Rojasringeling-Dissertation

Molecular regulation of mRNA stability and translation

Francisca Rojas Ringeling, MD

A dissertation submitted to Johns Hopkins University

in conformity with the requirements for the degree of

Doctor of Philosophy

Baltimore, Maryland

August 2017

c Francisca Rojas Ringeling, MD

The ﬂow of information from DNA to RNA to protein is a tightly regulated process, which ultimately determines the functional properties that each cell will possess. Defects in any of the multiple regulatory mechanisms that ensure that proper amounts of each protein are produced within a particular cell, may lead to dysregulation of cellular processes and disease.

This dissertation will deal with 2 different projects. The common theme between these projects is that we studied post-transcriptional regulation of gene expression at the level of messenger RNA (mRNA), and in both studies we took advantage of genome-wide sequencing techniques to develop insights and novel hypothesis within the realm of neurobiology.

Briefly, the first project is a study of the role of Cyfip1, a neuropsychiatric disease risk gene, in regulating the translation of its mRNA partners. We show that deletion and over-expression of Cyfip1 in the mouse brain, leads to diametric changes in protein translation of NMDAR subunits and postsynaptic components, and has consequences in behavior within these mouse models. The second study is an exploration of the role of m6A epitranscriptional modification in the regulation of cortical neurogenesis in-vivo, where we identified a critical and conserved role of m6A in the temporal control of mammalian cortical neurogenesis.

Advisor: Hongjun Song, PhD

Reader: Guoli Ming, MD, PhD

ii Acknowledgments

I could not have written this thesis without the help of many people. Most notably,

I thank my advisor Hongjun Song for his guidance and helpful advice throughout my doctoral studies. Hongjun has been a supportive and ﬂexible mentor, and he encouraged me to learn new things and go beyond my comfort zone.

I would also like to acknowledge Kijun Yoon, who has worked closely with me for the past 4 years. Kijun taught me almost everything I know about bench work

(which was very close to zero when I started), and he was always patient and kind with me. I also learned a lot from him on how to develop scientiﬁc hypothesis and think critically about experimental designs, not to mention many fun discussions about movies, music, and many other topics.

All the members of the Song Ming Lab have contributed to this thesis in one way or another. I really enjoyed working in our sometimes hectic, but always fun environment.

I would like to thank the faculty of the Human Genetics Pre-doctoral Training

Program, speciﬁcally Dr. David Valle for awarding me this wonderful opportunity.

Also our program administrator Sandy Muscelli for all the work she does for us students.

iii I am also much obliged to my thesis committee members Paul Worley, Guoli

Ming and Andrew McCallion for their thoughtful comments and fruitful discussions.

I would also like to express gratitude to my friends and colleagues who contributed to this thesis in many ways: Jessica Cassin, Caroline Vissers, Leire Abalde-

Atrista´ın, Stephanie Temme, Caroline Siebald, Kai Kammers, Ursula Smole, Juan

Calderon´ Giadrosic, Nam Nguyen, Ji Young Park, Dennisse Cyrus-Jimenez, Xinyuan

Wang, and many others. You made Baltimore my home, and I feel very lucky to be able to call you my friends.

I am deeply grateful to my wonderful parents and sister, who have always been supportive about everything I decide to do. I couldn’t be here if it weren’t for your relentless support.

Finally, I want to thank my husband Stefan. I can’t express in words how lucky

I am to have you. Thank you for being my partner in every aspect of our amazingly fun life.

iv Contents

1 Introduction 1

2 Cyﬁp1 4

2.1 Abstract ...... 4

2.2 Introduction ...... 6

2.3 Methods ...... 8

2.3.1 Experimental model and subject details ...... 8

2.3.2 Method details ...... 9

2.3.3 Behavior studies ...... 13

2.4 Results ...... 17

2.4.1 Deletion of Cyﬁp1 results in behavioral abnormalities re-

lated with schizophrenia ...... 17

2.4.2 Increased Cyﬁp1 dosage leads to ASD-like behavioral ab-

normalities ...... 19

iv 2.4.3 Cyﬁp1 interacts with mRNAs of synaptic and NMDA re-

ceptor related proteins in mouse hippocampus and human

cerebral cortex...... 22

2.4.4 Cyﬁp1 directly regulates mRNA translation of NMDAR

subunits and postsynaptic components ...... 26

2.4.5 Protein expression of NMDAR subunits and postsynaptic

components is altered within synaptosomes depending on

Cyﬁp1 dosages ...... 28

2.4.6 Bidirectional modulation of NMDAR signaling rescues be-

havioral abnormalities in cKO and cOE mice ...... 31

2.5 Discussion ...... 34

2.5.1 Cyﬁp1 mRNA targets are related to synaptic function and

neuropsychiatric diseases ...... 35

2.5.2 Diametric gene dosage effects of CNVs in neurodevelop-

mental diseases ...... 35

3 m6A controls cortical neurogenesis 37

3.1 Abstract ...... 37

3.2 Introduction ...... 39

3.3 Methods ...... 41

3.3.1 Experimental model and subject details ...... 41

3.3.2 Method details ...... 43

3.4 Results ...... 56

3.4.1 Nervous system Mettl14 deletion extends cortical neuroge-

nesis into postnatal stages ...... 56

3.4.2 Mettl14 deletion in neural progenitor cells leads to pro-

tracted cell cycle progression...... 59

3.4.3 Mettl3 regulates embryonic cortical neurogenesis . . . . . 61

v 3.4.4 m6A tags transcripts related to transcription factors, cell

cycle, and neurogenesis, and promotes their decay . . . . 64

3.4.5 Mettl14 deletion uncovers transcriptional pre-patterning for

normal cortical neurogenesis ...... 66

3.4.6 METTL14 regulates cell cycle progression of human corti-

cal NPCs ...... 71

3.4.7 m6A-seq of human forebrain brain organoids and fetal brain

reveals conserved and unique m6A landscape features com-

pared to mouse ...... 73

3.5 Discussion ...... 76

3.5.1 Transcriptional pre-patterning for cortical neurogenesis . . 76

3.5.2 Heightened transcriptional coordination of mammalian cor-

tical neurogenesis by m6A ...... 77

3.5.3 Conserved and unique features of human m6A landscape

during cortical neurogenesis ...... 78

A Supplemental Figures: Cyﬁp1 80

B Supplemental Figures: m6A 86

Curriculum 114

vi List of Figures

2.1 Cyﬁp1 mouse models...... 18

2.2 Cyﬁp1 cKO shows behavior abnormalities related to schizophrenia. 20

2.3 Cyﬁp1 cOE mice show ASD-like behavioral abnormalities. . . . . 23

2.4 Cyﬁp1 RIP-seq representative coverage plots...... 24

2.5 Gene ontology and disease ontology analysis of Cyﬁp1 target mR-

NAs...... 25

2.6 Conﬁrmation of CYFIP1 mRNA targets in mouse hippocampus

and human cerebral cortex...... 27

2.7 Cyﬁp1 regulates translation of NMDAR subunits and postsynaptic

components...... 29

2.8 Altered protein expression of NMDAR subunits and postsynaptic

components in synapses with differential Cyﬁp1 dosages...... 30

2.9 Memantine treatment in cKO rescued behavioral abnormalities. . . 32

2.10 D-cycloserin treatment in cOE rescued behavioral abnormalities. . 33

3.1 Nervous system Mettl14 deletion results in residual radial glia cells

and ongoing neurogenesis in the postnatal mouse cortex...... 57

vii 3.2 Mettl14−/− RGCs and NPCs exhibit prolonged cell cycle progres-

sion...... 60

3.3 Mettl14 cKO leads to depletion of m6A...... 62

3.4 Mettl3 regulates cell cycle progression of NPCs and maintenance

of embryonic cortical RGCs...... 63

3.5 m6A tags transcripts related to transcription factors, cell cycle, and

neuronal differentiation in the embryonic mouse brain...... 65

3.6 m6A promotes mRNA decay...... 67

3.7 Post-transcriptional regulation of pre-patterning gene levels by m6A

methylation in cortical neural stem cells ...... 68

3.8 Regulation of protein production of pre-patterning genes by m6A

methylation in cortical neural stem cells...... 70

3.9 METTL14 regulates cell cycle progression of human NPCs. . . . . 72

3.10 Conserved and unique features of m6A mRNA methylation in hu-

man forebrain organoids, human fetal brain and embryonic mouse

forebrain...... 75

A.1 Normal locomotor activity, motor coordination, nociception re-

sponse and repetitive behaviors of Cyﬁp1 cKO mice...... 81

A.2 Normal locomotor activity, novel object recognition, behavioral

despair and sensorimotor gating of Cyﬁp1 cOE mice, and impaired

maternal care of cOE mice...... 82

A.3 Cyﬁp1 RIP seq strategy...... 83

A.4 Bioinformatic analysis of Cyﬁp1 RIP seq experiment ...... 84

A.5 mRNA levels of Cyﬁp1 targets are unchanged in cKO and cOE mice. 85

B.1 Nervous system Mettl14 deletion in mice results in postnatal lethality. 87

viii B.2 Nervous system Mettl14 deletion in mice results in deﬁcits in timely

production of cortical neuron subtypes...... 88

B.3 Flow cytometry analysis reveals delayed cell cycle progression of

Mettl14−/− NPCs...... 89

B.4 Mettl3 is essential for m6A mRNA methylation and proper cell

cycle progression of mouse NPCs ...... 90

B.5 m6A-seq analysis of mouse embryonic forebrain...... 91

B.6 Expression of neuronal genes in RGCs of embryonic cortex in vivo. 92

B.7 Mettl14 regulates cell cycle progression of hNPCs...... 93

B.8 Comparison of m6A mRNA landscapes among human forebrain

organoids, fetal brain and mouse embryonic forebrain...... 94

ix CHAPTER 1

Introduction

The flow of information from DNA to RNA to protein is a tightly regulated process, which ultimately determines the functional properties that each cell will possess. Thus, it makes sense for the cell to have multiple levels of regulation of this flow of information, such as transcriptional regulation, differential isoform us- age, regulation by epitranscriptional modifications, regulation of messenger RNA

(mRNA) degradation and localization, and regulation at the level of protein translation. When any of these regulatory mechanisms goes astray, it can lead to dysregulation of cellular processes and disease.

The past couple of decades have seen a substantial improvement in our abil- ity to study the regulatory mechanisms involved in gene expression, owing mainly to the rapid advances in sequencing technologies and computational approaches to genomic data analysis. These technological advances have shed light into the highly complex and multidimensional mechanisms that have evolved to ensure the

1 precise and dynamic regulation of gene expression in mammalian systems.

The work that will be presented in this dissertation is composed of 2 sepa- rate projects. The common theme between these 2 projects, is that we looked at post transcriptional regulation of gene expression at the level of messenger RNA

(mRNA) using genome-wide approaches that led us to novel hypothesis within the realm of neurobiology.

The first project is a study of the role of CYFIP1, a neuropsychiatric disease risk gene, in the regulation of translation of its mRNA partners. Deletions and duplications encompassing CYFIP1 gene have been associated with neurodevelopmental disorders such as autism spectrum disorder (ASD), schizophrenia (SCZ), intellectual disability and epilepsy [13, 35, 33, 99]. We generated 2 mouse models, a conditional knock out of Cyfip1 which deletes Cyfip1 within the mouse brain, and a conditional over-expression which has a 1.5 fold higher expression of Cy-

fip1 within the mouse brain, and explored the behavioral consequences of these dosage imbalances of Cyfip1. We then took a genome-wide approach to identify the mRNAs that are bound by Cyfip1 within the mouse hippocampus, and further explored the consequences of Cyfip1 dosage imbalance in the regulation of translation efficiency of its mRNA targets. We found that mRNAs of N-Methyl-

D-aspartate receptor (NMDAR) subunits and postsynaptic proteins are bound by

Cyfip1 protein, and deletion or duplication of Cyfip1 leads to reciprocal alterations in translation of these mRNAs. Furthermore, we were able to rescue a subset of the behavior deficits we observed in our Cyfip1 dosage imbalance mouse models by pharmacologically modulating NMDAR signaling levels. Our study provides insights into how 15q11.2 CNVs may confer risk to various neurodevelopmental disorders

2 The second study is an exploration of the role of the epitranscriptional modiﬁ- cation m6A in the regulation of cortical neurogenesis in-vivo. N-6 methyladenosine

(m6A) is the most prevalent internal modification in mRNA. This modification is conserved across species and has been associated with multiple processes pertain- ing to mRNA function such as splicing, mRNA degradation, regulation of translation, and mRNA localization [62, 116]. m6A is a dynamic modification, with identified writers (writer complex is composed of Mettl3, Mettl14, Wtap, KIAA1429,

RBM15 and its paralogue RBM15B), erasers (Fto, Alkbh5) and reader proteins

(Ythdf proteins, HuR) [74, 62, 116]. In vitro studies have found that m6A depleted embryonic stem cells are “locked” in a self-renewal state and fail to differentiate, and genetically modiﬁed mice lacking Mettl3 have been shown to be embryonic lethal, dying at the post-implantation stage because of failure to undergo lineage differentiation [31, 3]. Given the importance of m6A in stem cell function, we decided to investigate the role of m6A in cortical neurogenesis in an in vivo model.

We used the Mettl14 conditional knockout mouse as a model to examine m6A function in embryonic cortical neurogenesis in vivo. We further investigated underlying cellular and molecular mechanisms. Finally, we extended our analysis to human embryonic cortical neurogenesis using induced pluripotent stem cell (iPSC)- derived forebrain organoids and compared m6A-mRNA landscapes between mouse and human cortical neurogenesis. Together, our results reveal critical epitranscriptomic control of mammalian cortical neurogenesis and provide novel insight into mechanisms underlying this highly coordinated developmental program.

3 CHAPTER 2

Cyﬁp1 regulates translation of NMDAR complex and impacts

on behavior in mouse models.

2.1 Abstract

Cyfip1 gene is located within a microdeletion/duplication region in 15q11.2. Dele- tion or duplication of this genomic region leads to increased risk for neurodevelopmental disorders such as schizophrenia (SCZ), autism spectrum disorder (ASD), intellectual disability and epilepsy. Within neurons, the CYFIP1/FMRP complex regulates translation of mRNA by impeding the assembly of the translation initiation complex. Here we show that dosage imbalance of Cyfip1 leads to behavior abnormalities in mice, where loss of Cyfip1 resulted in SCZ-like behavioral impairments, whereas the gain of Cyfip1 showed ASD-related behavioral phenotypes. We applied a genome-wide approach to identify the mRNAs that are bound by Cyfip1 within the mouse hippocampus, and further explored the consequences of Cyfip1 dosage imbalance in the translation efficiency of its mRNA targets. We found that

4 mRNAs of N-Methyl-D-aspartate receptor (NMDAR) subunits and postsynaptic proteins are bound by Cyfip1 protein, and deletion or over-expression of Cyfip1 leads to reciprocal alterations in translation of these mRNAs. Furthermore, we were able to rescue a subset of the behavior deficits we observed in our Cyfip1 dosage imbalance mouse models by pharmacologically modulating NMDAR signaling levels. Our findings support a model where changes in Cyfip1 dosage lead to an imbalance in NMDAR signaling due to altered translation efficiency of Cyfip1 mRNA targets.

5 2.2 Introduction

Mental disorders such as schizophrenia (SCZ) and autism spectrum disorder (ASD) are chronic and disabling. Although their etiology is largely unknown, cumulative evidence supports the view that both SCZ and ASD have a neurodevelopmental origin with high heritability [30, 107]. A large number of susceptibility genes for

ASD and SCZ have been identified by human genetic studies and various experimental models are being developed to explore the function of these susceptibility genes in regulating neuronal development. In addition to single-nucleotide polymorphisms or mutations in a single gene, submicroscopic variations in DNA copy number (CNVs) are also widespread in human genomes and specific CNVs have been identified as significant risk factors for ASD and SCZ [59]. Moreover, recent data in aggregate provide support for the polygenic inheritance and, importantly, genetic overlap between SCZ and ASD [9, 15, 44].

15q11.2 CNVs have emerged as prominent risk factors for various neuropsychiatric disorders, including ASD, SCZ, intellectual disability and epilepsy [13,

35, 33, 99]. Interestingly, 15q11.2 microduplications have been associated with

ASD [67, 98], whereas microdeletions of the same region have been identiﬁed as one of the three most frequent CNV risk factors for SCZ [90]. A recent study further showed that carriers of 15q11.2 microdeletions without an apparent phenotype have nevertheless minor but recognizable neurological alterations such as learning difﬁculties and dyslexia [91]. The same study showed reciprocal structural phenotypes in deletion and duplication carriers, including differences in the volume of gray matter in the perigenual anterior cingulated cortex and white matter in the temporal lobe. Together, these studies have established 15q11.2 CNVs as prominent dosage-sensitive genetic risk factors for neuropsychiatric disorders. However, it remains unclear how different doses of genes within 15q11.2 region may con-

6 tribute to the etiopathology underlying neuropsychiatric disorders.

CYFIP1 is one of four genes in 15q11.2 and encodes a protein that interacts with fragile X mental retardation protein (FMRP) and eukaryotic translation initiation factor 4E (eIF4E), and negatively regulates mRNA translation at synapses in an activity-dependent manner [17, 66]. Altered levels of CYFIP1 lead to abnormalities of dendrite complexity and spine morphology [69, 73], and synaptic plasticity [7]. Moreover, common SNPs in CYFIP1 have been associated with

SCZ [119] and ASD [100], and CYFIP1 mRNA expression is increased in ASD patients [101]. These ﬁndings highlight CYFIP1 as the most compelling neuropsychiatric disease risk gene within the 15q11.2 region. Thus, increase or decrease in

CYFIP1 levels may lead to diametric alterations in common signaling pathways, which underlie pathogenesis of 15q11.2 CNV-mediated disorders.

To resolve this question in vivo, we established both loss-of-function and gain- of-function mouse models of Cyfip1. Interestingly, loss of Cyfip1 model displayed distinct behavioral abnormalities related with SCZ, whereas gain of Cyfip1 model showed ASD-related behavioral phenotypes. Genome-wide screening with the

RIP-sequencing method identiﬁed novel Cyﬁp1-associated mRNA targets related with synaptic function, postsynaptic density and NMDA receptor complex.

Molecular characterization showed that our mouse models display diametric changes in translation efﬁciency of Cyﬁp1 mRNA targets encoding NMDAR subunits and postsynaptic proteins. Furthermore, bidirectional pharmacological ma- nipulations of NMDAR signaling were able to rescue a subset of the behavioral abnormalities observed in these mouse models. Our integrated analyses provide insight into how 15q11.2 CNVs may contribute to divergent neuropsychiatric dis-

7 orders.

2.3 Methods

2.3.1 Experimental model and subject details

Generation of Cyﬁp1 cKO and cOE mice

For generating a Cyﬁp1 ﬂoxed allele, the targeting vector was designed to insert loxP sequence in front of exon 2 and a positive selection marker (PGK promoter driven neomycin resistant gene) together with another loxP sequence next to exon

5 (Figure 2.1, A), and constructed by recombineering as described [57]. In detail, an 11.9 kb genome fragment containing exon 2 to exon 5 from 129Sv BAC clone

(bMQ182K14, Source Bioscience) was retrieved into PL253 plasmid containing a negative selection (MC1 promoter driven thymidine kinase gene) using homologous recombination. A loxP sequence and a FLPe-PGK-EM7-Neo-FLPe-loxP cassette were sequentially inserted into the engineered PL253, resulting in 6.0 kb and 1.0 kb homology arms. The targeting vector was linearized and electroporated into 129 S4/Sv Jae embryonic stem cells (The Transgenic Core Laboratory in Johns Hopkins School of Medicine), and homologous recombination was con-

firmed by PCR screening (PCR1 in Figure 2.1, A). Targeted clones were injected into C57BL/6J blastocysts, which were subsequently transferred into pseudopreg- nant foster mothers. Confirmation of germ-line transmission of floxed allele and routine genotyping were performed by PCR screening on tail genomic DNA (PCR2 in Figure 2.1, A; wt, 470 bp; floxed, 520 bp) using DNA primers as follows: 5-GCA

CCT CTC TGC ATT TCT GT-3 and 5-GCA CCA ATC AAG TGT TTT CC-3.

For generating a Cyﬁp1 conditional over-expression allele, the targeting vector was designed to knock a cassette that permits conditional expression of HA-tagged Cy-

ﬁp1 cDNA into Rosa26 locus (Figure 2.1, B) and constructed as described [120].

8 A CMV b-actin enhancer-promoter was placed upstream of the floxed PGK-EM7- neo-tPA sequence to drive Cyfip1 cDNA transcription once the floxed tPA (transcriptional stop) is removed by Cre recombinase. The targeting vector was electroporated into 129 S4/Sv Jae embryonic stem cells. Targeted clones were identified by PCR screening (PCR1 in Figure 2.1, B) and injected into C57BL/6J blastocysts.

Conﬁrmation of germ-line transmission of ﬂoxed allele and routine genotyping were performed by PCR screening on tail genomic DNA with two sets of primers:

WT forward, 5 -GCA CTT GCT CTC CCA AAG TC-3 and WT reverse, 5-GCG

GGA GAA ATG GAT ATG AA-3 (PCR2 in Figure 2.1, B, 596bp); KI forward,

5-GCA ACG TGC TGG TTA TTG TG-3 and KI reverse, 5-GGG GAA CTT CCT

GAC TAG GG-3 (PCR3 in Figure 2.1, B, 395bp). Both cKO and cOE mice were backcrossed to C57BL/6J at least 6 times before all the experiments. For the most of behavioral experiments, 3-4 month-old male mice were used. In marble burying assay, nest building assay and pup retrieval assay, 3-4 month-old female mice were used. All mouse work was done in accordance with the Animal Care and Use

Committee guidelines of Johns Hopkins University School of Medicine.

2.3.2 Method details

Preparation of synaptosomes

Synaptosomal fractions were prepared as previously described [4]. Brieﬂy, brain tissues were homogenized in 3 volumes of ice-cold homogenization buffer A (0.32

M sucrose, 10 mM HEPES pH7.4, 2 mM EDTA, Phosphatase Inhibitor Cocktail

(Cell Signaling), protease inhibitor cocktail (Sigma)) with 12 strokes at 900 rpm in a motor-driven 2 ml Potter-Elvehjem Tissue Grinders (Wheaton). After cleared by centrifugation at 1,400g for 10 min at 4 degrees Celsius, the supernatant was ultracentrifuged at 30,000g for 15 min at 4 degrees Celsius. The pellet was resus-

9 pended with buffer B (6mM Tris-base pH 8.0, 1mM NaHCO3, 0.32M sucrose), applied to 1.2M-1M-0.85M sucrose gradient, and ultracentrifuged at 82,500g for

2 hr at 4 degrees Celsius. Interface between 1.2M and 1M was collected, added with 4 volumes of buffer B, and ultracentrifuged at 48,200g for 20 min at 4 degrees

Celsius. The pellet was resuspended with buffer A and protein concentration was quantiﬁed. Samples were boiled for 5 min in Laemmli sample buffer (Biorad) and applied to Western blot analysis.

