Thesis Abstract

Post-transcriptional Regulatory Mechanisms Controlling Development of Murine Cerebral Cortical Precursors

Gianluca Amadei

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Molecular Genetics University of Toronto

Post-transcriptional Regulatory Mechanisms Controlling Development of Murine Cerebral Cortical Precursors

Gianluca Amadei

Doctor of Philosophy

Department of Molecular Genetics University of Toronto

2016 Thesis Abstract

The complex neural circuitry of the mammalian nervous system arises from a small pool of neural precursors that, during development, sequentially gives rise to neurons, astrocytes and oligodendrocytes. Several molecular mechanisms and environmental stimuli regulate the expansion of the neural precursor pool and the subsequent generation of differentiated progeny.

In this thesis, I sought to investigate whether post-transcriptional regulation plays a role in the development of mammalian neural precursors. In the first part of this thesis, I show that the double stranded RNA binding protein Staufen2 is part of a repressive complex, with Pumilio2 and DDX1, that prevents early differentiation of neural precursors into neurons by repressing the translation of neurogenic mRNAs such as prox1. In the second part of the thesis I show that

Smaug2, a translational repressor, also prevents early generation of neurons by repressing nanos1 mRNA, which encodes another RNA-binding protein. I also show that nanos1 repression occurs by its inclusion in a P-body like granule with 4E-T, another known repressor of mRNA translation. In the last chapter I identify mRNAs associating with Smaug2 and I suggest that

Smaug2 may have additional functions that are independent of nanos1 regulation and that unlike nanos1, which is regulated together by Smaug2 and 4E-T, the majority of Smaug2 mRNAs are regulated independently from 4E-T. Together, these studies suggest that several RNA-binding ii

proteins are crucial regulators of cortical development and that they perform this role by forming several, largely independent, repressive RNA-protein complexes. I suggest that these repressive complexes prime neural precursors to generate progeny at the appropriate time by repressing proneurogenic mRNAs until the appropriate developmental cue.

iii

Acknowledgments

Upon completion of this doctoral degree, I realize more than ever that this journey could not have been completed without the help of so many amazing people.

First and foremost, I would like to thank my supervisors Freda and David for all their help along this PhD. Your support, mentoring, and guidance meant so much to me and your passion for science has been a constant inspiration. Thank you for taking a chance on me and letting me study in your laboratory. It has been a terrific opportunity and a life changing experience!

I would like to thank my committee members Dr. Bret Pearson and Dr. Howard Lipshitz, who have constantly been evaluating my progress over the years. Your invaluable intellectual input and criticism was instrumental in guiding this thesis along and I am very grateful for all your help.

Many many thanks are also due to our wonderful collaborators, Dr. Craig Smibert and Dr. Jason Dumelie. When we were not sure where to go next, you were always able to offer advice and technical expertise and I am very grateful to you for that!

My gratitude also goes to my wonderful colleagues and friends in the Kaplan/Miller laboratory. I could not have asked for a more talented group of people to work with. Over the years, the incredible display of talent, dedication and passion for science that I have witnessed in each and every one of you is both humbling and inspiring. I feel every one of you, in his/her own way, has taught me something valuable and I hope to put these lessons to good use in my future work. I will always cherish our friendship and the memory of our years together.

A special place in this PhD journey belongs to two wonderful people, Peter and Sally Cant. Over the past few years I have been fortunate enough to benefit from the scholarship they established to honour the memory of their son, David S. Cant and I have also had the privilege of getting to know them better on a number of occasions. It is hard to find two people who are more kind, gentle, and generous than Sally and Peter and I will never forget their commitment to supporting brain research and students. It is also thanks to the contribution of people like them that research can move forward. Peter, Sally, I will never forget you and your generosity!

Last but not least, many thanks to my family and good friends. I do not know where I would be without your unwavering support. I am so fortunate to have you in my life!

To all of you, thank you, thank you, thank you, from the bottom of my heart!

Table of Contents

Acknowledgments ...... iv Table of Contents ...... v List of Abbreviations ...... viii List of Tables ...... xii List of Figures ...... xiii

Chapter 1 Literature Review ...... 1 1.1 Overview of thesis objectives: ...... 1 1.2 Overview of murine cortical development: ...... 2 1.3 Neural diversity results from a heterogeneous pool of cortical precursors: ...... 5 1.3.1 Neuroepithelial stem cells (NESCs): ...... 5 1.3.2 Radial precursor cells (RP): ...... 7 1.3.3 Basal or intermediate progenitor cells: ...... 8 1.3.4 Outer radial glial precursor cells: ...... 9 1.3.5 Oligodendrocyte precursor cells: ...... 10 1.4 Mechanisms regulating cortical neural precursor self-renewal versus differentiation: ...... 10 1.4.1 Mitotic polarity as a potential regulator of neural precursor cell fate: ...... 11 1.4.2 RNA-binding proteins as asymmetrically-segregated determinants in stem cell biology: .. 12 1.4.2.1 The RNA-binding protein Staufen2: ...... 15 1.4.2.2 The post-transcriptional regulator Smaug2: ...... 18 1.4.2.3 The function of mammalian Smaug: ...... 22 1.4.3 A potentially conserved mRNA repression pathway: ...... 27 1.4.3.1 Nanos: ...... 27 1.4.3.2 4E-T: ...... 28

Chapter 2 Experimental Procedures ...... 30

Chapter 3 An asymmetrically localized Staufen2-dependent RNA complex regulates maintenance of mammalian neural stem cells ...... 41 3.1 SUMMARY ...... 42 3.2 BRIEF INTRODUCTION AND RATIONALE ...... 42 3.3 RESULTS ...... 43 3.3.1 Staufen2 is apically localized in embryonic radial glial precursors in the developing murine cortex ...... 43 3.3.2 Staufen2 is part of an apical RNA complex in radial glial precursors ...... 46 3.3.3 Asymmetric localization of Staufen2 and prox1 mRNA in dividing radial glial precursors49 3.3.4 Knockdown of Staufen2 promotes neurogenesis and depletes radial glial precursors ...... 52 3.3.5 Knockdown of either Pumilio2 or DDX1 phenocopies the effects of Staufen2 knockdown ...... 58 3.3.6 Disruption of Staufen2-RNA interactions affects prox1 mRNA localization and expression and causes differentiation of radial glial precursors ...... 59 3.4 CONCLUSIONS ...... 64

Chapter 4 A Smaug2-based translational repression complex determines the balance between precursor maintenance versus differentiation during mammalian neurogenesis ...... 65 v

4.1 SUMMARY ...... 66 4.2 BRIEF INTRODUCTION AND RATIONALE ...... 66 4.3 RESULTS ...... 68 4.3.1 Smaug2 is expressed in embryonic cortical precursors during development ...... 68 4.3.2 Smaug2 regulates the genesis of cortical neurons ...... 68 4.3.3 Smaug2 is sufficient to maintain cortical precursors ...... 74 4.3.4 The mRNA encoding nanos1, but not nanos2 or nanos3, is a target of Smaug2 in the embryonic cortex ...... 77 4.3.5 Nanos1 promotes the genesis of neurons from cortical precursors ...... 81 4.3.6 Smaug2 and nanos1 mRNA are present in RNP granules containing the repressors Dcp1 and 4E-T ...... 85 4.3.7 Smaug2 and 4E-T may repress nanos1 by recruiting CNOT7, a conserved mammalian deadenylase ...... 90 4.3.8 Enhanced neurogenesis following Smaug2 knockdown is caused by derepression of nanos1 mRNA translation ...... 93 4.4 CONCLUSIONS ...... 96

Chapter 5 Smaug2 RIP-Chip identifies approximately 200 mRNAs associating with Smaug2 in cortical radial precursors in vivo...... 98 5.1 SUMMARY: ...... 98 5.2 BRIEF INTRODUCTION AND RATIONALE ...... 98 5.3 RESULTS ...... 100 5.3.1 Smaug2 RIP-Chip identifies mRNAs associating with Smaug2 in vivo ...... 100 5.3.2 Smaug2 RIP-Chip was validated by two independent methods ...... 104 5.3.3 Ingenuity Pathway Analysis highlights several Smaug2-associated mRNAs with known roles in migration ...... 111 5.3.4 PANTHER and DAVID analysis of the Smaug2-associated transcripts hints at additional Smaug2 roles in cortical development ...... 111 5.3.5 Comparison of Smaug2 and 4E-T associated transcripts show a small but significant overlap ...... 119 5.4 CONCLUSIONS ...... 122

