ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1622

Induced pluripotent stemcell (iPSC) modelling for theidentification of mechanismsbehind neurodevelopmentaldisorders

LOORA LAAN

ISSN 1651-6206ISBN 978-91-513-0833-3urn:nbn:se:uu:diva-398620

Dissertation presented at Uppsala University to be publicly examined in Room A1:111a,BMC, Husargatan 3, Uppsala, Friday, 7 February 2020 at 09:15 for the degree of Doctor ofPhilosophy (Faculty of Medicine). The examination will be conducted in English. Facultyexaminer: Associate Professor Jared Sterneckert (Dresden University of Technology, Centerfor Regenerative Therapies).

AbstractLaan, L. 2020. Induced pluripotent stem cell (iPSC) modelling for the identification ofmechanisms behind neurodevelopmental disorders. Digital Comprehensive Summaries ofUppsala Dissertations from the Faculty of Medicine 1622. 63 pp. Uppsala: Acta UniversitatisUpsaliensis. ISBN 978-91-513-0833-3.

Human induced pluripotent stem cells (iPSCs) have opened new possibilities to recapitulatedisease mechanisms and to model disorders in vitro. In the studies presented here, iPSCs wereestablished to model neural differentiation in Down syndrome (DS), caused by trisomy forchromosome 21 (T21); Dravet syndrome (DRS), caused by variants in the SCN1A gene; and anataxia syndrome, caused by a variant in the NFASC gene. The major aim has been to uncovermolecular and cellular mechanisms behind perturbed neurogenesis in the three disorders.

In Paper I, the analysis of transcriptomes and proteomes of the DS iPSC derived neuralmodel revealed several perturbed gene clusters with strong temporal dynamics along neuraldifferentiation, markedly down-regulated mitochondrial genes and a dysregulation of hubproteins. These results predict complex and genome-wide changes in T21 neural cells associatedwith prolonged cell cycle, reduced cell growth and a perturbed energy metabolism.

In Paper II, it was demonstrated that the transcriptional profile of iPSC based neuralmodel system for DS was enriched for differentially methylated genes and gene familieswhen compared to a corresponding euploid model. The differentially methylated genes wereenriched for transcriptional regulation and chromatin structure, suggesting novel mechanisticlinks between the genomic imbalance caused by T21 and the global transcriptional dysregulationin DS.

In Paper III, it was shown that DRS patient iPSCs differentiated into GABAergicinterneurons exhibit a dysregulated epilepsy gene network as well as an altered expressionof genes involved in chromatin remodelling, accompanied by abnormal electrophysiologicalproperties and increased stress sensitivity.

In Paper IV, it was shown that neural iPSCs, established from a patient with anataxia syndrome and a novel homozygous variant in the NFASC gene, lack a full-lengthneurofascin-186 important for cell adhesion. The patient derived neural iPSCs showed delayedneuronal differentiation, reduced sprouting, shorter neurites and altered electrophysiology.

The Papers I-IV show that patient derived neural iPSCs enable to identify molecularand cellular mechanisms associated with neuropathogenesis. Besides specific dysregulatedpathways and cellular defects in models of three developmental disorders, with shortlists ofnovel candidate disease biomarkers, the results are consistent with prior data and clinicalpresentation of patients. The knowledge gained is of paramount importance for translation intoclinical settings and a step towards development of novel therapies with the ultimate goal toalleviate symptoms of affected individuals.

Keywords: Induced pluripotent stem cells, Neurogenesis, Neural differentiation, Downsyndrome, Dravet syndrome, Ataxia

Loora Laan, Department of Immunology, Genetics and Pathology, Medicinsk genetik ochgenomik, Rudbecklaboratoriet, Uppsala University, SE-751 85 Uppsala, Sweden.

© Loora Laan 2020

ISSN 1651-6206ISBN 978-91-513-0833-3urn:nbn:se:uu:diva-398620 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-398620)

To my dearest mom, dad and sister

Mu kallile emmele, issile ja õele

Main Supervisor Niklas Dahl, PhD, Professor Dept. of Immunology, Genetics and Pathology, Uppsala University, Sweden Co-supervisors Jens Schuster, PhD, Associate Professor Dept. of Immunology, Genetics and Pathology, Uppsala University, Sweden Joakim Klar, PhD, Associate Professor Dept. of Immunology, Genetics and Pathology, Uppsala University, Sweden Anna Falk, PhD, Associate Professor Dept. of Neuroscience Karolinska Institutet, Sweden Faculty opponent Jared Sterneckert, PhD, Assistant Professor Center for Regenerative Therapies Technische Universität Dresden, Germany Review board members Finn Hallböök, PhD, Professor Dept. of Neuroscience Uppsala University, Sweden Joey Lau Börjesson, PhD, Associate senior lecturer Dept. of Medical Cell Biology Uppsala University, Sweden Henrik Ahlenius, PhD, Professor Dept. of Clinical Sciences Lund University, Sweden

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals. Reprints were made with permission from the re-spective publishers.

I Sobol M*, Klar J*, Laan L, Shahsavani M, Schuster J, Annerén

G, Konzer A, Mi J, Bergquist J, Nordlund J, Hoeber J, Huss M, Falk A, Dahl N. (2019) Transcriptome and proteome profiling of neural induced pluripotent stem cells from individuals with Down syndrome disclose dynamic dysregulations of key pathways and cellular functions. Molecular Neurobiology, 56:7113–7127.

II Laan L*, Klar J*, Sobol M, Hoeber J, Shahsavani M, Kele M, Zakaria M, Fatima A, Annerén G, Falk A, Schuster J, Dahl N. (2019) DNA methylation changes in Down syndrome derived neural iPSCs uncover co-dysregulation of ZNF and HOX3 fam-ilies of transcription factors. Submitted.

III Schuster J*, Laan L*, Klar J*, Jin Z, Huss M, Korol SV, Nor-addin FH, Sobol M, Birnir B, Dahl N. (2019) Transcriptomes of Dravet syndrome iPSC derived GABAergic cells reveal dysregulated pathways for chromatin remodeling and neurodevelopment. Neurobiology of Disease, 132:104583.

IV Kvarnung M*, Shahsavani M*, Taylan F, Moslem M, Breeu-wsma N, Laan L, Schuster J, Jin Z, Nilsson D, Lieden A, Ander-lid BM, Nordernskjöld M, Lundberg ES, Birnir B, Dahl N, Nordgren A, Lindstrand A, Falk A. (2019) Ataxia in patients with bi-allelic NFASC mutations and absence of full-length NF186. Frontiers in Genetics, 10:896.

* Shared authorship

Additional Publications

Schuster J, Fatima A, Sobol M, Norradin FH, Laan L, Dahl N. (2019) Generation of three human induced pluripotent stem cell (iPSC) lines from three patients with Dravet syndrome carrying distinct SCN1A gene mutations. Stem Cell Research, 39:101523. Schuster J, Sobol M, Fatima A, Khalfallah A, Laan L, Anderlid BM, Nordgren A, Dahl N. (2019) Mowat-Wilson syndrome: Generation of two human iPS cell lines (UUIGPi004A and UUIGPi005A) from siblings with a truncating ZEB2 gene variant. Stem Cell Research, 39:101518. Lam M, Moslem M, Bryois J, Pronk RJ, Uhlin E, Ellström ID, Laan L, Olive J, Morse R, Rönnholm H, Louhivuori L, Korol SV, Dahl N, Uhlén P, Anderlid BM, Kele M, Sullivan PF, Falk A. (2019) Single cell analysis of autism patient with bi-allelic NRXN1-alpha deletion reveals skewed fate choice in neural progenitors and impaired neuronal functionality. Experimental Cell Research, 383:111469. Schuster J, Fatima A, Schwarz F, Klar J, Laan L, Dahl N. (2019) Generation of human induced pluripotent stem cell (iPSC) lines from three patients with von Hippel-Lindau syndrome carrying distinct VHL gene mutations. Stem Cell Research, 38:101474.

Contents

Introduction ................................................................................................... 11 The human genome .................................................................................. 11 Genetic Variation ..................................................................................... 12 Identifying genes and gene variants for Mendelian and complex disorders ................................................................................................... 12 Model systems for genotype-phenotype correlations ............................... 13 Selected methods for read-out of model systems ..................................... 14

Induced pluripotent stem cell (iPSC) modelling for the identification of mechanisms behind neurodevelopmental disorders ...................................... 17

Development of the central nervous system ............................................. 17 Major groups of neurodevelopmental disorders ....................................... 19 Investigated neurodevelopmental disorders ............................................. 21 Modelling neurodevelopmental disorders with induced pluripotent stem cells .................................................................................................. 23 Electrophysiological changes as a readout for a disease .......................... 26

Experimental procedures .............................................................................. 29 Cell culture methods ................................................................................. 29 Molecular biology techniques .................................................................. 32

Aims of the study .......................................................................................... 35

Results and Discussion ................................................................................. 36 Paper I. Transcriptome and proteome profiling of neural induced pluripotent stem cells from individuals with Down syndrome disclose dynamic dysregulations of key pathways and cellular functions ............. 36 Paper II. DNA methylation changes in Down syndrome derived neural iPSCs uncover co-dysregulation of ZNF and HOX3 families of transcription factors .................................................................................. 39 Paper III. Transcriptomes of Dravet syndrome iPSC derived GABAergic cells reveal dysregulated pathways for chromatin remodeling and neurodevelopment .......................................................... 42 Paper IV. Ataxia in patients with bi-allelic NFASC mutations and absence of full-length NF186 ................................................................... 44

Conclusions ................................................................................................... 46

Future perspectives ....................................................................................... 48

Acknowledgement ........................................................................................ 50

References ..................................................................................................... 54

Abbreviations

AP BMP bp CNS CGI Ctrl DEP DiffNPC DMP DNA DS DRS EB EGF ESC FGF2 HSA21 ID iPSC Kv channels MBHB MS NDD NFASC NPC PNS PSC Nav channels RNA Seq SHH SNV TFs T21 3D 5hmC 5mC

Action potential Bone morphogenetic protein Base pair Central nervous system CpG island Control Differentially expressed protein Differentiated neural precursor cell Differentially methylated probe Deoxyribonucleic acid Down syndrome Dravet syndrome Embryoid body Epidermal growth factor Embryonic stem cell Fibroblast growth factor 2 Human chromosome 21 Intellectual disability Induced pluripotent stem cell Voltage-gated potassium channels Midbrain and hindbrain Mass spectrometer Neurodevelopmental disorders Neurofascin Neural precursor cell Peripheral nervous system Pluripotent stem cell Voltage-gated sodium channels Ribonucleic acid Sequencing Sonic hedgehog single nucleotide variant Transcription factors Trisomy 21 three-dimensional 5-hydrocymethylcytosine 5-methylcytosine

11

Introduction

Science is more than a body of knowledge. It is a way of thinking; a way of skeptically interrogating the universe with a fine

understanding of human fallibility. – Carl Sagan

The human genome In the mid-nineteenth century, remarkable scientific progress was made in ge-netic research by establishing the cellular basis of heredity. It was already un-derstood then that the cell nucleus contains hereditary information in thread-like structures, later termed chromosomes1. Almost a century later, the deox-yribonucleic acid (DNA), the molecular carrier of genetic information, was structurally solved2. DNA is an antiparallel double-stranded polynucleotide that determines the genetic code through the order of nucleotides guanine, ad-enine, thymine and cytosine2. These nucleotides base pair (bp) in a specific manner to build up the DNA sequence (adenine with thymine and cytosine with guanine). Most importantly, DNA is a template for transcribing ribonu-cleic acid (RNA), which is then translated into proteins.

The development of DNA sequencing methods enabled scientist to com-mence the Human Genome Project in 1990 with an aim to sequence the full set of human genes. The first draft of the human genome was published in 20013,4 and the complete sequence in 20045. This was followed by a revolution in the high-throughput sequencing technologies that resulted in affordable and relatively fast sequencing6. Thereafter, the 100 Genomes Project was launched7, and, to date, numerous countries have separately initiated the 100 000 Genomes Project (including the United Kingdom, Japan, China, Aus-tralia, Saudi Arabia, United States, Estonia, France, Dubai and Turkey)8. To date, over a million human genomes have been sequenced.

Today, we know that the human genome consists of approximately 3.1 bil-lion base pairs containing over 20 000 protein coding genes and numerous non-coding elements9. Strikingly, 99% of the total genome does not code for proteins and was considered as “junk” for a long time. However, with the En-cyclopedia of DNA Elements (ENCODE) project it was shown that 80% of the genome does serve a purpose10. These discoveries have revolutionised the field of genetics, deepened our understandings about diseases and opened a new era in medicine.

12

Genetic Variation Remarkably large genetic variation between individuals has made it compli-cated to understand the relationship between DNA sequence and phenotype. This diversity is the result of random alterations, known as mutations, that occur in our genes and non-genic functional regions7,11. A combination of en-dogenous (e.g. mistakes in DNA replication) and exogenous processes (e.g. tobacco smoke) are the drivers of mutation12. The majority of mutations have a neutral effect on genetic fitness, but occasionally they can be adaptive or deleterious13.

Mutations can be divided into types, including single nucleotide variants (SNVs), small nucleotide insertions or deletions (known as indels) and large rearrangements. SNVs are a substitution of a single nucleotide at a specific position in the DNA sequence, and these substitutions account for most ge-netic variation in humans14. Most SNVs have no effect, but missense or non-sense variants within a gene or in a regulatory element of a gene can cause disorders through amino acid alterations, premature stop codons, frameshifts or disruption of the splice sites15. Small insertions and deletions (indels) range from 1 to 10 000 bp in length and they are caused by repeat expansions, trans-poson insertions and random sequences16. Indels are distributed throughout the human genome, similarly to SNVs, and they can cause disorders when they occur at functionally important sites16. Another type of mutations are large rearrangements also known as structural variants. They include inver-sions, translocations and copy number variations that are larger than 1 kb17,18. Large rearrangements in protein coding or non-coding regions can result in disorders.

The described variations can be inherited within families as dominant or recessive alleles or they can occur de novo in germ cells (e.g. sperm and egg) or in the zygote state. Sometimes somatic mutations take place during devel-opment but these alterations are not transferred to the offspring19.

Identifying genes and gene variants for Mendelian and complex disorders Mutations that disrupt normal molecular processes can lead to either Mende-lian (also known as monogenic) disorders or complex, multifactorial disor-ders. Mendelian disorders are caused by a deleterious mutation that occurs at a single genetic locus20 while complex disorders are caused by multiple defec-tive genes in combination with environmental factors and lifestyle21. While monogenic disorders follow a Mendelian pattern of inheritance, complex dis-orders have no obvious inheritance pattern, making them difficult to study and treat.

