble those in vivo. The generation of cerebellar neurons from human iPSCs Purkinje cells, but also produces more mature looking PCs aims to recapitulate the complex in vivo molecular events remain to be addressed before researchers can fully exploit circuit formation during human embryogenesis . Early Despite this significant progress, a number of challenges thatprotocols direct attempted the specification cerebellar of differentiation cerebellar neurons, using15-17 either and neurons allow for close monitoring of their individual mouse or human embryonic stem cells (ESCs) . These developmentiPSC-based cerebellar throughout models. differentiation; Dissociated however, cultures their of approaches sought to mimic the in vivo signals18-21 involved maturation and survival is limited compared to neurons in early cerebellar development by delivering inductive signals, including bone morphogenic proteins, mitogens of appropriate support in in vitro cultures . Moreover, diseasedeveloping phenotypes in the embryonic affecting synapsebrain, likely formation, due26,27 to dendritic the lack differentiation process. However, the resulting cerebellar retraction and neuronal migration , which may only arise and neurotrophins, in a step-wise fashion throughout the from interactions between cells 28,29and their environment, and a relatively high number of cerebellar granule cells . cultures only contained a small proportion of Purkinje cells18,20 Muguruma et al. adopted a different approach, which mayA be way lost forward in these totwo-dimensional overcome this cultures.limitation might be in vitro structures patterning, by adding a combination of only three factors that include multiple cell types and mimic the in vivo aimed to induce endogenous signals mimicking cerebellar architecturethe use of organoids, of an organ self-organizing30. Despite the relative success 22,23. Followed (insulin, fibroblast growth factor 2 (Fgf2) and cyclopamine) cerebral cortex (reviewed in30,31), only one study has to floating aggregates of mouse or human ESCs thusin generating far reported organoids the generation mimicking of cerebellar development organoids. of the by dissociation, fluorescence-activated cell sorting for Muguruma et al. showed that the addition of Fgf19 and derivedKIRREL2/NEPH3-positive granule cell precursors, Purkinje this cell method progenitors, resulted and in subsequent co-culture with mouse primary rhombic lip- than had previously been achieved. The differentiated stromal cell-derived factor 1 (Sdf1) during differentiation the generation of Purkinje cells at a much higher efficiency promoted the self-organisation of human ESCs into polarized three-layer structures reminiscent of the human

Page 12 of 15 Wong MMK, Watson LM, Becker EBE. J Neurol Neuromedicine (2017) 2(7): 11-15 Journal of Neurology & Neuromedicine

Pluripotency factors

Reprogramming

Insights from cerebellar development sllec citamoS sllec sCSPi

Differentiation

Ataxia patients 2D cell culture 3D cerebellar organoids

Potential treatments Disease modelling

Drug screening Therapeutic development Phenotypic output

Figure 1: Schematic overview of iPSC-based disease modelling for cerebellar ataxias. Somatic cells from patients are reprogrammed into iPSCs by the introduction of pluripotency factors. These iPSCs can then be differentiated into disease-relevant cell types in either 2D- or 3D-format, following protocols that mimic cerebellar developmental cues in vitro. Cell culture models may then be used as a platform for the investigation of disease phenotypes, as well as for drug screening, with the ultimate goal of delivering effective therapies to patients. embryonic cerebellum23. To date, however, no studies have reported the generation of cerebellar organoids from in vivo. Indeed, transcriptomic analysis of Purkinje cells and their resemblance to adult human thePurkinje generated cells neurons are still in an embryonic state, maturehuman iPSCs.human Moreover, cerebellum it remains in vitro a question whether self- despiteother iPSC-derived the fact that individual neuronal modelsneurons hashave revealed been stained that isinduction needed foralone the maturationis sufficient of for multiple generating cell types a functionally that form 29). Such immature cells . While long-term culture for mature markers (reviewed in exchange,active neuronal cause cell networks death within in spatially organoids organized and therefore manner, cellsmay willbe beneficial be needed for to thestudy study the cerebellarof developmental degeneration aspects in limitthe lack culture of mechanical duration support,and maturity as well31. asTo nutrient overcome and this, gas ataxiasof disease. due However,to the late fully disease mature onset and in patients.functional Purkinje cerebellar organoids could be grown in spinning bioreactors Lastly, the reproducibility of the developed protocols to prolong their viability by improving nutrient and gas exchange32 none of the abovementioned studies have been reproduced could be used to generate progenitors of various cerebellar remains a challenge. Although revolutionary in the field, subtypes, . which Alternatively, could subsequently the self-induction be plated approach onto long differentiation times. Moreover, substantial differences biodegradable scaffolds to support their organization into betweenby other researchbatches groups, of iPSC likely differentiation due to their complexity experiments and 30. In such a system, . To allow for the microenvironmental niches surrounding progenitors 27 ahave layered been three-dimensional shown to promote structure terminal differentiation and isremain essential a general to explore issue ways in to the shorten field the differentiation neuronal circuit formation . process,extensive facilitate applications the ofmaturation iPSC-based of models,cerebellar therefore, neurons it 28 and reduce the variability of the protocols. Future studies information on the relative maturity of the generated should focus on detailed electrophysiology, as well as Another limitation of the studies so far is the lack of