Chapter 6 Discussion ...... 124 6.1 An asymmetrically localized, Staufen2-dependent RNA granules controls radial precursor maintenance and differentiation ...... 124 6.1.1 The companion paper published with this study (Kusek et al., 2012) supports our results ...... 127 6.1.2 Conserved developmental roles of the Staufen proteins? ...... 129 6.2 A Smaug2-based translational repression complex determines the balance between precursor maintenance versus differentiation during mammalian neurogenesis ...... 130 6.3 Smaug2 RIP-Chip identifies approximately 200 mRNAs associating with Smaug2 in cortical radial precursors in vivo...... 134 6.3.1 Smaug2 functions that are independent of nanos1 regulation...... 134 6.3.2 The relationship between Smaug2 and 4E-T ...... 137 6.3.2.1 Significance of the mRNAs bound by both 4E-T and Smaug2 ...... 139 6.3.3 Pairwise analysis of Smaug2 RIP-chip: pro and cons ...... 140 6.3.3.1 Limitations of the SRE score ...... 141 6.4 General Discussion and Future Directions ...... 142 vi

REFERENCES ...... 149

APPENDIX……………………………………………………………………………………..201

vii

List of Abbreviations

3’ UTR: 3’ untranslated region 4E-BP1: 4E Binding Protein 1 4E-T: 4E-transporter AGO1: Argonaute 1 AKT: Protein kinase B AMBRA1: Activating molecule in BECN1-regulated autophagy protein 1 AMPA: α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid APKC: Atypical protein kinase C ARHGAP: Rho GTPase-activating protein ASD: Autism spectrum disorder ASPM: Abnormal spindle-like microcephaly-associated protein BAX: BCL2-Associated X Protein bHLH: Basic helix-loop-helix BLBP: Brain-lipid binding protein BMP: Bone morphogenetic protein BP: Basal progenitor, also known as intermediate progenitor (IP) BRAF: Serine/threonine-protein kinase B-raf BRAT: Brain tumour protein CAF1: CCR4-associated factor 1 CAMKIIα: calcium/calmodulin-dependent protein kinase type II CAMKK2: Calcium/calmodulin-dependent protein kinase kinase 2 CCR4-NOT: Carbon catabolite repression 4 (CCR4)–negative on TATA-less (NOT) cDNA: Complementary DNA CDK: Cyclin-dependent kinase CDS: Coding sequence C. elegans: Caenorabditis elegans C-MYC: Proto-oncogene protein MYC CNOT6: CCR4-NOT transcription complex subunit 6 CNOT6L: CCR4-NOT transcription complex subunit 6-like CNOT7: CCR4-NOT transcription complex subunit 7 CNOT8: CCR4-NOT transcription complex subunit 8 CNS: Central nervous system CP: Cortical plate CPEB: Cytoplasmic polyadenylation element binding protein CUX1: Cut-Like Homeobox 1 CUX2: Cut-Like Homeobox 2 DAPI: 4', 6-diamidino-2-phenylindole DAVID: Database for Annotation, Visualization and Integrated Discovery DCLK: Doublecortin-like kinase DCP1: Decapping Enzyme 1 DDX1: DEAD box protein 1 DDX6: DEAD box protein 6 DAZL: Deleted in azoospermia-like DLX1: Distal-Less Homeobox 1 DNA: Deoxyribonucleic acid viii

D. melanogaster: Drosophila melanogaster E: Embryonic day. E.g.: E12, embryonic day 12. E. coli: Escherichia coli EGFP: Enhanced green fluorescent protein EIF4E: Eukaryotic initiation factor 4E EIF4G: Eukaryotic initiation factor 4G ELAVL3: Embryonic lethal abnormal vision-like 3, also known as HuC, human antigen C EMX2: Empty spiracles homeobox 2 ENC1: Ectoderm-neural cortex protein 1 ERK: Extracellular Signal-regulated Kinase EST: Expressed sequence tag FGF: Fibroblast growth factor FISH: Fluorescent in situ hybridization FMRP: Fragile X mental retardation protein G1: Gap1 G2: Gap2 S: S-phase S. cerevisiae: Saccharomyces cerevisiae GABA: γ-amino butyric acid GFAP: Glial fibrillary acidic protein GFP: Green fluorescent protein GIT1: G protein-coupled receptor kinase-interactor 1 GLAST: Astrocyte-specific glutamate transporter GO: Gene ontology GSH2: Glutathione synthetase HEK-293T: Human embryonic kidney 293T HES5: Hairy and enhancer of split 5 HSP83: Heat-shock protein 83 HUR: Human antigen R also known Elavl1, embryonic lethal abnormal vision-like 1 IGG: Immunoglobulin G INSC: Inscutable INM: Interkinetic nuclear migration IP: Intermediate progenitor also known as basal progenitor (BP) IPA: Ingenuity Pathway Analysis IZ: Intermediate zone JNK: C-jun N-terminal kinase LFC: Lbc’s first cousin LFNG: Lunatic fringe LHX6: LIM/homeobox protein 6 LSM1: U6 snRNA-associated Sm-like protein M7G: 7-methylguanosine cap MARK4: MAP/microtubule affinity-regulating kinase 4 MASH1: Mammalian achaete scute homolog-1 MEF2C: Myocyte-specific enhancer factor 2C MYH10: Myosin-10 miRNA: MicroRNA MLX: Max-like protein X mRNA: Messenger RNA ix

mTor: Mammalian target of rapamycin MXD1: Max dimerization protein 1 MXI1: Max-interacting protein 1 MW: Molecular weight MZT: Maternal to zygotic transition NESCs: Neuroepithelial stem cells NEUROD: Neurogenic differentiation factor 1 NGN2: Neurogenin 2 NHLH1: Nescient helix loop helix 1 NKX2.1: Thyroid transcription factor 1 NMDA: N-methyl-D-aspartate receptor OPC: Oligodendrocyte precursor cell ORB: Ovarian protein ORG: Outer radial glia OSVZ: Outer subventricular zone P-bodies: Processing bodies PB: Piggybac transposon PABP: Poly A binding protein PANTHER: Protein ANalysis THrough Evolutionary Relationships PAR3: Protease activated receptor 3 PAR6: Protease activated receptor 6 PAX6: Paired box protein 6 PCA: Principal Component Analysis PDGF: Platelet-derived growth factor PGCs: Primordial germline cells PHAT: Pseudo heat analogous topology PHTF1: Putative homeodomain transcription factor 1 PLA: Proximity Ligation Assay PPAP2B: Phospholipid phosphatase 3 PRDM8: PR domain zinc finger protein 8 PRKCD: Protein kinase C delta type PROX1: Prospero homeobox protein 1 PROS: Prospero PURα: Purine-rich single-stranded DNA-binding protein alpha RT-qPCR: Reverse transcriptase quantitative polymerase chain reaction RBP: RNA-binding protein RIP-Chip: RNA-binding protein immunoprecipitation – gene microarray hybridization RISC: RNA-induced silencing complex RNA: Ribonucleic acid RNP: Ribonucleoprotein particle RP: Radial precursor cell RT-PCR: Reverse transcriptase polymerase chain reaction S. cerevisiae: Saccharomyces cerevisiae S100β: S-100 protein beta chain S-foci: Smaug1 foci SAM: Sterile-α-motif containing domain SATB2: Special AT-rich sequence-binding protein 2 S.E.M.: Standard error of the mean x

SHH: Sonic hedgehog shRNA: Short hairpin RNA SOCS7: Suppressor of cytokine signalling 7 SOX2: Sex determining region Y-box 2 SRC: Proto-oncogene tyrosine-protein kinase Src SRE: Smaug recognition element SSR: Smaug similarity region STAU1: Staufen1 STAU2: Staufen2 SVET1: Subventricular expressed transcript 1 SVZ: Subventricular zone TBR2: T-brain gene 2 TLE4: Transducin-like enhancer protein 4 TRIM32: Tripartite motif-containing protein 32 TRKB: Tyrosine kinase receptor B TTP: Tristetraprolin v: Ventricle VTS1P: Protein vts1p, S. cerevisiae Smaug homologue VZ: Ventricular zone WNT: Wingless-related integration site XRN1: 5'-3' exoribonuclease 1 ZBP: Zipcode-binding protein ZO-1: Zona-occludens 1

List of Tables

Table 1: mRNAs significantly associated with Smaug2, as identified by Smaug2 RIP-Chip (In Appendix)

Table 2: SRE score of mRNA transcripts expressed at E12.5 in radial precursors, mRNAs significantly enriched in the control IgG RIP and mRNAs significantly enriched in the Smaug2 RIP (short version, full version in the attached file)

Table 3: Ingenuity Pathway Analysis (IPA) of the Smaug2 associated transcripts

Table 4: PANTHER GO enrichment analysis of the Smaug2-associated mRNAs

Table 5: Comparison of DAVID and PANTHER identification of the Smaug2-associated mRNAs (In Appendix)

Table 6: DAVID GO enrichment analysis of the Smaug2-associated mRNAs

Table 7: Smaug2-associated mRNAs highlighted by IPA, DAVID or PANTHER with a documented role in the CNS (In Appendix)

Table 8: Smaug2-associated mRNAs that were NOT highlighted by IPA, DAVID or PANTHER but with a documented role in the CNS

xii

List of Figures

Figure 1: Overview of neurogenesis in murine cortical development

Figure 2: Interkinetic nuclear migration

Figure 3: Smaug employs multiple mechanisms to repress and/or degrade target mRNA

