PCDH19 regulation of neural progenitor cell differentiation suggests asynchrony of neurogenesis as a mechanism contributing to PCDH19 Girls Clustering Epilepsy

Accepted Manuscript PCDH19 regulation of neural progenitor cell differentiation suggests asynchrony of neurogenesis as a mechanism contributing to PCDH19 Girls Clustering Epilepsy

Claire C. Homan, Stephen Pederson, Thu-Hien To, Chuan Tan, Sandra Piltz, Mark Corbett, Ernst Wolvetang, Paul Thomas, Lachlan A. Jolly, Jozef Gecz PII: DOI: Reference:

S0969-9961(18)30142-6 doi:10.1016/j.nbd.2018.05.004 YNBDI 4166

To appear in:

Neurobiology of Disease

Received date: Revised date: Accepted date:

23 January 2018 25 April 2018 9 May 2018

Please cite this article as: Claire C. Homan, Stephen Pederson, Thu-Hien To, Chuan Tan, Sandra Piltz, Mark Corbett, Ernst Wolvetang, Paul Thomas, Lachlan A. Jolly, Jozef Gecz , PCDH19 regulation of neural progenitor cell differentiation suggests asynchrony of neurogenesis as a mechanism contributing to PCDH19 Girls Clustering Epilepsy. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Ynbdi(2017), doi:10.1016/j.nbd.2018.05.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

PCDH19 regulation of neural progenitor cell differentiation suggests asynchrony of neurogenesis as a mechanism contributing to PCDH19 Girls Clustering Epilepsy. Claire C Homan1,2,3, Stephen Pederson4, Thu-Hien To4, Chuan Tan1,2, Sandra Piltz2,3,6, Mark Corbett1,2,3, Ernst Wolvetang5, Paul Thomas2,3,6, Lachlan A Jolly1,2*, Jozef Gecz1,2,3,6*

PT

Affiliations 1

School of Medicine, The University of Adelaide, Adelaide 5005, Australia Robinson Research Institute, The University of Adelaide, Adelaide 5006, Australia 3 School of Biological Sciences, The University of Adelaide, Adelaide 5005, Australia 4 Bioinformatics Hub, School of Biological Sciences, The University of Adelaide, Adelaide, 5005, Australia 5 Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, Queensland 4072, Australia 6 South Australian Health and Medical Research Institute, Adelaide 5000, Australia

NU

SC

RI

2

MA

* Corresponding Authors

PT E

Decelerations of interest: None

D

Lachlan A Jolly: [email protected], Jozef Gecz: [email protected]

Highlights:

Generation and study of the first PCDH19-GCE relevant induced pluripotent stem cell lines.



Loss of PCDH19 function promotes neurogenesis of mouse and human neural stem/progenitor cells.



Neural stem/progenitor cells lacking functional PCDH19 also display a loss of characteristic apical-basal polarity.



Our data implicate asynchronous neurogenesis (or mosaic cellular heterochrony) as a mechanism contributing to PCDH19-GCE.

AC

CE



Keywords: PCDH19; Epilepsy; Neurogenesis; Induced pluripotent stem cells, Polarity, Neural stem and progenitor cells

1

ACCEPTED MANUSCRIPT

Abstract PCDH19-Girls Clustering Epilepsy (PCDH19-GCE) is a childhood epileptic encephalopathy characterised by a spectrum of neurodevelopmental problems. PCDH19-GCE is caused by heterozygous loss-of-function mutations in the X-chromosome gene, Protocadherin 19 (PCDH19) encoding a cell-cell adhesion molecule. Intriguingly, hemizygous males are

PT

generally unaffected. As PCDH19 is subjected to random X-inactivation, heterozygous females are comprised of a mosaic of cells expressing either the normal or mutant allele,

RI

which is thought to drive pathology. Despite being the second most prevalent monogeneic cause of epilepsy, little is known about the role of PCDH19 in brain development. In this

SC

study we show that PCDH19 is highly expressed in human neural stem and progenitor cells (NSPCs) and investigate its function in vitro in these cells of both mouse and human origin.

NU

Transcriptomic analysis of mouse NSPCs lacking Pcdh19 revealed changes to genes involved in regulation of neuronal differentiation, and we subsequently show that loss of Pcdh19

MA

causes increased NSPC neurogenesis. We reprogramed human fibroblast cells harbouring a pathogenic PCDH19 mutation into human induced pluripotent stem cells (hiPSC) and employed neural differentiation of these to extend our studies into human NSPCs. As in

D

mouse, loss of PCDH19 function caused increased neurogenesis, and furthermore, we show

PT E

this is associated with a loss of human NSPC polarity. Overall our data suggests a conserved role for PCDH19 in regulating mammalian cortical neurogenesis and has implications for the pathogenesis of PCDH19-GCE. We propose that the difference in timing or ‘heterochrony’ of

CE

neuronal cell production originating from PCDH19 wildtype and mutant NSPCs within the same individual may lead to downstream asynchronies and abnormalities in neuronal network

activity.

AC

formation, which in-part predispose the individual to network dysfunction and epileptic

Introduction

PCDH19- Girls Clustering Epilepsy (PCDH19-GCE) is a predominately female-specific epileptic encephalopathy caused by heterozygous loss-of-function mutations in the Xchromosome gene Protocadherin 19 (PCDH19). The disorder is characterised clinically by the occurrence of seizures in girls at approximately 8-10 months of age followed by developmental regression with intellectual disability (ID) ranging from normal to profound (1-3). The disorder is also often associated with additional co-morbidities including Autism

2

ACCEPTED MANUSCRIPT Spectrum Disorders (ASD), language delay and other behavioural deficits. PCDH19 encodes a δ2-protocadherin, cell adhesion molecule, with a large highly conserved extracellular domain (among other protocadherins) and small cytoplasmic domain which shows variability among the protocadherin family (4, 5). Of more than 145 different (of 271 cases), reported PCDH19 disease causing variants, approximately 55% are nonsense, frameshift and splicing mutations which either ablate or severely truncate the PCDH19 protein, and the remaining

PT

45% are missense mutations, which cluster within the extracellular domain of the protein (Kolc et al. 2017, in review) (6). Based on generally shared clinical phenotype across all

RI

mutations, it is predicted that all identified mutations are loss of function to the protein.

SC

The inheritance pattern of PCDH19-GCE is an intriguing feature of the disorder. Unlike the majority of X-chromosome linked disorders, PCDH19-GCE presents only in carrier (i.e.

NU

heterozygous) females. Hemizygous males are generally spared from the disorder, although behavioural disturbances and/or ASD have been identified in some transmitting males (6). PCDH19 has been shown to be subject to random X-inactivation in females (5, 7), thus

MA

affected females are a mosaic of cells which express either the wildtype (WT) PCDH19 or mutant PCDH19 allele. It has been proposed that the pathogenesis of PCDH19-GCE is driven

D

by the cellular mosaicism within the brain, resulting in a partitioning of cell-cell connectivity and/or communication (8). Further support for this model comes from the identification of

PT E

now multiple affected males with postzygotic somatic mutation in PCDH19 (and hence mosaicism akin to heterozygous females) whom display typical presentations of PCDH19-

CE

GCE (8-10).

The cellular and molecular mechanisms that lead to the clinical presentations of PCDH19-

AC

GCE are still poorly understood. The first Pcdh19 knockout (KO) mouse models (produced by two independent groups, including our own) have been only recently reported (11-13). Neither models displayed an overt seizure phenotype in heterozygous females. Behavioural testing in one study revealed heterozygous animals had a decreased fear response, and both KO and heterozygous mice were hyperactive, particularly in response to stress.

Other

behavioural features including anxiety, social interaction, working and spatial memory were normal. Thus the mouse models may not faithfully recapitulate the severity and clinical features of the human condition. In both reported mouse models, no evidence of any gross morphological disturbances to brain structure was identified. However, it was discovered that KO neurons displayed increased migration rates in vitro (13), suggesting that more in-depth 3

ACCEPTED MANUSCRIPT analysis of the developmental processes and cellular architecture of the developing and adult brain is warranted. Indeed loss-of-function studies in zebrafish, has revealed more striking morphological phenotypes ranging from severe disruption of neural tube formation to milder disruption to only columnar architecture in the optic tectum depending on the technologies used (14, 15). It is intriguing that while the majority of patients have apparently normal Magnetic Resonance Imaging (MRI) reported, brain structural abnormalities including

PT

heterotopia in the left frontal cortex, atrophy of the hippocampus, cortical structural lesions and cortical dysplasia are evident in (at least) a small subset of patients (16-20). Collectively these studies highlight both the potential roles of PCDH19 in early embryonic brain

RI

developmental processes, and also the limitations of the model systems, both mouse and fish,

SC

in the study of PCDH19-GCE.

NU

Consistent with potential roles in early neural development, Pcdh19 expression has been shown to be expressed in the developing as well as adult brain (5, 13, 21). Pcdh19 expression is first detected in the neural stem and progenitor cells (NSPCs) of the neuroepithelium at

MA

embryonic day 9 (E9) in mouse, and is expressed throughout the neuroepithelium by the onset of neurogenesis at E12.5 (22). At E15.5, the peak period of neurogenesis in mouse,

D

Pcdh19 is highly expressed in both the NSPCs of the ventricular and sub-ventricular zones, the key proliferative zones of cortical development (13). At E10.5, heterozygous female mice

PT E

with mosaic expression of Pcdh19 were shown to establish segregation of WT and KO Pcdh19 cortical progenitors, suggesting this is likely a key developmental period for the establishment of PCDH19-GCE pathogenesis (12). Despite expression in NSPCs of the

CE

developing brain, and emerging evidence of the role of PCDH19 in early brain developmental processes, it remains unknown as to the involvement of Pcdh19 in the function NSPC

AC

populations during cortical development, which play critical primordial roles in establishing future adult brain architecture and overall function.

To identify evolutionary conserved roles of PCDH19 in mammalian cortical NSPCs, and the contribution these roles may play in the pathogenesis of PCDH19-GCE, we utilised both mouse Pcdh19 KO derived NSPCs and subsequently human patient derived (carrying loss-offunction missense mutation) NSPCs, and assayed their differentiation behaviours in vitro. In both instances, we also mixed WT cells with PCDH19 Mutant (MUT)/KO cells to mimic mosaicism as seen in PCDH19-GCE. In both models, loss of Pcdh19 caused premature neurogenesis of NSPCs. We show that in human NSPCs, this premature neuronal 4

ACCEPTED MANUSCRIPT differentiation is associated with evidence of reduced apical-basal polarity of NSPCs. Intriguingly, mosaic cultures of NSPCs in general behaved similar to cultures of MUT/KO NSPCs. Our data thus provides evidence that heterochronic neural development may occur in patients with PCDH19-GCE, and potentially contributes to pathogenesis. Through the extension of this model, we propose that the asynchronous timing of neuronal cell production in vivo can contribute to subtle abnormal neuronal circuit assembly and dysfunction,

PT

culminating in epileptic activity.

RI

Materials and Methods:

SC

Animals

The Pcdh19 KO mouse colony was used as described in (13). Unless otherwise described males and females were used in this study. Genotyping was performed as described in (13).

NU

All mice were maintained in same sex housing in a 12-hr day:night cycle with adequate sterile food and water under specific pathogen free conditions. The use of animals was

MA

approved under The University of Adelaide institutional ethics committee.