Western blot analysis

Hippocampus were quickly dissected out and homogenized in RIPA buffer (50 mM Tris pH 7.5, 120 mM NaCl, 1% Triton X-100, 0.5% Sodium Deoxycholate,

0.1% SDS, 5 mM EDTA, Phosphatase Inhibitor Cocktail (Cell Signaling), protease inhibitor cocktail (Sigma)). The lysates were incubated for 15 min on ice and centrifuged at 15,000g for 15 min at 4 degrees Celsius. The supernatant were collected and boiled for 5 min in Laemmli sample buffer (Biorad), resolved by SDS PAGE, transferred to PVDF membrane, and immunoblotted. Quantiﬁcation of bands was performed using ImageJ software.

Biotin-puromycin labeling and detection of newly synthesized proteins by West- ern blotting

Biotin-puromycin labeling experiment was performed as described with minor modiﬁcations [2]. First, 6 hippocampi from 3 mice were homogenized with ice- cold polysome buffer (50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 25 mM KCl, protease inhibitors and RNase inhibitor). Homogenized lysate were incubated on ice for 20 min with lysis buffer, cleared by centrifugation at 20,000g for 10 min at

10 4 degrees Celsius, loaded on top of sucrose solution, and pelleted by centrifugation at 48,600g for 3 hours at 4 degrees Celsius. The ribosome pellet was washed with

RNase-free water, recovered with polysome buffer and incubated with 5 Biotin-dC- puromycin (Jena Bioscience) for 20 min at 37 degrees Celsius. Biotin-puromycin labeled newly synthesized polypeptides were captured by streptavidin-HRP magnetic beads (Thermo Fisher Scientiﬁc) in PBS with 0.5% triton X-100 for overnight at 4 degrees Celsius, washed extensively with Urea/SDS buffer (50 mM Tris-HCl

(pH 7.5), 8 M urea, 2% SDS) for 2 hours at RT, collected by boiling with Laemmli sample buffer for 10 min at 95 degrees Celsius, and applied to Western blotting analysis.

RIP qPCR

RIP experiments were performed as described [118]. RIPs were performed on six hippocampi from three C57BL/6 mice using 10 µg of anti-rabbit CYFIP1 antibody

(AB6046; EMD Millipore) (RIP) or normal rabbit IgG (Invitrogen) (control). RNA was extracted with TRIzol (Thermo Fisher Scientiﬁc) and puriﬁed using RNeasy mini kit (Qiagen). Quantitative RT-PCR was performed using the StepOne Real-

Time PCR System (Applied Biosystems).

RIP seq experiments on mouse hippocampus

RIP-seq experiments were performed as previously described [118]. Brieﬂy, mouse hippocampi were quickly dissected out and homogenized in RIP lysis buffer. Lysates were then incubated with anti-HA antibody (H6908; SIGMA) or control normal rabbit IgG (Cell Signaling) for 2 hours at 4 degrees Celsius. After antibody incubation, RNA-protein complexes were immunoprecipitated with Dynabeads Protein

11 G (Thermo Fisher) for 2 hours, and beads were washed 5 times with lysis buffer before extracting RNA with Trizol (Invitrogen). cDNA libraries for pulled-down

RNA as well as input RNA were prepared using NEBNext Ultra RNA Library Prep

Kit for Illumina. cDNA libraries were sequenced on the Illumina Hiseq platform with 50 cycles single-end.

Bioinformatic analysis of RIP-seq data

All libraries were pre-processed using the Hannon-lab fastx toolkit, to remove poor quality reads (FASTQ quality ﬁlter), clip adaptors (FASTQ clipper) and ﬁlter du- plicated reads (FASTQ collapser). Pre-processed sequences were then aligned to the mouse mm10 refseq transcriptome using Tophat2 [45] with default options.

We predict statistically signiﬁcant RIP regions (peaks) indicative of a physical interaction of an RNA transcript with Cyﬁp1 using the software SICER [113]. While the union of RIP regions reported by SICER for all six RIP experiments, namely:

(1) knock-in1-HA (RIP) vs knock-in1-IgG (control), (2) knock-in1-HA (RIP) vs wild-type 1-HA (control), (3) knock-in1-HA (RIP) vs wild-type 2-HA (control)

(4) knock-in 2-HA (RIP) vs knock-in 2-IgG (control), (4) knock-in 2-HA (RIP) vs wild-type 1-HA (control),(6) knock-in2-HA (RIP) vs wild-type 2-HA (control) would yield the highest sensitivity, their intersection would provide the most con- servative estimate of the set of RIP regions. We decided on an intermediate strategy to achieve a balance in sensitivity and precision, by selecting genes for which

SICER reported at least one significant (FDR 0.05) RIP region in more than 50% of experiments (4 out of the 6 replicates). There were 1,570 genes that met these criteria, and we will refer to these genes as High Confidence Cyfip1 targets.

Gene ontology analysis on our High Conﬁdence Cyﬁp1 target list was performed using the ToppFun function from the Toppgene Suite [11] which implements a

12 hypergeometric probability mass function with Benjamini Hochberg FDR correction to identify signiﬁcant enrichment for GO terms. Disease association analysis was performed using WebGestalt [114], which uses a hypergeometric method and

Benjamini Hochberg FDR correction. Protein interaction network ﬁgures were generated using Cytoscape 3.3.0 [86], with the Reactome FI plugin.

Pharmacological treatment

Memantine hydrochloride (Sigma; 20 mg/kg/day) was administrated orally via drinking water into 10 weeks-old mice for 2 weeks, and was maintained during behavioral tests. The daily dosage was calculated based on a daily water intake of wild-type mice with similar ages. Water consumption was measured every two days throughout treatment to conﬁrm average daily memantine intake. There were no differences in water consumption between WT1 and cKO mice during the treatment. Oral dosing of memantine via drinking water at this level has been shown to produce a steady state plasma drug level of 1 µM in rodents [65], and proved to alleviate behavioral abnormalities in rodent models of Down syndrome and bipolar disorder [58, 28]. On the other hand, D-cycloserine (Abcam; 20 mg/kg) dissolved in saline or the same volume of saline were administrated into WT2 and cOE mice intraperitoneally, 30 min before three-chamber social interaction assay and amphetamine-induced hyperactivity assay.

2.3.3 Behavior studies

Pup retrieval assay [106]

Virgin female mice were isolated for four days before pup retrieval assays. Three wild-type pups (1-day old; C57BL/6N) were placed at three different corners of a home cage of a test female mouse, and the female mouse was allowed to retrieve

13 the pups for 10 min. Efﬁciency of pup retrieval was measured by the time taken to retrieve the ﬁrst, second, and third pups.

Locomotor activity and amphetamine-induced hyperactivity

Locomotor activity in the open ﬁeld was assessed for 60 min using the activity chambers with infrared beams (San Diego Instruments Inc., San Diego, CA, USA).

Horizontal and vertical activity, and center or peripheral activity were automati- cally recorded by beam breaks. For measurement of D-amphetamine-induced hyperactivity, each mouse was intraperitoneally administrated saline 30 min after the habituation to the test cage as a control of injection, then intraperitoneally administrated D-amphetamine (2.5mg/kg, Sigma) 30 min after saline injection. Locomotor activity was then measured for 120 min.

Sociability and social novelty test

Sociability and social novelty were performed in a standard mouse cage separated into three chambers by partitions. Each chamber measures 20 cm (length) x 40 cm

(width) x 22 cm (height). Each side chambers contained a cylindrical wire cage to enclose a live mouse. Before test, the test mouse was placed in the center of the empty three-chamber box for habituation. The mouse was allowed for 15 min to freely explore each chamber. In the ﬁrst session, an age and gender-matched

C57BL/6 mouse (stranger 1) that had never been exposed to the test mouse was placed in one of the two wire cages. The wire cage on the other side remained empty. In the second session, a second age and gender-matched C57BL/6 stranger mouse (stranger 2) that had never been exposed to the test mouse was placed in one wire cage, which previously served as an empty cage. The test mouse was placed

14 in the center and allowed to freely explore the chamber for 10 min for each session. Now the test mouse would have the choice between a mouse that was already familiar (stranger 1) and a new stranger mouse (stranger 2). The movement of the mouse was recorded by a video camera controlled by the AnyMaze software (San

Diego Instruments Inc., San Diego, CA, USA).

Acoustic startle response and prepulse inhibition

Experiments were performed using startle chamber (San Diego Instruments Inc.,

San Diego, CA, USA). Presentations of the acoustic stimuli were controlled by the SR-LAB software and interface system, which also rectiﬁed, digitized and recorded responses from the accelerometer. To measure acoustic startle response, each mouse was kept in the animal holder on a movement-sensitive platform for 5 min with 70 dB background white noise for acclimation. Then 10 startle stimuli of 120 dB were delivered every 20 s. During each PPI session, a mouse was exposed to the following types of trials: pulse-alone trial (120-dB, 100-ms), null trial

(no-stimulus trial), and ﬁve pre-pulse trials (74 dB, 78 dB, 82 dB, 86 dB, and 90 dB). Each session consisted of six presentations of each type of the trial presented in a pseudorandom order. Percent prepulse inhibition was the difference in startle responses in pulse alone trials and prepulse and pulse trials expressed as a percentage of the response in pulse alone trials.

Forced swim test

Experiment was performed in a glass cylinder (20 cm high x 13.5 cm diameter)

filled with water to a 12 cm depth. Mice were allowed to swim for 6 min, and total time spent immobile, defined as the absence of struggling and floating motionless

15 in water, and the latency to the ﬁrst immobile episode were counted during the last

5 min of the test.

Tail suspension test

Mice both acoustically and visually isolated by test cage were suspended by adhe- sive tape placed approximately 1 cm from the tip of their tails. The experiment was recorded for 6 min by video camera. The total time spent immobile, deﬁned as the absence of struggling, and the latency to the ﬁrst immobile episode was counted during the last 5 min of the test.

Hot plate assay

IITC hot plate analgesia meter was employed for assessment of nociceptive responses of mice. The animals were put on a hot plate that is maintained at a temperature of 32 degree Celsius for habituation. Response latency to jump or a hindpaw-lick was monitored utilizing an electronic-timer. If there is no response within 20 seconds, the test is terminated. The strength of stimulator is AI 40. We tested 7 trails for each mouse. The maximum and minimum data were excluded, the result measures the average of the last 5 trials.

Rotarod

Mice were loaded on the rotarod machine which is running at 5.0 RPM. After 1 minute habituation, the speed accelerated 1.0 RPM every 5 seconds. The duration of the test is 300 seconds. The test was repeated 4 times. Latency of each mouse to fall from the rotarod is recorded.

16 Novel object recognition

Mice were handled well before training and acclimated in the experiment box for

2 days, with 10 minutes habituation each day. On day 3, mice were put in the experiment box with two identical objects for 10 min. After 24 h, mice were put back into the same box for novel object recognition test. The interaction time of familiar and novel object was analyzed.

2.4 Results

2.4.1 Deletion of Cyﬁp1 results in behavioral abnormalities related with schizophrenia

CYFIP1 is highly expressed at excitatory synapses and regulates dendritic spine morphology and spine dynamics[17, 73]. However, in vivo effects of differential

Cyﬁp1 dosages on synaptic function have not yet been examined systematically. To address this question, we have generated a conditional knockout (Nestin-Cre+/Tg;

Cyﬁp1 f / f , cKO) and a conditional over-expression mouse model of Cyﬁp1 (Nestin-

Cr+/Tg; ROSA26 Cyfip1KI/KI, cOE) (Figure 2.1, A and B). cKO showed complete ablation of Cyfip1 protein in the forebrain lysates after E17.5 (Figure 2.1, C). On the other hand, homozygote conditional over-expression mice showed 1.5-2 fold increase of Cyfip1 expression in the hippocampus and the cortex of the brain (Fig- ure 2.1, D). We used homozygote conditional over-expression mice (cOE) for all the following analysis. cKO and cOE mice are fertile, have a normal appearance and expected Mendelian ratio of genotypes in their adulthood.

17 A C

B D

Figure 2.1: Cyfip1 mouse models. Genetic strategy for generation of conditional knock out (A) and conditional over-expression (B) mouse models. cKO shows complete ablation of Cyfip1 protein in forebrain lysates after E17.5. (C) Homozygote cOE shows 1.5-2 fold increase of Cyfip1 protein expression in forebrain lysates after E17.5

To investigate the consequences of Cyﬁp1 loss-of-function at the level of behavior, we applied a battery of behavioral tests related with human mental disorders on cKO mice. cKO mice displayed normal locomotor activity, motor coordination, nociception response, novel object recognition and repetitive behaviors

(Supplemental figure A.1, A-E). To test whether cKO mice have behavioral abnormalities related with negative symptoms in schizophrenia, we first performed a three-chamber social interaction assay [88]. cKO mice showed similar preference to explore a novel mouse (stranger 1) over an empty cage with littermate wild- type mice (Nestin-Cre+/+; Cyfip1 f / f , WT1). However, when the empty cage was replaced with a new novel mouse (stranger 2), cKO mice did not show significant preference to stranger 2 over stranger 1 while WT1 preferred to interact with stranger 2 (Figure 2.2, A and B), indicating impaired social novelty recognition.

Moreover, cKO mice exhibited impaired prepulse inhibition (PPI) and elevated startle response (Figure 2.2, C and D), suggesting deﬁcits in sensorimotor gating

18 common in schizophrenia patients. In addition, cKO mice showed elevated behavioral despair assessed by increased immobility in both tail suspension test (TST) and forced swim test (FST) compared with WT1 mice (Figure 2.2, E and F).

To model positive symptoms of schizophrenia, we measured amphetamine- induced hyperactivity in cKO and WT1 mice which is widely used in animal models of neuropsychiatric disorders [97]. Indeed, cKO mice showed increased locomotor activity after an acute injection of amphetamine as compared to WT1 mice

(Figure 2.2, G). cKO mice had normal locomotor activity, motor coordination, nociception response and repetitive behaviors (Supplemental ﬁgure A.1, A-G).

Nestin-Cre transgene did not affect social interaction, amphetamine-induced hyperactivity or prepulse inhibition (Supplemental ﬁgure A.1, H and I). In summary, these results suggest Cyﬁp1 loss-of-function in mice leads to multiple behavioral abnormalities related with negative and positive symptoms of human schizophrenia patients.

2.4.2 Increased Cyﬁp1 dosage leads to ASD-like behavioral abnormalities

Given that CYFIP1 mRNA expression is signiﬁcantly upregulated in ASD patients

[69, 97, 101], we hypothesized that over-expression of Cyfip1 would results in behavioral abnormalities related with ASD. cOE mice displayed normal locomotor activity and novel object recognition (Supplemental figure A.2, A and B). Inter- estingly, in the first stage of a three-chamber social interaction assay, cOE mice exhibited decreased interaction with stranger 1 compared with littermate wild-type mice (Nestin-Cre+/+; ROSA26 Cyfip1KI/KI, WT2) (Figure 2.3, A). Moreover, cOE mice did not prefer to interact with stranger 2 although WT2 interacted significantly more with stranger 2 than stranger 1 (Figure 2.3, B). These results suggest

19 A B C D

E F G

Figure 2.2: Cyﬁp1 cKO shows behavior abnormalities related to schizophrenia. (A and B) Normal social approach and impaired social novelty recognition of cKO mice in the three-chamber assay. n= 10 (WT1), 10 (cKO). (C and D) Abnormal prepulse inhibition and increased basal acoustic startle response (120dB) of cKO mice compared with WT1. n= 19 (WT1), 15 (cKO). (E and F) Elevated behavioral despair in cKO mice. Immobile time in both tail suspension test (TST) and forced swim test (FST) was increased in cKO mice compared with WT1 mice. n= 19 (WT1), 15 (cKO). (G) Enhanced amphetamine-induced hyperactivity in cKO mice compared with WT1 mice. Shown on the left is the trace of locomotor activity presented as number of beams broken during every 10 min. An arrow represents the time of amphetamine injection. Shown on the right is total number of beams broken after amphetamine injection. n = 11 (WT1), 14 (cKO). All values represent mean SEM (***p <0.001; **p <0.01; *p <0.05; n.s., not signiﬁcant; Students t test).

20 that cOE mice have impaired social approach as well as social novelty recognition.

Repetitive and stereotyped patterns of behavior are one of the core behavioral domains required for ASD diagnosis [78]. Therefore, we assessed repetitive behaviors of cOE mice by using marble burying assay and measuring digging behavior

[18]. cOE mice buried signiﬁcantly more marbles during the tests than WT2 mice

(Figure 2.3, C). Moreover, cOE mice also spent more time in digging behavior than

WT2 mice (Figure 2.3, D).

Maternal behaviors are frequently impaired in mouse models of ASD [81, 84,

109]. Interestingly, the survival rate of pups from cOE dams was markedly lower than that from WT2 dams (Figure 2.3, E). At postnatal day 3, some of the pups from cOE dams did not show milk in their stomach although all of the pups from

WT2 dams were ﬁlled with milk in their stomach (Supplemental ﬁgure A.2, C).

As a result, survived pups at postnatal day 7 from cOE dams were significantly smaller than pups from WT2 dams, indicating insufficient feeding during the first postpartum week (Supplemental figure A.2, D). Among the survived pups from cOE dams, the number of cOE pups was similar with the number of WT2 pups, suggesting that impaired maternal care is not dependent on the genotypes of the pups (data not shown). Moreover, cOE females showed impaired nest building behaviors (Figure 2.3, F) and less efficiency of pup retrieval compared with WT2 females (Figure 2.3, G), suggesting multiple traits of maternal behaviors are impaired with increased Cyfip1 dosage in a mouse model. On the other hand, cKO dams showed similar levels of pup survival rate with WT1 dams (data not shown).

In addition, Similar with cKO, cOE mice displayed increased amphetamine- induced hyperactivity as compared to WT2, indicating dysfunction in the dopamin-

21 ergic system of cOE mice (Figure 2H). Unlike with cKO mice, cOE mice did not display an increase in despair in TST (Supplemental ﬁgure A.2, E) or abnormal sensorimotor gating in PPI (Supplemental ﬁgure A.2, F).

Taken together, our behavioral analysis of cOE mice showed that increased Cy-

ﬁp1 dosage results in several behavioral abnormalities related with ASD, including social impairment, increased repetitive behavior and abnormal maternal behavior.

2.4.3 Cyﬁp1 interacts with mRNAs of synaptic and NMDA receptor related proteins in mouse hippocampus and human cerebral cortex.

Given the previous knowledge that Cyfip1 regulates translation in neurons [17, 66], we sought out to identify which mRNAs are binding partners of Cyfip1, and thus are potentially regulated by it. In order to do this, we took a genomewide approach, and performed RNA sequencing of immunopurofied ribonucleoproteins (RIP-seq)

[118]. These experiments were done in mouse hippocampus, and we took advantage of our cOE mouse model, which expresses a HA-tagged Cyﬁp1 (Figure 2.1,

B and supplemental ﬁgure A.3, A and B), allowing us to immunoprecipitate Cy-

fip1 using an anti-HA antibody. This approach allowed us to identify the mRNA species that co-precipitate with Cyfip1 in-vivo in our mouse model. We found 1,570 high confidence Cyfip1 targets (Figure 2.4 and supplemental figure A.4, B) which were identified in more than half of RIP experiments (4 out of 6) (Supplemental

ﬁgure A.4, A and B). Gene ontology analysis showed signiﬁcant enrichment of terms related to synaptic function and synaptic components (Figure 2.5, A and B).

Disease ontology analysis revealed enrichment for neuropsychiatric diseases like schizophrenia and autism among others (Figure 2.5, B).

22 Figure 2.3: Cyfip1 cOE mice show ASD-like behavioral abnormalities. (A and B) Reduced social approach and impaired social novelty recognition of cOE mice in the three-chamber assay. In the first stage, cOE mice showed significant lower preference to novel mouse (stranger 1). In the second stage, cOE mice showed no preference to stranger 2 over stranger 1. n = 13 (WT2), 13 (cOE). (C) Increased repetitive behavior of cOE mice in marble burying test. Number of marbles buried during 30 min test was counted. n = 16 (WT2, male), 11 (cOE, male), 8 (WT2, female), 13 (cOE, female). (D) Increased digging behavior of cOE mice. Time spent digging was measured during 10 min test. n = 11 (WT2), 10 (cOE). (E) Pups from cOE dam showed reduced survival rate in early postnatal period. n = 10 (WT2), 14 (cOE) litters. (F) cOE females displayed abnormal nest building behavior. Shown on the left is percentage weight of unshedded net and on the right the nest building score 24hr after new nest presentation. n = 15 (WT2), 12 (cOE). (G) cOE females showed impaired pup retrieval behavior. Retrieval latencies of three pups in pup retrieval test are shown. n = 13 (WT2), 11 (cOE). (H) Enhanced amphetamine-induced hyperactivity in cOE mice compared with WT2 mice. Shown on the left is the trace of locomotor activity presented as number of beams broken during every 10 min. Arrow represents the time of amphetamine injection. Shown on the right is total number of beams broken after amphetamine injection. n = 11 (WT1), 14 (cKO). All values represent mean SEM (***p <0.001; **p <0.01; *p <0.05; n.s., not significant; Students t test).

23 A eeerce nCfi1ple-onsmlscmae ihamc pull mock a with compared samples pulled-down Cyfip1 in enriched were NAs ihteND eetrcmlxadpssnpi est o ute validation: related further mRNAs for Shank1, density 5 postsynaptic chose and complex We receptor NMDA antibody. the with anti-CYFIP1 an using done endogenous was of Cyfip1 Immunoprecipitation (RIP-qPCR). PCR quantitative by hippocampus, followed mouse type per- wild we in analysis, immunoprecipitation genome-wide ribonucleoprotein formed our by unveiled targets mRNA interesting ically iue24 oeaeposfrom plots Coverage 2.4: Figure fCfi1tres(Shank1, targets Cyfip1 of RIPcontrol RIPcontrol RIPcontrol idepnl hwra oeae omlzdb irr ie rmCyfip1 from sizes library by normalized coverages read show panels middle uldw n oto irre,rsetvl,adbto aesso gene show panels bottom and respectively, libraries, control and pull-down nodrt ofimteitrcino yp rti ihasbe fbiolog- of subset a with protein Cyfip1 of interaction the confirm to order In Shank2, Dlg4/Psd95, rncit.Tpand Top transcripts. (Gapdh) target non and Shank2) Grin2a Cyfip1 structures. and 24 RIP-seq n nedfudta hi mR- their that found indeed and Grin2b, hwn ersnaieexamples representative showing A

B GO: Biological process GO: Celullar component Disease ontology Nervous system development Postsynaptic density Mental disorders Dendrite morphogenesis Synapse Mental retardation Actin cytoskeleton organization Neuron projection Schizophrenia Neuron migration Dendrite Bipolar disorder Synaptic plasticity Dendritic spine Epilepsy Learning Cytoskeleton Autistic disorder Synapse organization Excitatory synapse Nervous system diseases

0 5 10 15 0 10 20 30 40 50 0 5 10 15 -Log10 pvalue -Log10 pvalue -Log10 pvalue

Figure 2.5: Gene ontology and disease ontology analysis of Cyﬁp1 target mRNAs.