13

Currently, patients are diagnosed for discrete disease entities based on clin-ical presentation, but identification of disease-causing genes and gene variants is essential for mechanistically understanding the disease and developing treatments. However, discrimination between normal variation and disease-causing mutations is not straightforward. Often, it is assumed that variants found in highly conserved or functionally important regions are more likely pathogenic. Therefore, the first step to study a genetic disorder is to identify the location of the gene likely causing the disorder. This can be done by im-plementing first generation, next-generation and third-generation sequencing techniques. Identified genetic variants can be followed up by additional ex-perimentation using human samples or model systems, in order to link them to disease.

Sanger sequencing was the first precise sequencing method22. It is still in use to produce high quality reads less than one kb in length23. The method has limitations, such as low throughput and high cost, which led to the develop-ment of next-generation sequencing platforms that allow parallel sequencing of DNA and RNA (cDNA) fragments, making these technologies high throughput23. Whole genome sequencing and whole exome sequencing are widely used methods that implement next-generation sequencing. While whole genome sequencing targets both the coding and non-coding parts of the genome to identify genetic variations, whole exome sequencing only targets coding parts23. Recently, third- generation sequencing methods that produce long reads with unprecedented quality emerged and are under active develop-ment24.

Model systems for genotype-phenotype correlations Limited availability of human cells and tissues has promoted modelling of human disorders in other living organisms or in cell culture systems. Both approaches, knows as in vivo and in vitro model systems respectively, have been extremely important for understanding basic gene function, discriminat-ing between normal and pathogenic variations and identifying disturbed cel-lular processes caused by pathogenic mutations.

In vivo model systems Invertebrate organisms (e.g. nematode worm and fruit fly), non-mammalian vertebrates (e.g. zebrafish) and various mammals’ species (e.g. mice and rats) have been used in laboratory settings to help understand biological processes in health and disease. These model systems have relatively short life cycles and large number of offspring. In addition, many genes identified in these spe-cies are homologues to human genes25–29, allowing researchers to investigate the function of these genes without using human cells or subjects.

14

Despite some genetic conservation, the last common ancestor of humans and fruit flies can be placed around 780 million years ago30, humans and nem-atodes around 700 million years ago31, humans and zebrafish around 450 mil-lion years ago32 and humans and mice around 75 million years ago33. It is therefore unsurprising that significant genetic differences occur between these organisms. In disease modelling, the differences raise problems when trans-lating findings in model systems to humans, and limit what may be con-cluded27. This is also the reason why studies on in vivo model systems can fail to predict human reactions to pharmaceutical products and environmental agents34.

In conclusion, different in vivo model systems have been essential, albeit not perfect to investigate gene-phenotype correlations.

In vitro model systems The drawbacks of available in vivo model systems to study human disease make complementary in vitro systems important. Different immortalized and cancer cell lines are available for studying molecular pathways, genetic ma-nipulations (e.g. knockdown, knockout, protein and RNA overexpression) and drug response35,36, but culturing cells for generations changes their character-istics and can make them unsuitable representatives of their primary tissue cells37, and therefore less suitable for disease modelling.

Recently, pluripotent stem cells (PSCs) emerged as a tool to model differ-ent human tissues38–40. Pluripotent stem cells (PSCs), including embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs), have an unlim-ited capacity to self-renew in culture as well as the ability to undergo differ-entiation into almost any cell type found in the human body40–42. In disease modelling, PSCs derived from patients maintain the genetic signature of the patient, and thus enable linkage of an in vitro cell phenotype to a clinical presentation. ESCs are similarly suitable for disease modelling, but their usage is prevented by ethical issues and limited access to embryos with a desired genotype43. ESCs and iPSCs have been shown to be morphologically, tran-scriptionally and epigenetically similar41,44, therefore human iPSCs have emerged as a useful alternative to ESCs.

Selected methods for read-out of model systems Accurate disease models combined with suitable molecular biology tech-niques are extremely important for understanding disease pathogenesis and for developing novel therapeutics. Molecular biology techniques involve ma-nipulation and analysis of DNA, RNA and proteins.

15

Genome editing in biomedical research Combining model systems with gene editing enables research into both Men-delian and genetically complex disorders. Different methods with high effi-ciency and safety are available to change the DNA in cells and model organ-isms45. The biomedical research field has advanced enormously with the emer-gence of the clustered regulatory interspaced short palindromic repeat (CRISPR) system in combination with CRISPR associated (Cas) proteins (e.g. CRISPR/Cas9)46. This technique made it possible to generate isogenic cell lines from individuals that possess the same genotype or differ by only a single or a few genes. Thus, they are perfect control cell lines to study genetic disor-ders. Generation of such cell lines can be done by either introducing disease alleles into healthy cell lines or by repairing disease causing alleles in patient cells47. It has also been implemented to eliminate targeted chromosomes48, which could theoretically be used to generate control lines for trisomic disor-ders.

‘Omics’ approaches to disease Comprehensive assessment of a set of molecules, referred to as ‘omes’, offers opportunities to discover novel biomarkers and understand processes that un-derlie disorders49, even when a patient is in a pre-symptomatic state50. Each ‘omics’ dataset can provide a list of disease associated differences49. Today, there are many technologies which can capture entire pools of transcripts, pro-teins, metabolites, and genomes.

Epigenomics is used to characterize reversible modifications of DNA such as methylation49. Methylation is an essential DNA modification present in many animals, and it usually occurs on CpG sites in DNA, where it functions predominantly to silence gene transcription51. In many disease processes, the DNA methylation pattern changes52. Technological advances have opened possibilities to measure the genome-wide methylation at single-nucleotide resolution, enabling the study of methylation patterns in health and disease.

Transcriptomics enables researchers to study the full set of RNA transcripts expressed in cells (the transcriptome). RNA sequencing (RNA-seq) is a key technique used for identification and quantification of expressed transcripts. It is a next generation sequencing method adapted to sequence RNA and it relies on deep-sequencing technologies and computational methods53. RNA-seq has been shown to give more accurate results at measuring the levels of transcripts and their isoforms compared to other methods53. To date, this com-monly used method has revealed global transcriptional dysregulations associ-ated with many disorders.

Proteomics is used for proteome profiling and detecting the most abundant proteins found in a cell, tissue or organism54. For high-throughput analysis of proteins, mass spectrometry (MS) based proteomics is implemented55.

16

Despite the clear advantages provided by ‘omics’ techniques, their use also presents numerous challenges54. Both technology and software are continu-ously evolving, leading to results that might vary depending on the approach implemented. Further, the human proteome and metabolome are still not fully resolved and therefore limiting our understanding of results obtained by MS. Finally, several ethical issues are related to the storage of samples and ob-tained data. The high cost of ‘omics’ technologies also limits their usage at present.

17

Induced pluripotent stem cell (iPSC) modelling for the identification of mechanisms behind neurodevelopmental disorders

Essentially, all models are wrong, but some are useful. – George E.P. Box

Development of the central nervous system The human nervous system is divided anatomically and developmentally into central (CNS) and peripheral (PNS) components56 that comprise billions of nerve cells, or neurons, and supporting neuroglia, or shortly glia57. The CNS consists of the brain and the spinal cord while the PNS is made up from cranial and spinal nerves and glia outside of the CNS56. The development of the CNS begins in the third gestational week with specialisation of the embryonic mid-line ectoderm into neural ectoderm56 (Figure 1). This process is actively di-rected by the inhibition of bone morphogenetic protein (BMP) signalling58. Neural induction leads to formation of the neural plate, followed by neurula-tion where the edges of the neural plate lift upwards to form the neural groove59. During the fourth gestational week, the neural plate folds further inward and on embryonic day 30 it closes and segregates from the non-neural epithelium to form the neural tube58,59.

Figure 1. Early events during the development of the nervous system. The nervous system development is initiated by the inhibition of BMP signalling which leads to the formation of the neural plate. The neural plate folds further to from the neural groove that eventually will close and segregate from the non-neural epithelium to from the neural tube – the precursor of the CNS. Modified from Liu and Niswander 200558.

18

Subsequently, the neural stem cells within the neural tube start to proliferate, pattern and migrate to build up the nervous system60. These processes depend on the positional information and gradients of morphogens BMP, sonic hedge-hog (SHH), WNT and retinoic acid61. Binding of these signalling molecules to their receptors activates their downstream transcription factors (TFs) that regulate neural differentiation and define the order of different sub-types of neurons62. The order of generating neural cells in the developing CNS is widely conserved across vertebrate species and well exemplified by cortical development. The first wave of cells produced in the neural tube are neurons that form the cortical layers61. This is followed by the production of glioblasts that differentiate into astrocytes, oligodendrocytes and ependymocytes61. For normal development of this complex system, proliferation, differentiation and migration of specific neural subtypes must be balanced43.

After the formation of the neural tube, the presumptive future forebrain, midbrain and hindbrain become apparent in primary brain vesicles that later convert into five secondary brain vesicles (Figure 2)61. The prosencephalon (the future forebrain) brain vesicles develop into the cerebral hemisphere and structures underneath (thalamus, hypothalamus and epithalamus)63. This part of the brain is responsible for consciousness, sensorimotor transformation and sensory integration. Mesencephalon (the future midbrain) structures function mainly as a rely centre for sensory and motor impulses between forebrain and hindbrain structures. It is associated, for example, with vision, hearing, motor control and alertness. Rhombencephalon (the future hindbrain) secondary ves-icles develop into pons, cerebellum and medulla. The medulla regulates res-piration, heartbeat and reflex movements. The role of the pons is to relay sig-nals that link both the spinal cord and the cerebral cortex with the cerebellum. The cerebellum is important for information integration in order to fine-tune output important for balance control.63

At birth the human brain is only about 25% of its adult volume and the final size is reached at around 7 years of age61. The development of the brain and CNS continues into the mid-20s64 and is coordinated by genes. Disruption of these genes might lead to neurodevelopmental disorders (NDDs).

19

Figure 2. The presumptive brain subdivisions (the future forebrain, midbrain and hindbrain) in primary (upper left) and secondary (upper right) brain vesicles. Sagittal cut of the adult brain illustrating some of the forebrain, midbrain and hindbrain structures. Figure adapted from Wikimedia Commons65.

Major groups of neurodevelopmental disorders NDDs have their onset in childhood and are characterized by impaired brain function and worsened ability to process information, learn and to communi-cate. Currently, two major diagnostic classification systems for NDDs are in use (DSM-5 published by the American Psychiatric Association66 and the ICD-10 published by the World Health Organization67). Major groups of NDDs include intellectual disabilities (IDs), communication disorders, autism spectrum disorders, attention-deficit/hyperactivity disorders, specific learning disorders, motor disorders and other neurodevelopmental disorders66. Fre-quently, comorbidity between these disorders is examined. Moreover, they of-ten have polygenic and multifactorial etiologies68.

ID disorders are defined by deficits in intellectual and adaptive function-ing69. Often, patients with ID have difficulties in problem solving, planning, abstract thinking, judgement and learning. The severity of these symptoms can vary from mild to severe and profound. ID can be caused by environmental

20

factors such as exposure to toxic substances, nutritional deficiencies, brain in-fection, radiation, trauma and prenatal and either prenatal or postnatal compli-cations. Additionally, different genetic causes have been identified, however, many cases are of unknown etiology.69 Examples of ID disorders determined by genetics or inheritance are Down syndrome (DS)70 and Fragile X syn-drome71.

Individuals diagnosed with vocal communication disorder may have apha-sia, stuttering, verbal dyspraxia, articulation difficulties and slurry speech72. The etiology is not well understood, but as it seems to cluster in families, the genetic factors are likely involved72.

Autism spectrum disorders are a heritable group of conditions diagnosed based on a complex behavioural phenotype73. Commonly, patients exhibit im-paired socialisation and communication. Population studies suggest that au-tism spectrum disorders have multifactorial inheritance. Over the last decade, the diagnostic yield has grown rapidly because of the identification of various candidate genes74–76. Additionally, environmental factors play a role in autism spectrum disorders68.

Attention-deficit/hyperactivity disorders are characterised by inattention, impulsivity, hyperactivity and problems in social interaction and academic performance77. The etiology is complex and not well understood. Studies sug-gest that alterations in dopaminergic and noradrenergic function plays a role in attention-deficit/hyperactivity disorders. Also, environmental factors con-tribute to the disease.77

Specific learning disorders are characterized by difficulties in reading, writ-ing, and mathematics, in spite of normal intelligence and educational oppor-tunities78. It includes conditions such as dyslexia, dysgraphia and dyscalculia. Various susceptibility genes have been identified, but we are far from under-standing the etiology78,79.

Childhood-onset motor disorders (also known as movement disorders) are a diverse group characterized by either unwanted, excessive movements (hy-perkinetic disorders) or movement poverty (hypokinetic disorders)80. Abnor-mal involuntary movements, uncontrollable muscle contractions, jerks and tics are common in hyperkinetic disorders while spasticity, dystonia, and ri-gidity can be observed in patients with hypertonic disorders. In addition, pa-tients can exhibit epilepsy, poor motor control, lack of voluntary coordination of muscle movements (ataxia) and difficulties in activities requiring coordina-tion and movement80. The underlying etiology can be acquired or inherited81.

21

Investigated neurodevelopmental disorders Examples of genetically determined NDDs investigated in this thesis are Down syndrome (DS), Dravet disorder (DRS) and novel ataxia syndrome.

Down syndrome Down syndrome (DS), also known as trisomy 21 (T21) is a common cause of developmental delay and ID. It is estimated to affect one in every 750 live births70. Individuals diagnosed with DS have variable clinical presentations but a characteristic facial appearance and a small, hypocellular brain is present to some degree in every affected individual82. Brain abnormalities lead to im-paired cognition that is considered the main disabling feature in patients83–85.

In 95% of cases, DS is caused by a complete trisomy for chromosome 21. This occurs during gametogenesis when human chromosome 21 (HSA21) fails to separate86 and results in an extra copy in all the body cells. In a minor-ity of cases, the disorder is caused by Robertsonian chromosomal transloca-tion or through formation of an isochromosome or ring chromosome86. In a few cases, following fertilization the errors in cell division and chromosomal malsegregation lead to mosaicism of T2187. Chromosome 21, despite being the smallest human chromosome, encodes 233 coding genes, 423 non-coding genes and 188 pseudogenes88. Many of the HSA21 genes show increased ex-pression in individuals with DS, but also genome-wide dysregulations has been reported89,90. The connection between global dysregulation of genes and disrupted developmental processes are poorly understood.

Various mouse model systems have been developed to study DS, but in the mouse the HSA21 genes are distributed between regions of chromosomes 16, 17 and 1090, and the commonly used models (Ts65Dn91 and Ts1Cje92) only have partial trisomy of a region on chromosome 1693. These mice do exhibit some characteristics of DS, but their phenotype resembles more translocation DS.