Figure 4: Conservation between D. melanogaster and mammalian Smaug

Figure 5: S-foci respond to specific synaptic stimuli

Figure 6: Staufen2 is expressed by RPs and newborn neurons during cortical development

Figure 7: Staufen2 interacts and colocalizes with RNA complex proteins and target mRNAs

Figure 8: Staufen2 and its target mRNA prox1 are asymmetrically enriched in dividing ventricular precursors

Figure 9: Knockdown of Staufen2 in culture causes increased neurogenesis and depletion of cycling precursors

Figure 10: Depletion of Staufen2 in vivo increases genesis of neurons at the expense of RPs

Figure 11: Knockdown of either Pumilio2 or DDX1, components of the apically localized Staufen2-containing complex, phenocopies Staufen2 knockdown

Figure 12: Staufen2 mediates apical localization of prox1 mRNA by its RNA binding domain, and this RNA binding activity is essential for Staufen2 to promote RP maintenance

Figure 13: Smaug2 but not Smaug1 protein is expressed in apical precursors and newborn neurons during embryonic cortical neurogenesis

Figure 14: Smaug2 knockdown in culture and in vivo increases neurogenesis and depletes cycling precursors

Figure 15: Smaug2 overexpression in vitro and in vivo is sufficient to enhance cortical precursor self-renewal

Figure 16: nanos1 mRNA is a Smaug2 target in embryonic cortical precursors

Figure 17: Nanos1 is necessary and sufficient to promote neurogenesis in vivo

Figure 18: Smaug2 and nanos1 mRNA are associated with 4E-T in a P-Body-like granule in Pax6-positive apical precursors

xiii

Figure 19: Smaug2, 4E-T and nanos1 mRNA are associated with CNOT7, a conserved mammalian mRNA deadenylase protein

Figure 20: Knockdown of Smaug2 or 4E-T causes aberrant Nanos1 expression, and this is responsible for the Smaug2 knockdown-mediated increase in neurogenesis

Figure 21: RIP-Chip of endogenous Smaug2 isolated from E12.5 mouse cortices identifies 200 high confidence Smaug2 targets

Figure 22: Comparison of mRNAs associated with Smaug2 and 4E-T reveals a small but significant shared subset

xiv

Chapter 1 Literature Review

1.1 Overview of thesis objectives:

The adult mammalian cerebral cortex is an incredibly complex structure that arises from a small number of stem and progenitor cells (together called precursor cells in this thesis), that are somehow able to generate very different cell types such as neurons, astrocytes and oligodendrocytes (Miller and Gauthier, 2007; Kriegstein and Alvarez-Buylla, 2009). Each of these cell types arises in an orderly temporal fashion from either multipotent precursors or from progenitors with a more restricted potential that reside in the germinal zones along the lateral ventricles of the developing telencephalon (Malatesta et al., 2008; Taverna et al., 2014). From midgestation to birth, the precursors undergo a regulated number of cellular divisions to generate neurons, followed by astrocytes and oligodendrocytes, whose genesis is completed perinatally and postnatally, respectively. Such a feat implies that intrinsic changes within the precursors as well as extrinsic changes in the environment of the developing cortex regulate the generation of all these different cell types. The focus of this thesis is the investigation of the molecular mechanisms regulating events such as precursor fate determination to generate neurons. Neurons are the “computing units” of the central nervous system (CNS) and they are connected in a complex network that is able to receive and send signals (Boron and Boulpaep, 2009). Signals are transmitted electrically when a neuron generates and propagates an action potential, and chemically once an action potential reaches a synapse and triggers release of neurotransmitters, which will propagate the signal to other neurons. Different types of excitatory glutamatergic neurons are formed in the dorsal cortex and these are negatively regulated by inhibitory interneurons arising from the ventral telencephalon (Anderson et al., 2002). These different classes of neurons migrate away from their dorsal and ventral germinal zones and ultimately form the characteristic laminar structure of the cortex prior to establishing both local and subcortical connections with other CNS neurons (Molyneaux et al., 2007). Neurons, however, cannot perform their important function alone. Astrocytes in fact, play a critical role in providing support to the neurons by maintaining general homeostasis and by functioning at synapses where they reabsorb neurotransmitters and metabolize neurotransmitter precursors (Sofroniew and Vinters, 2010). As previously mentioned, astrocytes are generated after neurons.

Last but not least, efficient signal conduction in axons would not occur without the contribution of oligodendrocytes. In fact, oligodendrocytes are responsible for making the myelin sheath that isolates neuronal axons in the CNS, thereby promoting efficient action potential propagation and preventing interference (Purger et al., 2015). An increasing body of evidence indicates that neurological disorders can arise not only from defects within these cell types or from incorrect neuronal connections, but also if the physiological numbers of each of these cell types is altered. It has been known for a while that such alterations can derive from neuronal degeneration (Selkoe and Hardy, 2016) or loss throughout life, but more recently we have begun appreciating that early perturbations occurring during gestation can be equally deleterious. Examples of disorders that can arise from altered prenatal neural development include autism (DiCicco-Bloom et al., 2006), Down’s syndrome (Becker et al., 1991), epilepsy (Becker et al., 2006), schizophrenia (Raine, 2006; Mei and Xiong, 2008) and neurological cancers (Xie and Chin, 2008). Therefore, in order to understand developmental neurological disorders, a complete and integrated knowledge of neural precursor development and of fate decision regulation is needed. With this knowledge, it is a common hope in the scientific community but also in the general public that we can develop therapeutic tools to intervene and promote repair in situations such as stroke and spinal cord injury. Already in recent years the advances in technology and in our understanding have shed light on several exciting and novel mechanisms that control fundamental neural stem cell biology, and the therapeutic potential of some of these findings (Wang et al., 2012) is being evaluated in clinical trials.

1.2 Overview of murine cortical development:

There are several reasons why the murine cortex is a good model system for studying brain development. First, the timing of generation of neurons, astrocytes and oligodendrocytes is well- characterized (Kriegstein and Alvarez-Buylla, 2009). Second, cultured precursors largely recapitulate in vivo development, generating these same cell types in the same temporal order (Qian et al., 2000). Third, cortical development occurs during the latter part of gestation and during early postnatal life, and the cortex is superficially located within the embryo, making it accessible for electroporation and precursor dissection. Fourth, the availability of constitutive and inducible murine knock out models and inducible Cre recombinase drivers has allowed the study of thousands of gene for which a human homolog can be clearly identified (Nagy et al.,

2009; Abe and Fujimori, 2013). Lastly, the cortex is the seat of many higher cognitive functions and behaviours, meaning that altered behaviours, such as those observed in human autism spectrum disorder (ASD), can be modelled by behavioural tests in mice (Hulbert and Jiang, 2016). The cerebral cortex begins its formation when a small portion of the ectoderm is induced by signals from the mesoderm to acquire neural potential, and forms the neural plate (Boron and Boulpaep, 2009). Initially, the neural plate is a single layer of neuroectodermal cells, but their active proliferation results in the formation of neural folds first folding into a neural groove and then fusing dorsally to form the neural tube (Wilde et al., 2014). The lumen of the neural tube, the neural canal, eventually develops to form the four ventricles of the brain; two that will be mentioned throughout the thesis are those in the telencephalon, called the lateral ventricles. At embryonic day (E) 8 – 9, the cells in the neural tube are mitotically active and this thin layer of cells comprises the entire neuroepithelium. At this developmental stage gradients of morphogens such as Bone Morphogenetic Proteins (BMP), WNTs, fibroblast growth factor 8 (FGF8) and sonic hedgehog (Shh) diffuse through the neural tube and pattern it along an anterior-posterior as well as dorso-ventral axis (Molineaux et al., 2007; Briscoe and Small, 2015). The expression of key transcription factors in the prospective telencephalon is induced in response to these morphogens and different precursor populations acquire different identities. For instance, the transcription factors Pax6 and Emx2 specify dorsal cortical precursors while Dlx1, 2, 5, Gsh2, Mash1, Lhx6 and Nkx2.1, expressed specifically in the ventral telencephalon, specify subpallial precursors. Subsequent to these specification events, the majority of glutamatergic neurons are generated from the dorsal precursors, while most of the γ-amino butyric acid (GABA)-ergic interneurons come from the ventral precursors (Lee and Jessell, 1999; Campbell et al., 2003; Grove and Fukuchi-Shimogori, 2003). How is the task of creating cortical architecture accomplished by these various neural precursors? Extensive studies of precursor divisions have shown that cortical precursors undergo symmetric divisions to increase the precursor pool and asymmetric divisions to generate other types of precursors or differentiated progeny, such as neurons (Taverna et al., 2014). As the developing brain increases in size, two germinal zones form: the ventricular zone (VZ), which is the apical most region along the lateral ventricles (Kriegstein and Götz, 2003; Götz and Huttner, 2005), and the subventricular zone (SVZ), at the basal margins of the ventricular zone (Haubensak et al., 2004; Miyata et al., 2004; Noctor et al., 2004; Fietz et al., 2010; Hansen et al., 2010; Reillo and Borrell, 2012). These two germinal zones contain different 3 types of precursors which will be discussed in later sections. Basal to the SVZ there is an area of transition, called the intermediate zone, through which newborn neurons migrate to the most superficial region, the cortical plate, to ultimately integrate in the appropriate cortical layer (Figure 1).