Mouse NSPC culture

D

Mouse NSPCs were isolated from e14.5 embryonic mouse cortex (telencephalic vesicles) and

PT E

cultured as non-adherent neurospheres in the presence of 20ng/mL FGF and EGF as described (23). To generate the mosaic culture, neurospheres were passaged to single cells. Viable cell counts were determined using trypan blue and mixed (mosaic) cell lines formed

CE

by combining Pcdh19 WT (either fluorescently labelled or unlabelled) with Pcdh19 KO (either fluorescently labelled or unlabelled) NSPCs in a 1:1 ratio. All cultures were

AC

maintained for four passages (20 days) prior to use in neurosphere assays.

Generation of lentivirus The lentiviral LeGO vectors (LeGO-C2, LeGO-G2) expressing eGFP or mCherry were obtained from Addgene (MA, USA) and have been described previously (24). Cell-free viral supernatants were generated by transient transfection of HEK293T packaging cells a gift from Associate Professor Simon Barry (The University of Adelaide, South Australia, Australia) as previously described (25, 26)

5

ACCEPTED MANUSCRIPT Transduction of NSPCs was performed at a multiplicity of infection (MOI) of 5. Pooled biological triplicates were generated for viral transduction by combining three Pcdh19 WT (Pcdh19+/+) embryos derived NSPC lines and three Pcdh19 KO (Pcdh19-/-) embryo derived NSPC lines, in equal numbers, immediately after isolation. Pcdh19 WT NSPCs were transduced with LeGO-G2 virus expressing eGFP and Pcdh19 KO NSPCs were transduced with LeGO-C2 virus expressing mCherry. Cells were exposed to the virus for 20 hrs. Cells

PT

were expanded for a further passage before isolation of fluorescently labelled cells using the FACSCalibur (Becton Dickinson; BD, North Ryde, NSW, Australia) flow cytometry system

RI

and further culture as neurospheres.

SC

Mouse neurosphere differentiation assay

Neurospheres were passaged to single cells and plated at a density of 2.1x104 cells/cm2

NU

(immunofluorescence analysis) or 3.3x104 cells/cm2 (biochemical analysis) onto Poly-LLysine (Sigma) coverslips in neurosphere medium without growth factors. Cells were

MA

incubated at 37˚C in 5% CO2 and collected at four days of differentiation for analysis. Immunofluorescent staining and fluorescence microscopy was used for cell count analysis. Duplicate experiments were done in pooled biological triplicate and analysed in technical

PT E

D

triplicate with at least 1500 cells scored for each condition.

Direct differentiation of mouse NSPCs Mouse NSPCs isolated from the e14.5 telencephalic vesicles were plated directly onto PolyL-Lysine (Sigma) coverslips at a density of 2.1x104 cells/cm2 in neurosphere medium without

CE

growth factors. Cells were incubated at 37˚C in 5% CO2 and collected at seven days of differentiation for analysis. Immunofluorescent staining and fluorescence microscopy was

AC

used for cell count analysis. Duplicate experiments were done in pooled biological triplicate and analysed in technical triplicate with at least 1500 cells scored for each condition.

Immunofluorescence Cells were fixed with 4% PFA at room temperature for 15 minutes. Cells were initially permeablised in 0.2% Tween20 in PBS (PBST) (or for neural rosettes 0.5% TritonX in PBS) for 5 minutes. Cells were then blocked for 1 hour in 5% horse serum in PBST. Antibodies were diluted in a solution of 0.5% Horse Serum in PBST at the following dilutions; Mouse anti-PKCλ(1:500; BD transduction laboratories), mouse anti-OCT3/4(1:200; Santa Cruz),

6

ACCEPTED MANUSCRIPT goat anti-NANOG(1:500; R&D Systems), rabbit anti-PAX6 (1:300; Covance), rabbit antiSOX2(1:200; Millipore), rabbit anti-TBR2(1:200; Abcam), rabbit anti-TBR1(1:300; Abcam), mouse

anti-N-CADHERIN(1:300;

BD

transduction

laboratories),

rabbit

anti-

VGLUT1(1:300; Synaptic Systems), rabbit anti-CUX1(1:300; Santa Cruz), SATB2(1:100; Abcam), rat anti-CTIP2(1:300; Abcam), mouse anti-NESTIN(1:300; Millipore), mouse antiNEUN(1:1000;Millipore),

mouse

anti-GAD65/67(1:500;

Sapphire

bioscience),

Tubulin(1:300;

Sigma),guinea

MAP2(1:500;Millipore),

pig

Mouse

anti-DCX(1:300;

PT

VGLUT2(1:500; Millipore), mouse anti-βIII-Tubulin(1:300; Sigma), rabbit anti-βIIIMillipore),

anti-CNPase(1:1000;

Millipore),

chicken goat

antianti-

anti-rabbit

Alexafluor488,647(all

Alexafluor488(1:700;Invitorgen),

donkey

1:700;Invitrogen),

donkey

anti-rat

SC

donkey

RI

GFAP(1:700;Sigma), donkey anti-mouse Alexa fluor488,555,647(all 1:700;Invitorgen),

anti-chicken

Alexfluor555(1:500;Invitorgen),

NU

donkey anti-guinea pig Alexafluor555(1:500;Invitorgen). Primary antibody incubations were performed overnight in humid conditions at 4˚C. Secondary antibodies incubations were performed for 1hr at room temperature. When required cells were incubated with 0.1μM

MA

Phalloidin (Molecular Probes, Life Technologies) stain in 1% BSA in PBS for 30 minutes. Cells were mounted with ProLong® Gold Antifade mountant with DAPI (Life

PT E

D

Technologies).

Microarray

RNA was extracted from cells using Trizol Reagent (Invitrogen) and the RNeasy mini extraction kit (QIAGEN) as per manufacturer’s protocol. Samples were treated with the

CE

RNase-free DNase Kit (QIAGEN) to remove genomic contamination. RNA was quantitated using the Nanodrop 1000 photo-spectrometer (Thermo fisher Scientific, USA). RNA was

AC

diluted to 100 ng/µL. cDNA was hybridised to the Affymetrix MoGene 2.0 St array (Affymetrix , CA, USA). RNA quality was confirmed using the Bioanalyser. Each sample done in biological triplicate with a total of 18 arrays (three genotypes; WT, KO and Mosaic and two cell types; neurospheres and differentiated neurospheres).

Transcriptome analysis Data from the arrays was first inspected for the purpose of quality assessment using RLE and NUSE with no arrays showing significant departure from the average.

Data was then

background corrected using RMA (DOI: 10.1093/biostatistics/4.2.249) and quantile normalised using the R package affy (27). Non-specific binding properties of each probe 7

ACCEPTED MANUSCRIPT were determined using a modified MAT model (28) trained on the set of antigenomic probes using quantile normalised and optical-background corrected data. The parameters for the MAT model were estimated across the entire dataset, and fitted values for each probe broken into 20 equally-sized bins based in their fitted values. This gave approximately 847 probes in each bin, with empirical background distributions estimated within each bin. Fitted values were then obtained for perfect match (PM) probes based on their sequence content and the

PT

trained model. The fitted values were subtracted from the observed PM intensities on the log2 scale to provide the residuals from the MAT model giving an empirical p-value for detection of signal above background (DABG). Fisher's method was used to combine the p-values for

RI

each probe across the entire set of arrays and those receiving a DABG p-value < 0.05 across

SC

all arrays were retained. After DABG analysis, 51.4% of probes (293281 of 571035) were included for estimation of expression levels. Array-specific expression estimates were

NU

obtained for the remaining genes, with an expression set formed selecting only genes containing 9 or more retained probes. Differential expression was defined based on an FDRadjusted p-value of <0.05 and log2 fold change (logFC) outside the range ± 1. Genes with

MA

raw p-value < lx10-9 were considered as differentially expressed regardless of fold change.

D

Quantitative Real-Time PCR

RNA was extracted from cells using Trizol Reagent (Invitrogen) and the RNeasy mini

PT E

extraction kit (QIAGEN) as per manufacturer’s protocol. Samples were treated with the RNase-free DNase Kit (QIAGEN) to remove genomic contamination. cDNA was generated using superscript III® reverse transcriptase (Invitrogen) and random priming (Geneworks,

CE

Australia) as per manufacturer’s instructions. For mouse experiments; PCR reactions were prepared using SYBR green master mix (Biorad) and run on the StepOnePlus Real-Time

AC

PCR System (Applied Biosystems) using the following parameters; 95˚C for 5 minutes; 35 cycles of 95˚C for 3 seconds, 60˚C for 30 seconds; followed by a melt curve set as follows: 95°C for 15 seconds, 60°C for 1 minute followed by a 0.3% continuous ramp from 60 °C to 95 °C with data collected using the StepOne V2.2.2 Software (Applied Biosystems). For hiPSC experiments; PCR reactions were prepared using Fast SYBR green master mix (Applied Biosystems) run on the QuantStudio™ 7-flex real-time PCR system using the 384 well-plate format and the following cycling parameters; 95˚C for 20 seconds; 40 cycles of 95˚C for 3 seconds, 60˚C for 30 seconds; followed by a melt curve set as follows: 95°C for 15 seconds, 60°C for 1 minute followed by a 0.3% continuous ramp from 60 °C to 95 °C.

8

ACCEPTED MANUSCRIPT Data collected and analysed using the QuantStudio Real-Time PCR software v1.3 (Applied Biosystems).

Human cell culture All cell culture work was performed based on the informed consent of patients under the approval of the human research ethics committee (Adelaide Women’s and Children’s

PT

Hospital). Human iPSCs were maintained on irradiated mouse embryonic fibroblasts (MEFs) as described previously (29). hESCs were cultured in the presence of 10 ng/mL bFGF. hiPSCs were cultured in 80-100 ng/mL bFGF. MEFs were supplied by Stemcore Australia.

RI

Cultures were mechanically passaged in a biological safety cabinet class II (VWR) using a Feeder-free

SC

dissecting microscope (Olympus SZX16 stereo microscope, Tokyo, Japan).

culture of hPSCs on ECM (Sigma) in MEF conditioned media was carried out as described

NU

(30). Cells were incubated at 37˚C in 5% CO2 with medium changes daily. PCDH19-GCE individual’s fibroblasts were cultured in RPMI medium supplemented with 10% heat

MA

inactivated fetal bovine serum and 2 mM L-glutamine (all from sigma-Aldrich, Australia).

Episomal reprogramming of human fibroblasts

D

Reprogramming experiments were all performed by the commercial services of Stemcore Australia. Primary human skin fibroblasts (1x106 cells) were nucleofected with non-

PT E

integrating episomal vectors; pEP4 E02S ET2K & pEP4 E02S CK2M EN2L (31), using the Amaxa Human Dermal Fibroblasts Nucleofector kit, program U-023 (Lonza, VPD‐1001) as per manufacturer’s instructions. Immediately, following nucleofection cells were seeded on

CE

3x10 cm tissue culture dishes (0.76x104 cells/cm2) on a MEF feeder layer in human fibroblast culture medium. Dishes were either incubated at; 37˚C in 5% CO2 in atmospheric oxygen

AC

conditions or 37˚C in 5% CO2 and 5% O2 (low oxygen conditions). After 2 days of culture the medium was changed to hiPSC complete medium containing 100 ng/mL of bFGF, with daily medium changes. After 10 days of culture the medium was changed to hiPSC conditioned medium with 100 ng/mL of bFGF. Plates were monitored and as colonies emerged they were isolated mechanically and replated onto a MEF feeder layer in complete hiPSC medium and incubated at 37˚C in 5% CO2 in atmospheric oxygen conditions for all plates.