Imaging the brain of living individuals with T21, as well as post-mortem human brain specimens, has provided important information. These studies have shown that subjects with T21 have delayed prenatal growth of the cere-bral cortex and hippocampus94,95. These forebrain structures exhibit reduced proliferative rate, fewer neurons84,96 and synapses95,97,98, smaller dendritic ar-borizations99–101, defective myelination102 and disturbed glial cell fate103. Be-sides defective development of forebrain structures, children with T21 have smaller midbrain and hindbrain structures (e.g. cerebellum, pons and medulla oblongata) compared to control children104. A hypomorphic cerebellum is also present in adults with T21105 and in the mouse model for DS106. Hypomor-phism results from impaired neurogenesis and cell proliferation in these brain regions107.

22

The findings indicate that pathophysiological abnormalities associated with ID are established prenatally and earlier intervention would provide better out-come. Current treatment options are limited and there is a great need to find candidate targets for novel targets for drug development. However, inaccessi-bility of the relevant cell types and tissues hinders research.

Dravet syndrome Dravet syndrome (DRS) is a devastating form of epilepsy characterized by drug resistant seizures with an onset within the first year of life. Following, affected individuals show cognitive decline that may progress independently from the frequency or severity of seizures108. Patients suffer from ataxia, neu-rologic manifestations and are at high risk for sudden death109.

In approximately 70% of cases, DRS is caused by de novo mutations in the highly conserved SCN1A gene. This gene encodes the α-subunit of the volt-age-gated sodium channel Nav1.1110 mainly expressed in GABAergic inter-neurons of the CNS111. Interneurons partake in the generation and propagation of action potentials (APs). Numerous genetic mouse models try to replicate SCN1A loss-of-function112. These models have haploinsufficiency of voltage gated sodium channel Nav1.1, which leads to decreased firing of GABAergic interneurons and causes hyperexcitability and seizures113. Interestingly, the pleiotropic mutation can lead to both mild and severe DRS114.

Unfavourable long-term outcome in DRS patients makes it important to elucidate the neuropathophysiology behind this disorder. Precision medicine is limited by the phenotypic variability.

Ataxia Ataxia is a physical finding that refers to a lack or loss of movement coordi-nation, often indicating damage to (or dysfunction of) parts of the CNS re-sponsible for this function (e.g. cerebellum, spinal cord, brain stem, cerebral white matter, cortex)115–117. Ataxia is characterized by incoordination of bal-ance, unsteady gait, abnormal eye movements, errors in learning motor com-petency and dysarthria116. Ataxia varies from transient episodes to lifelong conditions that are especially debilitating for children as they are still in a pro-cess of learning motor competency118. Childhood ataxia has both hereditary and acquired causes and is estimated to affect approximately 26 out of 100 000 children (1:3800)115. Ataxias are inherited through autosomal recessive, auto-somal dominant, X-linked, and mitochondrial inheritance61. As ataxia can be the patients main complaint, or one of many symptoms116,117, the mechanisms underlying heritable ataxia syndromes are often unknown and need to be in-vestigated61.

In this thesis we introduce an early onset autosomal recessive ataxia syn-drome that results from the novel homozygous mutation in the NFASC gene encoding neurofascin (NFASC)119. The pre-mRNA of NFASC can be spliced into more than 50 different isoforms120, of which isoform NF186 (186 kDa) is

23

expressed in neuronal cells of the CNS121. NF186 is a cell adhesion molecule involved in the formation of nodes of Ranvier and recruitment of voltage gated ion channels into the nodes and axon initial segments121. More precisely, it interacts with axon initial segment master organizer ankyrin G to cluster volt-age-gated sodium (Nav) and voltage-gated potassium (Kv) channels121. In ad-dition, NF186 is essential for the maintenance of synapses and for tuning the output from Purkinje cells122–124. In cerebellum, dysfunction in signal tuning can affect motor coordination.

Modelling neurodevelopmental disorders with induced pluripotent stem cells Induced pluripotent stem cells (iPSCs) are pluripotent cells generated from adult cells. They own the capacity to self-renew and an ability to undergo differenti-ation into almost any cell type found in the human body (Figure 3b)40–42. The first iPSCs were generated by forced and transgenic expression of the transcrip-tion factors known as Yamanaka’s factors (i.e. Oct3/4, Sox2, c-Myc and Klf4) in dermal fibroblasts (Figure 3a)41. Since then, this technology has proven its universality for cellular reprogramming in different somatic cells125. For the pro-duction of the first human iPSCs, integrative retrovirus-vector were applied41. However, safer and more efficient delivery and expression of transgenes is en-sured with non-integrative vector-based approaches126,127.

During the reprogramming and long-term culture of iPSCs genomic alter-ations such as point mutations, CNVs, epigenetic and chromosomal aberra-tions may occur128–130. Besides, incomplete reprogramming, in which many colonies remain in an intermediate state, may take place. Therefore, charac-terization of genetic integrity, as well as morphology and molecular charac-teristics is needed for newly reprogrammed and long-term cultured iPSC lines. IPSCs should be able to self-renew unlimited times and independently of transgene expression. Morphologically and epigenetically, high quality iPSCs are similar to ESCs and exhibit a normal karyotype. On a molecular level, iPSCs express key pluripotency factors (Nanog, Oct4), ESC surface antigens (SSEA-3, SSEA-4, Tra-1-60, Tra-1-81) and down-regulate lineage-specific genes of the original somatic cell type. At the functional level, iPSCs must be able to differentiate into three germ layers127. True iPSC lines fulfil all previ-ously mentioned criteria and can be useful for modelling a disorder of inter-est39,131.

24

Figure 3. Induced pluripotent stem cells (iPSCs) derived from somatic cells can be differentiated into any cell type found in the human body. (a) Reprogramming of iPSC from the somatic cells (e.g. dermal fibroblasts) is achieved through overex-pression of different combinations of transcription factors. Figure adapted from Wikimedia Commons132. (b) It is characteristic for iPSCs to self-renew and undergo differentiation into any cell type of the three-germ layer (endoderm, mesoderm and ectoderm).

IPSCs are especially important for understanding brain disorders as access to brain specimens is limited and post-mortem tissue and imaging studies yield little insight into the pathogenesis43. Most importantly, it has been shown that differentiation of iPSCs into neural cells in vitro transit through stages that resemble events and cell populations present during in vivo neurogenesis (Fig-ure 4)133. For instance, the inner cell mass (ICM) of blastocyst that forms in early embryogenesis resembles iPSCs with their capacity to self-renew and proliferate41,133. In vitro differentiation of iPSCs into neuroepithelial neural precursor cells (NPCs) corresponds to the NPCs produced by the neural plate in vivo133. Further in vitro differentiation of NPCs into rosette-type NPCs re-sembles in vivo neurulation when the first waves of neurons are produced134. Further in vitro differentiation either in monolayer or three-dimensional (3D) culture leads to maturation of neural cells and this stage corresponds to in vivo fetal and adult neurogenesis.

25

Figure 4. Neural differentiation in vitro and in vivo. During the in vitro (upper row) differentiation of induced pluripotent stem cells (iPSCs) into neurons they transit through stages that resemble events and cell populations present during in vivo neu-rogenesis (lower row). Modified from Mertens et al 2016133.

Many human NDDs are caused by distortion in the abundance or functionality of neural cells, including specific neuronal subtypes or glial cells. Depending on the disease and the cell population of interest, differentiation of iPSCs into the appropriate cell lineage can be considered. Directed and undirected differ-entiation approaches can be employed. In the case of directed differentiation, cell-fate is controlled by growth-factors and small-molecule compounds that mimic developmental signals135. Undirected differentiation starts by establish-ing self-renewing pluripotent lineage-progenitor cells from iPSCs. These in-termediate cell types can be extensively expanded and thereafter differentiated into relevant cells using both directed and undirected differentiation ap-proaches4,129,136,137.

Numerous directed differentiation protocols to produce neurons or glial cells from iPSCs have been described135. Several of these protocols are adapted to produce cortical neurons that are dysfunctional in the pathophysi-ology in many NDDs. For example, DRS is caused by mutations in the SCN1A gene encoding a channel protein mainly expressed in cortical GABAergic in-hibitory interneurons. Studies of the effect of a mutated SCN1A gene would thus require directed differentiation of GABAergic interneurons. The differ-entiation to cortical and fore-brain neural cells can be initiated by induction with WNT and dual SMAD inhibition followed by patterning with SHH136. Modelling NDDs related to dysfunctional glutamatergic neurons would re-quire other approach137. Various publications have demonstrated that the use of morphogenetic factors normally expressed in the developing brain may be used in culture conditions in vitro to generate specific neuronal cell types126,136,137.

Neurogenesis can be modelled using undirected differentiation approaches implementing self-renewing pluripotent intermediates such as NPCs129,138. NPCs are captured at a defined stage in early neural commitment, making

26

them renewable, stable and homogeneous. It is characteristic for NPCs to ex-press neuronal progenitor markers and form morphologically distinct rosette-like structures that stay undifferentiated in the presence of epidermal growth factor (EGF) and fibroblast growth factor (FGF2)139. The removal of growth factors initiate differentiation of NPCs into neuronal and glial cells found in midbrain, hindbrain (MBHB) and spinal cord129. The NPCs are easily ex-panded and differentiated in the modelling of NDDs.

Evaluation of the cells generated is essential to ascertain that the appropri-ate cell fate was introduced. Importantly, the expression pattern in the cell-of-origin must be abolished and the expression of marker genes characteristic for the given cell type needs to be evident. Analysis of the morphology also en-sures a successful differentiation. Furthermore, assessment of transcription profiles and deoxyribonucleic acid (DNA) methylation profiles of generated cells can give important information. In addition, functionality tests such as electrophysiological measurements may help determine if differentiated cells exhibit similar characteristics as the cells in vivo. The strongest proof is to show that a differentiated cell type is able to functionally replace the same cell type in vivo.135

Moreover, there are now expectations that iPSCs may be used for replace-ment therapy to ameliorate or even cure disease135. Transplanting patient-de-rived iPSCs, in which mutations have been corrected by genome editing, is promising since it reduces the probability of host rejection. Therapeutic po-tential of iPSC derived neural derivatives to repair the damaged CNS has led to clinical trials. The targeted diseases include recurrent high-grade gliomas, amyotrophic lateral sclerosis (ALS), chronic spinal cord injury, stroke, lower limb ischemia, neuronal ceroid lipofuscinosis, cervical spinal cord injury, macular degeneration, thoracic spinal cord injury, Pelizaeus-Merzbacher dis-ease and Parkinson’s disease140. These studies are either on phase I or II and show that the patients tolerate the implantation of the cells and some patients even exhibit modest gain in their neurological function. This suggests that neural stem cell transplants can halt the advances of severe genetic disorders.

Electrophysiological changes as a readout for a disease Some of the cells in the human body are electrically excitable (neurons, mus-cle cells and some endocrine cells)141. Their electrical potential is achieved through different concentrations of ions, such as sodium (Na+) and potassium (K+), across the cell membrane. Because of the impermeable lipid bilayer, ions must cross the membrane only through specialized voltage-gated ion channels (e.g. Nav, Kv channels). These cells are excited by electrical or sensory inputs until they fire electrical potentials. In certain disorders, the electrical activity of the cells is distorted. Understanding the electrical properties of excitable

27

cells therefore has implications beyond the basic knowledge of organ func-tions, namely it helps us understand neurological disorders and then test the effects of pharmaceutical compounds. Moreover, it is a good method to eval-uate the properties of iPSC derived neurons, in order to assess whether disease influences electrical signaling.

Measuring electrophysiological changes in individual cells, usually in volt-age or current, became an option in the early 1970s with the development of the patch-clamp method142. This method implements an intracellular microe-lectrode to measure the electrical potential across the plasma membrane143. Today, different patch clamp configurations are available, such as cell-at-tached, inside-out, whole-cell and outside-out recording mode144 (Figure 5). In the cell-attached mode, measurements are made while the cell membrane is intact, while in inside-out mode they are made after separating the mem-brane patch from the rest of the cell (the inside-out mode). The most com-monly used configuration is the whole-cell recording mode that involves opening the interior of the recording pipette to the cytoplasm through brief but strong suction. The inside-out mode is a modified version of the whole-cell mode and it is suitable for measuring small patches or single channels. The whole-cell recording is used for measuring electrical potentials and currents in the entire cell. This can be done in the voltage-clamp or current-clamp con-figuration. In the voltage-clamp mode the current is recorded while the voltage is kept constant and in the current-clamp mode the membrane potential is rec-orded while the current is unchanged.144

28

Figure 5. Different patch-clamp configurations. In the cell-attached mode the seal between the recording pipette and the membrane is formed through mild suction and the membrane is left intact (cell-attached mode) or separated (inside-out mode) for recordings. In the case of the whole-cell patch clamp strong suction is applied to rupture the cell membrane and gain access to the cytoplasm. If the pipette is re-tracted, outside-out recordings can be performed. Figure adapted from Veitinger 2011145.

29

Experimental procedures

Ethical permits and informed consent The local ethics committee for human research in Uppsala, Sweden (Dnr 2016/209 for Papers I-III) or Stockholm, Sweden (Dnr 2016/430-31 for Pa-per IV) approved all studies prior to initiation. Written informed consent for publications were obtained from the donors of skin punch biopsies, or from their legal guardians.

Cell culture methods

Reprogramming In Paper I, III and IV, fibroblasts were reprogrammed using CytoTune™-iPS 2.0 Sendai Reprogramming Kit. This system applies nonintegrative Sen-dai virus to deliver four Yamanaka factors (hKlf4, hc-Myc, hSox2, hOct3/4)41 for efficient generation of iPSCs. After formation of iPSCs, colonies were manually picked for clonal expansion. Established iPSC lines were character-ized for pluripotency and genomic stability. The assays implemented included karyotyping, embryoid body (EB) formation assay and immunofluorescent staining against pluripotency markers (NANOG, SSEA4 and TRA-1-60).

Karyotyping Karyotyping allows detection of chromosomal abnormalities and assessment of chromosomal integrity. It is a method implemented in Paper I-IV. First, Colcemid was used to arrest the cells in the metaphase. This was followed by washing with hypotonic solution and fixing step. Finally, Giemsa-staining was performed to visualize and analyze chromosomes. Analysis was performed using Metafer slide scanning platform and IKAROS-software (both MetaSys-tems).