In the adult mammalian cortex there are six layers and different neuronal subtypes reside in each of these (Molyneaux et al., 2007). Cortical layers are generated in an inside-out fashion: early born neurons reside in the deeper layers and later born neurons migrate past the deeper- layer neurons to fill the outer, more superficial layers (Rakic, 1974). The earliest neurons are generated as early as E10.5 and migrate to the cortical plate to form the primordial plexiform layer. As development proceeds, the plexiform layer splits to form the marginal zone and the subplate (Molyneaux et al., 2007). In the mature cortex, the marginal zone becomes the most

4 superficial layer, layer I, right under the pial surface, while the subplate remains below the other cortical layers as they form (Takahashi et al., 1996, 1999). Both the VZ and SVZ are maintained and/or even expand during neurogenesis but once neurogenesis is complete at the end of gestation, the VZ disappears. However, while the SVZ decreases in size, it persists throughout adulthood to become part of the forebrain adult neural stem cell niche (Doetsch et al., 1999; Bonfanti and Peretto, 2007).

1.3 Neural diversity results from a heterogeneous pool of cortical precursors:

During cortical development, precursors face the challenge of generating the right type and number of daughter cells without exhausting themselves prematurely. Therefore, precise molecular mechanisms have evolved to allow for neural precursor pool expansion before neurogenesis, and to ensure the genesis of cortical neurons while setting aside enough precursor cells for the genesis of glial cells and adult neural stem cells (Götz and Huttner, 2005; Zhong and Chia, 2008). A diverse population of precursors plays a role at each of these steps, and will be discussed below.

1.3.1 Neuroepithelial stem cells (NESCs):

All of the neurons, astrocytes and oligodendrocytes of the brain, either directly or indirectly, descend from NESCs. NESCs are the cells of the mitotically active neuroepithelium at the earliest stages of neural development. At these early stages the neuroepithelium appears pseudostratified because NESCs undergo interkinetic nuclear migration (INM), whereby the position of a nucleus correlates with the status of the cell cycle. In particular, the nucleus of an NESC is closest to the ventricular surface during mitosis, after which it moves basally during G1 phase. The nucleus is furthest away from the ventricle at S-phase, following which the nucleus moves back towards the ventricle during G2 (Haubensak et al., 2004; Götz and Huttner, 2005) (Figure 2).

At early developmental points such as E8 and E9, NESCs are highly polarized along their apical-basal axis and they exhibit canonical markers of epithelial cells (Huttner and Brand, 1997). For instance, certain transmembrane proteins like prominin are selectively found in the apical plasma membrane (Weigmann et al., 1997; Corbeil et al., 2001), while receptors for basal lamina components such as integrin α6 are enriched in the basal plasma membrane, contacting the basal lamina (Wodarz et al., 2003). Tight and adherens junctions are found in the most apical portion of the lateral plasma membrane and this compact “barrier” formed by the apical endfeet constitutes the ventricular surface itself. Maintenance of the apical-basal polarity of NESCs requires the integrity of adherens junctions (Zhadanov et al., 1999). Clonal analysis performed by culturing NESCs at very low density (Qian et al., 2000) showed that NESCs are true multipotent neural progenitors because if they are isolated or labelled early during development they can sequentially give rise to neurons, astrocytes and oligodendrocytes. However, the majority of these differentiated progeny in vivo do not originate from the NESCs directly. Indeed, although NESCs can generate neurons and astrocytes via

6 asymmetric divisions in culture, in vivo they largely increase the precursor pool by symmetric divisions (Rakic, 1995; Chenn and McConnell, 1995) before giving rise to another type of precursor cell termed a radial glial precursor cell (called radial precursor or RP in this thesis), which then generates the majority of neurons and glia (Malatesta et al., 2003; Merkle et al., 2004; Spassky et al., 2005; Pinto and Götz, 2007).

1.3.2 Radial precursor cells (RP):

These cells were originally called radial glia because in some mammals (including humans but not mice), they expressed the astrocyte marker glial fibrillary acidic protein (GFAP) (Levitt and Rakic, 1980; Sancho-Tello et al., 1995), and because their long radial fibers, which extend from their apically localized cell bodies to the basal surface of the cortex, provided a scaffold for the migration of newborn neurons from the VZ/SVZ to the cortical plate. However, lineage tracing experiments conclusively showed that RPs not only provide the migratory scaffold, but that they also generate the neurons of the embryonic cortex (Malatesta et al., 2000; Hartfuss et al., 2001; Miyata et al., 2001; Noctor et al., 2001; Tamamaki et al., 2001). Later studies showed that RPs, as a population, are multipotent, sequentially giving rise to neurons and astrocytes, and that the majority of ventricular zone precursors are indeed radial precursors (Noctor et al., 2002). The story came full circle when it was discovered that during development NESCs undergo a transition to become RPs (Götz et al., 2002; Malatesta et al., 2003; Anthony et al., 2004). The transition from NESCs to RPs occurs immediately prior to the onset of neurogenesis at around E10 in the murine cortex (Hartfuss et al., 2001; Noctor et al., 2002) and NESCs downregulate some epithelial features such as tight junctions, with the exception of the tight junction protein zona-occludens 1 (ZO-1) (Aaku-Saraste et al., 1996). However, they maintain most of their epithelial characteristics, including adherens junctions at the apical-most portion of the lateral membrane and localization of the Par complex, formed by Par3/Par6 and aPKC, with the apical cell cortex (Chenn et al., 1998; Hartsock and Nelson, 2008). RPs also still undergo INM but the migration of the nucleus is restricted to the basal margin of the ventricular zone (Takahashi et al., 2002). Thus, like NESCs, RPs are polarized epithelial cells, although their basal process becomes greatly elongated as the cortex grows (Rakic, 2003). Nonetheless, RPs do start expressing a number of characteristics that were previously thought to be specific to astrocytes, including the presence of visible glycogen granules (Choi,

1981), and expression of "glial-specific" proteins such as astrocyte-specific glutamate transporter (GLAST) (Shibata et al., 1997; Malatesta et al., 2000; Hartfuss et al., 2001), the Ca2+-binding proteins S100β (Vives et al., 2003), vimentin (Schnitzer et al., 1981) and brain- lipid binding protein (BLBP) (Feng et al., 1994; Hartfuss et al., 2001). RPs also express proteins that are clearly not associated with astrocytes, including transcription factors like Pax6, which is detectable as early as E8.5 in the neural tube (Walther and Gruss, 1991; Grindley et al., 1995). A variety of approaches have been used to ask whether RPs maintain the multilineage potential of NESCs. Culture studies show that, at least in vitro, many RPs have the potential to generate both neurons and glia (Walsh and Cepko, 1988, 1992; Parnavelas et al., 1991; Williams et al., 1991; Reid et al., 1995). However, lineage tracing in vivo indicates that, while some RPs do generate both neurons and glia, most make just neurons or just glia (Luskin et al., 1988, 1993; Price and Thurlow, 1988; Grove et al., 1993; Krushel et al., 1993; Davis and Temple, 1994; Williams et al., 1995; Mione et al., 1997; McCarthy et al., 2001; Anthony et al., 2004; Wu et al., 2006; Battiste et al., 2007). The molecular mechanisms regulating these different in vivo fate choices are still unclear, and comprise one of the major questions addressed by this thesis. During the early stages of neurogenesis, RPs can divide symmetrically to give rise to more radial precursors or asymmetrically to generate a RP cell and a neuron or a RP cell and an intermediate progenitor cell (discussed below) (Haubensak et al., 2004; Noctor et al., 2004, 2007). At the end of neurogenesis, some RPs will divide symmetrically to yield two postmitotic neurons while others will give rise to glial cells (Misson et al., 1991; Noctor et al., 2008). Finally, some of them persist into adulthood and contribute to the adult neural stem cell pool (Gage, 2002; Alvarez-Buylla and Lim, 2004; Nottebohm, 2004; Ming and Song, 2005).