Screening of hiPSC clones

9

ACCEPTED MANUSCRIPT hiPSC clones were isolated and sub-cultured based on characteristic hiPSC morphology and growth characteristics. Clones were screened for the expression of pluripotency marker Tra1-60 using immunofluorescence (1:100). Standard G-banding chromosome karyotyping was carried out by Stemcore Australia, 15 metaphases analysed per sample. In vitro spontaneous differentiation assay and in vivo teratoma assays were both carried out by commercial

PT

services of Stemcore Australia.

X-inactivation screening of hiPSC

X-chromosome inactivation (XCI) was determined by both; Sanger sequencing of cDNA

RI

spanning the PCDH19 pathogenic mutation and the Human Androgen Receptor (HUMARA)

NU

Human iPSC cortical differentiation protocol

SC

PCR based assay (32).

Cortical differentiation of hiPSCs was performed as previously described (33) with minor modifications. Modifications included; hiPSCs were plated as a high density single cell

MA

monolayer prior to cortical induction. Cortical induction was initiated by the addition of 5 µM dorsomorphin and 10 µM SB431542 as described in (34). To select for cortical NSPCs,

D

rosettes were manually selected in a biological safety cabinet class II (VWR) using a dissecting microscope (Olympus SZX16 stereo microscope, Tokyo, Japan) and replated in 20

PT E

ng/mL FGF and 1 µM ROCK inhibitor for one day. The H9 control human embryonic stem

Microscopy

CE

cell line (35) was also used to validate the optimised differentiation protocol.

Fluorescence was observed using the Zeiss AxioImager M2 fluorescent microscope (Carl

AC

Zeiss, Germany). Fluorescent images were acquired with the AxioCam MRm high resolution camera and the AxioVision software Vs4.9.1.0 (Carl Zeiss, Germany). Neural Rosettes fluorescence was observed on Nikon Ti E live cell microscope with an automated stage (Adelaide microscopy, The University of Adelaide). Fluorescent images were acquired with a Photometrics camera and the NIS elements software (Nikon instruments Inc, NY, USA).

Neural rosette morphological analysis Neural rosette images acquired using the Zeiss fluorescent microscope were acquired sequentially at 100X magnification until the whole neural rosette colony was captured. Using the Image J plugin, stitching; Grid/collection (36), the neural rosette colony was 10

ACCEPTED MANUSCRIPT reconstructed as a single image for analysis. To avoid selection bias, neural rosette colonies were selected for imaging based on their position on the slide. Neural rosette images acquired using the Nikon live cell microscope were selected based on their position on the slide with 12 locations selected for imaging, using an 8x8 field of view grid at 100X magnification. Using image J software, the area of the neural rosette colony was determined using the freehand tracing tool to trace the outer edge of the neural rosette colony. The number of foci

PT

and neural rosette structures was then counted within each colony using the cell counter plugin based on phalloidin stain. Foci were identified based on a diameter <10µm and no visible central lumen. Polarised structures were identified as having a central lumen and

RI

being >10µm in diameter. The number of foci and polarised structures within each rosette

SC

colony was then determined relative to the area of the neural rosette colony as a measure of polarity. To measure the lumen area, the freehand tracing tool in ImageJ was used to trace

NU

around the periphery of the structures. The total area of the polarised structures was then determined relative to the area of the neural rosette colony.

Neurons were detected

immunofluorescently using a β-III tubulin antibody (as described above) and counterstained

MA

with DAPI to identify single cells. The number of neurons on the edge of the neural rosette colony was then determined by using the Image J cell counter plugin, counting the cells

D

which were co-labelled with β-III tubulin and DAPI. The number of neurons was determined

PT E

relative to the area of the neural rosette colony.

Results:

CE

Loss of Pcdh19 in mouse NSPC drives accelerated neurogenesis. To investigate the pathogenesis of PCDH19-GCE and the effect of loss of Pcdh19 function

AC

on neural development, and NSPCs in particular, we used a recently described Pcdh19 KO mouse (13). Mouse NSPCs (mNSPCs) were extracted from both WT and KO telencephalic vesicles at E14.5 and cultured ex-vivo in an undifferentiated state as non-adherent neurospheres (23). To create mosaicism observed in PCDH19-GCE, mNSPCs from WT and KO were also mixed in a 1:1 ratio to generate a Pcdh19 WT:KO (Mosaic) neurosphere culture (Figure 1A). To study the cell autonomous role of Pcdh19 in NSPC differentiation, mNSPCs from neurospheres were dissociated into single cells, and plated onto a poly-Llysine substrate in the absence of growth factor which induces differentiation of mNSPCs into mature cell types of the brain (namely, neurons, astrocytes and oligodendrocytes). Initially, we sought an unbiased analysis of the role of Pcdh19 in mNSPC differentiation: we

11

ACCEPTED MANUSCRIPT isolated mRNA from both undifferentiated and differentiated mNSPC cultures (WT, KO and Mosaic) and subjected them to a genome-wide gene expression analysis using Affymetrix MoGene 2.0 St microarrays (Figure 1A and Supplementary figure 1). As expected, Pcdh19 was one of the most significantly differentially expressed (DE) genes in all samples, with expression in the Mosaic cultures being intermediate to WT and KO cultures (Supplementary figure 1B). Principal component analysis (PCA) of data sets derived from the undifferentiated

PT

mNSPC cultures suggested that Pcdh19 is not having a significant impact on the overall transcriptional signature of undifferentiated mNSPCs, as all three culture types clustered together (Figure 1B). Data sets derived from cultures of differentiated mNSPCs all separated

RI

away from the undifferentiated mNSPC cluster, however the Mosaic and KO cultures

SC

clustered together, and separated from the WT cultures (Figure 1B). This suggested that differentiation of KO and Mosaic mNSPC cultures was altered compared to WT cultures.

NU

Given the Mosaic and the KO differentiated mNSPC cultures clustered together in the PCA, and no highly significant genes were identified as DE between the Mosaic and the KO cultures, we compared WT to KO cultures in subsequent analysis. To focus on the process of

MA

differentiation itself, we identified the genes that were differentially expressed between undifferentiated and differentiated cultures, and then compared this gene list between WT This analysis thus targets

D

and KO mNSPC cultures (Supplementary figure 1C-1E).

identification of potential pathways or cellular mechanisms regulated by Pcdh19 during

PT E

mNSPC differentiation. Filtering of the datasets (see material and methods) identified 920 significantly DE genes in WT and 904 significantly DE genes in KO, occurring across differentiation (Supplementary figure 1C-E). Of these 71.6% were conserved between the

CE

two genotypes, with 302 significantly DE genes identified as unique between WT and KO differentiation (Figure 1C). This unique gene list was submitted for enrichment analysis using

AC

the enrichr software (37), to identify significant gene ontology (GO) terms or pathways. The 15 most significantly enriched GO terms for biological processes, based on the enrichr combined score, included six neuronal process associated with neuronal differentiation and maturation (Figure 1D). This included neuronal migration, which we have previously shown to be accelerated in Pcdh19 KO neurons (13). Intriguingly, the most significantly enriched GO term for cell component, was cell-cell adherens (Supplementary figure 2C), which was of interest given Pcdh19 is a protocadherin itself, and has also been shown to interact with Ncadherin an integral component of adherens junctions in mNSPCs (38). Additionally, KEGG pathway analysis also showed enrichment of pathways known to be involved in cell fate decisions (e.g. proliferation vs differentiation) of mNSPCs, including the Hippo signalling 12

ACCEPTED MANUSCRIPT pathway (Figure 1E). In the top 15 enriched pathways both estrogen and oxytocin signalling pathways were also identified, which have been previously been implicated in PCDH19-GCE (39). Further analysis of the 302 DE genes identified clusters of genes with bonafide roles in cell-cell adhesion and/or cadherin signalling, neuronal differentiation and maturation, and negative regulators of cellular proliferation. (Figure 1F). In aggregate, the transcriptomic analysis suggests that (1) mNSPC differentiation in KO and Mosaic cultures were similar to

PT

each other and distinct from WT, and (2) implicate roles of Pcdh19 in mNSPC proliferation, neuronal differentiation and neuronal maturation.

RI

Given the above data, we next directly addressed if Pcdh19 effects the differentiation of

SC

mNSPCs. As per above, we again plated at low density undifferentiated WT and KO mNSPCs derived from neurospheres onto a Poly-L-Lysine substrate and allowed them to

NU

differentiate for four days. Cultures were then immunofluorescently stained using antibodies against cell type specific marker proteins to identify mNSPCs (Pax6) and derived postmitotic cell types of the three major neural cell lineages, namely neurons (β-III Tubulin),

MA

astrocytes (GFAP) and oligodendrocytes (CNPase). Cell counts were then performed and percentage of each cell type in the cultures determined (Figure 1G and H). There was a

D

significant increase in the percentage of neurons present in the KO cell population compared to WT cultures (59.7% compared to 50.4% respectively; Figure 1H). Given the sex-bias of

PT E

PCDH19-GCE we compared neuronal differentiation between male and female derived mNSPCs. Increased percentage of neurons in the KO cell population was observed regardless of the gender of the cells (Supplementary figure 3), suggesting no sex-specific effect on

CE

neuronal differentiation. This increase in neurons was accompanied by a decrease in the percentage of oligodendrocytes in the KO cells (6.18%) as compared to WT (10.44%). These

AC

results were supported further by RT-qPCR analysis of the mRNA expression of different cell type marker genes (Supplementary figure 4A). To track the behaviour of WT and KO cells in Mosaic cultures, we first fluorescently labelled the individual populations (KO cells labelled with lentivirus expressing mCherry, and WT cells likewise labelled with EGFP) prior to mixing to create the Mosaic culture. We found that in mosaic cultures, it was only the KO cells that prematurely differentiated into neurons (at similar rate to cultures of KO cells only) whilst WT cells behaved normally (similar to cultures of WT cells only). This data suggest that loss of Pcdh19 can impact neurogenesis in a cell autonomous manner (Supplementary 4B and C). We also explored if differences in the proportions of inhibitory or excitatory neuronal subtypes within WT and KO cultures existed, but found no evidence as such suggesting that 13

ACCEPTED MANUSCRIPT Pcdh19’s role in mNSPCs is not restricted to mNSPCs of either dorsal or ventral identify (from which excitatory and inhibitory neurons are derived respectively) (Supplementary figure 4D and E). These data reveal that Pcdh19 has a role in limiting the rate of neurogenesis of mNSPCs.

Generation of PCDH19-GCE hiPSC lines as a source of human NSPCs.