25 down with IgG (Figure 2.6, A), indicating those mRNAs are interacting partners of

Cyﬁp1.

After having identiﬁed and conﬁrmed the interaction of CYFIP1 with Shank1,

Shank2, Dlg4/Psd95, Grin2a and Grin2b in the mouse hippocampus, we decided to explore whether this interaction is conserved in the human brain. For that pur- pose, we did RIP-qPCR experiments on human cerebral cortex samples, which were obtained from surgical resection on epilepsy patients. We found signiﬁcant enrichment of these mRNAs in cerebral cortex samples pulled-down with anti-

CYFIP1 antibody compared to IgG pull down, which shows that the interaction of

CYFIP1 with these mRNAs is a conserved phenomenon (Figure 2.6, B).

2.4.4 Cyﬁp1 directly regulates mRNA translation of NMDAR subunits and postsynaptic components

CYFIP1 represses translation of mRNAs at the synapse by inhibiting the interaction between eIF4E and eIF4G at the 5′cap structure [17, 66, 72]. We measured mRNA levels of Cyfip1 targets Shank2, Grin2a, Grin2b, Dlg4/Psd95 and Homer1, and non-targets Grin1 and Syn1 in cKO and cOE mouse hippocampus (Supple- mental figure A.5, A and B), and found no changes as compared to wild type, suggesting that Cyfip1 may not regulate transcription or stability of mRNA targets.

To explore whether altered Cyfip1 dosages lead to dysregulated protein translation of Cyfip1 target mRNAs, we applied the recently developed PUNCH-P technique to monitor the amount of nascent peptides from cKO and cOE tissues in vivo (Fig- ure 2.7, A) [2]. To determine the effects of Cyfip1 ablation on the general translation rate, the amount of total biotin-puromycin labeled nascent peptides were assessed by streptavidin-HRP immunoblotting. Similar amounts of peptide were

26 A B

Figure 2.6: Confirmation of CYFIP1 mRNA targets in mouse hippocampus and human cerebral cortex.(A) Association of Cyfip1 protein with mRNAs of NMDAR subunits and postsynaptic components within mouse hippocampus. Top panel shows a representative immunoblot of Cyfip1 protein pulled-down by immunoprecipitation (IP) from wild type mice hippocampal lysates. Shown on the bottom are quantitative PCR results from co-IPed mRNAs by anti-CYFIP1 antibody compared with co-IPed mRNAs by control IgG. n= 4 independent experiments. (B) Association of CYFIP1 protein with mRNAs of NMDAR subunits and postsynaptic components within human cerebral cortex. Top panel shows a representative immunoblot of CYFIP1 protein pulled-down by immunoprecipitation (IP) from human cerebral cortex lysates. Shown on the bottom are quantitative PCR results from co-IPed mRNAs by anti-CYFIP1 antibody compared with co-IPed mRNAs by control IgG. n= 3 independent experiments. All values represent mean SEM (***p <0.001; **p <0.01; *p <0.05; n.s., not significant; Students t test).

27 being synthesized in cKO mice compared to WT1 mice, suggesting loss of Cyfip1 function does not change general translation rate (Figure 2.7, B). To monitor protein translation rate of specific target mRNAs, biotin-puromycin labeled peptides were captured and purified by streptavidin beads and examined by specific antibodies. We observed an increase in protein synthesis of Grin2a, Grin2b, Shank2 and Dlg4/Psd95 in cKO mice (Figure 2.7, C). Intriguingly, the reverse effect was observed in the cOE, where protein synthesis of these genes was decreased compared to wild type (Figure 2.7, D). No changes were observed for Cyfip1 target

Homer1 or for non-target Syn1.

2.4.5 Protein expression of NMDAR subunits and postsynaptic components is altered within synaptosomes depending on Cyﬁp1 dosages

To examine molecular changes in synapses of cKO and cOE mice, synaptosomes were isolated from hippocampi of cKO, cOE and wild-type littermates (Nestin-

Cre+/+; Cyfip1 f / f , WT1) and applied for Western blotting analysis. As expected, the expression of Cyfip1 was completely abolished in cKO and increased in cOE hippocampi (Figure 2.8, A and B). Expression levels of Cyfip1 targets Shank2,

Grin2b and Dlg4/Psd95 were significantly increased in synaptosomes from cKO compared to wild type (Figure 2.8, A). Conversely, decreased protein levels were found in cOE synaptosomes for these Cyfip1 targets (Figure 2.8, B). For Grin2a, which was also identified as Cyfip1 target in our study, we did not observe changes in protein levels in the cKO, but we did observe lower levels in cOE compared to wild type, similar to what we found for the other Cyfip1 targets that were assessed

(Figure 2.8, A and B). No changes were observed for Homer1 or Syn1 in either cKO or cOE synaptosomes (Figure 2.8, A and B).

28 Figure 2.7: Cyfip1 regulates translation of NMDAR subunits and postsynaptic components. (A) Schematic diagram of experimental design. Ribosomal fractions from hippocampal lysates were pelleted by ultracentrifugation, and then labeled with biotin-puromycin to release newly synthesized peptides. Released biotin-puromycin-tagged nascent peptides were directly applied to Western blotting with HRP-conjugated streptavidin (StAv-HRP) to monitor general translation rate (B), or pulled-down with magnetic beads conjugated with streptavidin and applied to Western blotting with specific antibody (C and D). (B) Normal general translation rate in cKO mice. Shown on the left is a representative immunoblot of total biotin-puromycin labeled peptide detected by StAV-HRP. Vertical line traces of HRP signal on each lane are shown on the right of the immunoblot. HRP signals were normalized with RPS6 signals and quantified. n = 3 independent experiments, 3 animals per group in each experiment. (C and D) Protein synthesis is increased in cKO mice and reduced in cOE mice. Shown on the left is a representative immunoblot of total pulled-down nascent peptides detected by StAV-HRP. The same samples were applied to Western blotting with specific antibodies as shown in the middle. Relative protein synthesis in cKO and cOE were normalized by StAV-HRP signals, quantified and compared with WT1 and WT2, respectively. n= 3 independent experiments, 3 animals per group in each experiment.

29 A B

Figure 2.8: Altered protein expression of NMDAR subunits and postsynaptic components in synapses with differential Cyﬁp1 dosages. (A) Increased expression of NMDAR subunits and postsynaptic components in cKO mice. Shown are representative immunoblots of synaptosomal fractions from hippocampal lysates of 3-month-old wild-type littermates (WT1) and cKO mice. (B) Decreased expression of NMDAR subunits and postsynaptic components in cOE mice. Shown are representative immunoblots of synaptosomal fractions from hippocampal lysates of 3-month-old wild-type littermates (WT2) and cOE mice. All data were normalized to Actin levels for loading control and then protein expression levels were normalized to that wild type and plotted as relative changes of expression level. n = 37 animals.

30 2.4.6 Bidirectional modulation of NMDAR signaling rescues behavioral abnormalities in cKO and cOE mice

Imbalance in NMDAR signaling has been implicated in multiple neuropsychiatric disorders [49, 51]. Our molecular characterization of our Cyﬁp1 mouse models revealed diametrical changes in translation efﬁciency and protein levels of NMDAR subunits and postsynaptic components. We thus hypothesized that reducing NM-

DAR signaling activity in cKO mice and enhancing NMDAR signaling activity in cOE mice, could rescue behavioral abnormalities in these models (Figure 2.9, A).

First, we treated cKO mice with NMDAR antagonist memantine and we observed that memantine treatment signiﬁcantly rescued the increased behavioral despair of cKO mice in both the tail suspension test and forced swim test (Figure 2.9, B and

C). In addition, increased amphetamine-induced hyperactivity of cKO mice was also normalized with memantine treatment (Figure 2.9, D). Impaired sensorimotor gating assessed by PPI in cKO mice was not improved with memantine treatment

(data not shown).

On the other hand, we took advantage of a partial agonist of NMDAR D- cycloserine (DCS), which has been shown to rescue ASD-related behavior in animal models with reduced NMDAR function [5, 109]. In a three-chamber social interaction assay, reduced social interaction of cOE was improved by DCS treatment (Figure 2.10, A), as well as impaired social novelty recognition (Figure 2.10,

B). Repetitive behaviors of cOE mice assessed by marble burying assay were not rescued by DCS treatment (data not shown). In addition, increased amphetamine- induced hyperactivity of cOE was restored by DCS treatment (Figure 2.10, C).

In conclusion, NMDAR function and downstream signaling was impaired in opposing directions in cKO and cOE mice, with the direction of the imbalance de-

31 Figure 2.9: Treatment with memantine in cKO rescued behavioral abnormalities. (A) Model of NMDAR dysfunction with different levels of Cyﬁp1. (B and C) Behavioral despair in cKO mice was rescued by memantine treatment. Immobile time in both TST and FST was restored in cKO mice after memantine treatment to the level of WT1 mice. TST, n= 10 (WT1+VEH), 9 (cKO+VEH), 8 (WT1+MEM), 8 (cKO+MEM); FST, n = 11 (WT1+VEH), 9 (cKO+VEH), 10 (WT1+MEM), 8 (cKO+MEM). (D) Level of amphetamine-induced hyperactivity in cKO mice was restored to the level of WT1 after memantine treatment. Shown on the left is the trace of locomotor activity. An arrow represents the time of amphetamine injection. Shown on the right is the total number of beams broken after amphetamine injection. n= 8 (WT1+VEH), 8 (cKO+VEH), 7 (WT1+MEM), 11 (cKO+MEM). All values represent mean SEM (***p <0.001; **p <0.01; *p <0.05; n.s., not signiﬁcant; Students t test).

32 A B

Figure 2.10: Treatment with D-cycloserin in cOE rescued behavioral abnormalities. (A and B) Social impairments of cOE mice were improved after DCS treatment. In the first stage, interaction time with stranger 1 in cOE mice after DCS treatment was significantly increased compared with cOE mice without treatment, similar with WT1 with or without treatment. In the second stage, cOE mice showed significant preference to stranger 2 after DCS treatment, whereas non-treated cOE showed no preference to stranger 2. n= 7 (WT2+VEH), 7 (cOE+VEH), 7 (WT2+DCS), 7 (cOE+DCS). (C) The level of amphetamine-induced hyperactivity in cOE mice was restored to the level of WT2 after DCS treatment. Shown on the left is the trace of locomotor activity. An arrow represents the time of amphetamine injection. Shown on the right is the total number of beams broken after amphetamine injection. n= 10 (WT2+VEH), 12 (cOE+VEH), 8 (WT2+DCS), 8 (cOE+DCS). All values represent mean SEM (***p <0.001; **p <0.01; *p <0.05; n.s., not significant; Students t test).

33 pending on the level of Cyﬁp1 in each mouse model.

2.5 Discussion

The etiology of most neurodevelopmental disorders, such as schizophrenia (SCZ) and autism spectrum disorder (ASD) is still largely unknown despite a large interest from the medical and scientific community. This is partly due to the high complexity of these diseases which is underscored by the large number of susceptibility genes so far ascribed to them. Specifically, SCZ and ASD have been shown to have polygenic inheritance, and recent studies support a significant genetic overlap between the two diseases [9, 15, 44]. In addition to single gene mutations or single-nucleotide polymorphisms, a number of copy number variants (CNVs) have also been associated with risk for SCZ and ASD [59]. 15q11.2 CNVs have emerged as a prominent risk factor for neuropsychiatric disorders including ASD and SCZ [35, 33, 99]. Here we show that dosage imbalance of Cyfip1, the critical gene within 15q11.2, leads to behavioral defects in mouse models. We found that deleting Cyfip1 within the mouse brain leads to behavior abnormalities related to

SCZ, while over-expression of Cyfip1 leads to a different set of behavior abnormalities that are related to ASD. We took a genome wide approach to identify mRNAs that are bound to Cyfip1 in the mouse hippocampus, and gene ontology analysis of these mRNAs showed enrichment for terms related to synaptic function and learning. Within the set of Cyfip1 targets we identified NMDAR subunits, and components of the postsynaptic density, and we found that Cyfip1 dosage imbalance

(deletion and over-expression) led to diametrical changes in protein translation and protein levels of Grin2a, Grin2b, Shank2 and Dlg4/Psd95. Furthermore, by pharmacologically modulating NMDAR signaling we were able to rescue a subset of

34 the behavioral abnormalities observed in our mouse models.

2.5.1 Cyﬁp1 mRNA targets are related to synaptic function and neuropsychiatric diseases

Our study provides a database of Cyﬁp1 mRNA binding partners within the mouse hippocampus. Previous work has shown that Cyﬁp1 interacts with fragile X mental retardation protein (FMRP), and eukaryotic translation initiation factor 4E (eIF4E), and negatively regulates mRNA at synapses in an activity dependent manner [66,

17]. We performed RIP seq experiments to identify those mRNAs that co-precipitate with Cyfip1, and thus could be implicated in the etiopathology of diseases associated with Cyfip1 imbalance. Interestingly, we found that the group of mRNAs that are Cyfip targets, are significantly enriched for gene ontology terms related to synaptic function, and disease ontology analysis revealed a strong enrichment for genes related to neuropsychiatric diseases like “mental retardation”, schizophrenia, bipolar disorder and autistic disorder. We have validated by RIP qPCR a subset of these targets in the mouse hippocampus, and we were also able to confirm the interaction of CYFIP1 with these mRNAs in human cerebral cortex. These results suggest that the risks for neuropsychiatric diseases associated with 15q11.2 could be related to a dysregulation of these Cyfip1 mRNA targets which are themselves associated with synaptic function and neuropsychiatric diseases.

2.5.2 Diametric gene dosage effects of CNVs in neurodevelopmental diseases

Gene dosage effects have been shown to be an important component in the etiopathology of neurodevelopmental disorders [14]. CNVs lead to such dosage changes,

35 where a heterozygous deletion of a CNV region leads to a decrease in the normal amount of gene product and conversely, a duplication of the same region will lead to higher amounts of gene product compared to normal. In recent years, a number of studies have shown opposing phenotypes in individuals harboring a deletion versus a duplication of a CNV region [91, 8, 32, 48]. Our study has shown that deletion and duplication of Cyfip1 leads to a SCZ-like set of behavioral defects and an ASD-like behavioral phenotype respectively in mice. Interestingly, we found reciprocal alterations in protein translation and protein levels of NMDAR subunits and postsynaptic components. These molecular alterations in opposite directions may lead to different phenotypes as observed in our mouse models. Finally we were able to rescue the behavioral defects observed in our Cyfip1 deletion mouse model by pharmacologically decreasing NMDAR signaling levels, supporting our hypothesis that Cyfip1 deletion results in de-repression of translation of NMDAR subunits and postsynaptic components, which in turns leads to inappropriately high levels of NMDAR signaling, leading to behavior defects. Conversely, by pharmacologically increasing NMDAR signaling in our Cyfip1 over-expression mouse, we rescued some of the behavior defects, again supporting the hypothesis of imbalance in NMDAR signaling brought upon by a defect in translation regulation by Cyfip1, where an excess of Cyfip1 would lead to higher repression of translation, with con- sequent lower levels of NMDAR subunits and postsynaptic components.

In summary, our study has shown that increase or decrease in Cyﬁp1 levels leads to divergent changes in signaling pathways important in synaptic function and plasticity. Our integrated analyses provides insights into how 15q11.2 CNVs may confer risk to various neurodevelopmental disorders.

36 CHAPTER 3

Temporal control of mammalian cortical neurogenesis by m6A

mRNA Methylation.

3.1 Abstract

N6-methyladenosine (m6A), installed by the Mettl3/Mettl14 methyltransferase complex, is the most prevalent internal mRNA modiﬁcation. Whether m6A regulates mammalian brain development is unknown. Here we show that Mettl14 deletion in the embryonic mouse brain diminishes m6A content, prolongs cell cycle progression of radial glia cells, and extends cortical neurogenesis into postnatal stages.

Mettl3 knockdown also prolongs neural progenitor cell cycle and promotes radial glia cell maintenance. m6A-sequencing of the embryonic mouse cortex reveals enrichment of mRNAs related to transcription factors, neurogenesis, cell cycle and neuronal differentiation, and m6A-tagging promotes their decay. No- tably, Mettl14−/− radial glia cells precociously express neuronal proteins. Further analysis uncovers previously unappreciated transcriptional pre-patterning in corti-

37 cal neural stem cells. Comparison of m6A-mRNA landscapes between mouse and human cortical neurogenesis reveals enrichment of human-speciﬁc m6A-tagging of transcripts related to brain disorder risk genes. Our study identiﬁes an epitranscriptomic mechanism in heightened transcriptional coordination during mammalian cortical neurogenesis.

38 3.2 Introduction

Proper development of the nervous system is critical for its function, and deﬁcits in neural development have been implicated in many brain disorders, such as microcephaly, autistic spectrum disorders, and schizophrenia [27, 42, 92]. In the embryonic mouse cortex, radial glia cells (RGCs) function as neural stem cells, sequentially giving rise to neurons residing in different cortical layers and then switching to glial production before their depletion during early postnatal stages

[47, 92]. Such a precise and predictable developmental schedule requires a highly coordinated genetic program [71]. Indeed, previous studies have revealed transcriptional cascades that orchestrate the dynamics of mammalian cortical neurogenesis [34, 37, 60, 64, 68, 80, 87, 94]. Recent discoveries of widespread mRNA chemical modiﬁcations [20, 116] raise the question of whether this mechanism plays any regulatory role in cortical neurogenesis.

Modiﬁed nucleotides in mRNAs were initially discovered over 40 years ago, but little was known about the extent, transcript identities, and potential functions of various reversible chemical modiﬁcations until very recently [20, 116].

High-throughput sequencing approaches have revealed a dynamic epitranscriptome landscape for many mRNA modiﬁcations in various organisms, including N6- methyladenosine (m6A), N1-methyladenosine (m1A), 5-methylcytosine (m5C), 5- hydroxymethylcytosine (hm5C), pseudouridine (ψ), and 2-O-methylnucleotides

[56]. Among these modifications, m6A is the most abundant internal modification in mRNAs of eukaryotic cells [19]. m6A profiling so far has mostly been performed with cell lines and bulk tissues due to the need of a substantial amount of input mRNAs. These studies revealed m6A sites in over 25% of human transcripts, with enrichment in long exons, and near transcription start sites and stop codons

[21, 43, 61, 63]. In mammals, m6A is installed by the methyltransferase complex

39 consisting of Mettl3 (methyltransferase-like 3), Mettl14, Wtap (Wilms tumour 1- associated protein), KIAA1429, RBM15 (RNA-binding motif protein 15) and its paralogue (RBM15B) [74], whereas its removal is mediated by demethylases Fto

(fat mass and obesity-associated) and Alkbh5 (alkB homolog 5) [62, 116]. Recent in vitro studies have identiﬁed multiple functions of m6A in mRNA metabolism, from processing in the nucleus to translation and decay in the cytoplasm [116].

The ﬁeld has just started to investigate physiological functions of m6A. For example, Mettl3 or Mettl14 knockdown reduces m6A levels and decreases self-renewal of primed mouse embryonic stem cells (mESCs) [104], whereas Mettl3 knockout na¨ıve mESCs exhibit improved self-renewal and impaired differentiation, due to dysregulated decay of m6A-tagged transcripts, such as Nanog [3, 31].

Identiﬁcation of the molecular machinery mediating m6A mRNA methylation provides an entry point to explore physiological functions of this pathway in vivo. Studies of Drosophila development showed that m6A methylation regulates sex determination and neuronal functions by modulating mRNA splicing

[38, 52]. In Zebraﬁsh embryos, m6A-tagging promotes clearance of maternal mR-

NAs and maternal-to-zygotic transition [117]. In mice, germline Mettl3 deletion results in early embryonic lethality [31]. The function of m6A methylation in the intact mammalian system remains elusive; almost nothing is known about its role in mammalian embryonic brain development. Here we used the Mettl14 conditional knockout mouse as a model to examine m6A function in embryonic cortical neurogenesis in vivo. We further investigated underlying cellular and molecular mechanisms. Finally, we extended our analysis to human embryonic cortical neurogenesis using induced pluripotent stem cell (iPSC)-derived forebrain organoids and compared m6A-mRNA landscapes between mouse and human cortical neurogenesis. Together, our results reveal critical epitranscriptomic control of mammalian

40 cortical neurogenesis and provide novel insight into mechanisms underlying this highly coordinated developmental program.

3.3 Methods

3.3.1 Experimental model and subject details

Animals

Exon 7, 8, and 9 of mouse Mettl14 were targeted by inserting a single loxP site in intron 6 and an FRT-ﬂanked neomycin resistance gene coupled with a loxP site in intron 9, with the consideration that they contain the DPWW active motif (Supplemental ﬁgure B.1, B). The targeting construct was electroporated into

129 mESCs, selected for neomycin resistance, screened for homologous recombination by Southern blotting, and selected mESC clones were used to generate chimeric mice by injection into C57BL/6J mouse blastocysts. Chimeric mice were bred to wild type C57BL/6J mice to test for germline transmission of the mutant allele, which was identiﬁed by PCR. The PCR-positive lines were crossed with a

β-actin promoter-driven Flp recombinase to remove the neomycin resistance gene via FRT site recombination. The neomycin cassette deleted mice were identiﬁed by

PCR, and the resultant Mettl14 f / f allele and Nestin-Cre+/Tg mice (Jackson Labo- ratory stock: 003771) [96] were used to generate Nestin-Cre+/Tg; Mettl14+/ f mice and Nestin-Cre+/+; Mettl14 f / f mice. WT and cKO mice were generated by cross- ing Nestin-Cre+/Tg; Mettl14+/ f males and Nestin-Cre+/+; Mettl14 f / f females.

For in utero electroporation analysis, timed-pregnant CD1 mice (Charles River

Laboratory) at E13.5 were used as previously described [112]. Timed pregnant mice were euthanized by cervical dislocation, and embryos were euthanized by decapitation before the dissection step. All animal procedures used in this study were performed in accordance with the protocol approved by the Institutional An-

41 imal Care and Use Committee of Johns Hopkins University School of Medicine.

Primary mouse NPCs

Mouse NPCs were isolated from Mettl14 WT and cKO mouse embryonic cortices and cultured in Neurobasal medium (Gibco BRL) containing 20 ng/ml FGF2,

20 ng/ml EGF, 5 mg/ml heparin, 2% B27 (v/v, Gibco BRL), Glutamax (Invitro- gen), Penicillin/Streptomycin (Invitrogen) on culture dishes pre-coated with Ma- trigel matrix (2%, Corning).