Embryoid body formation assay The capacity of the iPSC lines to differentiate into three germ layers was de-termined by EB formation assay in Paper I, III and IV. The EBs were stained against markers for ectoderm, mesoderm and endoderm. The DNA from EBs was used to run TaqMan hPSC scorecard assay (Thermo Fisher Scientific)

30

which evaluates pluripotency and potential to differentiate into three germ lay-ers.

Immunofluorescence staining Immunofluorescence staining was performed in Paper I, II, III and IV ac-cording to standard methodology. Briefly, cells were fixed on chosen timepoints with 4% paraformaldehyde, this was followed by primary and sec-ondary antibody staining. The cells were imaged using either a Zeiss AxioIm-ager or a Zeiss LSM700 confocal microscope.

Undirected neural differentiation Patient specific neural precursor cells (NPCs) derived from iPSCs were the basis for Paper I, II and IV (Figure 6). They were generated as previously described129,134. In brief, iPSCs were neuralized to generate rosettes which were manually isolated and cultured five to seven days without EGF and FGF2. Thereafter, neural rosettes were dissociated into single cells and plated in the presence of growth factors on poly-L-ornithine and laminin L2020 coated plates. A proportion of lines had the capacity to expand, exhibited dis-tinct rosette structures, could be frozen and recovered, expressed neural pre-cursor cell (NPC) markers (NESTIN and SOX2) and were able to differentiate into neuronal and glial cells after removal of growth factors. These lines were used for further experiments. To induce neural differentiation, FGF2 and EGF were removed from the culture medium.

Figure 6. Workflow describing undirected neural differentiation and sampling (day 0 and 30)

Directed neural differentiation Directed cortical differentiation of iPSCs was based on the protocol from Maroof et al 136 and the cell culture enriched with GABA positive neurons was the basis for Paper III. Briefly, neural differentiation was induced using dual SMAD inhibition by supplementing culture media with tankyrase inhibitor XAV939, ALK2/3 inhibitor LDN193189 and ALK4/5/7 inhibitor SB431542 (Figure 7). Subsequently, patterning towards the cortical lineage was accom-plished by addition of SHH and its agonist purmorphamine into culture media for nine days. Thereafter the progenitors matured into neurons for 65 days (iPSC-GABA).

31

Figure 7. Workflow describing directed neural differentiation and sampling (day 0, 19, 65)

Proliferation assay Long-term live cell imaging in the IncuCyte ZOOM™ live cell imaging sys-tem (Essen BioScience) was used to determine the proliferative rate of both trisomic and euploid NPCs in Paper I. Following the cell seeding, the images were taken after every second hour for 96 hours. IncuCyte software module for proliferations assay was implemented for analysis.

Neurite outgrowth assay Neurite outgrowth assay was used in Paper I and IV to estimate the capacity of iPSC derived neural cells to form functional networks. The NPCs were seeded and maintained in differentiation medium as monolayers cultures for 6 days either on coverslips or cell culture dish. In Paper I, this was followed by the immunofluorescent staining with β-III-tubulin (described under immu-nofluorescent staining), neurites visualisation and counting. In Paper IV, after 6 days of culture, cells were disassociated into neurospheres and kept 24 hours on a non-adherent cell culture dish. This was followed by embedding neuro-spheres on coverslips and another 24-hour incubation. Thereafter images of neurospheres were captured and number and length of neurites were meas-ured.

Reactive oxygen species assay Analysis of reactive oxygen species (ROS) levels was performed in Paper I and III to examine ROS production in the cells under normal culture condi-tions and under stress induced conditions. OxiSelect™ Intracellular ROS As-say Kit (Cell Biolabs) was applied following manufacturer’s protocols. Briefly, neuronal cells were loaded with 2′,7′-dichlorodihydrofluorescein di-acetate (DCFH-DA) non-fluorescent probe for 1 hour. Intracellularly occurs de-esterification of DCFH-DA into highly fluorescent 2′,7′-dichlorofluores-cein upon oxidation. The change in fluorescence (excitation 480 nm, emission 530 nm) was monitored for 120 minutes using VerioScan LUX plate reader (Thermo Fisher Scientific).

32

Whole-cell patch-clamp Electrophysiological studies were performed in Paper III and IV. Whole-cell patch-clamp method was used to characterize the neuronal biophysical activ-ity. Depending on the project, either current-clamp or voltage-clamp record-ings were carried out to measure resting membrane potential, sodium current density and spontaneous and evoked action potentials. Analysis and visuali-sation of the data was done by using Axon pCLAMP10.5 software, Excel and GraphPad Prism.

Molecular biology techniques

Exome sequencing Exome sequencing was implemented in Paper IV to identify pathogenic var-iant. DNA derived from blood was enriched with Agilent SureSelect Human All Exon 50M and thereafter analyzed by exome sequencing. Libraries were prepared from genomic DNA and enriched exome sequences. Constructed li-braries were fragmented and thereafter hybridized to probes that capture the coding regions of genome. Non-coding parts were washed away. Post-capture libraries were sequenced on HiSeq2000 instrument (Illumina). Reads were mapped to the human reference genome (CRGCh37/hg19). Genome Analysis Tool Kit (GATK v.3.2.0) and ANNOVAR were used for variants calling and annotation, respectively. Non-synonymous and rare variants (allele frequency less than 1%) were considered for further analysis. In silico prediction tools were used to score deleteriousness of SNVs.

Sanger sequencing Sanger sequencing was used in Paper III and IV to verify the mutations in fibroblast lines and iPSC samples. After isolation of genomic DNA from the cell cultures, standard PCR method, to amplify selected parts of genes using specifically designed primers, and Sanger sequencing were performed.

Transcriptome sequencing Transcriptome sequencing (RNA-Seq) was an essential method used for con-clusions in Paper I and III. Briefly, total RNA isolation was followed by de-pletion of the ribosomal RNA, fragmentation and reverse-transcription to cDNA. cDNA was ligated to adaptors for sequencing library preparation se-quenced on a HiSeq (Illumina). To be able to compare the abundance of tran-scripts between the samples and to reveal differentially expressed genes and dysregulated pathways, several bioinformatic tools were applied. The normal-ized gene expression was generated using the Spliced Transcripts Alignment to a Reference (STAR) software (http://star.mit.edu/)146. The differential gene

33

expression analysis was carried out using DESeq2 software147 which is avail-able as an R/Bioconductor package. Analysis of the gene sets generated was done with Enrichr (http://amp.pharm.mssm.edu/Enrichr/), PANTHER (http://www.pantherdb.org/) and DAVID (https://david.ncifcrf.gov/) tools.

Proteome analysis Proteome analysis was used in Paper I. Relative levels of proteins were quan-tified with mass spectrometry to identify dysregulated pathways. Samples were ionized by matrix assisted laser desorption/ionization (MALDI) source. Ions where separated by the mass analyser based on their mass-to-charge ratio (m/z). Generated mass spectrum contained information on the abundance of ions and their m/z. Proteome analysis was performed on patient and control samples at the neural progenitor cell and the neuronal stage.

DNA methylation analysis In Paper II, assessment of the DNA methylation of known CpG regions and promoters across the genome was done with Illumina HumanMethylation 450K BeadChip and Illumina HiScan 2000 (Figure 8). The initial quality con-trol and identification of signal intensities for each probe was performed using Illumina GenomeStudio Software. The methylation data was analyzed using the Minfi package by group-wise comparison between T21 and control sam-ples. Differentially methylated probes (DMPs) where identified with dmp-Finder function in R. To avoid small methylation differences between T21 and euploid lines due to stochastic variations, we only considered differences of >15% (average beta values of patients vs. controls) using an FDR adjusted p-value (q-value) and a cut off < 0.05. Enrichment in functional categories was performed using the Gene Ontology (GO) knowledgebase database (geneon-tology.org; PANTHER Overrepresentation Test (Released 20190711)) and the Molecular Function category. Gene expression data was retrieved for the neuronal stage from Paper I.

Figure 8. Infinium HumanMethylation450 BeadChip (450K array) workflow. Bisul-fite treatment is implemented to convert unmethylated cytosines into uracil. This is followed by whole-genomic DNA amplification using random hexamers. Amplified DNA products are enzymatically fragmented and purified. The fragments are dena-tured into single strands and hybridized to the chip. The chip contains beads for each CpG site per locus and corresponds either to the methylated cytosine locus or to the unmethylated cytosine locus. Fluorescence staining is used to measure the intensities of methylated and unmethylated bead types. Finally, analysis of methylation data is performed using proprietary software.

34

Bisulfite sequencing Targeted DNA bisulfite sequencing with EpiTYPER technology (Agena Bi-osceince) was implemented in Paper II to verify the results of Illumina 450K array. Genomic DNA was treated with sodium bisulfite using EZ-96 DNA Methylation kit (Zymo Research) and followed by PCR amplifying regions of interest by using T7-promoter-tagger reverse primers and in vitro transcrip-tion. The resulting transcripts were specifically cleaved at uracil residues and subjected to MALDI-TOF analysis (Agena Compact MassARRAY Ana-lyzer). The EpiTYPER software 1.2.22 was used to identify the mass-frag-ments and for quantification of DNA-methylation at single CpG-sites or of a CpG-unit.

Quantitative real-time PCR Quantitative real-time PCR (RT-qPCR) was applied in Paper I-IV. Sigma-Aldrich Ready-made FastStart Universal SYBR Green Master (ROX) mix was mixed with designed primer pairs and cDNA from samples. StepOnePlus Real-Time PCR System (Applied Biosystems) was used to run the RT-qPCR. Analysis was performed with StepOne software v2.2.2 and GraphPad Prism software. The expression levels were normalized against housekeeping genes (GAPDH or ACTB).

Western blot Western blot analysis was performed in Paper IV to examine the expression of neurofascin and ankytin G proteins in neural cells. Protein extraction was performed using Quiagen AllPrep RNA/Protein Kit. Protein concentrations was determined with ThermoFisher Scientific Qubit™ Protein Assay Kit and fluorometer. For western blot analysis, protein extracts were run on a 10% sodium dodecyl sulfate-polyacrylaide gel electrophoresis and electrotrans-ferred onto nitrocellulose membrane. Immunodetection was done with pri-mary and secondary antibodies, visualisation with LICOR ODYSSEY 91020 imaging system and analysis with Image Studie Lite V 5.2 software.

35

Aims of the study

The general aim of this thesis was to identify and characterize novel molecular pathways and key cellular mechanisms associated with abnormal neurodevel-opment and cognitive impairment of three genetic disorders – Down syndrome (DS), Dravet syndrome (DRS) and novel ataxia syndrome. Specific aims for Paper I: Establish self-renewing NPC lines from patients with DS and healthy in-

dividuals and enable robust, non-directed differentiation of neuronal and glial cells;

Assess established NPCs and differentiated NPCs with full T21 as a model for DS;

Identify biomarkers and neurodevelopmental perturbations in T21 neuro-genesis from the analysis of proteins and transcripts using high-through-put ‘omics’ technologies.

Specific aims for Paper II: Use model system established in Paper I to investigate the role of DNA

methylation on T21 neurogenesis. Specific aims for Paper III: Establish high quality iPSC lines from patients with DRS and healthy in-

dividuals and differentiate established lines into cell types relevant to the disorder (here cortical GABAergic interneurons);

Generate transcriptional profiles from neural precursors and GABAergic interneurons using RNA-seq and identify perturbed genes and networks associated with DRS;

Analyse electrophysiological properties of GABAergic interneurons with and without SCN1A mutations.

Specific aims for Paper IV: Establish self-renewing NPC lines from patient and healthy individual to

perform robust, non-directed differentiation of neuronal and glial cells with an aim to investigate NFASC as a novel ataxia causing gene.

36

Results and Discussion

Paper I. Transcriptome and proteome profiling of neural induced pluripotent stem cells from individuals with Down syndrome disclose dynamic dysregulations of key pathways and cellular functions Various studies have shown that iPSC based model systems recapitulate dif-ferent abnormalities identified in DS brain103,148–153. From these studies only few have implemented deep RNA-seq to identify pathophysiological path-ways related with T21 and even then either on undifferentiated iPSCs149,153 or on GABAergic interneurons148. Moreover, high throughput omics methods al-low investigating temporal dynamic abnormalities along with neurogenesis that was not covered by those papers. To identify druggable pathways along neural differentiation and investigate the transcriptome and the proteome as-sociated with T21, we developed a robust iPSC derived neural model system for DS.

Initially, fibroblasts from two patients with full T21 (DS1 and DS2) and three age-matched healthy individuals (Ctrl1, Ctrl2 and Ctrl9) were repro-grammed into iPSCs. Subsequently, stable self-renewing NPC lines that ex-press Pax6 and NESTIN were derived for all individuals (Figure 9a). Follow-ing 30 days of non-directed differentiation, the NPCs transitioned into more differentiated neural stage (DiffNPCs) where the cells were positive for TUBB3 (marker for immature neurons) and GFAP (marker for astrocytes) (Figure 9b).

The acquired culture is suggested to be suitable to model MBHB129, that are less investigated brain regions compared to forebrain in DS85. Hypocellu-larity caused by impaired neurogenesis and cell proliferation, is evident in all three regions84,107. This was also apparent in our trisomic NPCs that showed reduced proliferation rate compared to euploid NPCs.

37

Figure 9. Immunofluorescent staining of neural cells with full T21 (DS) and euploid controls (Ctrl) at two differentiation stages. (a) Representative images showed that both DS and Ctrl neural progenitor cells (NPCs) were uniformly immunopositive for NESTIN and Pax6. (b) Following 30 days of non-directed differentiation the cul-tures contained both TUBB3 and GFAP positive cells.

To investigate differences in global expression pattern between groups, we subjected euploid and trisomic samples at NPC and DiffNPC differentiation stages to transcriptome and proteome analysis. We first showed that transcrip-tomes of neural cultures correspond to early- and mid-gestational age, sug-gesting the iPSC-derived neural model to be relevant for early DS neurogen-esis. In addition, transcriptome analysis revealed genome wide alterations in numerous genes in trisomic neural cells. In NPC stage, we identified 922 dif-ferentially expressed genes (DEGs) (p<0.05) from which 79.6% were upreg-ulated compared to controls. In DiffNPC stage, we identified 879 DEGs (p<0.05) from which 72% were upregulated. Proteomics detected mainly pro-teins which expression in mRNA level was relatively high. In NPCs, 132 dif-ferentially expressed proteins (DEPs) were identified in T21 lines (p<0.05/724), and in DiffNPCs, 51 DEPs were identified (p<0.05/439). The findings indicate that T21 leads to extensive molecular abnormalities.

The analysis of RNA-seq data revealed functional clusters with strong tem-poral dynamics of HSA21 genes along neural differentiation and the expres-sion levels strongly deviated from an expected 3:2 dosage. Interestingly, RUNX1, a HSA21 gene recently reported to be responsible for certain neural phenotypes in DS154, had the strongest and most consistent upregulation at both differentiation stages. In addition, transcriptomics analysis identified down-regulation of 42% of mitochondrial genes.