PT

Whilst there are many conserved developmental mechanisms and features between mouse and human brain development, there are also many differences. In particular, whilst NSPCs

RI

have similar fundamental functions (e.g. self-renewal, multipotency, spatial and temporal fate

SC

restrictions), NSPCs in human must persist and function over a greatly extended period of time, and have greater diversity (e.g. outer radial glial cells) compared to mouse. Thus we sought evidence for conserved role of PCDH19 in neurogenesis of human NSPCs (hNSPCs)

NU

to both confirm and extend our findings in mouse. As a source of hNSPCs, we set out to generate relevant human induced pluripotent stem cells (hiPSCs), from which hNSPCs can be

MA

generated using directed neural differentiation techniques. As hiPSCs have yet to be generated, we employed our resources of skin fibroblast cell lines derived from individuals with PCDH19-CGE, with pathogenic (loss-of-function) mutations in PCDH19.

Three

D

unrelated individuals with different pathogenic PCDH19 mutations were selected for hiPSC

PT E

generation: Two females with PCDH19-GCE (Individual 1 and Individual 2) and an unaffected transmitting male (Individual 3). Individuals 1 and 3 have different pathogenic missense mutations in exon1 [(NM_001105243.1); c.1129G>C, p.Asn377His and

CE

c.1671C>G p.Asn557Lys, respectively] and Individual 2 has a nonsense mutation in the cytoplasmic domain [c.2342dupA, p.Ile781Asn fs*3]. Selected individual’s details are

AC

described (supplementary data) and summarised in figure 2.

Human skin fibroblast cell lines derived from each individual were reprogrammed using nonviral, non-integrative episomal reprogramming as previously described (31). Putative hiPSC clones were isolated based on morphology and growth characteristics over several passages, and expression of the pluripotent stem cell markers protein Tra-1-60 (as assessed via immunostaining). As PCDH19 is subjected to random X-inactivation in females, reprograming of the female lines provided an opportunity to generate both WT and mutant expressing isogenic clones. As such, the active X-chromosomes in putative hiPSC clones derived from the female individuals was identified using a combination of X-chromosome

14

ACCEPTED MANUSCRIPT inactivation assays (AR micro-satellite analysis), allele specific mRNA expression and sanger sequencing of PCDH19 cDNA species (Figure 2A and data not shown). Intriguingly we found that all putative female hiPSCs clones derived (n=25 in total tested) expressed exclusively the X-chromosome harbouring the PCDH19 WT allele. No clones were identified expressing the PCDH19 Mutant allele (Figure 2B). Thus we focussed on isolating mutant PCDH19 hiPSCs from the male (hemizygous) individual. After multiple rounds of

PT

reprogramming with increased plating density (1.14x104/cm2 in two 100mm dishes) and incubation at low oxygen conditions (5% O2), three putative hiPSC clones were isolated from the male individual (Individual 3). The reprogramming efficiency to isolate mutant PCDH19

RI

hiPSCs was much lower than for WT (0.0002%). These data suggest a selective advantage is

SC

conferred by expression of WT PCDH19, however hiPSCs can be generated with loss of PCDH19 function albeit at much reduced efficiency. Two of the three putative mutant hiPSC

NU

clones from Individual 3 were shown to maintain a normal karyotype, and subjected to further characterisation of their pluripotent state. Both lines expressed pluripotency marker proteins; OCT4, NANOG, SOX2 and Tra-1-60 (Figure 2C and D), and were shown to be able

MA

to differentiate into cells from all three embryonic germ layers using both an in vitro spontaneous differentiation assay and an in vivo teratoma assay (Figure 2E). All together,

D

these data suggest that whilst PCDH19 benefits the reprograming of fibroblasts into hiPSCs, bonafide hiPSC lines lacking PCDH19 function can be created, albeit at much lower

PT E

frequency. In summary, we were successful in creating two hiPSC harbouring hemizygous loss-of-function PCDH19 missense mutation (PCDH19Asn557Lys), which provided the vital

PCDH19-GCE.

CE

resources to study the role of PCDH19 in human NSPCs relevant to the pathogenesis of

AC

PCDH19 is highly expressed in cortical hNSPCs To generate hNSPCs and neurons from our hiPSC resources, we first optimised a previously published method of directed neural differentiation of human pluripotent stem cells (33). Our method employs initial plating hiPSCs as single cells at high density, dual-SMAD inhibition and ongoing adherent 2-dimensional culture. The method induces hNSPCs of dorsal cortical identity, and thus align with our previous studies of the same cells in mice. Chronological developmental hallmarks of the culture include induction of neuroepithelial sheet, generation of hNSPCs assembled as so called neural rosettes, hNSPC expansion, and with extended culture (up to 90 days), generation of both lower and upper layer excitatory cortical neurons

15

ACCEPTED MANUSCRIPT hence recapitulating the full neurogenic potential of in vivo NSPC equivalents; Figure 3A (33). Our optimised method was robust across several hiPSC and hESC lines. In the first instance we used this model of in vitro human cortical development to investigate the expression of PCDH19 mRNA. PCDH19 is increased upon neural induction, with expression peaking at the neural rosette stage and maintained through to 90-day cortical neuron stage, suggesting prominent roles of PCDH19 might exist early in cortical development, and in

PT

hNSPCs in particular (Figure 3B). Next, we employed neural differentiation of our control (C11 hiPSC line (29); WT) and PCDH19 Mutant (MUT) hiPSCs to investigate the function of PCDH19 in hNSCPs. Furthermore, to also model PCDH19-GCE, we also created mosaic

RI

cultures (Mosaic), by mixing WT and MUT hiPSC at a 1:1 ratio prior to being plated as

SC

single cells for neural induction (Supplementary figure 5). All three culture types were simultaneously subjected to neural induction and mRNA collected across the developmental

NU

hallmark stages of differentiation. Initially, we measured the expression of PCDH19 itself. The expression of PCDH19 was not significantly different between cultures across most developmental stages, but intriguingly, at the rosette stage, expression was significantly

MA

increased in both the MUT (fold change=4.97, P=0.0024) and Mosaic (fold change=4.17, P=0.0349) compared to WT cultures, suggesting a possible feedback response (Figure 3C).

D

Collectively, these results suggest that hNSPCs present at the rosette stage of hiPSC neural

PT E

differentiation may be particularly reliant on PCDH19 function.

CE

Pathogenic loss-of-function mutation in PCDH19 causes accelerated human NSPC differentiation and neuronal maturation.

AC

Next we used mRNA expression analysis of marker genes specific to the different representative cell types present in the cultures across the developmental stages to determine if PCDH19 regulated neural differentiation. PAX6 is a marker of dorsal cortical hNSPCs and expressed first in NSPCs of the neuroepithelial sheet, and maintained in hNSPCs at subsequent stages. We saw no difference in PAX6 expression at the neuroepithelial sheet stage suggesting PCDH19 has no effect on neural induction of hiPSCs (Figure 4A). In further support, we also saw no difference in expression of TBR2, a basal progenitor marker, and DCX, a pan-neuronal marker at this stage. At the neural rosette stage PAX6, TBR2 and NEUROG2 (marker of neural progenitors and radial glial cells) were all significantly decreased in both the MUT and the Mosaic culture types, suggesting loss of PCDH19

16

ACCEPTED MANUSCRIPT function caused a decrease in the number of hNSPCs (Figure 4B). Given our findings in mouse NSPCs, we considered if premature neurogenesis of NSPCs might explain the decrease in expression of hNSPC marker genes. The expression of the pan-immature neuronal maker genes DCX and TUBB3 was slightly (albeit not significantly) increased, as was the excitatory neuronal marker gene, VGLUT1, collectively providing evidence of increased neurogenesis (Figure 4B). No difference in GABAergic inhibitory neuronal marker

PT

gene GAD67 was identified. Increased expression of neuronal marker genes persisted at later stages of differentiation with an even greater increase in VGLUT1 and DCX expression observed in day 70 neurons in both the MUT and the Mosaic cultures compared to WT but

RI

neither reached significance (Figure 4C). Additional Glutamatergic marker genes GRIN1 and

SC

NEUROD6 showed similar increased expression, suggesting an increase in Glutamatergic excitatory neurons in the MUT and Mosaic cultures. A premature state of differentiation of

NU

the cultures, would also be consistent with a quicker change in expression from the early Glutamatergic excitatory neuronal marker, VGLUT2 to the mature Glutamatergic excitatory neuronal marker, VGLUT1. At the day 70 neurons stage, VGLUT2 expression was decreased

MA

between WT and MUT (fold change=0.112, P=0.0541) (Figure 4C). At day 90 stage (Figure 4D), this decrease reached significance with expression decreased between both WT and

D

MUT (fold change= 0.327, P=0.0018) and between WT and Mosaic (fold change= 0.329, P=0.0013). The level of expression in the Mosaic neurons was in-between WT and MUT.

PT E

Together, the marker gene analysis was consistent with a model whereby loss of PCDH19 function causes premature differentiation of hNSPCs first seen at the neural rosette stage.

CE

To more directly address if loss of PCDH19 results in increased numbers of neurons at the rosette stage of neural differentiation, we used immunofluorescent identification of newly

AC

born neurons (via immunostaining of βIII-tubulin) and cell counts. Neural rosette colonies are known to provide a spatial gradient of differentiation, in which newly born neurons migrate to the periphery and subsequently undergo maturation (40, 41). The number of peripheral βIII Tubulin positive cells were counted relative to the total area of the neural rosette colony (Figure 5A and B). Both the MUT and Mosaic cultures had a significant increase in the number of neurons relative to WT (4-fold and 3.82-fold respectively). We reasoned that if neurons are being born earlier in these cultures, their maturation (for example neurite growth) would also be advanced. Indeed, whilst the average primary neurite length for neurons in WT cultures was 155 µm, neurons in MUT and Mosaic cultures were significantly longer (248 and 238 µm respectively; Figure 5C and D). Thus, aligned with both our findings in mouse 17

ACCEPTED MANUSCRIPT NSPCs, and our marker gene analysis of human NSPCs, these data suggest that loss of PCDH19 function is associated with increased rate of NSPC differentiation.

Pathogenic loss-of-function mutation in PCDH19 diminishes the polarity of human NSPCs

PT

The assembly of hNSPCs into rosette structures arises through the establishment of apicalbasal polarity. This is evidenced initially by the focal accumulation (Foci) of apically

RI

localised polarity marker proteins in a few hNSCPs (41) (Figure 6A). With the subsequent expansion of polarised hNSPCs, a rosette structure is formed, featuring a central lumen

SC

demarcated by the apical surface of the cells, the size of which reflective of the number of apical-basal polarised hNSPCs (42). In opposition, differentiation of hNSPCs into neurons

NU

(which subsequently migrate radially outward to the colony periphery) is associated with loss of hNSPC polarity. This relationship between hNSPC polarity and hNSPC expansion (and likewise the inverse relationship of polarity with differentiation) is well established (43).