Human iPSC cultures and fetal brain sample

The human iPSC line used in the current study (C1) was fully characterized [108,

112]. They were cultured in stem cell medium, consisting of DMEM:F12 (In- vitrogen) supplemented with 20% Knockout Serum Replacer (Gibco), 1X Non- essential Amino Acids (Invitrogen), 1X Penicillin/Streptomycin (Invitrogen), 1X

2-Mercaptoenthanol (Millipore), 1X Glutamax (Invitrogen), and 10 ng/ml FGF-2

(Peprotech). Culture medium was changed every day. Human iPSCs were passaged every week onto a new plate pre-seeded with irradiated CF1 mouse embryonic ﬁbroblasts (Charles River Laboratory). Human iPSCs were detached from the plate by treatment of 1 mg/ml Collagenase Type IV (Invitrogen) for 1 hr. iPSC colonies were further dissociated into smaller pieces by manual pipetting. All studies were performed under approved protocols of Johns Hopkins University School of Medicine.

Human iPSCs were differentiated into primitive NPCs according to a previously published protocol [55]. Brieﬂy, iPSCs were passaged onto MEF feeders, and after 3 days, induction medium containing Advanced DMEM:F12 (50%) and Neu-

42 robasal medium (50%), CHIR99201 (4 µM, Cellagentech), SB431542 (3 µM, Cel- lagentech), Bovine serum albumin (5 µg/ml, Sigma), hLIF (10 ng/ml, Millipore),

Compound E (0.1 µM, EMD Millipore), Glutamax (Invitrogen), Pen/Strep, supplemented with N2 and B27 (Invitrogen), was added to the culture. After 6 days of differentiation, NPCs were dissociated with Accutase (Invitrogen) and plated, with the aid of a ROCK inhibitor (Y-27632, 3 µM, Cellagentech), onto culture dishes pre-coated with Matrigel matrix (2%, Corning).

The fetal human cortical tissue at 11 postconceptual weeks (PCW) was used for m6A-seq. All procedures used in this study were performed in accordance with the protocol approved by the Institutional Stem Cell Research Oversight Committee of

Johns Hopkins University School of Medicine and Liber Institute for Brain Devel- opment.

3.3.2 Method details

DNA constructs

For knockdown experiments for mouse genes, short hairpin RNA sequences were cloned into the retroviral vector expressing GFP under the control of the EF1 promoter and a specific shRNA under the control of human U6 promoter (pUEG) as previously described [29]. For knockdown experiments for human METTL14, short hairpin RNA sequence was cloned into the lentiviral vector expressing GFP under the control of the human ubiquitin C promoter and the specific shRNA under the control of human U6 promoter (cFUGW: Addgene plasmid 14883) as previously described [112]. The efficacy of each shRNA was confirmed in mouse B16-

F10 cells (ATCC), or human NPCs derived from C1 iPSC line.

43 In utero electroporation and Flash-Tagging

In utero electroporation was performed as described previously [112]. In brief, timed-pregnant CD1 mice (Charles River Laboratory) at E13.5 or E14.5 were anes- thetized and the uterine horns were exposed and approximately 1 to 2 µl of plasmid

DNA, 0.5 µg/µl pCAG-GFP (Addgene plasmid: 11150) and 2.5 µg/µl cFUGW plasmid with the control shRNA, or the shRNA against mouse Mettl3, Cnot1 and

Cnot7, was injected manually into the lateral ventricles of embryos using a calibrated micropipette. Five pulses (40 V, 50 ms in duration with a 950 ms interval) were delivered across the uterus with two 5-mm electrode paddles (CUY650-5,

Nepa Gene) positioned on either side of the head by a square wave electroporator (CUY21SC, Nepa Gene). After electroporation, the uterus was placed back in the abdominal cavity and the wound was sutured. Mouse embryos were analyzed at E17.5. For Flash-Tagging of RGCs, 1 µl of 10 µM of a carboxyﬂuorescein succinimidyl ester (CellTrace CFSE, ThermoFisher) was injected into the lateral ventricle of the E17.5 embryos using a calibrated micropipette. Mouse embryos were collected 3 hr later, ﬁxed with with 4% paraformaldehyde in PBS overnight at 4 degrees Celsius for analysis. All animal procedures were performed in accordance with the protocol approved by the Johns Hopkins Institutional Animal Care and Use Committee.

Immunohistology and confocal imaging

For EdU labeling, timed pregnant mice were injected with EdU (150 mg/kg body- weight, Invitrogen) at deﬁned time points before euthanasia. For immunostaining of tissue sections, brains were ﬁxed with 4% paraformaldehyde in PBS overnight at 4 degrees Celsius as previously described [112]. Samples were cryoprotected in

30% sucrose in PBS, embedded in OCT compound, and sectioned coronally (20

44 µm-thickness) on a Leica CM3050S cryostat. Brain sections were blocked and permeabilized with the blocking solution (5% normal donkey serum, 3% Bovine serum albumin, and 0.1% Triton X-100 in PBS) for 1 hr at room temperature, followed by incubation with primary antibodies diluted in the blocking solution at 4 degrees Celsius overnight. After washing, secondary antibodies diluted in blocking solution were applied to the sections for 1 hr at room temperature. Nuclei were visualized by incubating for 10 min with 0.1 µg/ml 4,6-diamidino-2-phenylindole

(DAPI, Thermo Fisher Scientiﬁc) in PBS. Stained sections were mounted with Pro-

Long Gold anti-fade reagents (Thermo Fisher Scientiﬁc) and analyzed.

Mouse and human NPC electroporation

Approximately 1.0 X 106 mouse or human NPCs were resuspended in 100 µL

Mouse Neural Stem Cell Nucleofector Solution from the Lonza Nucleofector Kit for Mouse Neural Stem Cells (Lonza, VAPG-1004). Additionally, 10 µg of the appropriate plasmid was added to the cell solution. The solution was then placed in a cuvette provided in the Nucleofector Kit and electroporated using a Lonza Nu- cleofector 2b device (LONZA). Next, the cells were resuspended in NPC media as described above with Rock Inhibitor (Y-27632, 3 µM, Cellagentech) to reduce cell death. Cells were allowed to grow for at least 3 days before analysis.

Human forebrain organoid culture

Protocols for generation of forebrain organoids were detailed previously [75]. Brieﬂy, human iPSCs were cultured in stem cell medium, consisting of DMEM:F12 (In- vitrogen) supplemented with 20% Knockout Serum Replacer (Gibco), 1X Non- essential Amino Acids (Invitrogen), 1X Penicillin/Streptomycin (Invitrogen), 1X

45 2-Mercaptoenthanol (Millipore), 1X Glutamax (Invitrogen), and 10 ng/ml FGF-2

(Peprotech) on irradiated CF1 mouse embryonic ﬁbroblasts (Charles River). On day 1, iPSC colonies were detached by treatment of 1 mg/ml Collagenase Type

IV (Invitrogen) for 1 hr and transferred to an Ultra-Low attachment 6-well plate

(Corning Costar), containing 3 ml of stem cell medium (without FGF-2), plus 2

µM Dorsomorphine (Sigma) and 2 µM A83-01 (Tocris). On days 5-6, half of the medium was replaced with induction medium consisting of DMEM:F12, 1X N2

Supplement (Invitrogen), 1X Penicillin/Streptomycin, 1X Non-essential Amino

Acids, 1X Glutamax, 1 µM CHIR99021 (Cellagentech), and 1 µM SB-431542

(Cellagentech). On day 7, organoids were embedded in Matrigel (Corning) and continued to grow in induction medium for 6 more days. On day 14, embedded organoids were mechanically dissociated from Matrigel and transferred to each well of a 12-well spinning bioreactor (SpinΩ) containing differentiation medium, consisting of DMEM:F12, 1X N2 and B27 Supplements (Invitrogen), 1X Peni- cillin/Streptomycin, 1X 2-Mercaptoenthanol, 1X Non-essential Amino Acids, 2.5

µg/ml Insulin (Sigma).

Forebrain organoid electroporation and analysis

Day 45 forebrain organoids were transferred into PBS solution in a 10 cm petri dish for electroporation. A mixture of 0.5 µl of plasmid DNA and 0.05% Fast green was injected into the lumen of neural tube structures in forebrain organoids using a calibrated micropipette. About 3-4 locations on one side of each forebrain organoid was targeted by the injection. The DNA-injected side of the organoid was placed toward the positive electrode in the middle of 5 mm gap of electrode paddles (CUY650-5, Nepa Gene). Five pulses (40 V, 50 ms in duration with a 950 ms interval) were delivered by a square wave electroporator (CUY21SC, Nepa Gene).

46 After electroporation, organoids were transferred back to the SpinΩ bioreactor for further culturing.

Analysis of cell cycle progression by EdU pulse labeling

Analyses of cell cycle progression of mouse NPCs, hNPCs, and dissociated human forebrain organoids were performed as described previously [105, 95]. In brief, mouse or human NPCs were pulsed by 10 µM EdU (ThermoFisher) for 30 min and washed thoroughly with NPC media. For human forebrain organoids, 10 µM

EdU directly applied to culture media and organoids were incubated in the SpinΩ bioreactor for 1 hr to ensure complete penetrance, then washed thoroughly with culture media. After deﬁned time points, cells were dissociated by Accutase, ﬁxed with 4% paraformaldehyde in PBS for 20 min at 4 degrees Celsius, stained with

Click-iT EdU Alexa 647 Flow Cytometry Kits (ThermoFisher) for Flow Cytome- try following manufacturers protocol. Cells were stained with Vybrant DyeCycle

Violet (ThermoFisher) or 7-AAD (ThermoFisher) for DNA content and applied to ﬂow cytometry using BD LSR II Flow Cytometer (BD Bioscience). EdU+ or

GFP+EdU+ cells were gated and DNA content of those cells was analyzed compared to that of whole cell population. Percentages of divided cells among EdU+ or GFP+EdU+ population (G1 or G0 phase determined by DNA content) during deﬁned time intervals were quantiﬁed from four independent experiments.

Time-lapse live imaging of mouse NPCs

96-well glass bottom microplates (655892, Geiner bio-one) were coated with phe- nol red-free Matrigel (356237, Corning). After electroporation of mNPCs with 10

µg CDK2-sensor plasmid (pPGK-H2B-mCherry-DHB(aa994-1087)-GFP), cells

47 were plated onto the microplates at a density of 3,000 cells per well and allowed to adhere overnight. Cells were imaged using a Nikon Eclipse Ti ﬂuorescent mi- croscope controlled by Metamorph microscopy automation software. Temperature

(37 degrees Celsius), CO2 (5%), and humidity were held constant throughout experiments. Five blank positions in a well containing Matrigel and media only were used to flat field mNPC images using custom software. ImageJ was used to merge the green and red channels. To quantify the total cell cycle length, time was measured from the first cell division to the next cell division of one or both daughters.

To quantify the G1 phase length, time was measured from one cell division to the time point of signiﬁcant reduction in the ratio of green/red intensity in the nucleus of the cell. S phase entry was quantitatively deﬁned as the time when the cytoplasmic/nuclear ratio of green/red was approximately 1, as previously described [89].

A nuclear marker, H2B-mCherry, was used in the plasmid sensor to accurately segment the cytoplasm and the nucleus. The time point from S phase entry through the second cell division was then quantiﬁed as S-G2-M length.

RNA puriﬁcation and quantitative RT-PCR analysis

For gene expression analysis, total RNA fraction was isolated from cultured NPC samples with RNeasy Mini Kit (Qiagen), treated with DNaseI and reverse-transcribed into the ﬁrst-strand cDNA with SuperScript III (Invitrogen). cDNAs were used for

SYBR-green based quantitative real-time PCR to measure the expression level of target genes with the T method (ABI).

Western blot analysis

Forebrains from E17.5 embryos were quickly dissected out and homogenized in

RIPA buffer (50 mM Tris pH 7.5, 120 mM NaCl, 1% Triton X-100, 0.5% Sodium

48 Deoxycholate, 0.1% SDS, 5 mM EDTA, Phosphatase Inhibitor Cocktail (Cell Sig- naling), protease inhibitor cocktail (Sigma). Lysates were incubated for 15 min on ice and centrifuged at 15,000g for 15 min at 4 degrees Celsius. Supernatant was collected and boiled for 5 min in Laemmli sample buffer (Biorad), resolved by

SDS PAGE, transferred to PVDF membrane, and immunoblotted. Quantiﬁcation of bands was performed using ImageJ software.

m6A dot blot assay mRNA was harvested from homogenized forebrains at embryonic stages E15.5 and E17.5 using Dynabeads mRNA Direct Puriﬁcation Kit (61011, Ambion). Four biological replicates were pooled for each sample to ensure sufﬁcient concentration of mRNA. Dots were applied to an Amersham Hybond-N+ membrane (GE

Healthcare) in duplicate as 100 ng mRNA per 1 µl dot. After complete drying, the mRNA was crosslinked to the membrane using a UV Stratalinker 2400 by running the auto-crosslink program twice. The membrane was then washed in PBST three times and blocked with 5% skim milk in PBST for 2 hr. The PBST wash was repeated and the membrane was incubated with primary anti-m6A antibody

(212B11, Synaptic Systems) at 1:1000 dilution for 2 hr at room temperature. After

3 washes in PBST, the membrane was incubated in HRP-conjugated anti-mouse

IgG secondary antibody for 2 hr at room temperature, then washed again 3 times in PBST. Finally, the membrane was visualized using SuperSignal West Dura Ex- tended Duration Substrate (34075, Thermo Scientific). To confirm equal mRNA loading, the membrane was stained with 0.02% methylene blue in 0.3 M sodium acetate (pH 5.2) and quantified m6A levels were normalized to amount of mRNA loaded. Four biological samples in technical duplicates for each time point were used.

49 m6A-sequencing m6A profiling was performed as previously described [21]. For m6A profiling of mouse developing brain, forebrains from WT E13.5 embryos were dissected to extract total RNA. For m6A profiling of human organoids, 25 to 30 forebrain organoids at day 47 were used for total RNA extraction using RNeasy Mini Kit

(Qiagen). For m6A proﬁling of PCW11 fetal human brain, cortex from 2 PCW11 fetuses were dissected and total RNA was extracted. mRNA was isolated using the

Dynabeads mRNA Puriﬁcation Kit (Invitrogen) and mRNA was fragmented via sonication to 100-200 base pairs. m6A pull-down was performed using a rabbit polyclonal anti-m6A antibody (Synaptic systems), and immunoprecipitation with protein G dynabeads (ThermoFisher). m6A-tagged mRNAs were competitively eluted from beads with free N6-methyladenosine. cDNA libraries from pulled- down RNA and input RNA were prepared using the NEBNext Ultra DNA Library

Prep Kit for Illumina. The experiment was performed with three technical replicates. For m6A proﬁling of day 47 human forebrain organoids, the same procedure was followed, with the exception that the experiment was performed with two technical replicates because of the amount of samples required.

m6A mRNA immunoprecipitation and Q-PCR

Total RNA from NPCs cultured from WT E13.5 mouse forebrain was extracted using RNeasy Mini Kit (Qiagen) and mRNA was isolated using the Dynabeads mRNA Puriﬁcation Kit (Invitrogen). 1% of input mRNA was reserved for reverse transcription. Full length m6A tagged transcripts were pulled-down using a rabbit polyclonal anti-m6A antibody (Synaptic systems) and a mock pull-down was done

50 with normal rabbit IgG (Cell Signaling Technologies). Immunoprecipitation was performed with protein G dynabeads (Thermo Fisher). m6A-tagged mRNAs were competitively eluted from beads with free N6-methyladenosine. Reverse transcription of input, m6A pull-down and mock pull-down mRNA was performed using the

SuperScript III First-Strand Synthesis System for RT-PCR (Thermo Fisher). cDNA was used for SYBR-green based quantitative real-time PCR. Enrichment of m6A tagged genes in m6A pull-down over input was calculated by comparing relative concentrations using Ct values (2−Ct ) and dividing each concentration by the relative concentration of the input. The concentrations of the immunoprecipitated

RNA was then divided by the concentration in the input RNA and multiplied by

100, to obtain the percentage of transcripts in the m6A immunoprecipitation relative to the input. This value was then normalized to enrichment in the mock (IgG) pull-down, which was also calculated using relative concentrations to determine a percentage of the input.

Bioinformatic analyses of m6A-seq cDNA libraries from input and m6A pull-down were sequenced on the Illumina

Nextseq platform, using a 50-cycle single-end run. Pre-processing of reads was performed using the FASTX toolkit, namely adapters were clipped, poor quality reads were ﬁltered out, and identical reads were collapsed. Pre-processed reads from E13.5 mouse forebrains were aligned to the mouse genome (build

GRCm38/mm10), and reads from the human organoids and fetal brain to the human genome (build GRCh37/hg19), using Tophat2 [45] with default settings. m6A- tagged regions were identiﬁed using the MACS2 peak calling algorithm [115], with the input library as background. For identifying high conﬁdence m6A regions, peaks were intersected in a pairwise fashion among all replicates using the Bed-

51 Tools package [76]. Peaks that overlap in at least 50% of their length among 2 or more samples were designated as high conﬁdence m6A regions.

For representative coverage plots of m6A and input libraries, RNA-seq read align- ments in bam format were transformed to bedGraph format and normalized for library size using the genomecov function from the BedTools package [76]. Anal- ysis of m6A peak enrichment was performed based on 5 non-overlapping transcript segments deﬁned as follows: Transcription start site (TSS) segment [TSS,

TSS+200bp], 5′UTR [TSS+201bp, CDS start-1bp], coding region (CDS) [CDS start, CDS stop-101bp], stop codon segment [CDS stop-100bp, CDS stop+100bp],

3′ UTR [CDS stop+101bp, TTS]. Each high conﬁdence peak was annotated to one of these regions using the BedTools package and fold enrichment was calculated from the ratio between observed peaks per region and expected number of peaks normalized by average region size. For analysis of correlation between gene expression levels and m6A peak fold change, we calculated RPKMs from input RNA seq libraries, using gene counts obtained with the htseq-count function from the

HTSeq python package [1] that were normalized by library size and gene length deﬁned as the length of its longest transcript. Fold changes for m6A peaks were obtained from MACS2 output.

Functional annotation and disease ontology

To assess enrichment of GO terms speciﬁc to biological process, the ToppFunn module of the ToppGene Suite [11] was used. A hypergeometric probability mass function with Benjamini Hochberg FDR correction was used to identify signiﬁcant enrichment for GO terms. Analysis of enrichment for Wikipathways terms was performed using ConsensusPathDB [39], which calculates enrichment p-values using the Wilcoxon’s matched-pairs signed-rank test, and Benjamini Hochberg FDR

52 correction. Disease association analysis was performed using WebGestalt [114], which uses a hypergeometric method and Benjamini Hochberg FDR correction.

Protein interaction network ﬁgures were generated using Cytoscape 3.3.0 [86], with the Reactome FI plugin.

RNA degradation assay cDNA libraries were prepared from cultured NPCs from E13.5 WT and Mettl14 cKO cortex, at 0 and 5 hr post Actinomycin D treatment, using the NEBNext Ul- tra RNA Library Prep Kit for Illumina. The experiment was performed with three replicates per condition. Sequencing was performed on the Illumina Nextseq platform, using a 100-cycle single-end run. Pre-processing of reads was performed using the FASTX toolkit. Gene expression levels were quantiﬁed using the RSEM package [54], which maps reads to the transcriptome using the aligner tool Bowtie2

[50]. Expected counts per gene per sample were combined into a count matrix, and this matrix was used as input for differential expression analysis using the EBSeq package [53], which uses empirical Bayesian methods to calculate the posterior probability of a gene being differentially expressed (PPDE). Posterior fold changes per gene between cKO and WT were obtained at time 0 and 5 hr after Actinomycin

D treatment. Fold changes at 5 hr were normalized by fold changes at 0 hr (no

Actinomycin D treatment) to speciﬁcally identify genes that degrade at a different slower rate in the cKO compared to WT, regardless of baseline changes in gene expression between two conditions. Genes with a normalized fold change higher than 2 in cKO over WT at 5 hr were considered as to be differentially degraded.

53 Half-life measurement of m6A-tagged transcripts

Mouse NPC were cultured in standard 6 well culture plates to approximately 80% conﬂuence. Actinomycin D (Sigma) was added at a concentration of 5 µM. Cells were collected at three time points after addition (0, 3 hr, 5 hr) by washing once with PBS, then lysing the cells in Buffer RLT from the RNeasy Kit (Qiagen) with

1% -Mercaptoethanol. A cell scraper was used to remove all cells from the well plate. Each sample was normalized for cell number by quantifying DNA content using a Quant-IT Pico Green dsDNA Assay Kit (ThermoFisher) according to manufacturer instructions. Equal amounts of cellular contents, as measured by

DNA quantity, were taken from each sample and 1 pg of luciferase control RNA

(Promega) was added to each sample before RNA puriﬁcation. Total RNA was then puriﬁed using an RNeasy Kit and reverse transcribed using the SuperScript

II First-Strand Synthesis System (Thermo Fisher). Real time PCR was done on a Step One Plus cycler from Applied Biosystems with Fast SYBR Green Master

Mix. Standard curves were generated by plotting CT values against the known initial concentration of luciferase control RNA, and then used to derive mRNA concentration of each target gene at each time point. The natural logarithm of mRNA concentrations at time point 0, 3 and 5 hr were then used to perform a lin- ear regression as a function of time, and identify the slope of said line as the decay rate (k). Half life was calculated with the following formula [10]: t1/2 = ln(2/k).

Metabolic labeling and puriﬁcation of nascent RNA

4sU labeling of nascent RNA was performed as previously described [24]. Mouse

NPC from E13.5 WT and Mettl14 cKO cortex were cultured in standard 6 well culture plates to approximately 80% conﬂuence, treated with 500 µM of 4sU (Car- bosynth) for 1 hr, washed with PBS, and harvested with TRIzol reagent (Ther-

54 moFisher). Samples were extracted by chloroform twice and precipitated with isopropanol. Biotinylation of 4sU-RNA were carried out in a total volume of 250

µl, containing 70 µg total RNA, 10 mM HEPES (pH 7.5), 1 mM EDTA, and 5

µg MTSEA biotin- XX (Biotium) freshly dissolved in DMF (ﬁnal concentration of DMF = 20%). Reactions were incubated at RT for 30 min in the dark, and excess biotin reagents were removed by chloroform extraction twice. Puriﬁed RNA was dissolved in 50 µl RNase-free water and denatured at 65 degrees Celsius for

10 min, followed by rapid cooling on ice for 5 min. Biotinylated RNA was separated from non-labeled RNA by incubating with 100 µl Streptavidin Magnetic

Beads (ThermoFisher) for 20 min at RT. Beads were washed twice with high-salt wash buffer (500 µl each, 100 mM Tris- HCl pH 7.4, 10 mM EDTA, 1 M NaCl, and 0.1% Tween-20). 4sU-RNA was eluted with 100 µl freshly prepared 100 mM

DTT followed by a second elution with an additional 100 µl 5 min later. RNA was recovered using the MinElute Spin columns (Qiagen) according to the instructions of the manufacturer, and applied for Q-PCR analysis.