Analysis of the proteome confirmed alterations in expression of proteins involved in energy metabolism, membrane functions and cytoskeleton. Inter-estingly, our trisomic NPCs also exhibited reduced cell proliferation while ROS were not altered. These results predict perturbed energy metabolism in studied neural cells and suggests an adaptation of the cells to energy shortage. These findings are consistent with previous studies showing that individuals with DS have dysregulated mitochondrial expression and energy shortage155.

To identify dysregulated pathways and altered molecular functions in T21 cells, we input DEGs and DEPs to Enrichr156. This analysis revealed that focal

38

adhesion, extracellular matrix function and TGF-beta signalling pathways are dysregulated in trisomic NPCs and DiffNPCs. Similar results were obtained with gene ontology (GO) enrichment analysis. Several aberrant pathways were associated either with NPCs or DiffNPCs. For instance, in trisomic NPCs, dysregulation of pathways for signalling that regulates pluripotency and synapse formation were observed, supporting perturbed neurogenesis. In trisomic DiffNPCs, perturbed glycolysis/gluconeogenesis, axon guidance and long-term depression of cerebellum pathways emerged.

We also investigated hub proteins highly connected with other proteins within a functional module or a pathway that increases the probability for a phenotypic association. We identified 23 hub proteins within significantly al-tered networks in trisomic NPCs and DiffNPCs and five of them were mark-edly upregulated (ITGB1, VCL, CAV1, IL7R and APP). All identified genes, except APP, encode proteins with interconnected functions in focal adhesion and ECM-receptor signalling. These are critical processes for cell cycle regu-lation, cell morphology, migration, axon guidance – processes disturbed in DS brains and DS models. Moreover, the upregulation of the hub gene APP also agrees with studies showing Alzheimer-like neurodegenerative processes in early DS brains157 and iPSCs with T21150,158.

In conclusion, we established a model of DS neurogenesis and showed that T21 leads to complex and genome-wide expression changes, accompanied by prolonged cell cycle and perturbed energy metabolism. Our finding identified transcriptional and protein imbalances which lead to abnormal neurogenesis in DS. Importantly, we identified several new candidate biomarkers and path-ways.

39

Paper II. DNA methylation changes in Down syndrome derived neural iPSCs uncover co-dysregulation of ZNF and HOX3 families of transcription factors The neural model system established in Paper I was implemented to clarify the role of differential methylation on T21 neurogenesis. Using the Illumina 450K array, DNA methylation of known CpG regions and promoters across the genome was examined at two stages of neural differentiation (NPCs and DiffNPCs). With this method, 500 stable and differentially methylated probes (DMPs) in T21 neural cells were identified.

The first notion was that the trisomic neural cells exhibited the tendency towards hypomethylation of DMPs (p<0.05). This slightly differs from the studies on DS fetal and adult cortex159,160. These results can be due to a com-position of cell types in our model of T21 neurogenesis that contains a con-siderable proportion of cells representing MBHB precursors129. Previously, large-scale genome methylation study revealed that the DNA methylation sig-natures markedly differ between cerebral cortex (forebrain structure) and cer-ebellum (hindbrain structure)161. More precisely, this study observed the cer-ebellum being more hypomethylated while cerebral cortex rather hypermeth-ylated, supporting our findings. So far, no methylation study has been per-formed in DS MBHB structures. Besides that, it has been suggested that three shifts in the methylation pattern occur during neuronal development and the phase of hypomethylation takes over when progenitor cells produce young neurons162,163. This is also the developmental stage we investigated in our study.

We also observed an uneven chromosomal distribution of DMPs, similarly to studies across variety of cells and tissues with T21159,160,164,165. Specifically, we identified significant hypomethylation of chromosomes 2, 8, 19, 21 and 22 whereas chromosomes 6 were enriched and 17 depleted with hypermethylated probes (p<0.05) (Figure 10). Various studies have shown hypomethylation of chromosome 21159,160, supporting our findings. The reason for uneven chro-mosomal distribution of DMPs is unclear. Nevertheless, it can be hypothe-sized that chromosomal organisation in 3D is affected by the genomic imbal-ance of T21.

40

Figure 10. Genomic distribution of hypomethylated (282) and hypermethylated (218) DMPs. We examined enriched hypomethylation (blue dots) of chromosomes 2, 8, 19, 21 and 22 (chromosomes highlighted in blue squares) whereas chromosomes 6 were enriched and 17 depleted with hypermethylated probes (green dots; chromosomes highlighted in green squares). Visualisation of location of DMPs was done in Pheno-Gram166.

To investigate the differential methylation in genomic context, we looked the distribution of DMPs in CG rich regions known as CpG islands (CGIs) and their flanking regions – shores and shelves (within 2 kb or 4kb of a CGI, re-spectively)167. These regions often co-localize with gene promoters and are more likely associated with transcriptional regulation168–171. Besides that, gene density is positively correlated with CG content, suggesting the importance of these regions. Interestingly, out of the 500 DMPs, 323 were either in CGIs or in proximity (within 4 kb). Out of 323 DMPs, 234 were annotated to 151 genes (either within 1.5 kb or 200 bp from a transcriptional start site, in the 5’UTR, 1st exon, gene body or 3’UTR, of a gene172). Among the identified genes, 37 were belonged to a DNA binding functional category (GO:0003677; Enrich-ment 2.23; FDR 8.22E-03). These 37 genes are factors of importance for tran-scriptional regulation and chromatin structure. These genes encode multiple members of the HOX proteins, HIST1 proteins, ZNF transcription factors, other transcription factors and other DNA binding proteins. We revisited the transcription data published in Paper I and showed that 12 out of 37 genes were differentially expressed in T21 neural cells (p<0.05). The identified genes and gene families suggest important novel links between the genomic

41

imbalance caused by T21 and the global transcriptional dysregulation in DS neurogenesis.

42

Paper III. Transcriptomes of Dravet syndrome iPSC derived GABAergic cells reveal dysregulated pathways for chromatin remodeling and neurodevelopment Even though previous studies have shed more light to DRS pathogenesis, the biological processes and molecular mechanisms are still not fully understood. To bring more clarity, we established an iPSCs from three DRS patients with different SCN1A variants and three healthy controls. As the mouse model113 and iPSC-based studies173,174 suggest that GABAergic neurons mediate the ab-normalities in DRS, we differentiated both patient and control iPSCs into NPCs (sampling day 19) and GABAergic inter-neuronal cells (sampling day 65). At day 65 of differentiation, we showed with immunofluorescent staining that 80% of the cells in the culture were GABA-positive.

We also assessed the electrophysiological properties of the neuronal cells with whole-cell patch-clamp recordings. We first evoked action potentials (APs) in response to step current injection and observed that the maximal fre-quency of the APs was decreased in DRS inter-neuronal cells compared to controls. In accordance with the shift in the maximal AP frequency, we also observed shift of the sodium current activation to more depolarized potentials in DRS inter-neuronal cells. These finding recapitulate electrophysiological hallmarks of the haploinsufficiency of Nav1.1 in DRS interneurons.

It is evidential that in DRS, seizures are often triggered by fever or other external stimuli175,176, indicating increased sensitivity to cellular stress in DRS neuronal cells. Besides that, it has been shown that oxidative stress contributes to the pathophysiology of epilepsy177. Therefore, we set out to investigate the intracellular levels of reactive oxygen species (ROS) in our cultures. Interest-ingly, cells displayed comparable levels of ROS until hydrogen peroxide in-duction markedly increased levels of ROS in DRS GABA lines. This suggests that inter-neuronal cells with mutated SCN1A gene exhibit impaired response to induced stress.

We also subjected the samples to transcriptomics analysis to investigate the capacity of this model system to capture neurodevelopment of DRS. Evalua-tion of the global gene expression pattern suggested that the cellular compo-sition of out model system is similar to that of prefrontal cortex, fetal brain, dentate gyrus and superficial dorsofrontal area, representing the brain regions of interest. Moreover, temporal changes in global transcriptomes between NPCs and GABAergic inter-neuronal cells, as expected, highlighted upregu-lation of genes (n=1210) in pathways for neuronal cell maturation and down-regulation of genes (n=780) in pathways negatively associated with differen-tiation. The time course analysis of the transcriptome data between DRS and control lines identified numerous transcripts that changed expression in a con-dition-specific manner over time in DRS neural lines. These DEGs were either markedly downregulated between day 19 and day 65 in DRS lines, whereas

43

in control lines their expression increased or retained, or opposite. Pathway analysis revealed that these DEGs belonged to a Histone Modification and Cell Cycle pathways.

Finally, we sought to identify genes that were dysregulated in DRS inter-neuronal cells at both differentiation stages. This resulted in ten genes that included one antisense transcript, five long non-coding RNA genes and four protein coding genes (DPP6, GSTM1, LYNX1 and ZXDA). Three of protein-coding these genes have an important neuronal function. Dipeptidyl peptidase VI encoded by DPP6 is essential for excitability and synaptic long-term po-tentiation178. Glutathione S-transferase encoded by GSTM1 is an epilepsy as-sociated gene that plays important role in oxidative stress control. It is also considered as a genetic factor influencing the response to anticonvulsant ther-apy.179 Finally, Ly6/neurotoxin encoded by LYNX1 is a gene regulating plas-ticity in the mature cortex180. These results prove the usefulness of our iPSC-based model system to capture Nav1.1 haploinsufficiency caused dysregula-tion relevant for DRS.

In conclusion, transcriptome analysis on GABAergic neuronal cultures with various SCN1A variants showed dysregulated genes for chromatin struc-ture, mitotic progression, neural function, exhibited shifted Na-current activa-tion and perturbed response to induced oxidative stress. This study, besides revealing transcriptional dysfunctions in DRS GABAergic cultures, provides a framework that overlaps with that of neurodevelopmental disorders and ep-ilepsies.

44

Paper IV. Ataxia in patients with bi-allelic NFASC mutations and absence of full-length NF186 Ataxia itself stands for a physical finding defined as lack or loss of voluntary muscle movements. It can be the patient’s main complaint or one of many symptoms116,117, thus the underlying etiology needs to be investigated. In this study, siblings with progressive ataxia and muscular weakness were exam-ined. As the main findings also included dysarthria, poor balance and vertigo, dysfunction of cerebellum was suggested. Exome sequencing identified a novel homozygous variant (c.3020-1G>A) at splice site in the NFASC gene as the disease causal mutation.

To investigate the underlying cause of symptoms in the patients, we first assessed the potential effect of the splicing variant at RNA level. We showed that the homozygous variant c.3020-1G>A leads to complete loss of NFASC full-length isoform NF186. Interestingly, this isoform NF186 is highly ex-pressed in the CNS121, and in animal models, loss of NF186 leads to dysfunc-tion of cerebellum and motor coordination181. Similarly to previous stud-ies181,182, we observed significantly reduced expression of ankyrin G in the absence of NF186. We therefore decided to investigate the effects of this mu-tation for human brain development further in an iPSC based neural model system.

To estimate the capacity of patient neural cells to form functional network during development, we performed a neurite outgrowth assay. Interestingly, the patient derived neural cells sprouted significantly fewer and shorter neu-rites and appeared to be delayed in their neuronal differentiation compared to neurons from control individual. A possible explanation is the dysregulation of genes that have a function in migration and neurite outgrowth. Previously, overexpression of axon guidance ligand SLIT3 was shown to cause shorter neurites183,184. Interestingly, overexpression of SLIT3 was also observed dur-ing neuronal differentiation of our patient cells compared to Ctrl cells.

To follow the differentiation of NPCs into more mature neural cells, we measured the expression of neuronal markers DCX and MAP2 and the elec-trophysiological activity of these cells. In patient cells, expression of both markers remained lower as compared to healthy control cells. This suggested that neurons lacking full-length NF186 display aberrant neurogenesis. This was also supported by the whole-cell patch-clamp recordings that showed that the patient derived neural iPSCs had altered electrophysiology. More specifi-cally, recorded inward sodium currents in response to depolarized membrane potentials showed that patient neuron Nav channels activated at about 10 mV more negative membrane potentials compare to the control neuron Nav chan-nels. This suggests that patient neurons need smaller membrane depolariza-tions to open, and thus, have lower opening threshold. In addition, the peak-

45

current amplitude for activation was lower in the patient derived neurons, in-dicating the presence of fewer Na+ channels in the patient neurons.

Within this study, we showed that neurons lacking full-length NF186 were delayed in their neuronal differentiation, sprouted fewer and shorter neurites and had altered electrophysiology. All these aberrant processes are likely det-rimental for the developing brain. This study illustrates the usefulness of hu-man iPSC neural model systems for the identification of key mechanisms be-hind ataxia.

46

Conclusions

We established cellular models of human neurogenesis to investigate dis-turbed mechanisms in three different neurodevelopmental disorders. Analysis of these models highlighted major disturbed molecular and cellular functions and shed more light on the neuropathogenesis of Down syndrome (DS), Dra-vet syndrome (DRS) and a novel ataxia syndrome. Furthermore, the findings presented in these papers suggest molecular targets for the development of future therapeutic strategies.

In Paper I, an iPSC-based neural model system for DS that recapitulates abnormalities identified in DS brain was established. It was shown that T21 leads to complex and genome-wide expression changes, accompanied by a prolonged cell cycle and a perturbed energy metabolism. Additionally, tran-scriptional and protein imbalances which may lead to abnormal neurogenesis were identified. Most importantly, several new candidate biomarkers and pathways were revealed within this paper. Moreover, this model system shows promise for that future understanding of mechanisms behind perturbed neuro-genesis at early stages in the DS fetal brain.

In Paper II, a trisomic neural model system was implemented to clarify the role of differential methylation on perturbed T21 neurogenesis. The iPSC-based model system for DS was enriched for differentially methylated genes and gene families (e.g. HOX, HIST1 and ZNF) of importance for transcrip-tional regulation and chromatin structure. These genes and gene families sug-gest important novel links between the genomic imbalance caused by T21 and the global transcriptional dysregulation in DS neurogenesis.

In Paper III, an iPSC-based model system for DRS that exhibits Nav1.1 haploinsufficiency, characteristic electrophysiological abnormalities and im-paired response to induced cellular stress was established. Transcriptome anal-ysis highlighted dysregulation of genes for chromatin components and its modifications as well as cell cycle regulators in DRS neural cells. Moreover, these cells exhibit co-dysregulation of genes essential for neuronal plasticity, excitability and connectivity.