MA

Given that PCDH19 (1) is a cell-cell adhesion molecule, (2) interacts with N-CADHERIN, (3) and was found to regulate signalling pathways involved in cell-cell adhesion in mouse

D

NSPCs, all of which have the potential to influence NSPC polarity, we questioned if the

PT E

premature differentiation of NSPCs lacking PCDH19 function might be related to decrease in NSPC polarity. To address this, we initially identified markers of polarity in our cultures. As with previous reports (42, 44), immunofluorescent staining of cultures with antibodies towards N-CADHERIN and AFADIN (AF6), reliably identified the apically localised cell-

CE

cell-adhesion structures (demarcating foci and rosette structure lumens), which were also identified with the filamentous actin stain Phalliodin (Figure 6B). In an initial attempt to

AC

capture any overt alteration to polarity, we compared the expression of the apical marker NCADHERIN mRNA across the culture types. N-CADHERIN, was slightly decreased (albeit not significantly) in both the MUT and Mosaic cultures compared to the WT (Figure 6C). Phalloidin stain was used next in a morphometric analysis to more accurately compare polarity between the three culture types. All three culture types (WT, MUT and Mosaic) contained colonies consisting of numerous foci and neural rosette structures (so called neural rosette colonies). No difference in the average rosette colony area was found between the three different culture types (Figure 6D). To identify changes in the ability of the cells to acquire polarity, the density of polarised foci and polarised neural rosette structures

18

ACCEPTED MANUSCRIPT (structures with a clear lumen) were quantified within neural rosette colonies (Figure 6E). Whilst there was no significant difference in the density of foci between the three culture types, there was a significant decrease in the density of neural rosette structures between both WT and MUT (63% decrease, P=0.007) and WT and Mosaic (34.4% decrease, P=0.0038) cultures (Figure 6F and G). The overall proportion of neural rosette structures to foci was also significantly decreased in the MUT cells relative to WT (31.4% decrease, P=0.0014, Figure

PT

6H). This trend was also observed in the Mosaic culture type but did not reach significance. Given the density of foci was not changed between the three culture types, this suggested that the reduction in neural rosettes structure density may be attributed to a reduced ability of

RI

hNSPCs to maintain polarity, rather than a reduced ability to acquire cell polarity.

SC

Collectively, a reduction in the number polarised hNSPCs in neural rosette structures would result, which can be inferred from measuring the lumen size of rosette structures (40). Indeed,

NU

rosette structures in both the MUT and the Mosaic cultures had a significant reduction in the average lumen area relative to WT (45.98% decrease, P=0.0142 and 58.79% decrease, P=0.0044 respectively, Figure 6I). Altogether, these data reveal that PCDH19 promotes the

MA

maintenance hNSPC polarity.

is

a

debilitating

PT E

PCDH19-GCE

D

Discussion

neurodevelopmental regression.

epileptic

disorder,

commonly

associated

with

No effective treatment is currently available for this

disorder, highlighting the need to better understand the roles of PCDH19 function during

CE

human brain development. Expression of PCDH19 in mNSPC populations of the developing brains of mice (13) provided initial evidence that PCDH19 may regulate the function of these

AC

cells, and in this study we extend this finding by showing expression elevated in human NSPCs also. We thus focussed our studies on NSPCs because they represent the earliest neural cell arising during development in which PCDH19 is likely to be required for function. Our major findings are that loss of PCDH19 function causes premature or accelerated neurogenesis of NSPCs, which was associated with a reduction in NSPC polarity. In addition, we show that in mosaic cultures of WT and MUT/KO NSPCs, which model the pathogenic cellular composition in PCDH19-CGE patients, NSPCs behaved similar to cultures of KO cells only. By interrogating PCDH19 function in both mouse and human NSPCs, our approach was designed at revealing conserved functions of PCDH19. In mouse, we used our Pcdh19 KO mouse to isolate mNSPCs from the dorsal cortex to address impact

19

ACCEPTED MANUSCRIPT of loss-of-function. In human, we employed hiPSC reprograming of patient derived fibroblasts, in which loss of function is implicated based on clinical presentation of PCDH19GCE. In a parallel study, we have extensively shown that the p.Asn557Lys mutation in the hiPSC lines carried forward in our studies is indeed loss-of-function using cell-reporter based assays, as well as in silico and protein modelling data (Pham et al, in preparation,(39)). It was intriguing that loss of PCDH19 function severely impacted our reprogramming efficiency,

PT

with strong selection bias during reprogramming in favour of an active WT X-chromosome in hiPSCs derived from female fibroblast cells. It is unclear if PCDH19 has a functional role in reprograming, or if this result is driven by another X-chromosomal feature, as has been

RI

previously reported in reprogramming of female cells, for example from female patients with

SC

(X-linked) MECP2 mutations in Rett Syndrome patients (45). Whilst this hindered our ability to derive female isogenic cell lines, we were able to eventually (after multiple rounds of

NU

reprograming) isolate and fully validate hiPSCs derived from a male individual with hemizygous loss-of-function PCDH19 missense mutation. Given that males with somatic PCDH19 mutation also present with PCDH19-GCE, and the daughter of this individual has

MA

PCDH19-GCE, we used this cell line to model the disorder and PCDH19 loss-of-function.

D

These are the first reported hiPSC lines created to study PCDH19-GCE.

To best compare across mouse and human, we used NSPCs of dorsal cortical identity,

PT E

isolated either by careful dissection (for mouse) or by a directed pluripotent stem cell differentiation protocol (for human). As validation, both mNSPCs and hNSPCs were found to express the cardinal dorsal cortical NSPC marker PAX6. Several lines of evidence supported

CE

a role PCDH19 in the suppression of NSPC differentiation, and neurogenesis (the production of neurons) in particular. Differentiation of mNSPCs lacking Pcdh19 resulted in gene

AC

transcriptional differences, albeit of modest magnitude which is consistent with subtle nature of the KO mouse phenotype (13). Nonetheless, gene ontology and pathway analysis of differentially expressed genes highlighted processes involved in the differentiation and maturation of neurons. This included telencephalic cell migration (which we have previously shown is accelerated in the Pcdh19 KO neurons (13)), as well as synaptic transmission and axiogenesis (for which protocadherins are known to be involved (46-48)). In particular, genes with known roles in neuronal differentiation of NSPCs, and neuronal maturation were primarily upregulated in the KO cultures, and downregulated in the WT cultures. More indepth analysis of these transcriptomic differences using RNAseq technologies may provide additional support and mechanisms underpinning the premature neurogenesis phenomena, for 20

ACCEPTED MANUSCRIPT example enabling the interrogation non-coding RNAs such as miRNAs which are known to also regulate neurogenesis (49). Encouraged by our transcriptomic analysis we directly quantified cell types present in differentiated cultures via immunofluorescence and identified that loss of Pcdh19 increased neurogenesis of mNSPCs. This phenomenon was conserved (in general) in hNSPCs, where again, both marker gene expression and quantification of neuronal cell types provided evidence of significantly increased neurogenesis. In further

PT

support, neurons in these cultures were also found to be at a more advanced stage of maturation (i.e. as reported by VGLUT1, NEUROD6 and GRIN1 expression and neurite growth) compared to their WT counterparts. Our findings that loss of PCDH19 function

RI

increased neurogenesis from both mouse and human cortical NSPCs are aligned with recent

SC

studies on mir-484, shown to promote mNSPC differentiation specifically through inhibition of Pcdh19 expression in mouse cortices (50). Thus we propose that these functions of

NU

PCDH19 in early NSPC populations may contribute to pathology in PCDH19-CGE by altering the developmental kinetics or trajectory of NSPCs (see below). It also very likely that functions of PCDH19 in other cell types, and at different developmental times will also be

MA

relevant to pathology. For example, whilst not studied for Pcdh19 directly, it is well established that protocadherins play critical roles in axon guidance and bundling, synaptic

D

recognition and connectivity, and overall neuronal circuitry. Conditional deletion models will be required to fully dissect the role of PCDH19 across different developmental processes, and

PT E

to best identify the dominant drivers of pathology in PCDH19-GCE.

During hiPSC neural differentiation, the expression of PCDH19 was significantly

CE

upregulated at the neural rosette stage compared to preceding neuroepithelial sheet stage. Intriguingly, in the absence of functional PCDH19 protein, we observed a significant up-

AC

regulation of PCDH19 mRNA (compered to WT cultures) specifically at this rosette stage of culture, indicative of a compensatory feedback mechanism. A major difference between hNSPCs at the neural rosette stage compared to the neuroepithelial stage is the establishment and maintenance of overt apical-basal polarity. The defining feature of polarised epithelia is the discrimination of apical from basolateral membranes, which is demarcated by the presence of cell-cell adhesion complexes including adherens and tight junctions. Pcdh19 has been shown to be localised to these structures in NSPCs, both during mouse cortical development in vivo (13), and in neural rosettes in vitro (51), and furthermore, has been shown to interact with N-cadherin (15), the major cell-cell adherens junctional protein in NSPCs. Our investigation on cells lacking functional PCDH19 indeed identified it as a 21

ACCEPTED MANUSCRIPT positive regulator of NSPC polarity. Specifically, cultures of hNSPCs lacking functional PCDH19 had significantly reduced densities of rosettes structures, and rosette structures themselves had reduced luminal size, all together indicative of fewer numbers of polarised NSPCs (42, 43). That MUT cultures had normal density of the polarity initiating ‘foci’ structures, suggested that PCDH19 might regulate maintenance rather than establishment of polarity, although this requires further interrogation. The dependency of NSPC self-renewal

PT

(i.e. proliferation in the undifferentiated state) on apical-basal polarity is well established but the molecular mechanisms are not yet fully understood. A classic illustration of this phenomena is provided by studies of mice lacking N-cadherin, where NSPC apical-basal

RI

polarity is lost, and is associated with cell-cycle exit, premature neuronal differentiation, and

SC

accelerated neuronal migration (52-54). Similar results have been observed for deletion of other adhesion junction proteins (e.g. β-catenin (55), AF6 (56), αE-Cadherin (57)) and

NU

polarity proteins (e.g. Par3 (58), aPKC (59), Cdc42 (60)). Our data provide further support, whereby hNSPCs lacking functional PCDH19 display both reduced polarity associated with evidence of reduced hNSPC numbers (e.g. reduced Pax6 expression) and premature

MA

neurogenesis. Whether PCDH19’s impact on polarity is the driver of altered neurogenesis (or vice versa) remains to be determined, but PCDH19 is a cell-cell adhesion molecule and

D

interactors of PCDH19, including N-cadherin and components of the actin remodelling wave regulatory complex (NAP1 and CYFIP2), are known regulators of NSPC polarity (61) (62).

PT E

Furthermore, our transcriptomic analysis identified a class of DE genes with roles in cell-cell adhesion and cadherin signalling including Ctnna2, Vcl and Pvrl3 which all participate in signalling from adhesion sites to the cytoskeleton (63-67). Collectively, these data support

CE

the involvement of PCDH19 in the promotion of NSPC polarity, with further interrogation

AC

required to elucidate the molecular mechanism.