Page 13 of 15 Wong MMK, Watson LM, Becker EBE. J Neurol Neuromedicine (2017) 2(7): 11-15 Journal of Neurology & Neuromedicine transcriptomic, proteomic and metabolomic analyses of demonstrated by a decrease in cell number and reduction in compare the characteristics of the generated cerebellar dendriticfor maturation arbours and in patient maintenance cells compared of Purkinje to controls cells, as. cellsiPSC-derived to available Purkinje data from cells. human Moreover, embryonic it will beand vital adult to 15,42

This study highlights the potential of patient iPSC-derived Disease modelling of cerebellar ataxias cerebellum to stage the iPSC-derived models appropriately. cellsPurkinje only cellsshowed to abnormal model developmental dendritic development and disease- upon stressrelevant but ataxia not under phenotypes. normal culture Of note, conditions, the SCA6 suggesting Purkinje pioneering opportunities to study and model the disease that environmental triggers might be crucial to causing processesThe iPSC-based underlying cerebellarcerebellar ataxias cultures in describedaffected human offer disease phenotypes. It will be interesting to see whether neuronal cell types. Interestingly, mounting evidence from mortem brain material as well as in SCA6 animal models, development, including impaired dendritic arborisation, includingSCA6 phenotypes calcium that overload, have been cytoplasmic observed in humanaggregation post- spinecell and development animal models andindicates synaptogenesis, that abnormal and Purkinje related cell of expanded Cav2.1 protein, nuclear localization of the functional changes, may contribute to the pathogenesis of ataxias, including neurodegenerative ataxias15. Thus, of gene expression it will be extremely interesting to investigate whether cleavedderived SCA6Cav2.1 model. C-terminus5,43 and the resulting perturbation , can be recapitulated in the iPSC- Notably, the observed developmental phenotype in the developmental abnormalities. In addition, these cultures iPSC-derived human cerebellar neurons display any thyroid releasing hormone and riluzole, demonstrating the SCA6 Purkinje cells could be rescued using two compounds, will allow us, for the first time, to unravel the processes that might underlie the cell-specific vulnerability of human to be used for drug screening and personalized therapy. To date, fewer than 20 studies have reported potential of patient-specific iPSC-derived cerebellar neurons cerebellar Purkinje cells. Conclusion ofinvestigations these studies into have cerebellar been performed ataxia, including in neurons FRDA, without A-T derived cerebellar neurons offer unprecedented and polyglutamine SCAs, using human iPSCs. The majority opportunitiesRecent breakthroughs to study the disease in generating mechanisms human underlying iPSC- (reviewed in ). None of these studies have yet successfully cerebellar ataxias. This exciting methodology will be regional specification,13,14 as well as in other affected cell types recapitulated the specific cerebellar neuronal dysfunction particularly powerful when combined with other cutting- and degeneration known to characterize these conditions. edge technologies including genome engineering, single- by an intronic repeat expansion in the FXN gene encoding cell sequencing and high-resolution microscopy. Together, In human iPSC-based studies of FRDA, which is caused understanding of the disease mechanisms underlying the cerebellariPSC-derived ataxias, cerebellar thereby models enabling are poised the development to advance our of Frataxin mRNA and protein levels as well as mitochondrial more effective treatment options for patients in the future. defectsFrataxin, were disease-relevant observed in cardiomyocytes phenotypes such and as peripheral reduced sensory neurons, two of the affected cell types in FRDA Acknowledgement 36 33-