In Paper IV, we identified a novel homozygous variant (c.3020-1G>A) in the neurofascin (NFASC) gene in siblings affected by progressive ataxia and muscular weakness. We showed that this variant leads to the complete loss of NFASC full-length isoform NF186 highly expressed in the CNS. Using a pa-tient derived iPSC neural model system, we showed that neurons lacking full-length NF186 exhibit altered electrophysiology and sprouted fewer and

47

shorter neurites. In addition, these neurons were delayed in their neuronal dif-ferentiation.

Overall, presented studies illustrate the usefulness of human iPSC neural model systems for the identification of key mechanisms behind neurodevel-opmental disorders.

48

Future perspectives

Most putative drug candidates face attrition during clinical trials due to safety issues or poor efficacy. This is often attributed to dependence on animal mod-els or immortalized and cancer cells for drug screening that poorly recapitulate human pathophysiology, thereby resulting in poor prediction of drug re-sponse185,186. Generally, these models have failed to serve as a platform for drug discovery186. To overcome the bottlenecks of the arduous drug discov-ery process, human derived iPSCs can be exploited. Drug screening using iPSCs provides unprecedented opportunities for bringing precision medicine to the clinic. Clinical trials utilizing iPSC model systems will help stratify pa-tients based on their response to the drugs. Moreover, molecular and cellular analyses of the iPSCs could help us identify specific biomarkers in the drug responders (corresponding to important patient subpopulations). This will fa-cilitate clinical trials as patients can be pre-selected based on their genotype for the subsequent phases38.

This thesis assessed the suitability of human iPSCs for modelling different neurodevelopmental disorders and within this thesis several potentially drug-gable candidate mechanisms were identified. For instance, in Paper I, we ob-served perturbation in energy metabolism in DS neurons apart from mitochon-drial copy number defects. Previous studies on iPSC derived neurons have been suitable to perform drug screening to target mtDNA disorders187. Thus, this suggest that the model system presented within this thesis should be fea-sible for similar screening. In Paper II, we uncovered co-dysregulation of various TFs, including ZNF and HOX3 families of TFs. As our understanding of transcription regulation has advanced and genetic engineering allows us to selectively modulate target genes, the development of transcription-based drugs188 for named TFs would be feasible. In Paper III, we showed that chro-matin remodeling factors could be promising drug targets for DRS. Besides that, one of the hallmarks of this disease is drug-resistant epilepsy. Recently, a zebrafish model system of DRS was employed to screen compound library molecules which helped shortlist drug compounds189. It would be interesting to employ our cellular model to screen these compounds. In Paper IV, we discovered impaired neuronal biophysical properties that could be tackled with different drug compounds. Besides that, we showed that the patient neu-ral cells exhibit reduced capacity to form functional network during develop-ment. Adopting the neurite outgrowth assay for drug screening owns a poten-tial to identify compounds that rescue neurite formation.

49

In addition to 2D cultures, organoids, a three-dimensional (3D) self-organ-izing cellular cluster derived from iPSCs, are an emerging model system for drug testing190. These structures recapitulate many features of brain develop-ment better than neuronal cells grown as a monolayer191. Developing organ-oids from iPSCs with T21 or mutation in SCN1A or NFASC would be instru-mental for future drug screening.

In summary, iPSCs have not just revolutionized how we understand dis-ease, but also present an untapped potential to ameliorate drug discovery and precision medicine.

50

Acknowledgement

Somewhere, something incredible is waiting to be known. – Carl Sagan

I want to express my sincere gratitude to all the people who have contributed to my PhD journey. All the patients and their families for their trust in our research.

My main supervisor, Professor Niklas Dahl who has created an amazing work environment and made this research possible. Thank you for investing your time and energy into mentoring me through my PhD, helping me to improve my strengths and weaknesses and inspiring me to apply for grants and schol-arships. It has not been an easy journey, but I have enjoyed these years. My co-supervisor, Jens Schuster, for teaching me everything about stem cells. For making sure that the lab runs smoothly, nitrogen tanks are full, and cells and kits are ready for experiments. Thank you for the scientific discus-sions and for taking care of me throughout these years. By the way, you also influenced me to become more ecological. My co-supervisor, Joakim Klar, our bioinformatic angel, without him so many figures would not be there. It took me some time to get the courage to bother him with all my requests, but now I am unstoppable. My co-supervisor, Anna Falk, for arranging retreats at KI, contributing to the papers, initiating collaborations and believing in my patch-clamp skills. To her lab members, especially Matti, Mastoureh, Mohsen, Harriet and Malin for being amazing researchers. Current and previous members of the Dahl Research Group who have made the research environment enjoyable and fun – Jan, Maria, Dorothea, Velin, Alberto, Lani, Ayda, Talia, Rein, Caro, Sanjay, Zafar and Zakaria. To Maria Sobol for trusting me to work with the Down syndrome project and for helping me to take the second project to another level. To Jan Hoeber for all the discussions that widened my perspectives and for helping me during stressful times. To Ambrin Fatima for always being little extra caring, for offering her expertise in CRISPR/Cas9 and for cooking amazing Pakistani food.

51

This thesis was carried out at the Department of Immunology, Genetics and Pathology (IGP). Throughout these years, I received enormous help from the IGP administration, especially Christina, Helene, Ulrica and Sara. To those carrying out their research at BMC corridor B11:4, big thank you Jan Dumanski, Marie Allen, Lars Feuk, Lars Forsberg, Claes Wadelius, Åsa Johansson, Marcel den Hoed and Marie-Louise Bondeson. Thank you for hiring amazing researchers to your research groups. Their presence has made the working environment extremely enjoyable and filled with experts. I am especially grateful to Hanna Davies and Biorad researchers who helped me set up the ddPCR experiments. I also want to thank everybody else in Upp-sala University whom I have met throughout these years. Thank you, Tanel Punga, for the crash course in cloning, and for offering your expertise in the field; especially for the discussions in our mother language. To Brynis Birnir and her incredible electrophysiology lab. Thank you Bryn-dis, Zhe, Sergy, Olga, Hayma, Atieh and Karin for warmly welcoming me to your lab. You gave me the most amazing ‘hobby’ in the world – patch-clamping. Learning this technique was only possible thanks to your instruc-tions, help and encouraging words. Every successful single cell recording took me closer to my goals! To the PhD students at the Department of Medical Cell Biology, especially Carmen, Carla, Emelie, Amanda and Kristel. Thank you, Carmen, for all the help and advice throughout these years. Thank you, Carla, for your effort to change the university language policy and enthusiastically keeping MDR alive. Thank you, Emelie, for taking over the MDR responsibilities when not so many would be willing to do so. Recent Christmas party was just a proof of your success. To the current and previous members of the IGP PhD council (Sanna, Jualia, Leonor, Svea, Marcus, Veronica, Anja, Bene, Ida, Tina, Emma, Sara, Linnea, Dijana, Phatu). Arranging events with you guys has been a pleasure. To the current and previous members of Ülo Langel research group, espe-cially Piret Arukuust and Ülo Langel for introducing me to amazing CPPs. To Liis Pärnpuu, Piia Jaksi and Anni Lember for all the years in Tartu and for being friends I can rely on. To Michael Williams for supervising me through the most awesome summer project with fruit flies. Also, to Linda, Lina, Daniel, Valerie and Henrik for all the laughs and good time in the lab, because science is fun!

52

To my Swedbank girls! Big thank you goes to Medhi Khadraoui for illustrating the cover of this the-sis. He made it perfectly! More of his work can be found https://brushtale.art/. To Carro, Malin, Mattias, Hampus, Oskar, Christoffer W, Simon, Cor-mac, Elias – thank you for the most awesome years in Flogsta dorm, waffle and pancake weekends, corridor crawls and movie nights. To all the people I met through the corridor people – Hampus, Anna, Hanna, Emma and An-ton. My Dota people (Matti, Rodrigo, Adam, Piers, Enea, Callum, Harry, Ben-jamin, Lukas, Niklas, Jesper, Dominic and Sebastian) for the good laughs and fun evenings. Thank you for letting me to do my thing and like you always remind me – teamwork makes the dream work. Thank you, Sebastian, for the time spent on teaching me this strategic game, it gave me a lot of friends. To my amazing Estonian girls in Uppsala – Kristel Parv, Madle Sirel, Liisi Blevins (Laaniste) and Kristi Krebs. I must thank you, Kristel, for too many things. I cannot list them all but thank you for inspiring me to join MDR, tak-ing care of Ellinor and Prinsen when needed, laughing at my jokes, forcing me to go out, taking me to hiking trips, etc. Thank you Madle for randomly coming over and spicing up our day, taking us to the farm and making us more active. I cannot forget the comparison of two of you as the hyenas from “The Lion King”. Thank you, Liisi, for fun Uppsala nights, scientific help and being inspirational scientist. Hugs to Kristi for being such a good friend! Still hop-ing you come back to Uppsala. Last years of my PhD were spiced up by our Halv åtta hos mig dinner parties, so big thank you to all the members for food and entertainment – Simon, Emma, Madle, Christoffer, Kristel, Cormac and that +1 who got to come. Besides that, big thank you to Christoffer for organizing BBQ evenings and dance battles! Hugs to Cormac for all the video calls with Kristel where I could join in. Hugs to Simon for being my fist Swedish best friend, that says all. Thank you, Sonny, for so many things, including introducing me to the proper midsummers and local festivals. Big thank you for the stress management offered by Emelie and David through Bilbo and Frig. They are amazing doggos. Without Frig the hike in Fulufjället would have been harder. Without Bilbo the hike would have been too easy for Kristel. To Yashraj and Kai for being the best course mates ever! #ebolacandy

53

Thank you, Olga Titova for all the lunch breaks which have been in a way like a stress management. You are not only an inspiring scientist, but also an amazing mother! To my Estonian friends – Ester, Kati, Milanna and Kadi. The time I have lived in Sweden (almost 6 years!) has made our relationship even stronger. The support I receive daily from you – I cannot put it into the words. I want to thank their families (Argo, Ahto, Aston, Mart, Artur, Mette) for letting them out when I come to Estonia. To Katrin and Ali for always treating us like family. Aitäh Katrin ja Ali, et te kohtlete meid kui oma lapsi. My most sincere gratitude goes to Sävstaholm Society! Your financial sup-port made these projects possible. Also, to Anna Maria Lundins, Rönnow and Letterstedt foundations for funding my scientific conferences. To my extended family – Aigi, Karmo-Gustav, Kai, Meljo, Karmen, Heimar, Uku, Priit and Tõnis. To my grandparents Endel, Ene, Viktor, Leida and Silva. Eriti suured kallistused vanaisale, kes on alati pidanud haridust oluliseks. Kallistused ka vanemale, kes inspireeris mind kiituskirjaga lõpetama. Soovin palju edu Aigile doktorantuuris ja Karmole ja Ukule koolis. These years would have been less warm without my little fluffy companions – Prinsen and Ellinor. If there is a disease that makes us love cats, then now I have it. Finally, and most importantly, this thesis is for my family – Eve, Raimo, Laura, Ketlin and Oliver. I cannot express how much I love you! Ma ei suuda sõnadesse panna kui palju ma teid armastan. Tänu teie toetusele on see kõik olnud võimalik.

54

References

1. O’Connor, C. & Miko, I. Developing the Chromosome Theory. Nat. Educ. 1, 44 (2008).

2. Watson, J. D. & Crick, F. H. C. Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid. Nature 171, 737–738 (1953).

3. Venter, J. C. et al. The sequence of the human genome. Science 291, 1304–1351 (2001).

4. Lander, E. S. et al. Initial sequencing and analysis of the human genome. Nature 409, 860–921 (2001).

5. International Human Genome Sequencing Consortium. Finishing the euchromatic sequence of the human genome. Nature 431, 931–945 (2004).

6. DNA Sequencing Costs: Data. National Human Genome Reseatch Institute (2019). Available at: https://www.genome.gov/about-genomics/fact-sheets/DNA-Sequencing-Costs-Data. (Accessed: 12th December 2019)

7. Abecasis, G. R. et al. A map of human genome variation from population-scale sequencing. Nature 467, 1061–1073 (2010).

8. Philippidis, A. 10 Countries in 100K Genome Club. Clinical OMICs (2018). Available at: https://www.clinicalomics.com/topics/biomarkers-topic/biobanking/10-countries-in-100k-genome-club/. (Accessed: 12th December 2019)

9. Human assembly and gene annotation: Statistics. Ensembl Release 98 (2019). Available at: https://www.ensembl.org/Homo_sapiens/Info/Annotation. (Accessed: 12th December 2019)

10. The ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012).

11. Abecasis, G. R. et al. An integrated map of genetic variation from 1,092 human genomes. Nature 491, 56–65 (2012).

12. Morley, A. A. & Turner, D. R. The contribution of exogenous and endogenous mutagens to in vivo mutations. Mutat. Res. 428, 11–15 (1999).

13. Fay, J. C., Wyckoff, G. J. & Wu, C. I. Positive and negative selection on the human genome. Genetics 158, 1227–1234 (2001).

14. Altshuler, D. M. et al. Integrating common and rare genetic variation in diverse human populations. Nature 467, 52–58 (2010).

15. Thomas, P. D. & Kejariwal, A. Coding single-nucleotide polymorphisms associated with complex vs. Mendelian disease: evolutionary evidence for differences in molecular effects. Proc. Natl. Acad. Sci. U. S. A. 101, 15398–15403 (2004).

16. Mullaney, J. M., Mills, R. E., Pittard, W. S. & Devine, S. E. Small insertions and deletions (INDELs) in human genomes. Hum. Mol. Genet. 19, R131-6 (2010).

17. Alkan, C., Coe, B. P. & Eichler, E. E. Genome structural variation discovery and genotyping. Nat. Rev. Genet. 12, 363–376 (2011).

55

18. Feuk, L., Carson, A. R. & Scherer, S. W. Structural variation in the human genome. Nat. Rev. Genet. 7, 85–97 (2006).

19. Bae, B. Il, Jayaraman, D. & Walsh, C. A. Genetic changes shaping the human brain. Dev. Cell 32, 423–434 (2015).

20. Chial, H. Mendelian genetics: Patterns of inheritance and single-gene disorders. Nat. Educ. 1, 63 (2008).

21. Mitchell, K. J. What is complex about complex disorders? Genome Biol. 13, 237 (2012).

22. Sanger, F., Nicklen, S. & Coulson, A. R. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. U. S. A. 74, 5463–5467 (1977).

23. Liu, L. et al. Comparison of next-generation sequencing systems. J. Biomed. Biotechnol. 2012, 251364 (2012).

24. van Dijk, E. L., Jaszczyszyn, Y., Naquin, D. & Thermes, C. The Third Revolution in Sequencing Technology. Trends Genet. 34, 666–681 (2018).