The major motivation of our study was to better understand the pathogenic mechanisms of PCDH19-GCE, and as such the principle challenge is to reconcile our data with the modes of PCDH19-GCE inheritance, in which only heterozygous females are affected, and hemizygous males are (largely) unaffected. This provides two posits for models of pathogenesis: (1) that any phenotype derived from KO cells should be considered as tolerated in normal human development, and (2) that mosaic cultures would be expected to have a phenotype that is distinct from WT and KO. Relating to the first posit, we propose that differences in NSPC behaviour (WT and MUT/KO) would indeed be tolerated, but only so long as cell behaviour (i.e. neurogenesis rate) is uniform / synchronised during development (i.e. as in either WT or 22

ACCEPTED MANUSCRIPT MUT/KO individuals). It is noteworthy that behavioural disturbances in some ‘carrier’ male individuals have been reported (6). In heterozygous individuals however, we postulate that the cellular mosaicism causes non-uniform / asynchronous NSPC behaviour, which can be viewed as a cellular heterochrony (differencing rates of development) of neurogenesis. The sequential order of neurogenesis is important for the formation of complex neural circuits with temporal and spatial precision (68). That the timing of birth of neurons dictates their

PT

identity, spatial location and downstream network connectivity and circuitry suggests that heterochrony of neurogenesis could result in abnormal cellular interactions and neuronal networks, predisposing the individual to network dysfunction and epileptic activity. This

RI

model is complementary to the proposed “cellular interference model’ which has been

SC

postulated to underlie PCDH19-GCE in that cellular heterochrony provides a mechanism by which the establishment of the abnormal neuronal cell interactions can arise. The model

NU

might also in-part explain why the Pcdh19 heterozygous mice lack the overt epileptic phenotype, as the modest changes in neurogenesis we observe might not have large impact over the developmental trajectory of NSPCs during the relatively short mouse neurogenic

MA

period (~8 days, E11-18), compared to the protracted neurogenic period in human (~150 days, GW8-30). Our data was however more challenging to reconcile with the second posit;

D

that mosaic cultures would be expected to have a phenotype that is distinct from WT and KO. Interestingly, in the majority of our assays (including transcriptomics, marker gene

PT E

expression, cell quantification and morphometric analysis), the mosaic cultures of NSPCs showed a phenotype similar to cultures consisting only of MUT/KO cells. We excluded any strong effect of ‘culture skewing’ towards mutant cells within mosaic cultures using X-

CE

Chromosome inactivation assays (data not shown), which were in the normal range, and consistent with random skewing seen in patients. Alternatively, PCDH19 may act non-cell

AC

autonomously to instruct the behaviour of neighbouring cells, for example, as is the case for other transmembrane protein based signalling pathways including Notch and Ephrin (69, 70). Cell tracing approaches (e.g. differential labelling of WT and KO) will be required to address this in future studies, preferably in vivo, or in NSPCs cultured at high density so as maintaining cell-cell contacts throughout differentiation. Interestingly, our initial mNSPC labelling experiment of dissociated single cell and low-density differentiation mosaic cultures revealed Pcdh19 can act cell-autonomously. However, this is not mutually exclusive with non-cell autonomous mechanism, and it is highly likely given the function of Pcdh19 in cellcell adhesion and communication that both can and will operate. Regardless of the mechanism, why Mosaic cultures behaved as KO cultures, and as such how this could be 23

ACCEPTED MANUSCRIPT pathogenic is puzzling, but perhaps relates to limitations of the in vitro approach we undertook which isolates cells from their native cellular architecture and environment. Indeed, in vivo NSPCs reside within specialised cellular environments, so-called stem cell niche’s, in which NSPCs are exposed to a milieu of instructive signalling events and behavioural cues (71). Our in vitro study design limits our ability to resolve the impact the native NSPC niche has on the behaviour of cells, the absence of which may have masked

PT

non-cell autonomous effects which potentially explain why the Mosaic cultures behaved as KO. As a case in point, during this study an additional Pcdh19 mouse model was established that enabled identification of KO and WT cells during mouse brain development, and showed

RI

segregation (or sorting) of KO and WT mNSPCs into contiguous blocks of only WT, or only

SC

KO cells just prior to, and during, neurogenesis (12). Our isolation of cells for in vitro analysis destroys this cellular niche architecture. Furthermore we were unable to recapitulate

NU

this sorting phenomena when WT and KO mNSPCs were fluorescently labelled and cultured in vitro, and instead found KO and WT cells mixed extensively with one-another and showed no signs of segregation (data not shown). The differences between the spatial arrangement of

MA

KO and WT NSPCs in their native in vivo niche and in vitro might be important and potentially contribute to in vitro artefact. For example, if indeed MUT/KO hNSPCs acts non-

D

cell autonomously to instruct the behaviour of neighbouring WT hNSPCs as we speculate above, this would have large impact in vitro, (where KO and WT cells often neighbour each-

PT E

other), whilst in vivo, this effect would be negligible (as vast majority of WT and KO cells are spatially separated). Ultimately, it will be most valuable to validate our model of cellular heterochrony of neurogenesis in vivo using this newly developed mouse in future endeavours,

CE

in which one would expect the segregated contiguous blocks of WT or KO NSPCs to be

AC

undergoing neurogenesis at different rates or times.

Conclusion

Taken together this study points to a mechanism whereby temporal asynchrony (cellular heterochrony) of WT and MUT/KO NSPC differentiation when present in the same individual could contribute to the pathogenesis of PCDH19-GCE. Normal cortical brain development occurs if the population of NSPCs is undergoing homogeneous synchronous differentiation (whether it be wildtype or slightly premature), whilst pathogenesis can occur when mosaic populations of cells undergo differentiation at different rates. This is not likely to be the only mechanism underlying PCDH19-GCE development, as recent studies have also shown Pcdh19 provides a highly specific combinatorial adhesion-code during cortical 24

ACCEPTED MANUSCRIPT development leading to cellular sorting of Pcdh19 KO and Pcdh19 WT cells in the mouse cortex (12). At a fundamental level, we have shown that WT and MUT/KO cells are essentially two different cell types as evidenced both at the transcript level and through differences in cellular behaviours (neuronal differentiation and migration). Thus this “cellular sorting” of WT and MUT/KO neuronal cells, through both cell autonomous effects (asynchrony of differentiation) and non-cell autonomous effects (altered cell-cell

PT

interactions) during cortical development is likely to result in the establishment of diverse neuronal networks. We can thus surmise PCDH19-GCE pathogenesis arises from the altered cortical development, converging on the role of X-inactivation leading to mosaicism and the

SC

RI

fundamental differences in the WT and MUT/KO neuronal cells.

Acknowledgements

NU

This work was supported by grants from the National Health and Medical Research Council of Australia: Program Grant (628952) and Research Fellowship (1041920) for J.G; Australian Research Council (DE160100620 to L.A.J); C.C.H was supported by an Australian

MA

Postgraduate Award from the Australian Government. The Authors would like to thank: Daniel Pederick for help with breeding of the Pcdh19 KO mouse colony; Patrick Fortuna and

D

Jane Sun with initial assistance with culture of hiPSCs; and Martin Lewis for guidance with

PT E

RT-qPCR.

References

AC

CE

1 Specchio, N., Marini, C., Terracciano, A., Mei, D., Trivisano, M., Sicca, F., Fusco, L., Cusmai, R., Darra, F., Bernardina, B.D. et al. (2011) Spectrum of phenotypes in female patients with epilepsy due to protocadherin 19 mutations. Epilepsia, 52, 1251-1257. 2 Duszyc, K., Terczynska, I. and Hoffman-Zacharska, D. (2015) Epilepsy and mental retardation restricted to females: X-linked epileptic infantile encephalopathy of unusual inheritance. J Appl Genet, 56, 49-56. 3 Tan, C., Shard, C., Ranieri, E., Hynes, K., Pham, D.H., Leach, D., Buchanan, G., Corbett, M., Shoubridge, C., Kumar, R. et al. (2015) Mutations of protocadherin 19 in female epilepsy (PCDH19FE) lead to allopregnanolone deficiency. Hum Mol Genet, 24, 5250-5259. 4 Wolverton, T. and Lalande, M. (2001) Identification and characterization of three members of a novel subclass of protocadherins. Genomics, 76, 66-72. 5 Dibbens, L.M., Tarpey, P.S., Hynes, K., Bayly, M.A., Scheffer, I.E., Smith, R., Bomar, J., Sutton, E., Vandeleur, L., Shoubridge, C. et al. (2008) X-linked protocadherin 19 mutations cause femalelimited epilepsy and cognitive impairment. Nat Genet, 40, 776-781. 6 van Harssel, J.J., Weckhuysen, S., van Kempen, M.J., Hardies, K., Verbeek, N.E., de Kovel, C.G., Gunning, W.B., van Daalen, E., de Jonge, M.V., Jansen, A.C. et al. (2013) Clinical and genetic aspects of PCDH19-related epilepsy syndromes and the possible role of PCDH19 mutations in males with autism spectrum disorders. Neurogenetics, 14, 23-34.

25

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

7 Cotton, A.M., Price, E.M., Jones, M.J., Balaton, B.P., Kobor, M.S. and Brown, C.J. (2015) Landscape of DNA methylation on the X chromosome reflects CpG density, functional chromatin state and X-chromosome inactivation. Hum Mol Genet, 24, 1528-1539. 8 Depienne, C., Bouteiller, D., Keren, B., Cheuret, E., Poirier, K., Trouillard, O., Benyahia, B., Quelin, C., Carpentier, W., Julia, S. et al. (2009) Sporadic infantile epileptic encephalopathy caused by mutations in PCDH19 resembles Dravet syndrome but mainly affects females. PLoS Genet, 5, e1000381. 9 Terracciano, A., Trivisano, M., Cusmai, R., De Palma, L., Fusco, L., Compagnucci, C., Bertini, E., Vigevano, F. and Specchio, N. (2016) PCDH19-related epilepsy in two mosaic male patients. Epilepsia, 57, e51-55. 10 Thiffault, I., Farrow, E., Smith, L., Lowry, J., Zellmer, L., Black, B., Abdelmoity, A., Miller, N., Soden, S. and Saunders, C. (2016) PCDH19-related epileptic encephalopathy in a male mosaic for a truncating variant. American journal of medical genetics. Part A, 170, 1585-1589. 11 Hayashi, S., Inoue, Y., Hattori, S., Kaneko, M., Shioi, G., Miyakawa, T. and Takeichi, M. (2017) Loss of X-linked Protocadherin-19 differentially affects the behavior of heterozygous female and hemizygous male mice. Sci Rep, 7, 5801. 12 Pederick, D., Richards, K., Piltz, S., Mandelstam, S., Dale, R., Scheffer, I., Gecz, J., Petrou, S., Hughes, J. and Thomas, P. (2017) Abnormal cell sorting underlies the unique X-linked inheritance of PCDH19 Epilepsy. bioRxiv, in press. 13 Pederick, D.T., Homan, C.C., Jaehne, E.J., Piltz, S.G., Haines, B.P., Baune, B.T., Jolly, L.A., Hughes, J.N., Gecz, J. and Thomas, P.Q. (2016) Pcdh19 Loss-of-Function Increases Neuronal Migration In Vitro but is Dispensable for Brain Development in Mice. Sci Rep, 6, 26765. 14 Cooper, S.R., Emond, M.R., Duy, P.Q., Liebau, B.G., Wolman, M.A. and Jontes, J.D. (2015) Protocadherins control the modular assembly of neuronal columns in the zebrafish optic tectum. J Cell Biol, 211, 807-814. 15 Biswas, S., Emond, M.R. and Jontes, J.D. (2010) Protocadherin-19 and N-cadherin interact to control cell movements during anterior neurulation. J Cell Biol, 191, 1029-1041. 16 Higurashi, N., Nakamura, M., Sugai, M., Ohfu, M., Sakauchi, M., Sugawara, Y., Nakamura, K., Kato, M., Usui, D., Mogami, Y. et al. (2013) PCDH19-related female-limited epilepsy: further details regarding early clinical features and therapeutic efficacy. Epilepsy Res, 106, 191-199. 17 Luebbig, A., Pieper, T., Kolodzieczyk, D., Kudernatsch, M., Bluemcke, I., Holinksi-Feder, E., Badinger, A., Karlmeier, A., Holthausen, H. and Staudt, M. (2013) Coincidence of duplication in the PCDH 19 gene and focal cortical dysplasia. Neuropediatrics, 44, PS22_1046. 18 Ryan, S.G., Chance, P.F., Zou, C.H., Spinner, N.B., Golden, J.A. and Smietana, S. (1997) Epilepsy and mental retardation limited to females: an X-linked dominant disorder with male sparing. Nat Genet, 17, 92-95. 19 Vincent, A.K., Noor, A., Janson, A., Minassian, B.A., Ayub, M., Vincent, J.B. and Morel, C.F. (2012) Identification of genomic deletions spanning the PCDH19 gene in two unrelated girls with intellectual disability and seizures. Clinical genetics, 82, 540-545. 20 Kurian, M., Korff, C.M., Ranza, E., Bernasconi, A., Lubbig, A., Nangia, S., Ramelli, G.P., Wohlrab, G., Nordli, D.R., Jr. and Bast, T. (2017) Focal cortical malformations in children with early infantile epilepsy and PCDH19 mutations: case report. Dev Med Child Neurol, in press. 21 Lin, J., Wang, C. and Redies, C. (2012) Expression of delta-protocadherins in the spinal cord of the chicken embryo. J Comp Neurol, 520, 1509-1531. 22 Gaitan, Y. and Bouchard, M. (2006) Expression of the delta-protocadherin gene Pcdh19 in the developing mouse embryo. Gene Expr Patterns, 6, 893-899. 23 Homan, C.C., Kumar, R., Nguyen, L.S., Haan, E., Raymond, F.L., Abidi, F., Raynaud, M., Schwartz, C.E., Wood, S.A., Gecz, J. et al. (2014) Mutations in USP9X are associated with X-linked intellectual disability and disrupt neuronal cell migration and growth. Am J Hum Genet, 94, 470-478.