1.3.3 Basal or intermediate progenitor cells:

As the cortex expanded during evolution, this required genesis of an increasingly large number of neurons. As one mechanism for accomplishing this expansion, the cortex contains a neurogenic transit-amplifying cell, the intermediate progenitor cell (also called a basal progenitor because of its location basal to the RPs). In contrast to NESCs and RPs, these intermediate progenitors are restricted to generating neurons, and appear around E11, at the onset of neurogenesis (Iacopetti et al., 1999; Haubensak et al., 2004). Intermediate progenitors are the products of asymmetric divisions of RPs and they reside basal to the ventricular zone, in what emerges as the subventricular zone during neurogenesis (Arnold et al., 2008; Sessa et al.,

2008). Intermediate progenitors do not extend processes to either the apical or basal surfaces of the embryonic cortex, and they express different genes from RPs, such as the non-coding RNA svet1 (Tarabykin et al., 2001), the transcription factors Cux1 and Cux2 (Conti et al., 2005), the t-box transcription factor Tbr2 (Englund et al., 2005), which is required for the transition from RPS to intermediate progenitors (Arnold et al., 2008), the vesicular glutamate transporter (Schuurmans et al., 2004) and at some stages, higher levels of the bHLH transcription factor Ngn2 (Miyata et al., 2004). Intermediate progenitors do not persist past neurogenesis and they contribute to neurons of all cortical layers, generating from 70% (Vasistha et al., 2015) to 80% (Kowalczyk et al., 2009) of all dorsal glutamatergic neurons in mice. Consistent with their role as transit-amplifying cells, murine intermediate progenitors have limited self-renewal capacity and undergo at most 2 rounds of division (Englund et al., 2005; Cappello et al., 2006). Intermediate progenitors are multipolar (Hansen et al., 2010; Pilz et al., 2013) and seem to follow the distribution of capillaries invading the cortex, suggesting that the vasculature in the developing cortex might comprise their niche (Javaherian and Kriegstein 2009; Stubbs et al., 2009).

1.3.4 Outer radial glial precursor cells:

When comparing the smooth murine lissencephalic cortex and the folded human gyrencephalic cortex, it is apparent that the two structures are anatomically quite different but that the basic laminar organization and the basic neuronal cell types are highly conserved. It is thought that this cortical folding or gyrification arose as an adaptation to increase the overall number of neurons, and thus potential computational output, without massively increasing the overall cranial size (LaMonica et al., 2012). This process was apparently accomplished by increasing the pool of precursor cells that can generate neurons. Thus, there are many more intermediate progenitor cells in the human cortex than in the murine cortex, thereby allowing a single RP to generate many more neurons via this transit-amplification step (Lui et al., 2011). A second example of an amplification strategy involves the recently described outer radial glia (oRGs). These are cells that have the same characteristics as RPs, but that reside in a region of the human cortex outside of the SVZ (the so-called outer SVZ or OSVZ) (Smart et al., 2002; Zecevic et al., 2005; Fish et al., 2008). In humans there are many of these oRGs cells, while only few of them exist in mice. Thymidine labelling experiments have shown that these oRGs generate neurons (Lukaszewicz et al., 2005), and that they express markers of both the RP and

9 intermediate progenitor phenotypes (Zecevic et al., 2005; Bayatti et al., 2008; Mo and Zecevic, 2008). Interestingly, they do not possess an apical fiber but only a basal one and they seem to migrate basally via mitotic somal translocation (Fietz et al., 2010; Hansen et al., 2010). These cells undergo several rounds of asymmetric cell division to generate intermediate progenitors. It is thought that, in humans, these oRGs and the increased number of intermediate progenitors, both of which are located basally, provide mechanisms for increasing the number of neurons generated per apical unit of the embryonic cortex (Shitamukai et al., 2011; Wang et al., 2011).

1.3.5 Oligodendrocyte precursor cells:

Gliogenesis, which occurs largely postnatally in the cortex, is not the focus of this thesis. Nonetheless, it is important to note that most of the oligodendrocytes of the mature cortex are derived from the same developing RPs that generate neurons and astrocytes. Intriguingly, during embryogenesis, the cortex is populated by newborn oligodendrocytes that migrate in from the adjacent ganglionic eminences at approximately the same time as cortical interneurons migrate in from these regions (Spassky et al., 2001; Tekki-Kessaris et al., 2001). However, these ventrally derived oligodendrocytes then die during later embryogenesis and are replaced by cortically derived oligodendrocytes (Kessaris et al., 2006). RPs are thought to generate these oligodendrocytes through a more committed glial precursor cell, called an oligodendrocyte precursor cell or OPC (Tripathi et al., 2011). OPCs persist throughout life in the cortex in both the gray and white matter (Dawson et al., 2003).

1.4 Mechanisms regulating cortical neural precursor self-renewal versus differentiation:

This thesis focuses upon the precursors of the embryonic murine cortex during the period of neurogenesis, at which time the RPs are undergoing both symmetric divisions to renew themselves and asymmetric divisions to generate newborn neurons and neurogenic intermediate progenitor cells. What then, are the mechanisms that determine this balance between self-renewal and differentiation? A large body of evidence indicates that the decision for an RP to self-renew versus differentiate is the result of an interplay between intrinsic developmental programs and environmental signals such as diffusible growth factors and cell- cell contact. I will not cover this body of literature in my thesis because of space constraints but instead refer the reader to the many excellent reviews on this topic (Götz and Huttner, 2005; 10

Miller and Gauthier, 2007; Rowitch and Kriegstein, 2010; Cremisi, 2013; Ninkovic and Götz, 2013; Guerout et al., 2014; Hippenmeyer, 2014; Paridaen and Huttner 2014; Tuoc et al., 2014; Taverna et al., 2014). Instead, I will focus this section on evidence suggesting that the polarity of neural precursor cell divisions might play an important role in regulating the decision to self- renew or differentiate, and how this work has led to the idea that post-transcriptional mechanisms might play a previously-uncharacterized role in cortical development.

1.4.1 Mitotic polarity as a potential regulator of neural precursor cell fate:

During cortical development, radial precursors shift from a symmetric and proliferative mode of division at the start of neurogenesis to an asymmetric mode of division that is predominantly pro-neurogenic a few days later (Kriegstein and Alvarez-Buylla, 2009). How does this shift occur? One attractive model system where a similar shift occurs involves D. melanogaster neuroblasts, which are thought to be the functional equivalent of radial precursors (Wang and Chia, 2005; Chia et al., 2008; Zhong and Chia, 2008). In this system, neuroblasts delaminate from the neuroepithelium and generate neurons by asymmetric division. With each division, the mother neuroblast generates two cells: a differentiating ganglion mother cell, which, like a cortical intermediate progenitor cell, then divides to generate two neurons, and a self-renewing neuroblast, which divides again in this fashion several more times. Intriguingly, this asymmetric division is accomplished by basal segregation of prodifferentiation factors and apical segregation of self-renewal factors (Doe and Skeath, 1996). During mitosis, orientation of the mitotic spindle along the apical-basal axis ensures inheritance of the basal factors to the differentiating cell and inheritance of the apical factors to the self-renewing neuroblast. Experiments where the segregation of these determinants was altered or the position of the mitotic spindle was randomized resulted in cell fate alterations (Neumuller and Knoblich, 2009). Therefore, the key feature of this model is that symmetric distribution of cell fate determinants corresponds to daughter cells of equal fate, whereas asymmetric distribution gives rise to different fates. However, while this is a very attractive model, evidence indicates that in the murine cortex the process is not as straightforward. In particular, although NESCs and radial precursors are clearly polarized along their apical-basal axis, the mitotic spindle is usually perpendicular to this axis (Smart, 1973; Zamenhof, 1985; Miyata, 2007; Noctor et al., 2008). Therefore, divisions with seemingly similar segregation of apical and basal components still result in the adoption of

11 different fates. Nonetheless a number of lines of evidence, reviewed in (Lesage et al., 2010; Culurgioni and Mapelli, 2013; Lancaster and Knoblich, 2013) suggest that mitotic spindle orientation does indeed play a role in fate determination in the embryonic cortex. For example, mutations in mitotic spindle regulators, such as LGN (Konno et al., 2008) and Inscutable (Postiglione et al., 2011), as well as the Rho GTPase regulator Lfc (Gauthier-Fisher et al., 2009), alter the proportion of cells being generated. How then can we reconcile these different observations? Several explanations have been suggested to account for this apparent discrepancy. Perhaps small and unappreciated variations in spindle orientation suffice to asymmetrically segregate cell fate determinants and thus generate different fates of the daughter cells (Kosodo et al., 2004; Marthiens and ffrench-Constant, 2009). Alternatively it is possible that different portions of the apical or basal side promote different fates, or maybe the amount of time during which a RP is exposed to pro-neurogenic or pro-self-renewal cues determines whether it will undergo a proliferative or neurogenic division (Calegari and Huttner, 2003; Calegari et al., 2005). Finally, perhaps post-transcriptional mechanisms are important. For example, translationally repressed mRNAs encoding differentiation proteins could be segregated equally to both daughter cells, but one daughter might encounter environmental cues that maintain repression while the other could encounter cues that cause rapid mRNA translation and subsequent cellular differentiation.