25. Pandey, U. B. & Nichols, C. D. Human disease models in Drosophila melanogaster and the role of the fly in therapeutic drug discovery. Pharmacol. Rev. 63, 411–436 (2011).

26. Sommer, R. J. & Streit, A. Comparative genetics and genomics of nematodes: genome structure, development, and lifestyle. Annu. Rev. Genet. 45, 1–20 (2011).

27. Perlman, R. L. Mouse models of human disease: An evolutionary perspective. Evol. Med. public Heal. 2016, 170–176 (2016).

28. Rossant, J. & McKerlie, C. Mouse-based phenogenomics for modelling human disease. Trends Mol. Med. 7, 502–507 (2001).

29. Howe, K. et al. The zebrafish reference genome sequence and its relationship to the human genome. Nature 496, 498–503 (2013).

30. Hedges, S. B., Dudley, J. & Kumar, S. TimeTree: a public knowledge-base of divergence times among organisms. Bioinformatics 22, 2971–2972 (2006).

31. Vanfleteren, J. R. et al. Molecular genealogy of some nematode taxa as based on cytochrome c and globin amino acid sequences. Mol. Phylogenet. Evol. 3, 92–101 (1994).

32. Kumar, S. & Hedges, S. B. A molecular timescale for vertebrate evolution. Nature 392, 917–920 (1998).

33. Waterston, R. H. et al. Initial sequencing and comparative analysis of the mouse genome. Nature 420, 520–562 (2002).

34. Bracken, M. B. Why animal studies are often poor predictors of human reactions to exposure. J. R. Soc. Med. 102, 120–122 (2009).

35. Domcke, S., Sinha, R., Levine, D. A., Sander, C. & Schultz, N. Evaluating cell lines as tumour models by comparison of genomic profiles. Nat. Commun. 4, 2126 (2013).

36. Niu, N. & Wang, L. In vitro human cell line models to predict clinical response to anticancer drugs. Pharmacogenomics 16, 273–285 (2015).

37. Nestor, C. E. et al. Rapid reprogramming of epigenetic and transcriptional profiles in mammalian culture systems. Genome Biol. 16, 11 (2015).

38. Inoue, H., Nagata, N., Kurokawa, H. & Yamanaka, S. iPS cells: a game changer for future medicine. EMBO J. 33, 409–417 (2014).

39. Sterneckert, J. L., Reinhardt, P. & Scholer, H. R. Investigating human disease using stem cell models. Nat. Rev. Genet. 15, 625–639 (2014).

40. Zhu, Z. & Huangfu, D. Human pluripotent stem cells: an emerging model in developmental biology. Development 140, 705–717 (2013).

41. Takahashi, K. et al. Induction of Pluripotent Stem Cells from Adult Human Fibroblasts by Defined Factors. Cell 131, 861–872 (2007).

56

42. Thomson, J. A. et al. Embryonic Stem Cell Lines Derived from Human Blastocysts. Science 282, 1145–1147 (1998).

43. Ardhanareeswaran, K., Mariani, J., Coppola, G., Abyzov, A. & Vaccarino, F. M. Human induced pluripotent stem cells for modelling neurodevelopmental disorders. Nat. Rev. Neurol. 13, 265–278 (2017).

44. David, L. & Polo, J. M. Phases of reprogramming. Stem Cell Res. 12, 754–761 (2014).

45. Porteus, M. H. A New Class of Medicines through DNA Editing. N. Engl. J. Med. 380, 947–959 (2019).

46. Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823–826 (2013).

47. Bassett, A. R. Editing the genome of hiPSC with CRISPR/Cas9: disease models. Mamm. Genome 28, 348–364 (2017).

48. Zuo, E. et al. CRISPR/Cas9-mediated targeted chromosome elimination. Genome Biol. 18, 224 (2017).

49. Hasin, Y., Seldin, M. & Lusis, A. Multi-omics approaches to disease. Genome Biol. 18, 83 (2017).

50. Higdon, R. et al. The Promise of Multi-Omics and Clinical Data Integration to Identify and Target Personalized Healthcare Approaches in Autism Spectrum Disorders. Omi. A J. Integr. Biol. 19, 197–208 (2015).

51. Suzuki, M. M. & Bird, A. DNA methylation landscapes: provocative insights from epigenomics. Nat. Rev. Genet. 9, 465–476 (2008).

52. Greenberg, M. V. C. & Bourc’his, D. The diverse roles of DNA methylation in mammalian development and disease. Nat. Rev. Mol. Cell Biol. 20, 590–607 (2019).

53. Wang, Z., Gerstein, M. & Snyder, M. RNA-Seq: a revolutionary tool for transcriptomics. Nat. Rev. Genet. 10, 57–63 (2009).

54. Horgan, R. P. & Kenny, L. C. ‘Omic’ technologies: genomics, transcriptomics, proteomics and metabolomics. Obstet. Gynaecol. 13, 189–195 (2011).

55. Dudley, E., Hassler, F. & Thome, J. Profiling for novel proteomics biomarkers in neurodevelopmental disorders. Expert Rev. Proteomics 8, 127–136 (2011).

56. Purves, D. et al. Studying the Nervous System. in Neuroscience. 4th edition 1–21 (2008).

57. von Bartheld, C. S., Bahney, J. & Herculano-Houzel, S. The search for true numbers of neurons and glial cells in the human brain: A review of 150 years of cell counting. J. Comp. Neurol. 524, 3865–3895 (2016).

58. Liu, A. & Niswander, L. A. Bone morphogenetic protein signalling and vertebrate nervous system development. Nat. Rev. Neurosci. 6, 945–954 (2005).

59. Schoenwolf, G., Bleyl, S., Brauer, P. & Francis-West, P. Third Week: Becoming Trilaminar and Establising Body Axes. in Larsen’s Human Embryology. 4th edition 69–100 (2009).

60. Temple, S. Division and differentiation of isolated CNS blast cells in microculture. Nature 340, 471–473 (1989).

61. Schoenwolf, G., Bleyl, S., Brauer, P. & Francis-West, P. Development of the Central Nervous System. in Larsen’s Human Embryology. 4th edition 247–295 (2009).

62. Okano, H. & Temple, S. Cell types to order: temporal specification of CNS stem cells. Curr. Opin. Neurobiol. 19, 112–119 (2009).

63. Ackerman S. Major Structures and Functions of the Brain. in Discovering the Brain (1992).

57

64. Silbereis, J. C., Pochareddy, S., Zhu, Y., Li, M. & Sestan, N. The Cellular and Molecular Landscapes of the Developing Human Central Nervous System. Neuron 89, 268 (2016).

65. OpenStax. Brain Vesicle. Wikimedia Commons (2016). Available at: https://upload.wikimedia.org/wikipedia/commons/2/2c/1302_Brain_Vesicle_DevN.jpg. (Accessed: 12th December 2019)

66. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM-5). (2013).

67. World Health Organization. The ICD-10 classification of mental and behavioral disorders. Clinical descriptions and diagnostic guidlines. (1992).

68. De Felice, A., Ricceri, L., Venerosi, A., Chiarotti, F. & Calamandrei, G. Multifactorial Origin of Neurodevelopmental Disorders: Approaches to Understanding Complex Etiologies. Toxics 3, 89–129 (2015).

69. Committee to Evaluate the Supplemental Security Income Disability Program for Children with Mental Disorders et al. Clinical Characteristics of Intellectual Disabilities. in Mental Disorders and Disabilities Among Low-Income Children (National Academies Press, 2015).

70. Parker, S. E. et al. Updated National Birth Prevalence estimates for selected birth defects in the United States, 2004-2006. Birth Defects Res. A. Clin. Mol. Teratol. 88, 1008–1016 (2010).

71. Coffee, B. et al. Incidence of fragile X syndrome by newborn screening for methylated FMR1 DNA. Am. J. Hum. Genet. 85, 503–514 (2009).

72. Kang, C. & Drayna, D. Genetics of speech and language disorders. Annu. Rev. Genomics Hum. Genet. 12, 145–164 (2011).

73. Miles, J. H. Autism spectrum disorders-a genetics review. Genet. Med. 13, 278–294 (2011).

74. Szatmari, P. et al. Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nat. Genet. 39, 319–328 (2007).

75. Pinto, D. et al. Convergence of genes and cellular pathways dysregulated in autism spectrum disorders. Am. J. Hum. Genet. 94, 677–694 (2014).

76. Schaefer, G. B. & Mendelsohn, N. J. Clinical genetics evaluation in identifying the etiology of autism spectrum disorders: 2013 guideline revisions. Genet. Med. 15, 399–407 (2013).

77. Sharma, A. & Couture, J. A review of the pathophysiology, etiology, and treatment of attention-deficit hyperactivity disorder (ADHD). Ann. Pharmacother. 48, 209–225 (2014).

78. Mascheretti, S. et al. Neurogenetics of developmental dyslexia: from genes to behavior through brain neuroimaging and cognitive and sensorial mechanisms. Transl. Psychiatry 7, e987 (2017).

79. McDowell, M. Specific learning disability. J. Paediatr. Child Health 54, 1077–1083 (2018).

80. Warner, T. Movement Disorders. in Practical Guide to Neurogenetics 102–136 (2009).

81. Cordeiro, D. et al. Genetic landscape of pediatric movement disorders and management implications. Neurol. Genet. 4, e265 (2018).

82. Stagni, F., Giacomini, A., Emili, M., Guidi, S. & Bartesaghi, R. Neurogenesis impairment: An early developmental defect in Down syndrome. Free Radic. Biol. Med. 114, 15–32 (2017).

83. Raz, N. et al. Selective neuroanatomic abnormalities in Down’s syndrome and their cognitive correlates: evidence from MRI morphometry. Neurology 45, 356–366 (1995).

58

84. Guidi, S. et al. Neurogenesis impairment and increased cell death reduce total neuron number in the hippocampal region of fetuses with Down syndrome. Brain Pathol. 18, 180–197 (2008).

85. Haydar, T. F. & Reeves, R. H. Trisomy 21 and early brain development. Trends Neurosci. 35, 81–91 (2012).

86. Antonarakis, S. E., Lyle, R., Dermitzakis, E. T., Reymond, A. & Deutsch, S. Chromosome 21 and down syndrome: from genomics to pathophysiology. Nat. Rev. Genet. 5, 725–738 (2004).

87. Papavassiliou, P., Charalsawadi, C., Rafferty, K. & Jackson-Cook, C. Mosaicism for trisomy 21: a review. Am. J. Med. Genet. A 167A, 26–39 (2015).

88. Chromosome 21: Chromosome summary - Homo sapiens. Ensembl Release 98 (2019).

89. Mao, R., Zielke, C. L., Zielke, H. R. & Pevsner, J. Global up-regulation of chromosome 21 gene expression in the developing Down syndrome brain. Genomics 81, 457–467 (2003).

90. Antonarakis, S. E. Chromosome 21: from sequence to applications. Curr. Opin. Genet. Dev. 11, 241–246 (2001).

91. Reeves, R. H. et al. A mouse model for Down syndrome exhibits learning and behaviour deficits. Nat. Genet. 11, 177–184 (1995).

92. Sago, H. et al. Ts1Cje, a partial trisomy 16 mouse model for Down syndrome, exhibits learning and behavioral abnormalities. Proc. Natl. Acad. Sci. U. S. A. 95, 6256–6261 (1998).

93. Davisson, M. T. et al. Segmental trisomy as a mouse model for Down syndrome. Prog. Clin. Biol. Res. 384, 117–133 (1993).

94. Golden, J. A. & Hyman, B. T. Development of the superior temporal neocortex is anomalous in trisomy 21. J. Neuropathol. Exp. Neurol. 53, 513–520 (1994).

95. Chakrabarti, L., Galdzicki, Z. & Haydar, T. F. Defects in embryonic neurogenesis and initial synapse formation in the forebrain of the Ts65Dn mouse model of Down syndrome. J. Neurosci. 27, 11483–11495 (2007).

96. Larsen, K. B. et al. Reduced cell number in the neocortical part of the human fetal brain in Down syndrome. Ann. Anat. 190, 421–427 (2008).

97. Petit, T. L., LeBoutillier, J. C., Alfano, D. P. & Becker, L. E. Synaptic development in the human fetus: a morphometric analysis of normal and Down’s syndrome neocortex. Exp. Neurol. 83, 13–23 (1984).

98. Weitzdoerfer, R., Dierssen, M., Fountoulakis, M. & Lubec, G. Fetal life in Down syndrome starts with normal neuronal density but impaired dendritic spines and synaptosomal structure. J. Neural Transm. Suppl. 59–70 (2001).

99. Becker, L. E., Armstrong, D. L. & Chan, F. Dendritic atrophy in children with Down’s syndrome. Ann. Neurol. 20, 520–526 (1986).

100. Benavides-Piccione, R. et al. On dendrites in Down syndrome and DS murine models: a spiny way to learn. Prog. Neurobiol. 74, 111–126 (2004).

101. Takashima, S., Ieshima, A., Nakamura, H. & Becker, L. E. Dendrites, dementia and the Down syndrome. Brain Dev. 11, 131–133 (1989).

102. Olmos-Serrano, J. L. et al. Down Syndrome Developmental Brain Transcriptome Reveals Defective Oligodendrocyte Differentiation and Myelination. Neuron 89, 1208–1222 (2016).

103. Chen, C. et al. Role of astroglia in down’s syndrome revealed by patient-derived human-induced pluripotent stem cells. Nat. Commun. 5, 1–18 (2014).

104. Fujii, Y. et al. A small pons as a characteristic finding in Down syndrome: A quantitative MRI study. Brain Dev. 39, 298–305 (2017).

105. Aylward, E. H. et al. Cerebellar volume in adults with Down syndrome. Arch. Neurol. 54, 209–212 (1997).

59

106. Baxter, L. L., Moran, T. H., Richtsmeier, J. T., Troncoso, J. & Reeves, R. H. Discovery and genetic localization of Down syndrome cerebellar phenotypes using the Ts65Dn mouse. Hum. Mol. Genet. 9, 195–202 (2000).

107. Guidi, S., Ciani, E., Bonasoni, P., Santini, D. & Bartesaghi, R. Widespread proliferation impairment and hypocellularity in the cerebellum of fetuses with down syndrome. Brain Pathol. 21, 361–373 (2011).

108. Chemaly, N. et al. The encephalopathy in dravet syndrome is not a direct consequence of epilepsy. Epilepsia 55, 213 (2014).

109. Akiyama, M., Kobayashi, K., Yoshinaga, H. & Ohtsuka, Y. A long-term follow-up study of Dravet syndrome up to adulthood. Epilepsia 51, 1043–1052 (2010).

110. Depienne, C. et al. Spectrum of SCN1A gene mutations associated with Dravet syndrome: analysis of 333 patients. J. Med. Genet. 46, 183–191 (2009).