Comparison between human and mouse m6A-seq datasets

For comparison of m6A sequencing data from day 47 human forebrain organoids,

PCW11 fetal human cortex, and mouse E13.5 forebrains, we restricted our analysis to genes with a one-to-one ortholog between species. For determining expressed genes, we calculated RPKMs (as stated above) from input libraries, and used a threshold of RPKM >1.

55 3.4 Results

3.4.1 Nervous system Mettl14 deletion extends cortical neurogenesis into postnatal stages

We ﬁrst investigated the expression pattern of molecular players mediating m6A signaling during mouse embryonic cortical neurogenesis. Mining the recently published single-cell RNA-seq dataset of RGCs and their progeny [94] revealed that

Mettl14 exhibits the highest expression in RGCs, whereas other m6A methyltransferase components (Mettl3, Wtap), demethylases (Fto, Alkbh5) and m6A readers

(Ythdf2, Ythdf3) exhibited similar levels during neurogenesis (Supplemental ﬁgure

B.1, A). To examine the functional role of m6A methylation during cortical development in vivo, we conditionally deleted Mettl14 in the developing mouse nervous system using the Nestin-Cre;Mettl14 f / f (cKO) model (Supplemental ﬁgure B.1,

B). We confirmed Mettl14 deletion at the protein level with Western blot analysis of E17.5 brains (Supplemental figure B.1, B). The cKO animals were smaller in size by P5 compared to wildtype (WT) littermates, and all cKO animals died before P25 (Supplemental figure B.1, C and D). Thus, the function of m6A molecular machinery in the nervous system is indispensable for life in the mammalian system.

We next examined cortical structures at P5. The cKO mice exhibited enlarged ventricles with an adjacent dense layer of cells that appeared to resemble the embryonic germinal zone (Figure 3.1, A). Immunohistological analysis showed the presence of Pax6+ and Nestin+ cells with radial ﬁbers along the ventricle in cKO mice, but not in WT mice (Figure 3.1, B). During mouse cortical development

Pax6+ RGCs are largely depleted by P5 [25], whereas a substantial number of

Pax6+ cells were present in cKO mice at P5 (Figure 3.1, C). Neurogenic Pax6+

RGCs give rise to intermediate progenitor cells (IPCs) expressing Tbr2/Eomes

56 Figure 3.1: Nervous system Mettl14 deletion results in residual radial glia cells and ongoing neurogenesis in the postnatal mouse cortex. (A-C) Presence of neurogenic RGCs in P5 Nestin-cre;Mettl14 f / f cKO cortices. Shown are sample confocal images (A, B) and quantifications (C). Regions in white boxes are shown at a higher magnification. Scale bars: 500 µm (A), 50 µm (A insert), 100 µm (B). Values in (C) represent mean + SEM (n = 4-7; ***: P <0.001; *: P <0.05; unpaired Students t-test). (D-E) Preserved IPCs in P5 cKO cortices. Shown are sample confocal images (D; scale bars: 100 µm) and quantification (E). Values represent mean + SEM (n = 6; ***: P <0.001; **: P <0.01; unpaired Students t-test). (F-G) Ongoing neurogenesis in P5 cKO cortices. P5 pups were injected with EdU and analyzed 48 hr later. Shown in (F) are sample confocal images of the ventricular side of the primary somatosensory cortex. Arrows indicate Pax6+EdU+ cells (top) and Tbr2+TuJ1+EdU+ cells (bottom). Scale bars: 100 µm. Quantification of EdU+ cells with different markers is shown in (G). Values represent mean + SEM (n = 6; ***: P <0.001; unpaired Students t-test). (H-K) Reduced production of upper-layer neurons and astrocytes in cKO cortices. Pregnant mice were injected with EdU at E15.5 and analyzed at P5. Shown are sample confocal images (H, J; scale bars: 100 µm) and quantification (I, K). Values represent mean + SEM (n = 6; ***: P <0.001; unpaired Students t-test).

57 [26]. The presence of Pax6+ cells in cKO mice was accompanied by Tbr2+ IPCs, which were absent in WT mice by P5 (Figure 3.1, D and E). To conﬁrm that cortical neurogenesis continued postnatally, we pulsed animals with EdU at P5 and analyzed 2 days later. Signiﬁcant numbers of EdU+Pax6+ proliferating RGCs,

EdU+Tbr2+ IPCs, and EdU+Tbr2+TuJ1+ neuroblasts were present in cKO mice, but very few in WT littermates (Figure 3.1, F and G). These results indicate that cKO mice maintain neurogenic RGCs with extended cortical neurogenesis into postnatal stages.

To further characterize the impact of Mettl14 deletion on cortical development, we examined neuronal subtype and glia production. We pulsed animals with EdU at E15.5 and examined them at P5. Compared to WT littermates, cKO mice exhibited a signiﬁcantly decreased number of EdU+Satb2+ neurons, suggesting a deﬁcit in producing late-born upper-layer neurons (Figure 3.1, H-I). Direct measurement of the number of different cortical neuron subtypes also showed a reduced number of Satb2+ upper-layer neurons, but comparable numbers of Tbr1+ and

Ctip2/Bcl11b+ lower-layer early-born neurons in P5 cKO mice (Supplemental ﬁg- ure B.2, A and B). On the other hand, analysis of Ctip2+ neurons at E17.5 showed a reduction in cKO mice, suggesting a delay in production of neuron subtypes of different cortical layers, instead of a defect in differentiation (Supplemental ﬁgure

B.2, C and D). In addition, we observed a signiﬁcant decrease in the number of s100β+ astrocytes in cKO mice at P5 (Figure 3.1, J and K). Together, these results indicate that Mettl14 function is critical for proper temporal progression of neurogenesis and gliogenesis during mouse cortical development in vivo.

58 3.4.2 Mettl14 deletion in neural progenitor cells leads to protracted cell cycle progression.

Given the well-deﬁned temporal progression of cortical neurogenesis from RGCs

[71], we suspected that there could be RGC deﬁcits during embryonic stages in cKO mice. Interkinetic nuclear migration (INM), the periodic movement of the cell nucleus in phase with cell-cycle progression, is a common feature of developing neuroepithelia [36, 92]. We pulsed animals with EdU at E17.5 to label cells in S-phase and followed positions of nuclei in EdU+Pax6+ RGCs (Figure 3.2, A).

While there was no difference at 0.5 hr, nuclei of labeled RGCs were positioned further away from the ventricular surface at 6 hr in cKO mice compared to WT

(Figure 3.2, B), suggesting delayed INM and potential cell cycle deﬁcits. To directly examine the S to M phase transition, we analyzed expression of phospho-

Histone 3 (pH3), an M phase marker, 2 hr after EdU labeling (Figure 3.2, C). We found a signiﬁcant decrease in the percentage of EdU+pH3+Pax6+ cells among all pH3+Pax6+ cells in cKO mice, suggesting a prolonged S to M phase transition of RGCs (Figure 3.2, D). To examine cell cycle exit of proliferating neural progenitors, we analyzed expression of Ki67, a proliferation marker, 24 hr after EdU labeling (Figure 3.2, E). We found a signiﬁcant decrease in the percentage of Ki67 negative cells among EdU+ cells in cKO mice, indicating a delay in cell cycle exit

(Figure 3.2, F).

To address the cell intrinsic effect of Mettl14 deletion on cell cycle progression, we performed time-lapse imaging of individual cortical neural progenitor cells

(NPCs) cultured from E13.5 mouse cortex. We used a dual reporter system with nuclear localized H2B-mCherry and a GFP-tagged Cdk2 substrate, DNA Helicase

B (DHB) [89]. Cdk2 becomes active during the G1-S transition and phosphorylates DHB-GFP, which is then translocated from nucleus to cytoplasm. Therefore,

59 Figure 3.2: Mettl14−/− RGCs and NPCs exhibit prolonged cell cycle progression. (A-B) Abnormal INM of RGCs in Mettl14 cKO cortices. Pregnant mice were injected with EdU at E17.5 and analyzed 0.5 or 6 hr later. Shown are sample confocal images (A; scale bars: 50 µm) and quantification of the distance from Pax6+EdU+ nuclei to the ventricular surface (B). Values for the percentages of nuclei in each 20 µm bin represent mean + SEM (n = 4; ***: P <0.001; **: P <0.01; *: P <0.05; unpaired Students t-test). (C-D) Delayed S to M phase transition of RGCs in Mettl14 cKO mice. Pregnant mice were injected with EdU at E17.5 and analyzed 2 hr later. Shown in (C) are sample confocal images. Arrowheads point to Pax6+pH3+EdU+ cells and arrows point to Pax6+pH3+EdU- cells. Scale bar: 50 µm. Shown in (D) is the quantification of the percentage of Pax6+pH3+EdU+ cells, representing cells proceeded from S to M phase during the 2 hr chase, among total Pax6+pH3+ cells. Values represent mean + SEM (n = 5 for WT and n = 8 for cKO; ***: P <0.001; unpaired Students t-test). (E-F) Delayed cell cycle exit of neural progenitors in Mettl14 cKO mice. Pregnant mice were injected with EdU at E17.5 and analyzed 24 hr later. Shown in (E) are sample confocal images. Arrowheads point to Ki67-EdU+ cells and arrows point to Ki67+EdU- cells. Scale bar: 50 µm. Shown in (F) is the quantification of the percentage of Ki67-EdU+ cells, representing cells exited from cell cycle, among total EdU+ cells. Values represent mean + SEM (n = 6; ***: P <0.001; unpaired Students t-test). (G-J) Time-lapse imaging analysis of mouse NPCs showing prolonged S-G2-M phase length in the absence of Mettl14. Shown in (G) are sample time-lapse images with time stamps. Scale bars: 10 µm. Also shown are box plots of quantifications for the total cell cycle length (H; n = 38 for WT and n = 30 for cKO), G1 phase length (I; n = 20), and S-G2-M phase length (J; n = 20). Each dot represents data from one NPC (***: P <0.001; unpaired Students t-test).

60 the presence of GFP in the mCherry+ nucleus indicates cells in G1 phase, whereas translocation to the cytoplasm indicates the initiation of the S phase, and continual buildup of cytoplasm GFP occurs until mitosis (Supplemental ﬁgure B.3, A).

Quantiﬁcation of the length between sequential mitoses showed an increase of the total cell cycle length in Mettl14−/− NPCs (Figure 3.2, G and H). Further analysis of different cell cycle phases revealed a speciﬁc increase of the S-G2-M phases in the absence of Mettl14, but no difference in the G1 phase (Figure 3.2, I and J).

To quantify cell cycle characteristics at the population level, we pulsed NPCs with EdU for 30 min and performed flow cytometry analysis 0 or 5 hr later (Supple- mental figure B.3, B). We found a significant decrease in the percentage of EdU+ cells that divided in Mettl14-/- NPCs compared to WT at 5 hr, indicating a delay in cell cycle progression (Supplemental figure B.3, C and D).

3.4.3 Mettl3 regulates embryonic cortical neurogenesis

Consistent with the ﬁnding that Mettl14 is a critical component of the m6A methyltransferase complex [102], Mettl14 deletion led to a signiﬁcant reduction of m6A levels in mRNAs from both embryonic mouse cortex in vivo and cultured cortical

NPCs (Figure 3.3, A and B). To further assess our model that m6A methylation regulates cortical neurogenesis, we compared the phenotype of Mettl14 cKO to knockdown of Mettl3, another critical component of the m6A methyltransferase complex [102].

We first confirmed effective Mettl3 knockdown (KD) with Q-PCR and dimin- ished m6A content in mRNAs from Mettl3 KD cells with dot blot analysis (Sup- plemental figure B.4, A-C). We next performed population cell cycle analysis with

61 Figure 3.3: (A) Depletion of m6A-tagging on mRNAs purified from E15.5 and E17.5 Mettl14 cKO mouse forebrain. Shown in the left panels are sample images of m6A dot blot and methylene blue staining (for loading controls). Data were normalized to the averaged levels of WT samples and quantification is shown in the right panel. Values represent mean + SEM (n = 3; **: P <0.01; unpaired Students t-test).(B) Depletion of m6A-tagging on mRNAs purified from Mettl14 cKO NPCs. Values represent mean + SEM (n = 3; **: P <0.01; unpaired Students t-test).

EdU pulse-chase and flow cytometer quantification (Supplemental figure B.4, D).

We found a significant reduction in the percentage of GFP+EdU+ NPCs that divided upon Mettl3 KD (Figure 3.4, A and B), similar to the effect of Mettl14 cKO (Supplemetal figure B.3, C-D). To examine the impact of Mettl3 KD on RGC behavior in vivo, we electroporated plasmids co-expressing GFP and the shRNA against mouse Mettl3, or the control shRNA, in utero at E13.5 and analyzed GFP+ cells at E17.5. Newborn neurons normally migrate toward the cortical plate (CP) through the intermediate zone (IZ), whereas self-renewing RGCs remain in the ventricular zone (VZ) and subventricular zone (SVZ) [92]. Compared to the control group, GFP+ cells with Mettl3 KD were more abundant in the VZ and SVZ and less abundant in the CP (Figure 3.4, C and D), similar to the result of EdU fate mapping in Mettl14 cKO mice (Figure 3.1, H). There was also a significant increase in the percentage of GFP+Pax6+ cells among all GFP+ cells with Mettl3

KD compared to the control group (Figure 3.4, E).

Together, these results indicate that decreasing m6A levels by either Mettl14

62 A B

C D E

Figure 3.4: Mettl3 regulates cell cycle progression of NPCs and maintenance of embryonic cortical RGCs. (A-B) Flow cytometry analysis of cell cycle status of mouse NPCs. NPCs were electroporated to co-express GFP and the control shRNA, or the shRNA against Mettl3. At day 4, NPCs were pulse-labeled with EdU (10 µM) for 30 min, cultured for 9 hr, followed by EdU and DNA content (DyeCycle Violet) staining and flow cytometry analysis. Shown are sample histograms of DNA content from GFP+EdU+ cells and the total cell population (as a reference (A) and quantification (B). Values in (B) represent mean + SEM (n = 4; **: P <0.01; unpaired Students t-test). (C-E) Embryonic mouse cortices were electroporated in utero at E13.5 to co-express GFP and shRNA-control, or GFP and shRNA-Mettl3, and analyzed at E17.5. Shown in (C) are sample confocal images. Scale bars: 50 µm. The distribution of GFP+ cells in each zone (D) and the percentage of GFP+Pax6+ cells among total GFP+ cells (E) were quantified. VZ: ventricular zone; SVZ: subventricular zone; IZ: intermediate zone; CP: cortical plate. Values represent mean + SEM (n = 4; ***: P <0.001; **: P <0.01; unpaired Students t-test).

63 cKO or Mettl3 KD leads to consistent phenotypes of protracted cell cycle progression of cortical NPCs and reduced differentiation of RGCs during mouse embryonic cortical neurogenesis.

3.4.4 m6A tags transcripts related to transcription factors, cell cycle, and neurogenesis, and promotes their decay

To gain insight into the molecular mechanism underlying m6A regulation of cortical neurogenesis, we performed m6A-seq of mouse forebrain at E13.5, a stage enriched for neural stem cells. We identified 4,055 high confidence m6A peaks corresponding to 2,059 gene transcripts (Supplemental figure B.5, A). Similar to previous findings from cell lines [21, 43, 61, 63], our in vivo analysis showed enriched distribution of m6A sites near stop codons (Supplemental figure B.5, B).

We found no correlation between transcript levels and m6A-tagging (Supplemental

ﬁgure B.5, C). Notably, many transcripts encoding transcription factors were m6A- tagged, such as Pax6, Sox1, Sox2, Emx2, and Neurog2/Neurogenin 2 (Figure 3.5,

A and B). Gene ontology (GO) and Wikipathways analyses of m6A-tagged transcripts revealed enrichment of genes related to cell cycle, stem cell, and neuronal differentiation (Figure 3.5, A-C). We observed similar m6A-tagging for a selected group of transcripts in cortical NPCs derived from E13.5 mouse cortex (Supple- mental ﬁgure B.5, D).

To determine the functional consequence of m6A tagging on mRNAs, we explored whether Mettl14 deletion affects decay of m6A-tagged transcripts with an

RNA stability assay [31, 103]. Cortical NPCs derived from E13.5 WT and Mettl14 cKO mice were treated with Actinomycin D to halt de novo transcription, and

RNA-seq was performed 0 and 5 hr later to obtain the ratio of mRNA levels for

64 Figure 3.5: m6A tags transcripts related to transcription factors, cell cycle, and neuronal differentiation in the embryonic mouse brain. (A) Coverage plots from m6A-seq of E13.5 mouse forebrains showing representative examples of m6A-tagged (Sox1, Emx2, and Cdk9) and non m6A-tagged (Rad17) transcripts. Top and middle panels show read coverages normalized by library sizes from m6A pulled-down and input libraries, respectively, and bottom panels show gene structures (arrows point to the direction of transcription; S.C.: stop codon). (B-C) GO analysis of m6A-tagged genes reveals enrichment for biological process terms related to transcription factors, neurogenesis, cell cycle, and stem cell differentiation. Also shown is Wikipathways gene set enrichment analysis. FDR: false discovery rate.

65 each gene in order to measure their stability (Supplemental ﬁgure B.5, E). Across the transcriptomes, m6A-tagged transcripts exhibited signiﬁcantly lower stability compared to non m6A-tagged transcripts in the WT NPCs, and this difference was reduced in cKO NPCs (Figure 3.6, A). Direct comparison of WT and Mettl14 cKO

NPCs showed that m6A-tagged transcripts exhibited larger increase in their stability compared to non-tagged transcripts upon Mettl14 deletion; one m6A tag per transcript was sufficient to increase stability and there was minimal additional effect for more tagging sites (Figure 3.6, B). It should be noted that our m6A-seq method could not determine whether multiple sites are simultaneously methylated in the same transcript. We confirmed our result with the direct measurement of the half-life of a selected group of transcripts (Figure 3.6, C and supplemental figure

B.5, F).

All together, these results support a model that m6A methylation of mRNAs related to cell cycle and neurogenesis confers their rapid turnover during the dynamic progress of cortical neurogenesis; a lack of m6A methylation attenuates the decay of these mRNAs, resulting in deﬁcits in temporal speciﬁcation and cell cycle progression of NPCs.

3.4.5 Mettl14 deletion uncovers transcriptional pre-patterning for normal cortical neurogenesis

Among the 2,059 m6A-tagged genes in the E13.5 mouse cortex, two major GO terms were generation of neurons and neuronal differentiation (Figure 3.5, C).

For example, IPC marker Tbr2 and Neurog2, and neuronal markers Neurod1 and

Neurod2 [40], were m6A-tagged (Figure 3.7, A). Q-PCR analysis of total mRNA showed increased levels of Tbr2, Neurog2, Neurod1, and Neurod2, but not non

66 A B C

6 Figure 3.6: m A promotes mRNA decay. (A) Cumulative distribution of log2 of gene expression ratios at time 5 hr post ActD over time 0 hr for m6A tagged genes (purple line) and non-m6A tagged genes (black line) for WT and Mettl14 cKO NPCs. D = value of Kolmogorov Smirnov test statistic corresponding to maximum difference between methylated and non-methylated distributions. (B) Cumulative distribution of log2 fold change in ratios of gene expression at 5 hr post ActD treatment over time 0 hr upon Mettl14 deletion. Top panel shows cumulative distribution for non-targets (black line) and transcripts with 1-1.9 m6A sites on average (bright red line), or 2 or more sites on average (dark red line). Bottom panel shows cumulative distribution for non-targets (black line) and m6A-tagged transcripts with 1-2, 2-3, 3-4, and 4 or more sites on average (red, yellow, blue, and green lines, respectively). (F) Summary of half-life of a selected group of transcripts in WT and Mettl14 cKO NPCs. Values represent mean + SEM (n = 4; ***: P <0.001; **: P <0.01; unpaired Students t-test).

67 tagged Rad17, in Mett14−/− NPCs compared to WT NPCs (Figure 3.7, B). This result raised the possibility that neuronal lineage genes are already expressed in neural stem cells and their levels are actively suppressed post-transcriptionally by m6A-dependent decay; alternatively, Mettl14 deletion may transcriptionally upreg- ulate these neuronal genes.

Figure 3.7: Post-transcriptional regulation of pre-patterning gene levels by m6A methylation in cortical neural stem cells. (A) Coverage plots from m6A-seq of E13.5 mouse forebrains showing representative examples of m6A-tagged IPC (Tbr2 and Neurog2) and neuronal (Neurod1 and Neurod2) genes. (B) Q-PCR analysis of total mRNA and 4sU-puriﬁed nascent mRNA from WT and Mettl14 cKO NPCs. All Ct values were ﬁrst normalized to Gapdh control (not m6A tagged), which were similar in both WT and cKO NPCs. The ratio (cKO over WT) was calculated for each experiment and values represent mean + SEM (n = 3 cultures; ***: P <0.001; **: P <0.01; *: P <0.05; unpaired Students t-test).

To differentiate between these two possibilities, we performed two sets of experiments. First, we quantiﬁed the levels of nascent mRNA using the metabolic labeling approach with 4-thiouridine [24, 77] . We found comparable and even lower levels of nascent mRNA of multiple neuronal genes, such as Tbr2, Neurog2, and Neurod2, in Mettl14−/− NPCs in comparison to WT NPCs (Figure 3.7, B). The lower levels of nascent mRNA observed for some neuronal genes in Mettl14−/−

NPCs could be explained by a negative feedback loop at the level of transcription, originated from elevated expression of stem cell genes, such as Emx2 and Sox1

68 (Figure 3.7, B). Second, we quantified the levels of pre-mRNA with Q-PCR and again found comparable levels of neuronal genes in Mettl14−/− NPCs in comparison to WT NPCs (Supplemental figure B.6, A), suggesting that the increase in the total mRNA of neuronal lineage genes in Mettl14−/− NPCs is not due to transcriptional upregulation. Together, these results support that neuronal lineage genes are already expressed in neural stem cells under normal conditions. Consistent with our result, mining the published single-cell RNA-seq dataset [94] revealed expression of neuronal genes, such as Tbr2, Neurog2, Neurod6 and Tubb3/Tuj1, in individual RGCs in the embryonic mouse cortex in vivo (Supplemental figure B.6,

B).

We next examined Tbr2 and Neurod1 protein levels in RGCs in vivo. Pax6+Tbr2+ cells were localized in the SVZ in WT at E17.5, but extended into the VZ in

Mettl14 cKO mice (Figure 3.8, A and B). Pax6+Neurod1+ cells were rare, but detectable just above the SVZ in WT cortices. In contrast, cKO mice exhibited a signiﬁcantly increased number of Pax6+Neurod1+ cells with a much broader distribution, including in the SVZ and VZ (Figure 3.8, C and D). To speciﬁcally examine expression in RGCs, we pulse-labeled juxtaventricular newborn cells by FlashTag

[94]. We found a signiﬁcantly increased number of FlashTag+Pax6+Tbr2+ and

FlashTag+Pax6+Neurod1+ cells in Mettl14 cKO cortex compared to those in WT

3 hr after labeling (Supplemental ﬁgure B.6, C and D). Given that FlashTag+ cells at 3 hr upon labeling are exclusively undifferentiated RGCs [94], these results suggest that Mettl14 regulates neuronal gene expression directly in RGCs.