. In the case of A-T, chromosomal instability and cell Society, the Wellcome Trust, the European Commission ATM Research in the Becker laboratory is funded by the Royal cycle checkpoint defects, resulting from mutations in the gene encoding ataxia-telangiectasia mutated (ATM) (Marie Skłodowska-Curie Actions), the Rosetrees Trust, protein, have been found to37 reduce the efficiency of the induced to differentiate into neuronal cells , it remains to ConflictBrAshA-T andof Interest the John StatementFell OUP Research Fund. reprogramming of A-T iPSCs . Although A-T37 iPSCs can be be seenA handful whether of studies they are have able focused to generate on the characterizationPurkinje cells. ReferencesWe declare no conflict of interests. . To 1. Rossi M, Perez Lloret S, Doldan L, et al. Autosomal dominant cerebellar ofdate, iPSCs only orone iPSC-derived study has examined neurons the from disease patients phenotype25,38-41 with polyglutaminein the context of SCAs, the affected including neuronal SCA2, cell 3, 6 type. and Ishida 7 et al. ataxias: a systematic review of clinical features. Eur J Neurol. 2014; 2. 21(4): 607-615. cells from SCA6 patients that harbour a polyglutamine CerebellarBürk K. Clinical Disorders. Scales Dordrecht:of Cerebellar Springer Ataxias. Netherlands; In: Manto M, 2013.Gruol DL,pp. expansionrecently reported in the calcium the differentiation channel Cav2.1 of iPSCs25 into Purkinje Schmahmann JD, et al, editors. Handbook of the Cerebellum and cells were more vulnerable to depletion of thyroid hormone 3. 1783-1798.Sun YM, Lu C, Wu ZY. Spinocerebellar ataxia: relationship between triiodothyronine (T3), an important growth. Patient hormone Purkinje phenotype and genotype a review. Clin Genet. 2016; 90(4): 305-314.

Page 14 of 15 Wong MMK, Watson LM, Becker EBE. J Neurol Neuromedicine (2017) 2(7): 11-15 Journal of Neurology & Neuromedicine

Sandford E, Burmeister M. Genes and genetic testing in hereditary

4. Generated from Spinocerebellar Ataxia Type 6 Patient-Derived iPSCs. 5. Matillaataxias. Genes.Duenas 2014; A, Ashizawa 5(3): 586-603. T, Brice A, et al. Consensus paper 26. Cell Rep. 2016; 17(6): 1482-1490. pathological mechanisms underlying neurodegeneration in of GABA/Glutamate Neuron Differentiation in Autism Spectrum Mariani J, Coppola G, Zhang P, et al. FOXG1-Dependent Dysregulation

6. spinocerebellarMeera P, Pulst SM, ataxias. Otis TS. Cerebellum. Cellular and 2014; circuit 13(2): mechanisms 269-302. underlying Disorders. Cell. 2015; 162(2): 375-390. 27. Quadrato G, Nguyen T, Macosko EZ, et al. Cell diversity and network spinocerebellar ataxias. J Physiol. 2016; 594(16): 4653-4660. dynamics in photosensitive human brain organoids. Nature. 2017; spinocerebellar ataxias treatment based on common mechanisms of 545(7652): 48-53. 7. Bushart DD, Murphy GG, Shakkottai VG. Precision medicine in 28. Bir59.ey F, Andersen J, Makinson CD, et al. Assembly of functionally disease. Ann Transl Med. 2016; 4(2): 25. integrated human forebrain spheroids. Nature. 2017; 545(7652): 54- 29. 8. Hekman KE, Gomez CM. The autosomal dominant spinocerebellar pluripotent stem cells for modelling neurodevelopmental disorders. ataxias: emerging mechanistic themes suggest pervasive Purkinje cell Ardhanareeswaran K, Mariani J, Coppola G, et al. Human induced 9. vulnerability. J Neurol Neurosurg Psychiatry. 2015; 86(5): 554-561. 30. Nat Rev Neurol. 2017; 13(5): 265-278. Cendelin J. From mice to men lessons from mutant ataxic mice. 10. Cerebellum Ataxias. 2014; 1: 4. Dutta D, Heo I, Clevers H. Disease Modeling in Stem Cell-Derived 3D 31. Organoid Systems. Trends Mol Med. 2017; 23(5): 393-410. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Kelava I, Lancaster MA. Dishing out mini-brains: Current progress and 11. Cell. 2006; 126(4): 663-676. future prospects in brain organoid research. Dev Biol. 2016; 420(2): 32. 199-209. Takahashi K, Tanabe K, Ohnuki M, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007; Qian X, Nguyen HN, Song MM, et al. Brain-Region-Specific Organoids 12. 131(5): 861-872. Using Mini-bioreactors for Modeling ZIKV Exposure. Cell. 2016; 33. Eigentler165(5): 1238-1254. A, Boesch S, Schneider R, et al. Induced pluripotent stem Yu J, Vodyanik MA, Smuga Otto K, et al. Induced pluripotent stem cell cells from friedreich ataxia patients fail to upregulate frataxin during lines derived from human somatic cells. Science. 2007; 318(5858): in vitro differentiation to peripheral sensory neurons. Stem Cells Dev. 13. 1917-19120. technology for modelling and therapy of cerebellar ataxia. Open Biol. Watson LM, Wong MMK, Becker EBE. Induced pluripotent stem cell 2013; 22(24): 3271-3282. cardiomyocytes derived from induced pluripotent stem cells as a 2015; 5(7): 150056. 34. Hick A, Wattenhofer Donze M, Chintawar S, et al. Neurons and hiPSC Disease Modeling of Rare Hereditary Cerebellar Ataxias. 14. Lukovic D, Moreno Manzano V, Rodriguez Jimenez FJ, et al. model for mitochondrial defects in Friedreich’s ataxia. Dis Model 35. Mech. 2013; 6(3): 608-621. 15. Neuroscientist. 2016; 1073858416672652. pluripotent stem cell lines from Friedreich ataxia patients. Stem Cell Liu J, Verma PJ, Evans Galea MV, et al. Generation of induced Leto K, Arancillo M, Becker EBE, et al. Consensus Paper: Cerebellar 16. Development. Cerebellum. 2016; 15(6): 789-828. 36. Rev. 2011; 7(3): 703-713. Butts T, Green MJ, Wingate RJ. Development of the cerebellum: simple Du J, Campau E, Soragni E, et al. Role of mismatch repair enzymes Andreu A, Crespo Enriquez I, Echevarria D. Molecular events directing steps to make a ‘little brain’. Development. 2014; 141(21): 4031-4041. in GAA.TTC triplet-repeat expansion in Friedreich ataxia induced Nayler S, Gatei M, Kozlov S, et al. Induced pluripotent stem cells from 17. pluripotent stem cells. J Biol Chem. 2012; 287(35): 29861-29872. the patterning and specification of the cerebellum. Eur J Anat. 2014; 37. 18(4): 245-252. ataxia-telangiectasia recapitulate the cellular phenotype. Stem Cells MorinoTransl Med. H, Matsuda 2012; 1(7): Y, Muguruma523-535. K, et al. A mutation in the low 18. Erceg S, Ronaghi M, Zipancic I, et al. Efficient differentiation of human embryonic stem cells into functional cerebellar-like cells. Stem Cells 38. properties of the channel, causing spinocerebellar ataxia. Mol Brain. 19. Dev. 2010; 19(11): 1745-1756. voltage-gated calcium channel CACNA1G alters the physiological neurons from human pluripotent stem cells. Curr Protoc Stem Cell Erceg S, Lukovic D, Moreno Manzano V, et al. Derivation of cerebellar Biol. 2012; Chapter 1: Unit 1H.5. 39. 2015; 8: 89. pluripotent stem cells to model spinocerebellar ataxia type 2 in vitro. 20. Salero E, Hatten ME. Differentiation of ES cells into cerebellar neurons. Xia G, Santostefano K, Hamazaki T, et al. Generation of human-induced