1.4.2 RNA-binding proteins as asymmetrically-segregated determinants in stem cell biology:

While all of the aforementioned possibilities might be valid in certain circumstances, perhaps the most important lesson from the D. melanogaster system is that mechanisms of post- transcriptional regulation are very important for cell fate determination. Indeed, while post- transcriptional regulation has been shown to be crucial in stem cells of model organisms like C. elegans (Nousch and Eckmann, 2013) and D. melanogaster (Lasko, 2012), very little is known about the role of RNA-binding proteins, and translational regulators especially, in the context of mammalian stem cell development or mammalian cortical development. And yet, translational regulators are a very attractive class of proteins because they could provide several advantages to a developing radial precursor, especially in the context of a quickly changing niche such as the developing mammalian cortex. For example, post-transcriptional regulation would allow a radial precursor to transcribe and translationally repress mRNAs encoding diverse cell fate

12 determinants, such as those regulating neurogenesis, thereby priming the cell for differentiation while at the same time maintaining it in an undifferentiated, proliferative state. As discussed above, this "primed" state could provide a mechanism for asymmetric precursor divisions because even though both daughter cells would be inheriting these mRNAs, only the daughter cell receiving pro-neurogenic environmental signals would derepress these complexes, leading to translation of neurogenic fate determinants and differentiation. Moreover, this might provide a platform for RP heterogeneity, since these repressed mRNAs could be diversified in different RPs, thereby allowing rapid changes in the cell types that are generated over relatively short developmental timeframes. The fact that this concept has not been previously explored in the developing mammalian cortex is somewhat surprising, especially considering that there are many examples of RNA-binding proteins regulating the biology of postmitotic mammalian neurons. In one example, ribonucleoprotein particles (RNPs) comprised of repressed mRNAs and RNA binding proteins are known to localize at dendritic spines and mature synapses. Several different RNPs can be present for any given synapse and they are known to respond specifically to different kinds of synaptic stimulation (Doyle and Kiebler, 2011; Fernandez-Moya et al., 2014; Thomas et al., 2014). Upon stimulation, the mRNA contained in these RNPs is translated and the new proteins can then modulate or alter the properties of the synapse, ultimately causing permanent changes in neural circuitry (Matsuzaki et al., 2004; Holt and Schuman, 2013). As a second example, RNPs are known to be localized at the tips of the neuronal growth cones. These growth cones encounter environmental cues that promote or inhibit growth, in part by regulating local translation of the mRNAs associated with these RNPs (Jung et al., 2014). Of course though, one may argue that these mechanisms are essential for neurons because these are large, polarized cells that extend over long distances, making it impractical for them to rely only on de novo transcription whenever a quick change is required. Instead, cell types like radial precursors may not have the need for this kind of regulation. Is there a precedent, then, for RNA-binding proteins regulating mammalian stem cell development? One example of an RNA binding protein playing a key role in mammalian stem cell biology involves Musashi, a protein initially characterized in D. melanogaster (Okabe et al., 1997) but also conserved in mice and humans. There are two Musashi homologues in mammals, Musashi1 and Musashi2 (Msi1 and Msi2) and they play important roles in neural precursor and stem cell biology (Sakakibara et al., 2001, Okano et al., 2005). Musashi1 is expressed at high levels in radial precursors and adult neural stem cells, where it represses Numb, a critical 13 negative regulator of Notch function, and in so doing promotes radial precursor maintenance (Okano et al., 2005). How does Musashi do this? It can recognize specific sequences in the 3’ untranslated region (UTR) of its target mRNAs and it then interacts with PABP to compete for eIF4G binding, thus repressing translation (Imai et al., 2001). Phosphorylation of Musashi1 abrogates its binding to its target mRNAs and translation then occurs (MacNicol et al., 2015). Appearance of Musashi1 phosphorylation correlates with conditions inducing differentiation or cell cycle arrest. Additional Musashi1 targets include p21, an important promoter of cell cycle arrest (Battelli et al., 2006) and doublecortin, an early marker of neuronal differentiation (Horisawa et al., 2009). In neural precursors it appears that Musashi1 and 2 are somewhat redundant because double but not single Musashi1 knockouts reduce self-renewal (Sakakibara et al., 2002). Another RNA binding protein that functions in mammalian stem cells is Pumilio, which is conserved from yeast to humans (Datla et al., 2014). Pumilio family members are structurally conserved in different species, both with respect to their RNA-binding recognition motif and their function (Miller et al., 2011). Pumilio has been characterized in C. elegans (Kimble et al., 2007) and D. melanogaster (Joly et al., 2013) as a factor important for the maintenance of germline stem cells and this role seems to be conserved in mammals, which have two homologues, Pumilio1 and Pumilio2 (Lai and King, 2013). Pumilio2 is highly expressed in the mouse germline where it interacts with other proteins important for proper germline development. By analogy to other organisms, mammalian Pumilio2 is thought to translationally repress mRNAs in the germline. Support for this idea comes from evidence showing that Pumilio2 localizes to the chromatoid body, a centre of RNA processing in germline stem cells, together with Nanos1, which is also a conserved binding partner in other model organisms (Spik et al., 2006; Ginter-Matuszewska et al., 2009; 2011). A role for Pumilio1 in mammalian stem cells is just beginning to be uncovered but during oocyte maturation it seems that Pumilio1 binds and represses cyclinB1 messenger RNA (Kotani et al., 2013). Furthermore, Pumilio1 loss prevents implantation of early mouse embryos during development, suggesting a crucial role early on in development (Zhang et al., 2015). Pumilio1 and Pumilio2 also seem to be highly expressed in hematopoietic stem cells, hence they may also play a role in their maintenance (Spassov and Jurecic, 2003). Together, these findings suggest that post-transcriptional mechanisms involving RNA binding proteins might be as important for mammalian RPs as they are for D. melanogaster neuroblasts. Indirect evidence that this might be the case comes from observations that radial 14 precursors may express mRNAs important for neuronal differentiation at times when they do not express the protein (Edri et al., 2015). In this thesis, I have tested the hypothesis that RNA- binding proteins and post-transcriptional regulation are key determinants of cortical neurogenesis, focusing on two RNA-binding proteins, Staufen2, a protein known to be important for mRNA localization, and Smaug2, a protein that is mainly involved in mRNA translational repression and degradation.

1.4.2.1 The RNA-binding protein Staufen2:

Staufen is a protein which binds double stranded RNA and its function is crucial at multiple developmental stages. In D. melanogaster, Staufen function has been characterized during early oocyte patterning and at later stages during nervous system development. During oocyte formation, Staufen is required for the localization, anchoring and translation of patterning transcripts bicoid and oskar. bicoid mRNA is localized and translated at the anterior of the oocyte; there, high levels of Bicoid protein are required for the development of anterior structures (Lipshitz and Smibert, 2000). oskar mRNA instead is transported to and anchored at the posterior pole, where Oskar protein is then first detected. Oskar protein is necessary and sufficient to regulate the formation of the polar plasm, whose components are required for the development of abdominal segments and primordial germline stem cells (Lipshitz and Smibert, 2000). nanos mRNA is one of the factors translated at the posterior pole in response to Oskar protein accumulation (Zaessinger et al., 2006). During nervous system development, Staufen regulates the localization of equally important, albeit different, transcripts in D. melanogaster neuroblasts, which, as discussed above, are functionally-similar to mammalian RPs. Neuroblasts undergo asymmetric division by segregating differentiation complexes basally and self-renewal complexes apically (Schober et al., 1999). Orienting the mitotic spindle along this apical-basal axis then ensures that the apical complex is inherited by the self-renewing neuroblast and the basal complex is inherited by the neurogenic ganglionic mother cell. Staufen plays a critical role in this process because it binds directly to prospero mRNA (Li et al., 1997; Wodarz et al., 1999), which encodes a transcription factor with well established roles in cell-cycle exit and differentiation (Knoblich, 1997; Stergiopoulos et al., 2015). Since Staufen is part of the basal complex, it segregates prospero mRNA to the differentiating daughter cell.

However, the importance of Staufen is not restricted to this developmental function because a role for Staufen in the mature D. melanogaster nervous system as a regulator of synaptic plasticity has also been described (Dubnau et al., 2003; Timmerman et al., 2013). Unsurprisingly, Staufen is conserved across metazoans; in mice and humans there are two orthologs, Staufen 1 and 2, and multiple isoforms of each exist. Just like D. melanogaster Staufen, they perform important functions in developing and in mature tissues. Staufen1 is ubiquitously expressed, while Staufen2 is predominantly expressed in the nervous system (Duchaîne et al., 2002). The function of Staufen1 and 2 during early development is not as well- characterized as it is in the more mature nervous system but one report showed that Staufen1 knockdown impairs early embryonic stem cell differentiation in the preimplantation mouse embryo (Gautrey et al., 2008). A more recent report instead showed that Staufen1 may be important in neuronal differentiation by regulating several microRNAs (miRNAs) (Peredo et al., 2014) and in skeletal muscle, Staufen1 appears to regulate myogenic differentiation by controlling c-Myc translation (Ravel-Chapuis et al., 2014). Therefore, Staufen plays important developmental roles both in D. melanogaster and mammals. Staufen function in the mammalian mature nervous system was initially characterized in rat hippocampal neurons; particles containing Staufen and mRNAs were assembled and transported to neuronal dendrites in a microtubule-dependent fashion (Köhrmann et al., 1999). It was suggested that these ribonucleoprotein particles (RNPs) were repressive in nature because they were devoid of translation initiation factors, but upon synaptic stimuli such as depolarization, RNPs at synapses quickly changed structure to permit association of their mRNA with polysomes and de novo local protein synthesis (Krichevsky and Kosik, 2001; Ferrari et al., 2007). Transport of Staufen RNPs to dendrites relies on molecular motors, such as kinesin, which have been shown to co-immunoprecipitate with these RNPs (Kanai et al., 2004). Many of these important early studies do not distinguish between Staufen1 and Staufen2 and/or the different Staufen isoforms, but the availability of better reagents has allowed this distinction to be made in more recent studies. For example, in rodent mature hippocampal dendrites, Staufen1 and Staufen2 are both present in similar distributions, but they largely do not colocalize, suggesting that they form different RNPs (Duchaîne TF et al., 2002). Similar complexes with human Staufen were also observed in differentiated SHSY5Y neuroblast cells (Villacé et al., 2004). Functional studies at mature synapses indicate that Staufen is important for neuronal function and connectivity. In particular, Staufen2 knockdown impairs long-term depression 16