To further assess our model that mRNA decay regulates neuronal gene expression in RGCs, we performed in vivo knockdown experiments for the components of CCR4-NOT complex (Cnot7 and Cnot1), a major cytoplasmic mRNA deadeny-

69 A B

C D

Figure 3.8: Regulation of protein production of pre-patterning genes by m6A methylation in cortical neural stem cells. (A-D) Precocious expression of Tbr2 and Neurod1 proteins in RGCs in E17.5 Mettl14 cKO mice in vivo. Shown are sample confocal images (A, C; scale bars: 50 µm) and quantiﬁcations of the percentage of Tbr2+Pax6+ cells (B), or Neurod1+Pax6+ cells (D), among total Pax6+ cells (left panels, n = 6) and the density distribution of Tbr2+Pax6+ (B), or Neurod1+Pax6+ cells (D), from the ventricular surface (right panels, n = 4). Values represent mean + SEM (***: P <0.001; **: P <0.01; *: P <0.05; unpaired Students t-test).

70 lase complex responsible for mRNA decay [22, 85]. Both Cnot7 KD and Cnot1

KD led to increased number of Tbr2+Pax6+ and Neurod1+Pax6+ cells with the lo- cation closer to the ventricular surface compared to control shRNA electroporation

(Supplemental ﬁgure B.6, E and F), a phenotype resembling Mettl14 cKO (Figure

3.8, A-D).

Taken together, our results suggest a heightened transcriptional coordination and a previously unappreciated transcriptional pre-patterning mechanism for mammalian cortical neurogenesis, in which late IPC and neuronal genes are already transcribed in cortical neural stem cells and these transcripts are down regulated post-transcriptionally by m6A-dependent decay.

3.4.6 METTL14 regulates cell cycle progression of human cortical NPCs

We next examined whether m6A function is conserved in human cortical neurogenesis. Using a previously developed protocol [112], we differentiated human iPSCs into a highly pure population of NESTIN+SOX2+ NPCs (hNPCs; 96.4 ±

1% among all cells; n = 5; Supplemental ﬁgure B.7, A). We co-expressed GFP and the validated shRNA against human METTL14 KD in these hNPCs (Supplemental

figure B.7, B). After 4 days, we labeled cells with EdU for 30 min and performed cell cycle analysis with flow cytometer quantification 14 hr later (Supplemental figure B.7, C). Similar to results from Mettl14−/− mouse NPCs (Supplemental figure

B.3, C and D), we found a signiﬁcant decrease in the percentage of GFP+EdU+ hNPCs that divided, indicating a delayed cell cycle progression (Figure 3.9, A and

B).

We recently developed a human iPSC-derived forebrain organoid model, which

71 Figure 3.9: METTL14 regulates cell cycle progression of human NPCs. (A-B) Flow cytometry analysis of cell cycle progression of hNPCs with METTL14 KD. Human NPCs were electroporated to co-express GFP and shRNA-control, or shRNA-METTL14. After 4 days, hNPCs were pulse-labeled with EdU (10 µM) for 30 min, incubated for 14 hr, followed by EdU and DNA content (DyeCycle Violet) staining and ﬂow cytometry analysis, similarly as in Figure 3.4, A. Values represent mean + SEM (n = 4; **: P <0.01; unpaired Students t-test). (C-D) Flow cytometry analysis of cell cycle progression with METTL14 KD in human forebrain organoids. Day 45 forebrain organoids were electroporated to co-express GFP and shRNA-control, or shRNA-METTL14. After 7 days, forebrain organoids were pulse-labeled with EdU (10 µM) for 1 hr, cultured further for 14 hr, followed by dissociation and analysis similarly as in Figure 3.4 A and B. Values represent mean + SEM (n = 4; ***: P <0.001; unpaired Students t-test).

72 exhibits transcriptome profiles similar to fetal human cortex along development up to the second trimester [75]. Around day 47, these forebrain organoids resemble mouse cortical neurogenesis at E13.5 (Supplemental figure B.7, D). We microin- jected plasmids co-expressing GFP and the shRNA against human METTL14, or the control shRNA, into the lumen of day 45 forebrain organoids and performed electroporation to transfect RGCs (Supplemental figure B.7, E). After 7 days, we pulsed organoids with EdU for 1 hr and performed cell cycle analysis of GFP+ cells 14 hr later (Supplemental figure B.7, F). Similar to findings from monolayer hNPC cultures, we observed a significant decrease in the percentage of GFP+EdU+ cells that divided (Figure 3.9, C and D). Together, these results indicate that m6A mRNA methylation plays a conserved role in regulating cortical NPC cell cycle progression in both mouse and human.

3.4.7 m6A-seq of human forebrain brain organoids and fetal brain reveals conserved and unique m6A landscape features compared to mouse

Finally, we performed m6A-seq of day 47 human forebrain organoids. We detected

11,994 high conﬁdence m6A peaks associated with 4,702 transcripts (Supplemen- tal ﬁgure B.8, A). Our previous systematic RNA-seq analyses of human forebrain organoids at different stages revealed that transcriptomes of organoids around day

47 were similar to human fetal cortex at 8-12 post-conception weeks (PCW) [75].

We further performed m6A-seq of PCW11 fetal human brain and identified 10,980 high confidence peaks associated with 5,049 transcripts (Supplemental figure B.8,

C). m6A sites were enriched near transcription start sites and stop codons for both human samples (Supplemental figure B.8, B and D). Furthermore, m6A profiles from both samples showed significant overlap (Figure 3.10, B). GO analysis of

73 common m6A-tagged transcripts in both samples showed enrichment of genes related to neurogenesis, neuronal differentiation and development (Figure 3.10, C).

Many recently identiﬁed risk genes for schizophrenia and autistic spectrum disorders have been shown to be dynamically expressed and play critical roles during mammalian embryonic brain development [70, 93]. Interestingly, disease ontology analysis of these m6A-tagged genes in humans showed enrichment related to mental disorders, mental retardation, schizophrenia and bipolar disorder (Figure 3.10,

C).

We further performed comparison among m6A landscapes during mouse and human cortical neurogenesis. About 19.3%, 34.7% and 31.4% of detected transcripts exhibited m6A-tagging in E13.5 mouse brain, day 47 human forebrain organoids, and PCW11 human fetal brain, respectively (Supplemental ﬁgure B.8,

E). Therefore, there appeared to be prevalent m6A mRNA methylation in human compared to mouse in these samples. Among transcripts expressed in all three samples, 856 transcripts were commonly m6A-tagged (Figure 3.10, D). These commonly m6A-tagged transcripts are enriched for genes related to neurogenesis and neuronal differentiation (Supplemental ﬁgure B.8, F). Notably, 1,173 transcripts were expressed in both species, but only m6A-tagged in human (Figure 3.10, D).

Ontology analysis of these human-speciﬁc m6A-tagged transcripts showed enrichment of transcripts related to mental disorders and mental retardation (Figure 3.10,

E and F). In contrast, analysis of the gene set of m6A-tagged transcripts shared between mouse and human showed enrichment for oncogenic processes (Figure

3.10, E). Notably, among genes associated with the 108 loci recently identiﬁed for genetic risk of schizophrenia [79], 60 genes were m6A-tagged in human and 21 genes were uniquely tagged in both human forebrain organoids and fetal brain, but not in mouse E13.5 forebrain.

74 Figure 3.10: Conserved and unique features of m6A mRNA methylation in human forebrain organoids, human fetal brain and embryonic mouse forebrain. (A) Representative plots of two human-speciﬁc m6A-tagged transcripts in day 47 human forebrain organoids, PCW11 human fetal brain and mouse E13.5 forebrain. (B) Venn diagram showing shared m6A-tagged transcripts between day 47 human forebrain organoids and PCW11 fetal human brain. (C) GO and disease ontology analyses of shared m6A-tagged genes in day 47 human forebrain organoids and PCW11 human fetal brain. (D) Venn diagram showing shared and unique m6A-tagged transcripts among mouse E13.5 forebrain, day 47 human forebrain cortex, and PCW11 fetal human brain. Only ortholog genes expressed in all three samples were used for analysis. (E-F) Disease ontology analysis of transcripts uniquely m6A-tagged in human shows enrichment for mental disorders and mental retardation, whereas disease ontology analysis of commonly m6A-tagged transcripts showed enrichment for oncogenic processes.

75 3.5 Discussion

From ﬂies to mammals, neurogenesis is a highly coordinated process with sequential waves of gene expression [46]. Here we revealed a critical role of m6A mRNA methylation in this process in the mammalian system in vivo. Our results suggest a model that m6A-tagging of transcripts related to neural stem cells, cell cycle, and neuronal differentiation confers their rapid turn-over to control the transcriptome composition at different phases of the dynamic cortical neurogenesis process.

The observation of RGCs expressing markers thought to be expressed only in late

IPCs and post-mitotic neurons in Mettl14 cKO mice in vivo led to the discovery of transcriptional pre-patterning in normal cortical neurogenesis and identifies m6A mRNA methylation as a key mechanism to prevent precocious expression of genes of later lineage states at the protein level in neural stem cells. We also provide the emerging epitranscriptomic field with databases of m6A mRNA landscapes of mouse and human cortical neurogenesis and identify intriguing human-specific features.

3.5.1 Transcriptional pre-patterning for cortical neurogenesis

The concept of pre-patterning initially came from analysis of chromatin states within multipotent progenitors to regulate the fate choice for liver and pancreas

[110]. Recent genome-wide mapping studies have suggested that epigenetic pre- patterning is important for spatio-temporal regulation of gene expression and may be a widespread phenomenon in cell fate decision [12]. Our study suggests, for the ﬁrst time, transcriptional pre-patterning in normal cortical neural stem cells in

76 vivo. Consistent with our model, Pax6 has been shown to bind and activate both

Tbr2 and Neurod1 promoters [82]. We showed that pre-patterning transcripts are tagged with m6A and subjected to rapid decay, therefore most of them are present in low levels among the bulk mRNA preparation, resulting in little protein under normal conditions a likely reason why such a mechanism has escaped detection in previous studies. While epigenetic mechanisms play a key role in transcriptional regulation during neurogenesis [41, 111], epitranscriptomic regulation as a post- transcriptional mechanism could provide speed and additional specificity, while maintaining plasticity of gene expression. By working in concert, the epigenetic landscape can permit transcription of certain genes, such as genes defining late lineages, while the epitranscriptome prevents aberrant protein production. Future studies are needed to investigate whether transcriptional pre-patterning is a general mechanism in fate specification of other stem cells during development.

3.5.2 Heightened transcriptional coordination of mammalian cortical neurogenesis by m6A

Our study provides the ﬁrst in vivo evidence in the mammalian system to support the emerging notion that m6A methylation plays a critical role in developmental fate transition. The precise and predictable developmental schedule of cortical neurogenesis requires rapid, tightly controlled changes in gene expression [71].

Our results suggest that epitranscriptomic m6A-tagging, via regulation of mRNA decay, provides a key mechanism for temporal control of dynamic gene expression, which in turn regulates cell cycle progression of cortical neural stem cells in both mouse and human.

There are three major categories of m6A-tagged transcripts in the embryonic

77 mouse brain. First, many classic transcription factors involved in neural stem cell maintenance and neurogenesis, such as Pax6, Sox2, Emx2, and Tbr2, are m6A- tagged and subject to rapid decay. Second, cell cycle-related transcripts, such as

Cdk9, Ccnh/Cyclin H, and Cdkn1C/p57, are m6A-tagged. Functionally, the loss of m6A methylation leads to prolonged cell cycle progression of cortical NPCs, resulting in delayed generation of different neuronal subtypes, extension of cortical neurogenesis into postnatal stages and deﬁcits in astrocyte generation in vivo.

Third, transcripts that were generally thought to be expressed only in later IPCs and post-mitotic neurons, such as Neurod1 and Neurod2, are m6A-tagged and expressed in neural stem cells. While expression of transcription factors are known to overlap during different stages of mammalian cortical neurogenesis [40], our

ﬁnding suggests a greater degree of transcriptional coordination than previously thought. On the other hand, expression of these neuronal genes at the protein level in a signiﬁcant number of RGCs located in the SVZ in the absence of Mettl14 high- lights the critical role of the epitranscriptomic mechanism in preventing precocious gene expression during the normal process of mammalian cortical neurogenesis.

3.5.3 Conserved and unique features of human m6A landscape during cortical neurogenesis

Our study provides databases of m6A mRNA landscapes during mouse and human cortical neurogenesis. Consistent with a similar role for m6A mRNA methylation in regulating cell cycle progression of cultured human NPCs and mouse NPCs in vitro and in vivo, the shared m6A-tagged transcripts in our mouse and human samples are enriched with genes related to neural stem cells, cell cycle, and neurogenesis. Notably, many genes associated with genetic risk for mental disorders, such as schizophrenia and autistic spectrum disorders, are only m6A-tagged in hu-

78 mans, but not in mice, raising the possibility that epitranscriptomic dysregulation may contribute to these human brain disorders. So far, one association study found evidence of ALKBH5 in conferring genetic risk for major depression disorder [23], and two studies identiﬁed association of FTO mutations with growth retardation and developmental delay [6, 16].

In summary, our study identiﬁes a critical and conserved role of an m6A epitranscriptomic mechanism in the temporal control of mammalian cortical neurogenesis via promotion of mRNA decay of transcripts related to transcription factors, neural stem cells, cell cycle, and neuronal differentiation. Future studies will address how this epitranscriptomic mechanism interacts with various epigenetic mechanisms to coordinate dynamic transcriptomes during cortical neurogenesis, and how dysregulation of epitranscriptomic mechanisms may contribute to developmental brain disorders.

79 APPENDIX A

Supplemental Figures: Cyﬁp1

80 A B C D

E F G

Figure A.1: Normal locomotor activity, motor coordination, nociception response and repetitive behaviors of Cyfip1 cKO mice. (A) Open field locomotor activity of WT1 and cKO mice. Locomotor activity was assessed by number of beams broken in open field box for 1hr. n = 12 (WT1), 11 (cKO). (B) Rotarod test of WT1 and cKO mice. Latency to fall from the rotarod was quantified in each trial. n = 9 (WT1), 14 (cKO). (C) Hot plate assay of WT1 and cKO mice. n = 6 (WT1), 6 (cKO). (D) Novel object recognition of WT1 and cKO mice. Discrimination index was calculated as the ratio of time spent exploring the novel object over total time exploring. n = 8 (WT1), 8 (cKO). (E) Marble burying assay of WT1 and cKO mice. Number of marbles buried during 30 min test was counted. n = 10 (WT1), 10 (cKO). (F and G) Normal social approach and social novelty recognition of Nestin-Cre transgenic mice (Nestin-Cre+/Tg; Cyfip1+/+) in the three-chamber assay, compared with wild-type mice (Nestin-Cre+/+; Cyfip1+/+). n = 10 (Wild-type), 7 (Nestin-Cre transgenic). (H) Normal level of amphetamine-induced hyperactivity in Nestin-Cre transgenic mice compared with wild-type mice. Shown on the left is the trace of locomotor activity presented as number of beams broken during every 10 min. An arrow represents the time of amphetamine injection. Shown on the right is the total number of beams broken after amphetamine injection. n = 9 (Wild-type), 8 (Nestin-Cre transgenic). (I) Normal PPI in Nestin-Cre transgenic mice compared with wild-type mice. n = 11 (Wild-type), 11 (Nestin-Cre transgenic). All values represent mean + SEM (***p <0.001; n.s., not significant; Students t test).

81 Tail suspention test PPI Test

Figure A.2: Normal locomotor activity, novel object recognition, behavioral despair and sensorimotor gating of cOE mice, and impaired maternal care of Cyfip1 cOE mice. (A) Open field locomotor activity of WT2 and cOE. (B) Novel object recognition of WT2 and cOE mice. (C) Representative images of pups nursed by WT2 female and cOE females at P3. Note that stomachs of pups were filled with milk (arrows) in WT2 dams, but not in cOE dams. Scale bar, 10mm. (D) Representative images of pups nursed by WT2 female and cOE female at P7. Scale bar, 10mm. (E) Normal level of behavioral despair of cOE mice in tail suspension test. n = 9 (WT2), 12 (cOE). (F) Normal prepulse inhibition of acoustic startle response in cOE mice compared with WT2 mice. n = 7 (WT2), n = 8 (cOE). All values represent mean + SEM (n.s., not significant; Students t test).

82 A

Figure A.3: Cyfip1 RIP seq strategy. (A) Schematic view of RIP seq strategy. Hippocampus tissue from Cyfip1 cOE mice was used for RIP seq experiments. Cyfip1 cOE mice have an HA peptide attached to Cyfip1, which allows to immunoprecipitate using anti-HA antibody. (B) Immunoblot showing pulldown of Cyfip1 in RIP (ki1 and ki2 pulled down with HA antibody) and control samples. ki1: knock in 1, ki2: knock in 2, wt1: wild type 1.

83 A

Figure A.4: Bioinformatic analysis of Cyﬁp1 RIP seq experiment. (A) Schematic view of bioinformatic approach for RIP-seq analysis. (B) Heatmap depicting overlap of genes identiﬁed in each RIP experiment. k1h-k1i: knock in 1-HA antibody (RIP) vs knock in1-IgG (control), k1h-w1h: knock in 1-HA antibody (RIP) vs wild type 1-HA antibody (control), k1h-w2h: knock-in 1-HA antibody (RIP) vs wild type 2-HA antibody (control), k2h-k2i: knock in 2-HA antibody (RIP) vs knock in 2-IgG (control), k2h-w1h: knock in 2-HA antibody (RIP) vs wild type 1-HA antibody (control), k2h-w2h: knock in 2-HA antibody (RIP) vs wild type 1-HA antibody (control).

84 Figure A.5: mRNA levels of Cyﬁp1 targets are unchanged in cKO and cOE mice. (A) Total mRNA levels of synaptic genes in 3-month-old WT1 and cKO hippocampi. All data were normalized to mRNA level of Actin. Values represent mean + SEM. n = 3 animals for each genotype. (B) Total mRNA levels of synaptic genes in 3-month-old WT2 and cOE hippocampi. All data were normalized to mRNA level of Actin. Values represent mean + SEM. n = 3 animals for each genotype.

85 APPENDIX B

Supplemental Figures: m6A

86 Figure B.1: Nervous system Mettl14 deletion in mice results in postnatal lethality. (A) Expression of molecular players mediate m6A signaling, based on a published single-cell RNA-seq dataset of embryonic mouse cortical neurogenesis (Telley et al., 2016). Shown are the expression proﬁles of selected genes as violin plots, generated using the Seurat package of R [83]. AP: Apical progenitors/RGCs; BP: daughter basal progenitors/IPCs; EN: early neurons; LN: late neurons. (B) Depletion of Mettl14 protein in the forebrain of Nestin-cre;Mettl14 f / f cKO mice. Shown are genetic deletion strategy (left) and sample Western blot images from WT or cKO E17.5 forebrain lysates (right). Because Mett14 was only deleted in the nervous system, the minor non-neural cells contributed to the residual Mettl14 proteins (faint bands). (C) Appearance of WT and cKO pups at P5 and P14. Note the impairment in the P14 cKO pup to maintain body balance. Scale bars: 1 cm. (D) Survival curve of WT (n = 45), Het (n = 23) and cKO (n = 22) pups.

87 A B

C D

Figure B.2: Nervous system Mettl14 deletion in mice results in deficits in timely production of cortical neuron subtypes. (A and B) Deficits in the production of upper-layer neurons in cKO cortices at P5. Shown in (A) are sample confocal images of staining for Satb2 (layer 2/3), Ctip2 (layer 5) and DAPI, or Ctip2 (layer 5), Tbr1 (layer 6) and DAPI. Scale bars: 100 m. Quantification is shown in (B). Values represent mean + SEM (n = 6; **: P <0.01; unpaired Students t-test). (C and D) Deficits in the production of lower-layer neurons in cKO cortices at E17.5. Shown in (C) are sample confocal images of staining for Ctip2 and DAPI. Scale bar: 100 m. Quantification is shown in (D). Values represent mean + SEM (n = 6; **: P <0.01; unpaired Students t-test).

88 Figure B.3: Flow cytometry analysis reveals delayed cell cycle progression of Mettl14−/− NPCs. (A) Schematic diagrams of the dual reporter system used to track cell cycle status by time-lapse imaging. Nuclear localized H2B-mCherry and a GFP-tagged Cdk2 substrate DHB are co-expressed in the individual cell. Cdk2 becomes active during the G1-S transition and phosphorylates DHB-GFP, which is then translocated from the nucleus to the cytoplasm. The presence of GFP in the mCherry+ nucleus indicates cells in the G1 phase, whereas translocation to the nucleus indicates the initiation of the S phase, and continual buildup of nuclear GFP occurs until mitosis. (B-D) Flow cytometry analysis of cell cycle progression of WT and Mettl14 cKO NPCs. NPCs were pulse-labeled with EdU (10 M) for 30 min, cultured for 0 or 5 hr, followed by EdU and DNA content (7AAD) staining and flow cytometry analysis. Shown in (B) are sample dot plots at 0 and 5 hr after EdU pulsing. Cells in a specific cell cycle phase were marked in a box. Note that EdU+ cells (S phase at 0 hr) were segregated into divided (G1*) and non-divided (S/G2*/M*) populations. Shown in (C) are sample histograms of DNA content from EdU+ cells and the total cell population (as a reference). Quantification is shown in (D). Values represent mean + SEM (n = 4; ***: P <0.01; unpaired Students t-test).

89 Figure B.4: Mettl3 is essential for m6A mRNA methylation and proper cell cycle progression of mouse NPCs. (A) Efficacy of the shRNA against mouse Mettl3. Mouse B16F10 cells were transfected with shRNA-control and shRNA-Mettl3. The amount of Mettl3 mRNA was assessed by Q-PCR 3 days later. Values represent mean + SEM (n = 3; ***: P <0.001; unpaired Students t-test). (B-C) Depletion of m6A mRNA methylation by Mettl3 KD. Shown are sample images of m6A dot blot and methylene blue staining (as loading controls (B) and quantification (C). Data were normalized to the averaged levels of WT samples. Values represent mean + SEM (n = 3; ***: P <0.01; unpaired Students t-test). (D) Flow cytometry analysis of cell cycle status of mouse NPCs. Mouse NPCs were electroporated to co-express GFP and shRNA-control, or shRNA-Mettl3. After 4 days, NPCs were pulse-labeled with EdU (10 M) for 30 min, cultured for 9 hr, followed by EdU and DNA content (DyeCycle Violet) staining and flow cytometry analysis. GFP+ and GFP- cells were gated separately and shown as dot plots. Note that GFP+ cells with Mettl3 KD showed accumulation of non-divided (S/G2*/M*) population.