J Mol Neurosci. 2013; 51(2): 237-248. 21. Proc Natl Acad Sci U S A. 2007; 104(8): 2997-3002. neuron precursors from embryonic stem cells. Dev Biol. 2006; 290(2): 40. Koch P, Breuer P, Peitz M, et al. Excitation-induced ataxin-3 aggregation Su HL, Muguruma K, Matsuo Takasaki M, et al. Generation of cerebellar in neurons from patients with Machado-Joseph disease. Nature. 2011; Luo480(7378): Y, Fan Y, 543-546. Zhou B, et al. Generation of induced pluripotent stem 22. 287-296. 41. Muguruma K, Nishiyama A, Ono Y, et al. Ontogeny-recapitulating cells from skin fibroblasts of a patient with olivopontocerebellar generation and tissue integration of ES cell-derived Purkinje cells. Nat 23. atrophy. Tohoku J Exp Med. 2012; 226(2): 151-159. Neurosci. 2010; 13(10): 1171-1180. cell dendritic development via the thyroid hormone receptor alpha1. polarized cerebellar tissue in 3D culture of human pluripotent stem 42. Heuer H, Mason CA. Thyroid hormone induces cerebellar Purkinje Muguruma K, Nishiyama A, Kawakami H, et al. Self-organization of Taroni F, Chiapparini L, Mariotti C. Autosomal Dominant Wang S, Wang B, Pan N, et al. Differentiation of human induced J Neurosci. 2003; 23(33): 10604-10612. cells. Cell Rep. 2015; 10(4): 537-550. Spinocerebellar Ataxias and Episodic Ataxias. In: Manto M, Gruol 43. 24. 2015; 5: 9232. Cerebellar Disorders. Dordrecht: Springer Netherlands; 2013. pp. pluripotent stem cells to mature functional Purkinje neurons. Sci Rep. 25. DL, Schmahmann JD, et al, editors. Handbook of the Cerebellum and

Ishida Y, Kawakami H, Kitajima H, et al. Vulnerability of Purkinje Cells 2193-2267.

Page 15 of 15