(Lebeau et al., 2011a) while Staufen1 knockdown alters long-term potentiation (Lebeau et al., 2011b), which are both forms of synaptic plasticity. During dendritic spine morphogenesis, a process required for the formation of new synapses, knockdown of either protein affects formation and maturation of new spines. Hence, some functions seemingly overlap while others differ. Staufen1 knockout mice do not show overt phenotypes with respect to learning, memory or fear and anxiety but they display locomotion deficits compared to wildtype (Vessey et al., 2008). While in this system Staufen2 was not upregulated in mice without Staufen1, compensation cannot be excluded a priori because in other instances they have been shown to heterodimerize (Park and Maquat, 2013). Staufen2 knockout mice have not yet been generated but they would be informative to test whether Staufen2 ablation in the brain is more critical than Staufen1. Further characterization of the composition of Staufen2 RNPs in rodent neurons indicates that they contain many additional RBPs involved in mRNA repression, including FMRP, Purα, DDX6, Pumilio2 and members of the RISC complex (Fritzsche et al., 2013). Staufen1 has also been observed in similar RNPs (Kanai et al., 2004) and it is interesting to note that while heterogeneity has been observed with respect to the protein composition of these RNPs, some components are shared (Fritzsche et al., 2013). The presence of so many translational repressors and the absence of translation initiation factors provide further evidence that Staufen-containing RNPs are repressive granules. Finally, Staufen2 granules contain the nuclear cap-binding protein 80, which is normally exchanged with eIF4E after the pioneering round of translation (Fritzsche et al., 2013). This would imply that these mRNAs have undergone no translation at all and that they are translationally stalled. While all of these observations suggest that Staufen mediates its effects on mRNA translation via other protein partners, recent evidence suggests that the Staufen proteins might themselves be direct translational regulators. For instance, a recent study showed that in HEK293T cells Staufen1 binds to parts of the coding sequence (CDS) of poorly translated mRNAs and that Staufen1 knockdown increases their translation (Sugimoto et al., 2015). In the same study, however, transcripts with Staufen1 bound to the 3’ UTR were stabilized upon Staufen1 knockdown, but their translation efficiency did not change, which is consistent with work indicating that Staufen1 binding to the 3’UTR promotes decay (Kim et al., 2005). Interestingly, Staufen2 has also been reported to be able to promote mRNA decay in human HEK293T cells (Park et al., 2013). However, in rodent neurons, Staufen2 knockdown caused downregulation of its mRNA targets suggesting that in this system Staufen2 promotes mRNA 17 stability (Heraud-Farlow et al., 2013). Thus, Staufen proteins may trigger different outcomes depending upon their cellular context. One open question in the field has been how Staufen recognizes its targets in vivo. A recent study has provided insights into this issue (Laver et al., 2013). In this study, RIP-Chip experiments for endogenous and GFP-tagged D. melanogaster Staufen allowed the identification of unique features of Staufen mRNA binders, such as a particularly long 3’UTR (3-4 fold longer compared to non-binders) and three specific types of secondary structures which are also enriched in the 3’UTRs of Staufen targets (Laver et al., 2013). This binding preference appears to be conserved between D. melanogaster and mammals, because Staufen2 mRNA targets in rodent neurons were also enriched for these structures (Heraud-Farlow et al., 2013). Hopefully the characterization of the Staufen binding preference will aid in the complete identification of Staufen1 and Staufen2 mRNA targets in vivo, since this would help unravel the different phenotypes observed upon their knockdown. To date, a study in HEK293T cells using overexpressed Staufen1 and Staufen2 isoforms indicated that the mRNAs they bind largely do not overlap (Furic et al., 2008), but of course these findings have to be taken with caution because the proteins were overexpressed. In summary, all these studies indicate that Staufen plays many important roles in the developing and mature nervous system with respect to RNA localization, stability and translation. However, a role for Staufen in mammalian stem cells was not known when I started the work described in this thesis, and, based upon the D. melanogaster literature, it provided an attractive candidate for potential asymmetric segregation of cell fate determinants in RPs during neurogenesis. The third chapter of my thesis describes the work that I did to address this possibility. Importantly, at the time I was doing this work, Sally Temple's laboratory was pursuing related studies (Kusek et al. 2012), and we co-published the resultant papers at the same time. I will discuss the complementary relationship between these two studies in my final thesis chapter.

1.4.2.2 The post-transcriptional regulator Smaug2:

Establishment of the D. melanogaster body plan during embryonic development relies on mRNA localization and localized translation of body patterning transcripts (Lasko, 2009) but an extensive summary of all the known examples and mRNA transcripts involved in these