26

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

24 Weber, K., Bartsch, U., Stocking, C. and Fehse, B. (2008) A multicolor panel of novel lentiviral "gene ontology" (LeGO) vectors for functional gene analysis. Molecular therapy : the journal of the American Society of Gene Therapy, 16, 698-706. 25 Brown, C.Y., Sadlon, T., Gargett, T., Melville, E., Zhang, R., Drabsch, Y., Ling, M., Strathdee, C.A., Gonda, T.J. and Barry, S.C. (2010) Robust, reversible gene knockdown using a single lentiviral short hairpin RNA vector. Human gene therapy, 21, 1005-1017. 26 Jolly, L.A., Homan, C.C., Jacob, R., Barry, S. and Gecz, J. (2013) The UPF3B gene, implicated in intellectual disability, autism, ADHD and childhood onset schizophrenia regulates neural progenitor cell behaviour and neuronal outgrowth. Hum Mol Genet, 22, 4673-4687. 27 Gautier, L., Cope, L., Bolstad, B.M. and Irizarry, R.A. (2004) affy--analysis of Affymetrix GeneChip data at the probe level. Bioinformatics, 20, 307-315. 28 Kapur, K., Xing, Y., Ouyang, Z. and Wong, W.H. (2007) Exon arrays provide accurate assessments of gene expression. Genome Biology, 8, R82. 29 Briggs, J.A., Sun, J., Shepherd, J., Ovchinnikov, D.A., Chung, T.L., Nayler, S.P., Kao, L.P., Morrow, C.A., Thakar, N.Y., Soo, S.Y. et al. (2013) Integration-free induced pluripotent stem cells model genetic and neural developmental features of down syndrome etiology. Stem Cells, 31, 467478. 30 Xu, R.H., Peck, R.M., Li, D.S., Feng, X., Ludwig, T. and Thomson, J.A. (2005) Basic FGF and suppression of BMP signaling sustain undifferentiated proliferation of human ES cells. Nat Methods, 2, 185-190. 31 Yu, J., Hu, K., Smuga-Otto, K., Tian, S., Stewart, R., Slukvin, II and Thomson, J.A. (2009) Human induced pluripotent stem cells free of vector and transgene sequences. Science, 324, 797801. 32 Kopp, P., Jaggi, R., Tobler, A., Borisch, B., Oestreicher, M., Sabacan, L., Jameson, J.L. and Fey, M.F. (1997) Clonal X-inactivation analysis of human tumours using the human androgen receptor gene (HUMARA) polymorphism: a non-radioactive and semiquantitative strategy applicable to fresh and archival tissue. Molecular and cellular probes, 11, 217-228. 33 Shi, Y., Kirwan, P. and Livesey, F.J. (2012a) Directed differentiation of human pluripotent stem cells to cerebral cortex neurons and neural networks. Nat Protoc, 7, 1836-1846. 34 Nayler, S., Gatei, M., Kozlov, S., Gatti, R., Mar, J.C., Wells, C.A., Lavin, M. and Wolvetang, E. (2012) Induced pluripotent stem cells from ataxia-telangiectasia recapitulate the cellular phenotype. Stem Cells Transl Med, 1, 523-535. 35 Thomson, J.A., Itskovitz-Eldor, J., Shapiro, S.S., Waknitz, M.A., Swiergiel, J.J., Marshall, V.S. and Jones, J.M. (1998) Embryonic stem cell lines derived from human blastocysts. Science, 282, 1145-1147. 36 Preibisch, S., Saalfeld, S. and Tomancak, P. (2009) Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics, 25, 1463-1465. 37 Chen, E.Y., Tan, C.M., Kou, Y., Duan, Q., Wang, Z., Meirelles, G.V., Clark, N.R. and Ma’ayan, A. (2013) Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics, 14, 128. 38 Kadowaki, M., Nakamura, S., Machon, O., Krauss, S., Radice, G.L. and Takeichi, M. (2007) Ncadherin mediates cortical organization in the mouse brain. Developmental Biology, 304, 22-33. 39 Pham, D.H., Tan, C., Homan, C.C., Kolc, K., Corbett, M., McAninch, D., Fox, A., Thomas, P., Kumar, R. and Gecz, J. (2017) Protocadherin 19 (PCDH19) interacts with paraspeckle protein NONO to co-regulate gene expression with estrogen receptor alpha (ERalpha). Hum Mol Genet, in press. 40 Edri, R., Yaffe, Y., Ziller, M.J., Mutukula, N., Volkman, R., David, E., Jacob-Hirsch, J., Malcov, H., Levy, C., Rechavi, G. et al. (2015) Analysing human neural stem cell ontogeny by consecutive isolation of Notch active neural progenitors. Nat Commun, 6, 6500. 41 Main, H., Radenkovic, J., Jin, S.B., Lendahl, U. and Andersson, E.R. (2013) Notch signaling maintains neural rosette polarity. PLoS One, 8, e62959.

27

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

42 Elkabetz, Y., Panagiotakos, G., Al Shamy, G., Socci, N.D., Tabar, V. and Studer, L. (2008) Human ES cell-derived neural rosettes reveal a functionally distinct early neural stem cell stage. Genes Dev, 22, 152-165. 43 Banda, E., McKinsey, A., Germain, N., Carter, J., Anderson, N.C. and Grabel, L. (2015) Cell polarity and neurogenesis in embryonic stem cell-derived neural rosettes. Stem Cells Dev, 24, 10221033. 44 Jolly, L.A., Taylor, V. and Wood, S.A. (2009) USP9X enhances the polarity and self-renewal of embryonic stem cell-derived neural progenitors. Molecular biology of the cell, 20, 2015-2029. 45 Pomp, O., Dreesen, O., Leong, D.F., Meller-Pomp, O., Tan, T.T., Zhou, F. and Colman, A. (2011) Unexpected X chromosome skewing during culture and reprogramming of human somatic cells can be alleviated by exogenous telomerase. Cell Stem Cell, 9, 156-165. 46 Tissir, F., Bar, I., Jossin, Y., De Backer, O. and Goffinet, A.M. (2005) Protocadherin Celsr3 is crucial in axonal tract development. Nat Neurosci, 8, 451-457. 47 Hoshina, N., Tanimura, A., Yamasaki, M., Inoue, T., Fukabori, R., Kuroda, T., Yokoyama, K., Tezuka, T., Sagara, H., Hirano, S. et al. (2013) Protocadherin 17 regulates presynaptic assembly in topographic corticobasal Ganglia circuits. Neuron, 78, 839-854. 48 Weiner, J.A., Wang, X., Tapia, J.C. and Sanes, J.R. (2005) Gamma protocadherins are required for synaptic development in the spinal cord. Proc Natl Acad Sci U S A, 102, 8-14. 49 Lang, M.-F. and Shi, Y. (2012) Dynamic Roles of microRNAs in Neurogenesis. Frontiers in Neuroscience, 6, 71. 50 Fujitani, M., Zhang, S., Fujiki, R., Fujihara, Y. and Yamashita, T. (2016) A chromosome 16p13.11 microduplication causes hyperactivity through dysregulation of miR-484/protocadherin-19 signaling. Mol Psychiatry, in press. 51 Compagnucci, C., Petrini, S., Higuraschi, N., Trivisano, M., Specchio, N., Hirose, S., Bertini, E. and Terracciano, A. (2015) Characterizing PCDH19 in human induced pluripotent stem cells (iPSCs) and iPSC-derived developing neurons: emerging role of a protein involved in controlling polarity during neurogenesis. Oncotarget, 6, 26804-26813. 52 Zhang, J., Woodhead, G.J., Swaminathan, S.K., Noles, S.R., McQuinn, E.R., Pisarek, A.J., Stocker, A.M., Mutch, C.A., Funatsu, N. and Chenn, A. (2010) Cortical neural precursors inhibit their own differentiation via N-cadherin maintenance of beta-catenin signaling. Dev Cell, 18, 472-479. 53 Zhang, J., Shemezis, J.R., McQuinn, E.R., Wang, J., Sverdlov, M. and Chenn, A. (2013) AKT activation by N-cadherin regulates beta-catenin signaling and neuronal differentiation during cortical development. Neural Dev, 8, 7. 54 Miyamoto, Y., Sakane, F. and Hashimoto, K. (2015) N-cadherin-based adherens junction regulates the maintenance, proliferation, and differentiation of neural progenitor cells during development. Cell Adhesion & Migration, 9, 183-192. 55 Woodhead, G.J., Mutch, C.A., Olson, E.C. and Chenn, A. (2006) Cell-Autonomous β-Catenin Signaling Regulates Cortical Precursor Proliferation. The Journal of neuroscience : the official journal of the Society for Neuroscience, 26, 12620-12630. 56 Rakotomamonjy, J., Brunner, M., Juschke, C., Zang, K., Huang, E.J., Reichardt, L.F. and Chenn, A. (2017) Afadin controls cell polarization and mitotic spindle orientation in developing cortical radial glia. Neural Dev, 12, 7. 57 Stocker, A.M. and Chenn, A. (2009) Focal Reduction of αE-catenin Causes Premature Differentiation and Reduction of β-catenin Signaling During Cortical Development. Developmental biology, 328, 66-77. 58 Costa, M.R., Wen, G., Lepier, A., Schroeder, T. and Gotz, M. (2008) Par-complex proteins promote proliferative progenitor divisions in the developing mouse cerebral cortex. Development, 135, 11-22. 59 Tischfield, D.J., Kim, J. and Anderson, S.A. (2017) Atypical PKC and Notch Inhibition Differentially Modulate Cortical Interneuron Subclass Fate from Embryonic Stem Cells. Stem Cell Reports, 8, 1135-1143.