90 Figure B.5: m6A-seq analysis of mouse embryonic forebrain. (A) Venn diagram showing intersection among m6A peaks identified in 3 independent m6A-seq experiments. 4,055 high confidence peaks shared by 2 out of 3 replicates, corresponding to 2,050 genes, were used for downstream analysis. (B) Enrichment of m6A peaks in 5 non-overlapping transcript segments. Pie chart shows percentage of peaks annotated to each segment. Bar plot shows fold enrichment of peaks for each segment, normalized for the segment length. (C) m6A peaks do not correlate with transcript expression levels. Scatter plot shows gene expression levels (lnRPKM) of m6A-tagged genes plotted against m6A peak intensity (ln f oldchange). Histogram shows distribution of gene expression levels (lnRPKM) for all transcripts detected in 3 RNA-seq input libraries. (D) Validation of m6A-tagging in specific transcripts in cortical NPCs. The enrichment of m6A-tagged transcripts by IP with anti-m6A antibody over lgG was quantified by Q-PCR. Values represent mean + SEM (n = 3; ***: P <0.01; **: P <0.01; *: P <0.05; unpaired Students t-test). (E) Representative coverage plots from the RNA-seq analysis at 0 or 5 hr after treatment of ActD showing increased stability of m6A-tagged genes (Sox1 and Emx2), but not a non m6A-tagged gene (Rad17) in Mettl14−/− NPCs compared to WT NPCs. (F) Representative plot for calculating half-life of transcripts in WT and cKO NPCs. Data for Emx2 is plotted as an example.

91 Figure B.6: Expression of neuronal genes in RGCs of embryonic cortex in vivo. (A) Q-PCR analysis of pre-mRNA from WT and Mettl14 cKO NPCs. Ct values were normalized to Actin control (not m6A tagged). Values represent mean + SEM (n = 4 cultures; **: P <0.01; *: P <0.05; unpaired Students t-test). (B) Single-cell transcriptome analysis reveals expression of neuronal lineage genes in mouse embryonic cortical RGCs in vivo. (C-D) Increased expressions of neuronal lineage genes in FlashTag+ (FT+) RGCs 3 hr after pulse labeling. Shown are sample confocal images (C; scale bars: 20 m) and quantiﬁcations of the percentage of FT+Tbr2+Pax6+ cells, or FT+Neurod1+Pax6+ cells (D), among total FT+ cells. Values represent mean + SEM (n = 5 sections from 2 animals; ***: P <0.01; unpaired Students t-test). (E-F) Precocious expression of Tbr2 and Neurod1 proteins in RGCs upon knockdown of mRNA deadenylase components in vivo. (E) Sample confocal images (scale bars: 20 m). (F) quantiﬁcations of the percentage of GFP+Tbr2+Pax6+ cells, or GFP+Neurod1+Pax6+ cells, among total GFP+Pax6+ cells and the density distribution of GFP+Tbr2+Pax6+, or GFP+Neurod1+Pax6+ cells from the ventricular surface (F). Values represent mean + SEM (n = 5 sections from 3 animals; ***: P <0.001; **: P <0.01; *: P <0.05; unpaired Students t-test).

92 Figure B.7: Mettl14 regulates cell cycle progression of hNPCs. (A) Validation of hNPC differentiation from human iPSC. Shown is a sample confocal image. Scale bar: 50 m. (B) Efficacy of the shRNA against METLL14. Human NPCs were electroporated to co-express GFP and shRNA-control, or shRNA-METLL14, and dissociated 3 days later. Amount of METLL14 mRNA in FACS-purified GFP+ cells was assessed by Q-PCR. Values represent mean + SEM (n = 3; ***: P <0.01; unpaired Students t-test). (C) Flow cytometry analysis of cell cycle progression of hNPCs with METTL14 KD. Similarly as in Figure B.4 B. (D) Comparison of neuronal differentiation among day 47 human forebrain organoids and embryonic mouse cortical development at E13.5, E15.5 and E17.5. Shown are confocal images of immunostaining for CTIP2 and SATB2 and DAPI. Scale bar: 50 m. Note that day 47 human forebrain organoids exhibit differentiation pattern most similar to E13.5 mouse cortex. (E) Electroporation of human forebrain organoid with shRNA expressing plasmid. Day 45 forebrain organoids were electroporated to co-express GFP and shRNA-control, or shRNA-METLL14, by microinjection into the lumen of organoids. After 7 days, organoids were pulse-labeled with EdU (10 M) for 30 min, and cultured further for 14 hr. Shown are sample confocal images at day 52. Scale bar: 100 m.(F) Flow cytometry analysis of cell cycle progression with METTL14 KD in human forebrain organoids. Shown are sample dot plots 14 hr after EdU pulse. Cells in a specific cell cycle phase were marked within a box.

93 A B C D

E F

Figure B.8: Comparison of m6A mRNA landscapes among human forebrain organoids, fetal brain and mouse embryonic forebrain. (A) Venn diagram showing intersection between m6A peaks identified in 2 independent m6A-seq of day 47 human forebrain organoids. 11,994 high confidence peaks corresponding to 4,702 genes were used for downstream analysis. (B) Venn diagram showing intersection between m6A peaks identified in 3 independent m6A-seq of PCW11 human fetal brain. 10,980 high confidence peaks corresponding to 5,049 genes were used for downstream analysis. (C-D) Enrichment of m6A peaks in 5 non-overlapping transcript segments for day 47 human forebrain organoids (C) and PCW11 fetal human brain (D). Same as in Figure B.5, B. (E) Pie charts showing the percentage of m6A-tagged genes among all expressed genes in each samples. (F) GO analysis for m6A-tagged genes shared between human forebrain organoids and fetal brain, but not in mouse E13.5 forebrain, and disease ontology analysis of m6A-tagged genes shared among all three samples.

94 Bibliography

[1] Simon Anders, Paul Theodor Pyl, and Wolfgang Huber. Htseqa python

framework to work with high-throughput sequencing data. Bioinformatics,

31(2):166–169, 2015.

[2] Ranen Aviner, Tamar Geiger, and Orna Elroy-Stein. Genome-wide identi-

ﬁcation and quantiﬁcation of protein synthesis in cultured cells and whole

tissues by puromycin-associated nascent chain proteomics (punch-p). na-

ture protocols, 9(4):751, 2014.

[3] Pedro J Batista, Benoit Molinie, Jinkai Wang, Kun Qu, Jiajing Zhang,

Lingjie Li, Donna M Bouley, Ernesto Lujan, Bahareh Haddad, Kaveh

Daneshvar, et al. m 6 a rna modiﬁcation controls cell fate transition in mam-

malian embryonic stem cells. Cell stem cell, 15(6):707–719, 2014.

[4] Craig D Blackstone, Stephen J Moss, Lee J Martin, Allan I Levey, Donald L

Price, and Richard L Huganir. Biochemical characterization and localization

of a non-n-methyl-d-aspartate glutamate receptor in rat brain. Journal of

neurochemistry, 58(3):1118–1126, 1992.

95 [5] Jacqueline Blundell, Cory A Blaiss, Mark R Etherton, Felipe Espinosa, Kat-

suhiko Tabuchi, Christopher Walz, Marc F Bolliger, Thomas C Sudhof,¨ and

Craig M Powell. Neuroligin-1 deletion results in impaired spatial memory

and increased repetitive behavior. Journal of Neuroscience, 30(6):2115–

2129, 2010.

[6] Sarah Boissel, Orit Reish, Karine Proulx, Hiroko Kawagoe-Takaki, Barbara

Sedgwick, Giles SH Yeo, David Meyre, Christelle Golzio, Florence Moli-

nari, Noman Kadhom, et al. Loss-of-function mutation in the dioxygenase-

encoding fto gene causes severe growth retardation and multiple malforma-

tions. The American Journal of Human Genetics, 85(1):106–111, 2009.

[7] Ozlem Bozdagi, Takeshi Sakurai, Nathan Dorr, Marion Pilorge, Nagahide

Takahashi, and Joseph D Buxbaum. Haploinsufﬁciency of cyﬁp1 produces

fragile x-like phenotypes in mice. PLoS One, 7(8):e42422, 2012.

[8] Nicola Brunetti-Pierri, Jonathan S Berg, Fernando Scaglia, John Belmont,

Carlos A Bacino, Trilochan Sahoo, Seema R Lalani, Brett Graham, Bren-

dan Lee, Marwan Shinawi, et al. Recurrent reciprocal 1q21. 1 deletions and

duplications associated with microcephaly or macrocephaly and develop-

mental and behavioral abnormalities. Nature genetics, 40(12):1466–1471,

2008.

[9] Liam S Carroll and Michael J Owen. Genetic overlap between autism,

schizophrenia and bipolar disorder. Genome medicine, 1(10):102, 2009.

[10] Chyi-Ying A Chen, Nader Ezzeddine, and Ann-Bin Shyu. Messenger rna

half-life measurements in mammalian cells. Methods in enzymology, 448:

335–357, 2008.

[11] Jing Chen, Eric E Bardes, Bruce J Aronow, and Anil G Jegga. Toppgene

96 suite for gene list enrichment analysis and candidate gene prioritization. Nu-

cleic acids research, 37(suppl 2):W305–W311, 2009.

[12] Taiping Chen and Sharon YR Dent. Chromatin modiﬁers: regulators of

cellular differentiation. Nature reviews. Genetics, 15(2):93, 2014.

[13] Devin M Cox and Merlin G Butler. The 15q11. 2 bp1–bp2 microdeletion

syndrome: A review. International journal of molecular sciences, 16(2):

4068–4082, 2015.

[14] Bernard Crespi. Diametric gene-dosage effects as windows into neuroge-

netic architecture. Current opinion in neurobiology, 23(1):143–151, 2013.

[15] Bernard Crespi, Philip Stead, and Michael Elliot. Comparative genomics

of autism and schizophrenia. Proceedings of the National Academy of Sci-

ences, 107(suppl 1):1736–1741, 2010.

[16] Hussein Daoud, Dong Zhang, Fiona McMurray, Andrea Yu, Stephanie M

Luco, Jason Vanstone, Olga Jarinova, Nancy Carson, James Wickens, Shi-

fali Shishodia, et al. Identiﬁcation of a pathogenic fto mutation by next-

generation sequencing in a newborn with growth retardation and develop-

mental delay. Journal of medical genetics, pages jmedgenet–2015, 2015.

[17] Silvia De Rubeis, Emanuela Pasciuto, Ka Wan Li, Esperanza Fernandez,´

Daniele Di Marino, Andrea Buzzi, Linnaea E Ostroff, Eric Klann, Fried JT

Zwartkruis, Noboru H Komiyama, et al. Cyﬁp1 coordinates mrna transla-

tion and cytoskeleton remodeling to ensure proper dendritic spine formation.

Neuron, 79(6):1169–1182, 2013.

[18] RM Deacon. Digging and marble burying in mice: simple methods in vivo

identiﬁcation of biological impace. Nat Protocols, 1:122–24, 2006.

97 [19] Ronald C Desrosiers, Karen H Friderici, and Fritz M Rottman. Charac-

terization of novikoff hepatoma mrna methylation and heterogeneity in the

methylated 5’terminus. Biochemistry, 14(20):4367–4374, 1975.

[20] Veronica Dezi, Chavdar Ivanov, Irmgard U Haussmann, and Matthias Soller.

Nucleotide modiﬁcations in messenger rna and their role in development and

disease. Biochemical Society Transactions, 44(5):1385–1393, 2016.

[21] Dan Dominissini, Sharon Moshitch-Moshkovitz, Schraga Schwartz, Mali

Salmon-Divon, Lior Ungar, Sivan Osenberg, Karen Cesarkas, Jasmine

Jacob-Hirsch, Ninette Amariglio, Martin Kupiec, et al. Topology of the

human and mouse mˆ sup 6ˆ a rna methylomes revealed by mˆ sup 6ˆ a-seq.

Nature, 485(7397):201, 2012.

[22] Hao Du, Ya Zhao, Jinqiu He, Yao Zhang, Hairui Xi, Mofang Liu, Jinbiao

Ma, and Ligang Wu. Ythdf2 destabilizes m6a-containing rna through direct

recruitment of the ccr4-not deadenylase complex. Nature communications,

7, 2016.

[23] Tingfu Du, Shuquan Rao, Lin Wu, Ning Ye, Zeyue Liu, Huiling Hu, Jianbo

Xiu, Yan Shen, and Qi Xu. An association study of the m6a genes with

major depressive disorder in chinese han population. Journal of affective

disorders, 183:279–286, 2015.

[24] Erin E Duffy, Michael Rutenberg-Schoenberg, Catherine D Stark, Robert R

Kitchen, Mark B Gerstein, and Matthew D Simon. Tracking distinct rna

populations using efﬁcient and reversible covalent chemistry. Molecular

cell, 59(5):858–866, 2015.

[25] Noelle D Dwyer, Bin Chen, Shen-Ju Chou, Simon Hippenmeyer, Lau-

rent Nguyen, and H Troy Ghashghaei. Neural stem cells to cerebral cor-

98 tex: emerging mechanisms regulating progenitor behavior and productivity.

Journal of Neuroscience, 36(45):11394–11401, 2016.

[26] Chris Englund, Andy Fink, Charmaine Lau, Diane Pham, Ray AM Daza,

Alessandro Bulfone, Tom Kowalczyk, and Robert F Hevner. Pax6, tbr2, and

tbr1 are expressed sequentially by radial glia, intermediate progenitor cells,

and postmitotic neurons in developing neocortex. Journal of Neuroscience,

25(1):247–251, 2005.

[27] Fred H Gage and Sally Temple. Neural stem cells: generating and regener-

ating the brain. Neuron, 80(3):588–601, 2013.

[28] Yonglin Gao, Ralphiel S Payne, Avital Schurr, Tyler Hougland, Joshua Lord,

Laura Herman, Zhenmin Lei, Pradeep Banerjee, and Rif S El-Mallakh. Me-

mantine reduces mania-like symptoms in animal models. Psychiatry re-

search, 188(3):366–371, 2011.

[29] Shaoyu Ge, Eyleen LK Goh, Kurt A Sailor, Yasuji Kitabatake, Guo-li Ming,

and Hongjun Song. Gaba regulates synaptic integration of newly generated

neurons in the adult brain. Nature, 439(7076):589, 2006.

[30] Daniel H Geschwind. Advances in autism. Annual review of medicine, 60:

367–380, 2009.

[31] Shay Geula, Sharon Moshitch-Moshkovitz, Dan Dominissini, Abed AlFa-

tah Mansour, Nitzan Kol, Mali Salmon-Divon, Vera Hershkovitz, Eyal Peer,

Nofar Mor, Yair S Manor, et al. m6a mrna methylation facilitates resolu-

tion of na¨ıve pluripotency toward differentiation. Science, 347(6225):1002–

1006, 2015.

[32] Christelle Golzio, Jason Willer, Michael E Talkowski, Edwin C Oh,

Yu Taniguchi, Sebastien´ Jacquemont, Alexandre Reymond, Mei Sun, Akira

99 Sawa, James F Gusella, et al. Kctd13 is a major driver of mirrored neu-

roanatomical phenotypes associated with the 16p11. 2 cnv. Nature, 485

(7398):363, 2012.

[33] Hannah M Grayton, Cathy Fernandes, Dan Rujescu, and David A Collier.

Copy number variations in neurodevelopmental disorders. Progress in neu-

robiology, 99(1):81–91, 2012.

[34] LF Custo Greig, Mollie B Woodworth, Maria J Galazo, Hari Padmanabhan,

and Jeffrey D Macklis. Molecular logic of neocortical projection neuron

speciﬁcation, development and diversity. Nature Reviews. Neuroscience, 14

(11), 2013.

[35] Detelina Grozeva, Donald F Conrad, Chris P Barnes, Matthew Hurles,

Michael J Owen, Michael C O’Donovan, Nick Craddock, George Kirov,

et al. Independent estimation of the frequency of rare cnvs in the uk popu-

lation conﬁrms their role in schizophrenia. Schizophrenia research, 135(1):

1–7, 2012.

[36] Sabrice Guerrier and Franck Polleux. The ups and downs of neural progen-

itors: Cep120 and taccs control interkinetic nuclear migration. Neuron, 56

(1):1–3, 2007.

[37] Francois Guillemot, Zoltan´ Molnar,´ Victor Tarabykin, and Anastassia

Stoykova. Molecular mechanisms of cortical differentiation. European

Journal of Neuroscience, 23(4):857–868, 2006.

[38] Irmgard U Haussmann, Zsuzsanna Bodi, Eugenio Sanchez-Moran, Nigel P

Mongan, Nathan Archer, Rupert G Fray, and Matthias Soller. m6a poten-

tiates sxl alternative pre-mrna splicing for robust drosophila sex determina-

tion. Nature, 540(7632):301–304, 2016.

100 [39] Ralf Herwig, Christopher Hardt, Matthias Lienhard, and Atanas Kamburov.

Analyzing and interpreting genome data at the network level with consen-

suspathdb. Nature protocols, 11(10):1889–1907, 2016.

[40] Robert F Hevner, Rebecca D Hodge, Ray AM Daza, and Chris Englund.

Transcription factors in glutamatergic neurogenesis: conserved programs in

neocortex, cerebellum, and adult hippocampus. Neuroscience research, 55

(3):223–233, 2006.

[41] Yusuke Hirabayashi and Yukiko Gotoh. Epigenetic control of neural precur-

sor cell fate during development. Nature reviews. Neuroscience, 11(6):377,

2010.

[42] Saumya S Jamuar and Christopher A Walsh. Genomic variants and varia-

tions in malformations of cortical development. Pediatric clinics of North

America, 62(3):571–585, 2015.

[43] Shengdong Ke, Endalkachew A Alemu, Claudia Mertens, Emily Conn

Gantman, John J Fak, Aldo Mele, Bhagwattie Haripal, Ilana Zucker-Scharff,

Michael J Moore, Christopher Y Park, et al. A majority of m6a residues are

in the last exons, allowing the potential for 3 utr regulation. Genes & devel-

opment, 29(19):2037–2053, 2015.

[44] EM Kenny, P Cormican, S Furlong, E Heron, G Kenny, C Fahey, E Kelleher,

S Ennis, D Tropea, R Anney, et al. Excess of rare novel loss-of-function

variants in synaptic genes in schizophrenia and autism spectrum disorders.

Molecular psychiatry, 19(8):872, 2014.

[45] Daehwan Kim, Geo Pertea, Cole Trapnell, Harold Pimentel, Ryan Kelley,

and Steven L Salzberg. Tophat2: accurate alignment of transcriptomes in

101 the presence of insertions, deletions and gene fusions. Genome biology, 14

(4):R36, 2013.

[46] Minoree Kohwi and Chris Q Doe. Temporal fate speciﬁcation and neural

progenitor competence during development. Nature Reviews. Neuroscience,

14(12):823, 2013.

[47] Arnold Kriegstein and Arturo Alvarez-Buylla. The glial nature of embryonic

and adult neural stem cells. Annual review of neuroscience, 32:149–184,

2009.

[48] Melanie Lacaria, Pradip Saha, Lorraine Potocki, Weimin Bi, Jiong Yan,

Santhosh Girirajan, Brooke Burns, Sarah Elsea, Katherina Walz, Lawrence

Chan, et al. A duplication cnv that conveys traits reciprocal to metabolic

syndrome and protects against diet-induced obesity in mice and men. PLoS

genetics, 8(5):e1002713, 2012.

[49] Shaheen E Lakhan, Mario Caro, and Norell Hadzimichalis. Nmda receptor

activity in neuropsychiatric disorders. Frontiers in psychiatry, 4, 2013.

[50] Ben Langmead and Steven L Salzberg. Fast gapped-read alignment with

bowtie 2. Nature methods, 9(4):357–359, 2012.

[51] Eun-Jae Lee, Su Yeon Choi, and Eunjoon Kim. Nmda receptor dysfunction

in autism spectrum disorders. Current opinion in pharmacology, 20:8–13,

2015.

[52] Tina Lence, Junaid Akhtar, Marc Bayer, Katharina Schmid, Laura Spindler,

Cheuk Hei Ho, Nastasja Kreim, Miguel A Andrade-Navarro, Burkhard

Poeck, Mark Helm, et al. m6a modulates neuronal functions and sex de-

termination in drosophila. Nature, 540(7632):242–247, 2016.

102 [53] Ning Leng, John A Dawson, James A Thomson, Victor Ruotti, Anna I Riss-

man, Bart MG Smits, Jill D Haag, Michael N Gould, Ron M Stewart, and

Christina Kendziorski. Ebseq: an empirical bayes hierarchical model for

inference in rna-seq experiments. Bioinformatics, 29(8):1035–1043, 2013.

[54] Bo Li and Colin N Dewey. Rsem: accurate transcript quantiﬁcation from

rna-seq data with or without a reference genome. BMC bioinformatics, 12

(1):323, 2011.

[55] Wenlin Li, Woong Sun, Yu Zhang, Wanguo Wei, Rajesh Ambasudhan, Peng

Xia, Maria Talantova, Tongxiang Lin, Janghwan Kim, Xiaolei Wang, et al.

Rapid induction and long-term self-renewal of primitive neural precursors

from human embryonic stem cells by small molecule inhibitors. Proceed-

ings of the National Academy of Sciences, 108(20):8299–8304, 2011.

[56] Xiaoyu Li, Xushen Xiong, and Chengqi Yi. Epitranscriptome sequencing

technologies: decoding rna modiﬁcations. Nature methods, 14(1):23–31,

2017.

[57] Pentao Liu, Nancy A Jenkins, and Neal G Copeland. A highly efﬁcient

recombineering-based method for generating conditional knockout muta-

tions. Genome research, 13(3):476–484, 2003.

[58] Jason Lockrow, Heather Boger, Heather Bimonte-Nelson, and Ann-

Charlotte Granholm. Effects of long-term memantine on memory and neu-

ropathology in ts65dn mice, a model for down syndrome. Behavioural brain

research, 221(2):610–622, 2011.

[59] Dheeraj Malhotra and Jonathan Sebat. Cnvs: harbingers of a rare variant

revolution in psychiatric genetics. Cell, 148(6):1223–1241, 2012.

103 [60] Ben Martynoga, Daniela Drechsel, and Franc¸ois Guillemot. Molecular con-

trol of neurogenesis: a view from the mammalian cerebral cortex. Cold

Spring Harbor Perspectives in Biology, 4(10):a008359, 2012.

[61] Kate D Meyer and Samie R Jaffrey. The dynamic epitranscriptome: N6-

methyladenosine and gene expression control. Nature reviews. Molecular

cell biology, 15(5):313, 2014.

[62] Kate D Meyer and Samie R Jaffrey. Rethinking m6a readers, writers, and

erasers. Annual Review of Cell and Developmental Biology, (0), 2017.