111. Lorincz, A. & Nusser, Z. Molecular Identity of Dendritic Voltage-Gated Sodium Channels. Science 328, 906–910 (2010).

112. Griffin, A. et al. Preclinical Animal Models for Dravet Syndrome: Seizure Phenotypes, Comorbidities and Drug Screening. Front. Pharmacol. 9, 573 (2018).

113. Yu, F. H. et al. Reduced sodium current in GABAergic interneurons in a mouse model of severe myoclonic epilepsy in infancy. Nat. Neurosci. 9, 1142–9 (2006).

114. Salgueiro-Pereira, A. R. et al. A two-hit story: Seizures and genetic mutation interaction sets phenotype severity in SCN1A epilepsies. Neurobiol. Dis. 125, 31–44 (2019).

115. Musselman, K. E. et al. Prevalence of ataxia in children: a systematic review. Neurology 82, 80–89 (2014).

116. Brunberg, J. A. Ataxia. AJNR. Am. J. Neuroradiol. 29, 1420–1422 (2008). 117. Ashizawa, T. & Xia, G. Ataxia. Continuum (Minneap. Minn). 22, 1208–1226

(2016). 118. Konczak, J. & Timmann, D. The effect of damage to the cerebellum on

sensorimotor and cognitive function in children and adolescents. Neurosci. Biobehav. Rev. 31, 1101–1113 (2007).

119. Rathjen, F. G., Wolff, J. M., Chang, S., Bonhoeffer, F. & Raper, J. A. Neurofascin: a novel chick cell-surface glycoprotein involved in neurite-neurite interactions. Cell 51, 841–849 (1987).

120. Hassel, B., Rathjen, F. G. & Volkmer, H. Organization of the neurofascin gene and analysis of developmentally regulated alternative splicing. J. Biol. Chem. 272, 28742–28749 (1997).

121. Nelson, A. D. & Jenkins, P. M. Axonal Membranes and Their Domains: Assembly and Function of the Axon Initial Segment and Node of Ranvier. Front. Cell. Neurosci. 11, 136 (2017).

122. Ango, F. et al. Ankyrin-based subcellular gradient of neurofascin, an immunoglobulin family protein, directs GABAergic innervation at purkinje axon initial segment. Cell 119, 257–272 (2004).

123. Kriebel, M., Wuchter, J., Trinks, S. & Volkmer, H. Neurofascin: a switch between neuronal plasticity and stability. Int. J. Biochem. Cell Biol. 44, 694–697 (2012).

124. Desmazieres, A. et al. Differential stability of PNS and CNS nodal complexes when neuronal neurofascin is lost. J. Neurosci. 34, 5083–5088 (2014).

125. Omole, A. E. & Fakoya, A. O. J. Ten years of progress and promise of induced pluripotent stem cells: historical origins, characteristics, mechanisms, limitations, and potential applications. PeerJ 6, e4370 (2018).

60

126. Chhabra, A. Derivation of Human Induced Pluripotent Stem Cell (iPSC) Lines and Mechanism of Pluripotency: Historical Perspective and Recent Advances. Stem Cell Rev. 13, 757–773 (2017).

127. Maherali, N. & Hochedlinger, K. Guidelines and Techniques for the Generation of Induced Pluripotent Stem Cells. Cell Stem Cell 3, 595–605 (2008).

128. Soldner, F. & Jaenisch, R. iPSC disease modeling. Science 338, 1155–1156 (2012).

129. Falk, A. et al. Capture of neuroepithelial-like stem cells from pluripotent stem cells provides a versatile system for in vitro production of human neurons. PLoS One 7, 1–13 (2012).

130. Sobol, M. et al. Methods of Reprogramming to Induced Pluripotent Stem Cell Associated with Chromosomal Integrity and Delineation of a Chromosome 5q Candidate Region for Growth Advantage. Stem Cells Dev. 24, 2032–2040 (2015).

131. Schuster, J. et al. Transcriptome Profiling Reveals Degree of Variability in Induced Pluripotent Stem Cell Lines: Impact for Human Disease Modeling. Cell. Reprogram. 17, 327–337 (2015).

132. Tambe, Y. Induction of iPS cells. Wikimedia Commons (2007). Available at: https://upload.wikimedia.org/wikipedia/commons/0/09/Induction_of_iPS_cells.svg. (Accessed: 12th December 2019)

133. Mertens, J., Marchetto, M. C., Bardy, C. & Gage, F. H. Evaluating cell reprogramming, differentiation and conversion technologies in neuroscience. Nat. Rev. Neurosci. 17, 424–437 (2016).

134. Koch, P., Opitz, T., Steinbeck, J. A., Ladewig, J. & Brüstle, O. A rosette-type, self-renewing human ES cell-derived neural stem cell with potential for in vitro instruction and synaptic integration. Proc. Natl. Acad. Sci. U. S. A. 106, 3225–3230 (2009).

135. Cohen, D. E. & Melton, D. Turning straw into gold: Directing cell fate for regenerative medicine. Nat. Rev. Genet. 12, 243–252 (2011).

136. Maroof, A. M. et al. Directed Differentiation and Functional Maturation of Cortical Interneurons from Human Embryonic Stem Cells. Cell Stem Cell 12, 559–572 (2013).

137. Cao, S. Y. et al. Enhanced derivation of human pluripotent stem cell-derived cortical glutamatergic neurons by a small molecule. Sci. Rep. 7, 1–11 (2017).

138. Tailor, J. et al. Stem Cells Expanded from the Human Embryonic Hindbrain Stably Retain Regional Specification and High Neurogenic Potency. J. Neurosci. 33, 12407–12422 (2013).

139. Amit, M. et al. Clonally derived human embryonic stem cell lines maintain pluripotency and proliferative potential for prolonged periods of culture. Dev. Biol. 227, 271–278 (2000).

140. Trounson, A. & McDonald, C. Stem Cell Therapies in Clinical Trials: Progress and Challenges. Cell Stem Cell 17, 11–22 (2015).

141. Kornreich, B. G. The patch clamp technique: principles and technical considerations. J. Vet. Cardiol. 9, 25–37 (2007).

142. Neher, E. & Sakmann, B. Single-channel currents recorded from membrane of denervated frog muscle fibres. Nature 260, 799–802 (1976).

143. Purves, D. et al. Electrical Signals of Nerve Cells. in Neuroscience. 5th edition 25–40 (Sinauer Associates, Inc, 2012).

144. Cahalan, M. & Neher, E. Patch clamp techniques: an overview. Methods Enzymol. 207, 3–14 (1992).

61

145. Veitinger, S. The Patch-Clamp Technique. Leica Microsystems (2011). Available at: https://www.leica-microsystems.com/science-lab/the-patch-clamp-technique/. (Accessed: 10th December 2019)

146. Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).

147. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 1–21 (2014).

148. Huo, H.-Q. et al. Modeling Down Syndrome with Patient iPSCs Reveals Cellular and Migration Deficits of GABAergic Neurons. Stem cell reports 10, 1251–1266 (2018).

149. Hibaoui, Y. et al. Modelling and rescuing neurodevelopmental defect of Down syndrome using induced pluripotent stem cells from monozygotic twins discordant for trisomy 21. EMBO Mol. Med. 6, 259–277 (2014).

150. Murray, A. et al. Brief report: isogenic induced pluripotent stem cell lines from an adult with mosaic down syndrome model accelerated neuronal ageing and neurodegeneration. Stem Cells 33, 2077–2084 (2015).

151. Jiang, J. et al. Translating dosage compensation to trisomy 21. Nature 500, 296–300 (2013).

152. Briggs, J. A. et al. Integration-free induced pluripotent stem cells model genetic and neural developmental features of down syndrome etiology. Stem Cells 31, 467–478 (2013).

153. Letourneau, A. et al. Domains of genome-wide gene expression dysregulation in Down’s syndrome. Nature 508, 345–350 (2014).

154. Halevy, T., Biancotti, J.-C., Yanuka, O., Golan-Lev, T. & Benvenisty, N. Molecular Characterization of Down Syndrome Embryonic Stem Cells Reveals a Role for RUNX1 in Neural Differentiation. Stem cell reports 7, 777–786 (2016).

155. Conti, A. et al. Altered expression of mitochondrial and extracellular matrix genes in the heart of human fetuses with chromosome 21 trisomy. BMC Genomics 8, 268 (2007).

156. Kuleshov, M. V. et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 44, W90–W97 (2016).

157. Wiseman, F. K. et al. A genetic cause of Alzheimer disease: mechanistic insights from Down syndrome. Nat. Rev. Neurosci. 16, 564–574 (2015).

158. Shi, Y. et al. A human stem cell model of early Alzheimer’s disease pathology in Down syndrome. Sci. Transl. Med. 4, 124ra29 (2012).

159. El Hajj, N. et al. Epigenetic dysregulation in the developing Down syndrome cortex. Epigenetics 11, 563–578 (2016).

160. Lu, J. et al. Global hypermethylation in fetal cortex of Down syndrome due to DNMT3L overexpression. Hum. Mol. Genet. 25, 1714–1727 (2016).

161. Xin, Y. et al. Genome-wide divergence of DNA methylation marks in cerebral and cerebellar cortices. PLoS One 5, e11357 (2010).

162. McGowan, H. Bridging the Gap between DNA Methylation, DNA Methylation Readers, and Neurodevelopmental Disorders. J. Neurosci. 36, 6851–6853 (2016).

163. Sharma, A., Klein, S. S., Barboza, L., Lohdi, N. & Toth, M. Principles Governing DNA Methylation during Neuronal Lineage and Subtype Specification. J. Neurosci. 36, 1711–1722 (2016).

164. Henneman, P. et al. Widespread domain-like perturbations of DNA methylation in whole blood of Down syndrome neonates. PLoS One (2018).

165. Bacalini, M. G. et al. Identification of a DNA methylation signature in blood cells from persons with Down syndrome. Aging (Albany. NY). (2015).

62

166. Wolfe, D., Dudek, S., Ritchie, M. D. & Pendergrass, S. A. Visualizing genomic information across chromosomes with PhenoGram. BioData Min. 6, 18 (2013).

167. Gardiner-Garden, M. & Frommer, M. CpG islands in vertebrate genomes. J. Mol. Biol. 196, 261–282 (1987).

168. Deaton, A. M. & Bird, A. CpG islands and the regulation of transcription. Genes Dev. 25, 1010–1022 (2011).

169. Jones, P. A. Functions of DNA methylation: islands, start sites, gene bodies and beyond. Nat. Rev. Genet. 13, 484–492 (2012).

170. Ziller, M. J. et al. Charting a dynamic DNA methylation landscape of the human genome. Nature 500, 477–481 (2013).

171. Ioshikhes, I. P. & Zhang, M. Q. Large-scale human promoter mapping using CpG islands. Nat. Genet. 26, 61–63 (2000).

172. Kent, W. J. et al. The human genome browser at UCSC. Genome Res. 12, 996–1006 (2002).

173. Higurashi, N. et al. A human Dravet syndrome model from patient induced pluripotent stem cells.

174. Jiao, J. et al. Modeling Dravet syndrome using induced pluripotent stem cells (iPSCs) and directly converted neurons. Hum. Mol. Genet. (2013).

175. Desnous, B., Goujon, E., Bellavoine, V., Merdariu, D. & Auvin, S. Perceptions of fever and fever management practices in parents of children with Dravet syndrome. Epilepsy Behav. 21, 446–448 (2011).

176. Ragona, F. et al. Dravet syndrome: early clinical manifestations and cognitive outcome in 37 Italian patients. Brain Dev. 32, 71–77 (2010).

177. McElroy, P. B., Liang, L.-P., Day, B. J. & Patel, M. Scavenging reactive oxygen species inhibits status epilepticus-induced neuroinflammation. Exp. Neurol. 298, 13–22 (2017).

178. Maussion, G. et al. Implication of LRRC4C and DPP6 in neurodevelopmental disorders. Am. J. Med. Genet. A 173, 395–406 (2017).

179. Liu, C.-S. & Tsai, C.-S. Enhanced lipid peroxidation in epileptics with null genotype of glutathione S-transferase M1 and intractable seizure. Jpn. J. Pharmacol. 90, 291–294 (2002).

180. Miwa, J. M. et al. lynx1, an endogenous toxin-like modulator of nicotinic acetylcholine receptors in the mammalian CNS. Neuron 23, 105–114 (1999).

181. Zonta, B. et al. A critical role for Neurofascin in regulating action potential initiation through maintenance of the axon initial segment. Neuron 69, 945–956 (2011).

182. Alpizar, S. A., Baker, A. L., Gulledge, A. T. & Hoppa, M. B. Loss of Neurofascin-186 Disrupts Alignment of AnkyrinG Relative to Its Binding Partners in the Axon Initial Segment. Front. Cell. Neurosci. 13, 1 (2019).

183. Shahsavani, M. et al. An in vitro model of lissencephaly: expanding the role of DCX during neurogenesis. Mol. Psychiatry 23, 1674–1684 (2018).

184. Brose, K. & Tessier-Lavigne, M. Slit proteins: key regulators of axon guidance, axonal branching, and cell migration. Curr. Opin. Neurobiol. 10, 95–102 (2000).

185. Moffat, J. G., Vincent, F., Lee, J. A., Eder, J. & Prunotto, M. Opportunities and challenges in phenotypic drug discovery: An industry perspective. Nat. Rev. Drug Discov. 16, 531–543 (2017).

186. Horvath, P. et al. Screening out irrelevant cell-based models of disease. Nature reviews. Drug discovery 15, 751–769 (2016).

187. Lorenz, C. et al. Human iPSC-Derived Neural Progenitors Are an Effective Drug Discovery Model for Neurological mtDNA Disorders. Cell Stem Cell 20, 659-674.e9 (2017).

63

188. Butt, T. R. & Karathanasis, S. K. Transcription factors as drug targets: opportunities for therapeutic selectivity. Gene Expr. 4, 319–336 (1995).

189. Dinday, M. T. & Baraban, S. C. Large-Scale Phenotype-Based Antiepileptic Drug Screening in a Zebrafish Model of Dravet Syndrome. eNeuro 2, ENEURO.0068-15.2015 (2015).

190. Lee, C. T., Bendriem, R. M., Wu, W. W. & Shen, R. F. 3D brain Organoids derived from pluripotent stem cells: Promising experimental models for brain development and neurodegenerative disorders Julie Y.H. Chan. J. Biomed. Sci. 24, 1–12 (2017).

191. Gabriel, E. & Gopalakrishnan, J. Generation of iPSC-derived Human Brain Organoids to Model Early Neurodevelopmental Disorders. J. Vis. Exp. (2017).

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1622

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-398620

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020