28

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

60 Cappello, S., Attardo, A., Wu, X., Iwasato, T., Itohara, S., Wilsch-Brauninger, M., Eilken, H.M., Rieger, M.A., Schroeder, T.T., Huttner, W.B. et al. (2006) The Rho-GTPase cdc42 regulates neural progenitor fate at the apical surface. Nat Neurosci, 9, 1099-1107. 61 Chen, B., Brinkmann, K., Chen, Z., Pak, C.W., Liao, Y., Shi, S., Henry, L., Grishin, N.V., Bogdan, S. and Rosen, M.K. (2014) The WAVE regulatory complex links diverse receptors to the actin cytoskeleton. Cell, 156, 195-207. 62 Yoon, K.J., Nguyen, H.N., Ursini, G., Zhang, F., Kim, N.S., Wen, Z., Makri, G., Nauen, D., Shin, J.H., Park, Y. et al. (2014) Modeling a genetic risk for schizophrenia in iPSCs and mice reveals neural stem cell deficits associated with adherens junctions and polarity. Cell Stem Cell, 15, 79-91. 63 Vite, A., Li, J. and Radice, G.L. (2015) New Functions for Alpha-Catenins in Health and Disease: From Cancer to Heart Regeneration. Cell and tissue research, 360, 773-783. 64 Stocker, A.M. and Chenn, A. (2006) Differential expression of alpha-E-catenin and alpha-Ncatenin in the developing cerebral cortex. Brain Res, 1073-1074, 151-158. 65 Satoh-Horikawa, K., Nakanishi, H., Takahashi, K., Miyahara, M., Nishimura, M., Tachibana, K., Mizoguchi, A. and Takai, Y. (2000) Nectin-3, a new member of immunoglobulin-like cell adhesion molecules that shows homophilic and heterophilic cell-cell adhesion activities. J Biol Chem, 275, 10291-10299. 66 Kurita, S., Yamada, T., Rikitsu, E., Ikeda, W. and Takai, Y. (2013) Binding between the junctional proteins afadin and PLEKHA7 and implication in the formation of adherens junction in epithelial cells. J Biol Chem, 288, 29356-29368. 67 Ishiyama, N., Tanaka, N., Abe, K., Yang, Y.J., Abbas, Y.M., Umitsu, M., Nagar, B., Bueler, S.A., Rubinstein, J.L., Takeichi, M. et al. (2013) An autoinhibited structure of alpha-catenin and its implications for vinculin recruitment to adherens junctions. J Biol Chem, 288, 15913-15925. 68 Pinto-Teixeira, F., Koo, C., Rossi, A.M., Neriec, N., Bertet, C., Li, X., Del-Valle-Rodriguez, A. and Desplan, C. (2018) Development of Concurrent Retinotopic Maps in the Fly Motion Detection Circuit. Cell, 173, 485-498.e411. 69 Schwieterman, A.A., Steves, A.N., Yee, V., Donelson, C.J., Bentley, M.R., Santorella, E.M., Mehlenbacher, T.V., Pital, A., Howard, A.M., Wilson, M.R. et al. (2016) The Caenorhabditis elegans Ephrin EFN-4 Functions Non-cell Autonomously with Heparan Sulfate Proteoglycans to Promote Axon Outgrowth and Branching. Genetics, 202, 639-660. 70 Kawamata, S., Du, C., Li, K. and Lavau, C. (2002) Notch1 Perturbation of Hemopoiesis Involves Non-Cell- Autonomous Modifications. The Journal of Immunology, 168, 1738. 71 Regalado-Santiago, C., Juárez-Aguilar, E., Olivares-Hernández, J.D. and Tamariz, E. (2016) Mimicking Neural Stem Cell Niche by Biocompatible Substrates. Stem Cells International, 2016, 1513285. 72 Huang da, W., Sherman, B.T. and Lempicki, R.A. (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc, 4, 44-57. 73 Huang da, W., Sherman, B.T. and Lempicki, R.A. (2009) Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res, 37, 1-13. 74 Kanehisa, M. and Goto, S. (2000) KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res, 28, 27-30.

29

ACCEPTED MANUSCRIPT Figure Legends:

NU

SC

RI

PT

Figure 1: Loss of Pcdh19 enhances neurogenesis of mNSPCs. A) Schematic of the experimental approach used to create cultures of Pcdh19 wildtype (WT), Knockout (KO) and mosaic cultures of mNSPCs. B) Principal component analysis (PCA) of gene expression profiles of undifferentiated and differentiated cultures of mNSPCs (n=3 cultures per type for both undifferentiated and differentiated states). Dashed circles show the clustering of WT away from the Mosaic and KO cultures in the differentiated mNSPC populations. C) Identification of genes displaying differential expression between the undifferentiated and differentiated mNSPC states, and their comparison between WT and KO cultures: 761 differentially expressed genes were shared, and 302 DE genes were unique between the WT and KO cultures. D and E) The 302 DE genes unique to WT and KO mNSPC cultures were subjected to in-silico analysis to derive (D) enriched gene ontology terms for biological processes (DAVID(72, 73)) and (E) enriched pathways (KEGG(74)). F) Expression changes of unique DE genes with highly relevant roles in mNSPC differentiation and neuronal maturation. G and H). G. Representative immunofluorescent images used to quantitate percentage of neurons in culture using antibodies directed towards the neuronal specific βIIItubulin. H. Quantification of neural cell types. Graph values represent the mean values of n≥3 biological replicates per culture type. Error bars=SEM. * p<0.05, Student’s T-test.

CE

PT E

D

MA

Figure 2: Generation of hiPSCs expressing pathogenic PCDH19 alleles. A) Schematic of the pipeline used to reprogram fibroblasts into iPSCs. Fibroblasts are derived from both females with PCDH19-CGE, and a male with hemizygous pathogenic PCDH19 mutation. B) Summary of results from the reprogramming experiments. Note that karyotypically normal iPSCs expressing mutant PCDH19 alleles could only be derived from fibroblasts of the male donor (Individual 3). C-D) Characterisation of male iPSC clones. C) Representative immunofluorescence images hiPSC clone stained with antibodies against specific pluripotency marker proteins NANOG and OCT4. D) Representative metaphase spread showing normal karyotype; 46,XY. E) Representative hematoxylin and eosin stained sections of teratomas derived from male iPSC clones indicating the presence of tissue derivatives of the three embryonic germ layers (mesoderm: hyaline cartilage; endoderm: columnar epithelium; and ectoderm: pigmented neural epithelium).

AC

Figure 3: PCDH19 is highly expressed in iPSC derived hNSPCs. A) Immunofluorescences images showing the major developmental stages of the iPSC cortical differentiation protocol. Cultures were stained with antibodies against proteins specific for dorsal cortical hNSPC (PAX6, NESTIN), cell polarity markers (AF6 and N-CADHERIN) and neurons (NEUN, βIII-Tubulin, DCX and MAP2). B) RT-qPCR analysis of PCDH19 mRNA expression across stages of cortical differentiation. Data from differentiation of both human embryonic stem cells (hESCs) and hiPSCs. Data combined from n=8 independent differentiation experiments. C) PCDH19 mRNA expression is upregulated in PCDH19 mutant and mosaic cultures specifically at the neural rosette stage of cortical differentiation. iPSCs derived occurs only at the neural rosette stage of cortical differentiation. Error bars=SEM. * P< 0.05, ** P<0.01. by ordinary one-way ANOVA with Turkey’s multiple comparisons. Data combined from n=5 independent differentiation experiments per culture type (WT, MUT and Mosaic).

30

ACCEPTED MANUSCRIPT Figure 4: Profiling PCDH19 hiPSC neuronal differentiation with cell-specific marker gene expression. A-D) RT-qPCR analysis of mRNA expression of cell type specific markers during hiPSC neural differentiation. A) Neuroepithelial sheet stage. B) Neural rosette stage. C) Day 70 neurons stage. D) Day 90 neurons stage. n=3 independent experiments / culture type. * P< 0.05, ** P<0.01 by RM One-way ANOVA with the Greenhouse-Gessier correction and a Turkey’s multiple comparison test. Graph values are the mean ± SEM.

NU

SC

RI

PT

Figure 5: Pathogenic PCDH19 mutation results in premature hNSPC neuronal differentiation. A-B) Increased neuronal cell numbers in MUT and Mosaic cultures of hNSPCs at neural rosette stage of differentiation. A) Representative images of cultures stained with neuronal specific antibody βIII-Tubulin (Green), Phalloidin (Red), and DAPI (Blue). B) Quantification of the mean number of βIII-tubulin positive cells around the periphery of the neural rosette colonies relative to the area. n>10 replicates per genotype across 3 biological experiments. (C-D) Primary neurite length of neurons is increased in MUT and Mosaic cultures. C) Representative immunofluorescent images of neurons (β-III Tubulin (Green), DAPI (Blue)). D) Quantitation of the average primary neurite length. n>329 neurons scored across 5 biological replicates per culture type. * P< 0.05, ** P<0.01 by RM One-way ANOVA with the Greenhouse-Gessier correction and a Turkey’s multiple comparison test. Graph values are the mean ± SEM.

AC

CE

PT E

D

MA

Figure 6: Pathogenic PCDH19 mutation yields reduced polarity in hNSPCs. A) Schematic depicting the formation and disassembly of neural rosette structures as differentiation proceeds. B) Representative immunofluorescence highlighting co-localisation of enriched phalloidin staining with cell adhesion marker proteins N-CADHERIN (Green) and AFADIN (AF6: Cyan). Scale bar is 50μm. C) RT-qPCR analysis of mRNA expression of the NSPC cell adhesion / polarity marker gene N-CADHERIN. Error bars=SEM. n=5 independent experiments/culture type. D) Average area of neural rosette colonies is consistent between WT, MUT and Mosaic cultures. Error bars=SEM. n>10 replicates per culture type across 3 biological experiments. * P< 0.05, ** P<0.01 by ordinary one-way ANOVA with Turkey’s multiple comparisons test. E) Representative immunofluorescent images of neural rosette colonies; Phalloidin (Red), DAPI (Blue). Box shows the enlarged image indicating the classification of neural rosette structure and foci. F). Quantification of the mean density of neural rosette structures in neural rosette colonies C) Quantification of the mean density of foci in neural rosette colonies D) Proportion of neural rosette structures to foci in the neural rosette colonies. Experiments F-G preformed in biological duplicate and analysed in technical triplicate (n>10 rosette colonies per culture type per triplicate). E) Quantification of the mean total area of the central lumen (polarised structures) relative to the area of the rosette colony. Experiment preformed in biological triplicate (n >13 rosette colonies per culture type per triplicate). Graph values represent the mean ± SEM. * P< 0.05, ** P<0.01 by One-way ANOVA with Greenhouse-Gessier correction and a Turkey’s multiple comparison test. n≥60 neural rosette colonies for each culture type.

31

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6