[63] Kate D Meyer, Yogesh Saletore, Paul Zumbo, Olivier Elemento, Christo-

pher E Mason, and Samie R Jaffrey. Comprehensive analysis of mrna

methylation reveals enrichment in 3 utrs and near stop codons. Cell, 149

(7):1635–1646, 2012.

[64] Jeremy A Miller, Song-Lin Ding, Susan M Sunkin, Kimberly A Smith, Ly-

dia Ng, Aaron Szafer, Amanda Ebbert, Zackery L Riley, Kaylynn Aiona,

James M Arnold, et al. Transcriptional landscape of the prenatal human

brain. Nature, 508(7495):199, 2014.

[65] Rimante Minkeviciene, Pradeep Banerjee, and Heikki Tanila. Cognition-

enhancing and anxiolytic effects of memantine. Neuropharmacology, 54

(7):1079–1085, 2008.

[66] Ilaria Napoli, Valentina Mercaldo, Pietro Pilo Boyl, Boris Eleuteri,

Francesca Zalfa, Silvia De Rubeis, Daniele Di Marino, Evita Mohr, Marzia

Massimi, Mattia Falconi, et al. The fragile x syndrome protein represses

activity-dependent translation through cyﬁp1, a new 4e-bp. Cell, 134(6):

1042–1054, 2008.

104 [67] Yuhei Nishimura, Christa L Martin, Araceli Vazquez-Lopez, Sarah J

Spence, Ana Isabel Alvarez-Retuerto, Marian Sigman, Corinna Steindler,

Sandra Pellegrini, N Carolyn Schanen, Stephen T Warren, et al. Genome-

wide expression proﬁling of lymphoblastoid cell lines distinguishes different

forms of autism and reveals shared pathways. Human molecular genetics,

16(14):1682–1698, 2007.

[68] Alex S Nord, Kartik Pattabiraman, Axel Visel, and John LR Rubenstein. Ge-

nomic perspectives of transcriptional regulation in forebrain development.

Neuron, 85(1):27–47, 2015.

[69] A Oguro-Ando, C Rosensweig, E Herman, Y Nishimura, D Werling,

BR Bill, JM Berg, F Gao, G Coppola, BS Abrahams, et al. Increased cy-

ﬁp1 dosage alters cellular and dendritic morphology and dysregulates mtor.

Molecular psychiatry, 20(9):1069, 2015.

[70] Kazutaka Ohi, Takamitsu Shimada, Yusuke Nitta, Hiroaki Kihara, Hiroaki

Okubo, Takashi Uehara, and Yasuhiro Kawasaki. Speciﬁc gene expression

patterns of 108 schizophrenia-associated loci in cortex. Schizophrenia re-

search, 174(1):35–38, 2016.

[71] Hideyuki Okano and Sally Temple. Cell types to order: temporal speciﬁ-

cation of cns stem cells. Current opinion in neurobiology, 19(2):112–119,

2009.

[72] Debabrata Panja, Justin W Kenney, Laura DAndrea, Francesca Zalfa, Anni

Vedeler, Karin Wibrand, Rikiro Fukunaga, Claudia Bagni, Christopher G

Proud, and Clive R Bramham. Two-stage translational control of dentate

gyrus ltp consolidation is mediated by sustained bdnf-trkb signaling to mnk.

Cell reports, 9(4):1430–1445, 2014.

105 [73] M Pathania, EC Davenport, J Muir, DF Sheehan, G Lopez-Dom´ enech,´ and

JT Kittler. The autism and schizophrenia associated gene cyﬁp1 is critical

for the maintenance of dendritic complexity and the stabilization of mature

spines. Translational psychiatry, 4(3):e374, 2014.

[74] Deepak P Patil, Chun-Kan Chen, Brian F Pickering, Amy Chow, Constanza

Jackson, Mitchell Guttman, and Samie R Jaffrey. m6a rna methylation

promotes xist-mediated transcriptional repression. Nature, 537(7620):369,

2016.

[75] Xuyu Qian, Ha Nam Nguyen, Mingxi M Song, Christopher Hadiono,

Sarah C Ogden, Christy Hammack, Bing Yao, Gregory R Hamersky, Fadi

Jacob, Chun Zhong, et al. Brain-region-speciﬁc organoids using mini-

bioreactors for modeling zikv exposure. Cell, 165(5):1238–1254, 2016.

[76] Aaron R Quinlan and Ira M Hall. Bedtools: a ﬂexible suite of utilities for

comparing genomic features. Bioinformatics, 26(6):841–842, 2010.

[77] Bernd Radle,¨ Andrzej J Rutkowski, Zsolt Ruzsics, Caroline C Friedel, Ul-

rich H Koszinowski, and Lars Dolken.¨ Metabolic labeling of newly tran-

scribed rna for high resolution gene expression proﬁling of rna synthesis,

processing and decay in cell culture. Journal of visualized experiments:

JoVE, (78), 2013.

[78] Isabelle Rapin and Roberto F Tuchman. Autism: deﬁnition, neurobiology,

screening, diagnosis. Pediatric Clinics of North America, 55(5):1129–1146,

2008.

[79] Stephan Ripke, Benjamin M Neale, Aiden Corvin, James TR Walters, Kai-

How Farh, Peter A Holmans, Phil Lee, Brendan Bulik-Sullivan, David A

106 Collier, Hailiang Huang, et al. Biological insights from 108 schizophrenia-

associated genetic loci. Nature, 511(7510):421, 2014.

[80] John Rubenstein and Pasko Rakic. Patterning and Cell Type Speciﬁcation in the Developing CNS and PNS: Comprehensive Developmental Neuro-

science, volume 1. Academic Press, 2013.

[81] Tetsushi Sadakata, Miwa Washida, Yoshimi Iwayama, Satoshi Shoji, Yumi

Sato, Takeshi Ohkura, Ritsuko Katoh-Semba, Mizuho Nakajima, Yukiko

Sekine, Mika Tanaka, et al. Autistic-like phenotypes in cadps2-knockout

mice and aberrant cadps2 splicing in autistic patients. The Journal of clinical

investigation, 117(4):931, 2007.

[82] Stephen N Sansom, Dean S Grifﬁths, Andrea Faedo, Dirk-Jan Kleinjan,

Youlin Ruan, James Smith, Veronica Van Heyningen, John L Rubenstein,

and Frederick J Livesey. The level of the transcription factor pax6 is es-

sential for controlling the balance between neural stem cell self-renewal and

neurogenesis. PLoS genetics, 5(6):e1000511, 2009.

[83] Rahul Satija, Jeffrey A Farrell, David Gennert, Alexander F Schier, and Aviv

Regev. Spatial reconstruction of single-cell gene expression data. Nature

biotechnology, 33(5):495–502, 2015.

[84] Yasushi Satoh, Shogo Endo, Takahiro Nakata, Yasushi Kobayashi,

Kazuyuki Yamada, Toshio Ikeda, Atsuya Takeuchi, Takeshi Hiramoto, Ya-

suhiro Watanabe, and Tomiei Kazama. Erk2 contributes to the control of

social behaviors in mice. Journal of Neuroscience, 31(33):11953–11967,

2011.

[85] Daniel R Schoenberg and Lynne E Maquat. Regulation of cytoplasmic mrna

decay. Nature reviews. Genetics, 13(4):246, 2012.

107 [86] Paul Shannon, Andrew Markiel, Owen Ozier, Nitin S Baliga, Jonathan T

Wang, Daniel Ramage, Nada Amin, Benno Schwikowski, and Trey Ideker.

Cytoscape: a software environment for integrated models of biomolecular

interaction networks. Genome research, 13(11):2498–2504, 2003.

[87] Mikihito Shibata, Forrest O Gulden, and Nenad Sestan. From trans to cis:

transcriptional regulatory networks in neocortical development. Trends in

Genetics, 31(2):77–87, 2015.

[88] Jill L Silverman, Mu Yang, Catherine Lord, and Jacqueline N Crawley. Be-

havioural phenotyping assays for mouse models of autism. Nature Reviews

Neuroscience, 11(7):490–502, 2010.

[89] Sabrina L Spencer, Steven D Cappell, Feng-Chiao Tsai, K Wesley Overton,

Clifford L Wang, and Tobias Meyer. The proliferation-quiescence decision

is controlled by a bifurcation in cdk2 activity at mitotic exit. Cell, 155(2):

369–383, 2013.

[90] Hreinn Stefansson, Dan Rujescu, Sven Cichon, Olli PH Pietilainen,¨ An-

dres Ingason, Stacy Steinberg, Ragnheidur Fossdal, Engilbert Sigurdsson,

Thordur Sigmundsson, Jacobine E Buizer-Voskamp, et al. Large recurrent

microdeletions associated with schizophrenia. nature, 455(7210):232, 2008.

[91] Hreinn Stefansson, Andreas Meyer-Lindenberg, Stacy Steinberg, Brynja

Magnusdottir, Katrin Morgen, Sunna Arnarsdottir, Gyda Bjornsdottir,

G Bragi Walters, Gudrun A Jonsdottir, Orla M Doyle, et al. Cnvs con-

ferring risk of autism or schizophrenia affect cognition in controls. Nature,

505(7483):361, 2014.

[92] Elena Taverna, Magdalena Gotz,¨ and Wieland B Huttner. The cell biology

of neurogenesis: toward an understanding of the development and evolution

108 of the neocortex. Annual review of cell and developmental biology, 30:465–

502, 2014.

[93] Andrew TN Tebbenkamp, A Jeremy Willsey, Matthew W State, and Nenad

Sestan.ˇ The developmental transcriptome of the human brain: implications

for neurodevelopmental disorders. Current opinion in neurology, 27(2):149,

2014.

[94] Ludovic Telley, Subashika Govindan, Julien Prados, Isabelle Stevant, Serge

Nef, Emmanouil Dermitzakis, Alexandre Dayer, and Denis Jabaudon. Se-

quential transcriptional waves direct the differentiation of newborn neurons

in the mouse neocortex. Science, 351(6280):1443–1446, 2016.

[95] Nicholas HA Terry and R Allen White. Flow cytometry after bromod-

eoxyuridine labeling to measure s and g2+ m phase durations plus doubling

times in vitro and in vivo. Nature protocols, 1(2):859, 2006.

[96] Franc¸ois Tronche, Christoph Kellendonk, Oliver Kretz, Peter Gass, Katrin

Anlag, Paul C Orban, Rudolf Bock, Rudiger¨ Klein, and Gunther¨ Schutz.¨

Disruption of the glucocorticoid receptor gene in the nervous system results

in reduced anxiety. Nature genetics, 23(1):99–103, 1999.

[97] Maarten van den Buuse. Modeling the positive symptoms of schizophre-

nia in genetically modiﬁed mice: pharmacology and methodology aspects.

Schizophrenia bulletin, 36(2):246–270, 2009.

[98] Bert van der Zwaag, Wouter G Staal, Ron Hochstenbach, Martin Poot,

Henk A Spierenburg, Maretha V de Jonge, Nienke E Verbeek, Ruben van’t

Slot, Michael A van Es, Frank J Staal, et al. A co-segregating microduplica-

tion of chromosome 15q11. 2 pinpoints two risk genes for autism spectrum

109 disorder. American Journal of Medical Genetics Part B: Neuropsychiatric

Genetics, 153(4):960–966, 2010.

[99] Clemence´ Vanlerberghe, Florence Petit, Valerie´ Malan, Catherine Vincent-

Delorme, Sonia Bouquillon, Odile Boute, Muriel Holder-Espinasse, Bruno

Delobel, Ben´ edicte´ Duban, Louis Vallee, et al. 15q11. 2 microdeletion (bp1–

bp2) and developmental delay, behaviour issues, epilepsy and congenital

heart disease: a series of 52 patients. European journal of medical genetics,

58(3):140–147, 2015.

[100] Regina Waltes, Eftichia Duketis, Michael Knapp, Richard JL Anney, Guil-

laume Huguet, Sabine Schlitt, Tomasz A Jarczok, Michael Sachse, Laura M

Kampfer,¨ Tina Kleinbock,¨ et al. Common variants in genes of the postsynap-

tic fmrp signalling pathway are risk factors for autism spectrum disorders.

Human genetics, 133(6):781–792, 2014.

[101] Jian Wang, Yu Tao, Fan Song, Yue Sun, Jurg Ott, and David Saffen. Com-

mon regulatory variants of cyﬁp1 contribute to susceptibility for autism

spectrum disorder (asd) and classical autism. Annals of human genetics,

79(5):329–340, 2015.

[102] Xiang Wang, Jinbo Huang, Tingting Zou, and Ping Yin. Human m6a writ-

ers: Two subunits, 2 roles. RNA biology, 14(3):300–304, 2017.

[103] Xiao Wang, Zhike Lu, Adrian Gomez, Gary C Hon, Yanan Yue, Dali Han,

Ye Fu, Marc Parisien, Qing Dai, Guifang Jia, et al. m6a-dependent regula-

tion of messenger rna stability. Nature, 505(7481):117, 2014.

[104] Yang Wang, Yue Li, Julia I Toth, Matthew D Petroski, Zhaolei Zhang, and

Jing Crystal Zhao. N6-methyladenosine modiﬁcation destabilizes develop-

110 mental regulators in embryonic stem cells. Nature cell biology, 16(2):191,

2014.

[105] Zhaoqing Wang, Ming Fan, Demet Candas, Tie-Qiao Zhang, Lili Qin,

Angela Eldridge, Sebastian Wachsmann-Hogiu, Kazi M Ahmed, Brett A

Chromy, Danupon Nantajit, et al. Cyclin b1/cdk1 coordinates mitochon-

drial respiration for cell-cycle g2/m progression. Developmental cell, 29(2):

217–232, 2014.

[106] Zhenshan Wang and Daniel R Storm. Maternal behavior is impaired in fe-

male mice lacking type 3 adenylyl cyclase. Neuropsychopharmacology, 36

(4):772, 2011.

[107] Daniel R Weinberger. Implications of normal brain development for the

pathogenesis of schizophrenia. Archives of general psychiatry, 44(7):660–

669, 1987.

[108] Zhexing Wen, Ha Nam Nguyen, Ziyuan Guo, Matthew A Lalli, Xinyuan

Wang, Yijing Su, Nam-Shik Kim, Ki-Jun Yoon, Jaehoon Shin, Ce Zhang,

et al. Synaptic dysregulation in a human ips cell model of mental disorders.

Nature, 515(7527):414, 2014.

[109] Hyejung Won, Hye-Ryeon Lee, Heon Yung Gee, Won Mah, Jae-Ick Kim,

Jiseok Lee, Seungmin Ha, Changuk Chung, Eun Suk Jung, Yi Sul Cho, et al.

Autistic-like social behaviour in shank2-mutant mice improved by restoring

nmda receptor function. Nature, 486(7402):261, 2012.

[110] Cheng-Ran Xu, Philip A Cole, David J Meyers, Jay Kormish, Sharon Dent,

and Kenneth S Zaret. Chromatin prepattern and histone modiﬁers in a fate

choice for liver and pancreas. Science, 332(6032):963–966, 2011.

111 [111] Bing Yao, Kimberly M Christian, Chuan He, Peng Jin, Guo-li Ming, and

Hongjun Song. Epigenetic mechanisms in neurogenesis. Nature Reviews

Neuroscience, 17(9):537–549, 2016.

[112] Ki-Jun Yoon, Ha Nam Nguyen, Gianluca Ursini, Fengyu Zhang, Nam-Shik

Kim, Zhexing Wen, Georgia Makri, David Nauen, Joo Heon Shin, Young-

bin Park, et al. Modeling a genetic risk for schizophrenia in ipscs and mice

reveals neural stem cell deﬁcits associated with adherens junctions and po-

larity. Cell Stem Cell, 15(1):79–91, 2014.

[113] Chongzhi Zang, Dustin E Schones, Chen Zeng, Kairong Cui, Keji Zhao,

and Weiqun Peng. A clustering approach for identiﬁcation of enriched

domains from histone modiﬁcation chip-seq data. Bioinformatics, 25(15):

1952–1958, 2009.

[114] Bing Zhang, Stefan Kirov, and Jay Snoddy. Webgestalt: an integrated sys-

tem for exploring gene sets in various biological contexts. Nucleic acids

research, 33(suppl 2):W741–W748, 2005.

[115] Yong Zhang, Tao Liu, Clifford A Meyer, Jer´ omeˆ Eeckhoute, David S John-

son, Bradley E Bernstein, Chad Nusbaum, Richard M Myers, Myles Brown,

Wei Li, et al. Model-based analysis of chip-seq (macs). Genome biology, 9

(9):R137, 2008.

[116] Boxuan Simen Zhao, Ian A Roundtree, and Chuan He. Post-transcriptional

gene regulation by mrna modiﬁcations. Nature reviews. Molecular cell bi-

ology, 18(1):31, 2017.

[117] Boxuan Simen Zhao, Xiao Wang, Alana V Beadell, Zhike Lu, Hailing Shi,

Adam Kuuspalu, Robert K Ho, and Chuan He. m6a-dependent maternal

112 mrna clearance facilitates zebraﬁsh maternal-to-zygotic transition. Nature,

542(7642):475–478, 2017.

[118] Jing Zhao, Toshiro K Ohsumi, Johnny T Kung, Yuya Ogawa, Daniel J Grau,

Kavitha Sarma, Ji Joon Song, Robert E Kingston, Mark Borowsky, and Jean-

nie T Lee. Genome-wide identiﬁcation of polycomb-associated rnas by rip-

seq. Molecular cell, 40(6):939–953, 2010.

[119] Qian Zhao, Tao Li, XinZhi Zhao, Ke Huang, Ti Wang, ZhiQiang Li, Jue Ji,

Zhen Zeng, Zhao Zhang, Kan Li, et al. Rare cnvs and tag snps at 15q11. 2 are

associated with schizophrenia in the han chinese population. Schizophrenia

bulletin, 39(3):712–719, 2012.

[120] Jia Zou, Liang Zhou, Xiao-Xia Du, Yifei Ji, Jia Xu, Junlong Tian, Wanxiang

Jiang, Yi Zou, Shouyang Yu, Lingxue Gan, et al. Rheb1 is required for

mtorc1 and myelination in postnatal brain development. Developmental cell,

20(1):97–108, 2011.

113 F. Rojas Ringeling Curriculum Vitae

Curriculum Vitae

Personal Information

Full Name Francisca Rojas Ringeling, MD Place of Birth Santiago, Chile Citizenship Chilean, German Phone +49 160 96820574 Email [email protected]

Research Areas

Human genetics, Genomics, Bioinformatics, Neurogenetics, RNA biology

Education

2013-present Human Genetics Graduate Training Program Johns Hopkins University School of Medicine, Baltimore, Maryland Advisor: Hongjun Song, PhD Expected graduation date: September 2017 2010-2013 Residency Program in Medical Genetics Escuela de Medicina Universidad de Chile , Santiago, Chile Degree: Specialist in Medical Genetics, July 2013 2001-2009 Medical School Escuela de Medicina Universidad de Chile , Santiago, Chile Degree: Medical doctor, July 2013

Professional Experience

2013-present PhD student Laboratory of Dr. Hongjun Song, Institute for Cell Engineering, Johns Hopkins University School of Medicine, Baltimore, USA 2010-2013 Resident of Medical Genetics Hospital Clínico Universidad de Chile, Santiago, Chile 2009 Primary health care physician Centro de Salud Familiar Lo Barnechea, Santiago, Chile

114 F. Rojas Ringeling Curriculum Vitae

Awards and Scholarships

2014 Full scholarship for Riken Brain Science Institute Summer Program: “Disentangling Mental Disorders: from Genes to Circuits”, Riken BSI, Saitama, Japan 2013 Specialist in Medical Genetics title awarded with maximal distinction, Grade 6.9 (grade scaling system: approved= 4 - 4.9, approved with distinction= 5 - 5.9, approved with maximal distinction= 6 - 7), Escuela de Medicina, Universidad de Chile, Santiago, Chile 2009 Medical Doctor degree awarded with maximal distinction. Grade 6.1 (grade scaling system: approved= 4 - 4.9, approved with distinction= 5 - 5.9, approved with maximal distinction= 6 - 7), Escuela de Medicina, Universidad de Chile, Santiago, Chile

Poster presentations

2014 “Understanding the Role of CYFIP1 in Neuronal Development and Synaptic Function”, Riken Brain Science Institute Summer Program: “Disentangling Mental Disorders: from Genes to Circuit”, Riken BSI, Saitama, Japan

Languages

✄ Spanish (mother tongue) ✄ English (ﬂuent written and spoken) ✄ French (good knowledge) ✄ German (basic knowledge)

Publications

PUBLICATIONS IN PROGRESS

1. Yoon K*, Rojas Ringeling F*, Vissers C*, Jacob F, Pokrass M, Jimenez-Cyrus D, Su Y, Kim N, Zhu Y, Zheng L, Kim S, Wang X, Hyde TM, Weinberger DR, Jin P, Zhuang X, Regot S, Canzar S, He C, Ming G, and Song H. *These authors contributed equally to this work Temporal Control of Mammalian Cortical Neurogenesis by m6A mRNA Methylation. (Accepted for publication, Cell)

JOURNAL ARTICLES

2. Yoon K, Song G, Qian X, Pan J, Xu D, Rho H, Kim N, Habela C, Zheng L, Jacob F, Zhang F, Lee E, Huang W, Rojas Ringeling F, Vissers C, Li C, Yuan L, Kang K, Kim S, Yeo J, Cheng Y, Liu S, Wen Z, Qin C, Wu Q, Christian K, Tang H, Jin P, Xu Z, Qian J, Zhu H, Song H and Ming G. Zika-Virus-Encoded NS2A Disrupts Mammalian Cortical Neurogenesis by Degrading Adherens Junction Proteins. Cell Stem Cell, 2017.

115 F. Rojas Ringeling Curriculum Vitae

3. Unger S, Górna MW, Le Béchec A, Do Vale-Pereira S, Bedeschi MF, Geiberger S, Grigelioniene G, Horemuzova E, Lalatta F, Lausch E, Magnani C, Nampoothiri S, Nishimura G, Petrella D, Rojas-Ringeling F, Utsunomiya A, Zabel B, Pradervand S, Harshman K, Campos-Xavier B, Bonafé L, Superti-Furga G, Stevenson B, Superti-Furga A. FAM111A mutations result in hypoparathyroidism and impaired skeletal development. American Journal of Human Genetics, 2013 Jun 6;92(6):990-5.

4. Suazo J, Taucher SC, Rojas F, Martin LM, Santos JL, Rotter K, Solar M, Tapia E, Pardo RA. Family Based Association Study Between SLC2A1, HK1 and LEPR Polymorphisms with Spina Biﬁda in Chile. Reproductive Sciences, 2013 Oct;20(10):1207-14.

5. Passalacqua C, Melo C, Martín LM, Rojas F, Sanz P, Taucher SC, Araníbar L. Research Letter: A Pigmentary Skin Defect is a New Finding in Marshall-Smith Syndrome. American Journal of Medical Genetics A, 2011 Aug; 155A(8):2015-7.

116