18 developmental processes is beyond the scope of this thesis. Instead, I refer the reader to these excellent reviews for more information (Becalska and Gavis, 2009; Lasko, 2009, 2011; Lai and King, 2013). Suffice it to say, however, that it is in this context that D. melanogaster Smaug was initially characterized as the translational repressor of a transcript called nanos, whose tight translational control is paramount for correct embryonic development (Lasko, 2009). nanos mRNA encodes an RNA-binding protein that serves several important functions. First, Nanos is required for the correct specification and development of D. melanogaster primordial germline cells (PGCs) (Kobayashi et al., 1996; Forbes and Lehmann, 1998). Second, Nanos represses anterior patterning genes in the posterior of the D. melanogaster embryo, thus allowing for the correct development of the abdomen and posterior structures (Dahanukar and Wharton, 1996; Smibert et al., 1996, 1999). Smaug was initially characterized as one of the factors regulating the balance of nanos translation because it repressed nanos mRNA in the bulk of the embryo. nanos mRNA at the posterior pole, instead, was somehow protected from Smaug and was translated, thus establishing a Nanos protein gradient (Wang and Lehmann, 1991; Wharton and Struhl, 1991; Gavis and Lehmann, 1994; Dahanukar and Wharton, 1996; Smibert et al., 1996). A second Smaug target, Hsp83 mRNA, was subsequently discovered and, unlike nanos mRNA, Smaug did not affect its translation but instead promoted its degradation (Semotok et al., 2005). Over time it became clear that not only was Smaug a key protein in D. melanogaster PGCs development and posterior patterning but that it also played a crucial role earlier, during the D. melanogaster maternal to zygotic transition (MZT) (Tadros and Lipshitz, 2009; Walser and Lipshitz, 2011). During early metazoan development, the embryonic genome is transcriptionally silent and early development is controlled by maternally encoded mRNAs that are deposited in the oocyte before fertilization (Langley et al., 2014). During the MZT, the zygotic genome is finally actively transcribed and takes control of all subsequent development. In the D. melanogaster embryo, the MZT marks the transition from a syncytial, multinucleated embryo to a multicellular one and it is the first process entirely controlled by the zygotic genome (Tadros and Lipshitz, 2009; Walser and Lipshitz, 2011). Smaug mutant embryos are lethal at this stage due to several aberrations in development, including failed cellularization, failed DNA replication checkpoint activation, failed zygotic transcription activation and altered cell cycle timing (Benoit et al., 2009). Why does this happen? Studies in the mature D. melanogaster egg indicate that Smaug is responsible for the decay of approximately 2/3 of unstable maternal transcripts and that many of these transcripts encode genes important for cell cycle, cell proliferation and DNA replication/chromosome cycle (Tadros et al., 2007). A study 19 on the modes of Smaug repression indicated that 71% of the mRNAs that are degraded by Smaug are also translationally repressed and that 46% of the genes that are translationally repressed are also degraded (Chen et al. 2014). However, although many Smaug interactors have been identified, it is not known what characteristics of a given transcript promote Smaug- dependent repression or degradation, or both. How does Smaug repress or degrade mRNAs? A large body of work has shown that Smaug regulates its mRNA targets in several different ways after first binding to the relevant target mRNA consensus sequence called a Smaug Recognition Element or SRE. SREs are stem loop structures with the consensus loop sequence CNGGN(0-3), where N can be any nucleotide. The sequence of the stem is not important as long as Watson-Crick pairing is present (Aviv et al., 2003). SREs can be located in any part of a mRNA transcript and still be functional. In D. melanogaster Smaug, the RNA-binding domain (RBD) comprises amino acids (aa) 596 to 764 and has the shape of a muscular bent arm (Green et al., 2003). The RBD can be divided in two distinct domains, set at approximately 90 degree angles, both of which are entirely composed of α-helices. The former is the Sterile-α-motif containing (SAM) domain and the latter is the pseudo heat analogous topology (PHAT) domain. Unlike the SAM domain, which is evolutionarily conserved in other species, the PHAT domain is not. To determine conclusively whether the SAM domain is solely responsible for RNA binding, Vts1p, which is the yeast, PHAT-less homolog of D. melanogaster Smaug, was shown to be able to bind reporter RNAs containing the nanos SRE (Aviv et al., 2003). Interestingly, homology searches in other species showed that the SAM domain is highly conserved across species and that the Smaug mRNA binding consensus sequence is, if not identical, very similar (Aviv et al., 2003; Green et al., 2003). The SAM domain thus establishes a family of conserved post-transcriptional regulators comprising 11 members. Eight of these 11 members not only share homologous SAM domains, but they also share a high degree of homology in the two regions of unknown function at the N-terminus, called Smaug Similarity Regions (SSR) 1 and 2. How then does Smaug prevent translation of its target mRNAs once it binds to the SRE? First, Smaug can recruit the co-repressor Cup which is able to interact with eIF4E. This interaction can then compete with eIF4G for binding to eIF4E, ultimately preventing the formation of the 43S pre-initiation complex and inhibiting translation (Nelson et al., 2004). Second, Smaug is also able to recruit the CCR4-NOT deadenylase complex, one of the main conserved complexes involved in mRNA degradation (Semotok et al., 2005; Semotok et al., 2008). Specifically, Smaug is able to interact with one of the subunits, the deadenylase Caf1 and 20 this association is RNA-independent (Zaessinger et al., 2006). In most cases, shortening of the poly-A tail is the first step in directing mRNA towards the RNA degradation pathway. Poly-A tail shortening is usually followed by mRNA decapping and then degradation of the mRNA (Parker, 2012). Genetic studies where CCR4-NOT function was reduced or absent showed that transcripts that were normally degraded by Smaug were stabilized and ectopic protein expression occurred (Semotok et al., 2005; Zaessinger et al., 2006; Semotok et al., 2008). Other mechanisms that Smaug employs to repress its target nanos mRNA involve small RNAs and members of the Argonaute family. In a study by Rouget et al., 2010, they showed that Smaug and subunits of the CCR4-NOT partially colocalized with Aubergine and Argonaute3 in cellular foci and that Aubergine co-immunoprecipitated with Smaug, CCR4 and Argonaute3 in the absence of RNA. Moreover, they showed that this interaction was important for nanos repression. Argonaute 3, however, is not the only Argonaute member involved in nanos mRNA repression because a recent study showed that Smaug also associates with Argonaute1 to repress nanos mRNA in the bulk of the embryo. Intriguingly, miRNAs are not required for this interaction to occur, making it distinct from the usual mode of Argonaute1 action (Pinder and Smibert, 2013a; Pinder and Smibert, 2013b) (Figure 3).

In summary, Smaug is a critical mRNA regulator during D. melanogaster development. Smaug binds to target mRNAs via the SRE, and this leads to mRNA repression and/or degradation via mechanisms that are still being studied. Therefore, I chose to ask whether Smaug is equally important in mammalian cortical RPs, based upon the important role that it plays in this model developmental system and my interest in post-transcriptional regulation in mammalian neural precursor cells. So what is known about mammalian Smaug?

1.4.2.3 The function of mammalian Smaug:

In mammals there are two homologues of Smaug, Smaug1 and Smaug2 (Baez et al., 2005) (Figure 4).

Neither of these two proteins has been very well studied, although there is somewhat more known about Smaug1. Studies in rat hippocampal neurons have shown that, like other RNA-binding proteins such as Staufen or FMRP, Smaug1 is found in neuronal cell bodies and dendrites (Baez et al., 2011). However, Smaug1 foci are apparently distinct from other characterized RNP granules containing Staufen or FMRP. Smaug1 foci are also apparently different from P-bodies, since knockdown of P-body components disrupted P-bodies in neurons, but did not alter the foci that contained Smaug. Approximately 50% to 70% of hippocampal synapses have Smaug1 foci, although they are not the only kind of RNPs present at any given synapse (Baez et al., 2011). Interestingly enough, Smaug1 foci and FMRP foci respond to different synaptic stimuli. Smaug1 foci dissolve following activation of NMDA receptors and FMRP foci dissolve following activation of AMPA receptors. Dissolution and re-assembly dynamics of Smaug1-foci are dependent upon polysome function. For instance, polysome stabilization with cycloheximide

23 or emetine caused Smaug1-foci dissolution, but impairment of polysome assembly by treatment with either of two drugs (puromycin or hippuristanol) abrogated Smaug1-foci dissolution. Smaug1-foci dissolution, however, was not affected by inhibition of the proteasome and Smaug1-foci reassembly did not require novel Smaug1 synthesis, suggesting that Smaug1-foci dynamics rely on existent Smaug1 (Baez et al., 2011) (Figure 5).

What are the mRNA targets of Smaug in mammals? While this is largely an unanswered question, one Smaug1 in vivo target that was identified in the above study is camkIIα mRNA, which encodes the α-subunit of calcium/calmodulin-dependent protein kinase type II. Smaug1 immunofluorescence and fluorescence in situ hybridization for camkIIα mRNA indicated that approximately half of camkIIα mRNA in the dendritic shaft and synapses co-localized with Smaug1. Upon NMDA stimulation, it was observed that co-localization was significantly reduced both in synapses and along the shaft, while the proportion of synapses associating with camkIIα mRNA was significantly increased and Smaug1-foci dissolved. Although it has to be shown directly that camkIIα mRNA at the synapse was actively translated, significant increases in CamkIIα protein foci and foci intensity were observed by immunofluorescence at synapses, which would support this interpretation (Baez et al., 2011). Do alterations in Smaug1 function affect hippocampal neurons? This appears to be the case since Smaug1 regulates hippocampal synaptogenesis. Smaug1 knockdown at the onset of this process caused a large reduction in synaptic size but increased synapse number. Furthermore, these synapses displayed an immature morphology and impaired function. While more work is needed to determine whether increased translation of camkIIα mRNA is responsible for the observed phenotypes and whether camkIIα mRNA repression is SRE- dependent, two studies in cell lines have shown that overexpressed Smaug1 can repress translation of reporter mRNA containing SREs sites in a dose dependent manner, without degrading the mRNA itself (Baez et al., 2005; Fernandez-Alvarez et al., 2016). Additionally, a mutation in the N-terminus of Smaug1 has been reported to induce leanness in mice because of mTor pathway deregulation (Chen et al., 2014b). Immunoprecipitation studies, however, indicated that Smaug1 did not associate with any member of the mTor pathway itself, suggesting that this phenotype might be linked in some as-yet-undefined way to post- transcriptional regulation (Chen et al., 2014b). Finally, a recent report also indicated that smaug2 mRNA is expressed in cultured rodent hippocampal neurons and its expression is highest during synaptogenesis, similar to what has been described for smaug1 and that expression of Smaug2 in cell lines forms cytosolic bodies (Fernandez-Alvarez et al., 2016). In summary, two Smaug isoforms exist in mammals and although we have just scratched the surface of their function, these studies suggest that they may be involved in translational repression.

1.4.3 A potentially conserved mRNA repression pathway:

Since Smaug is clearly conserved in mammals, this poses the question of whether it is part of a conserved pathway of mRNA repression across metazoans. If this is the case, then one might predict that some Smaug targets, such as nanos mRNA, and other proteins interacting with Smaug, such as Cup, could also be conserved and interacting with mammalian Smaug. In the following section, I will discuss this idea and describe our current knowledge of some of these potential Smaug-interacting factors in mammals, focusing on two of particular relevance to this thesis, Nanos and 4E-T.

1.4.3.1 Nanos:

In mice there are three Nanos homologues, Nanos1, Nanos2 and Nanos3 and, like D. melanogaster Nanos, they play important roles in regulating mammalian primordial germ cells, likely by functioning as repressive RNA-binding proteins. Their function has been mostly characterized in the context of murine germline stem cells, where they play both redundant and non-redundant roles to regulate their maintenance and differentiation. One of the major conclusions from the work on Nanos is that this family of proteins appears to have a conserved role in germline maintenance in organisms as diverse as D. melanogaster, zebrafish, mouse and humans. In mammals, each of the Nanos family members is important within this context. Nanos3, which is perhaps the best-characterized of the three, is expressed in primordial germ cells and its ablation results in their loss (Suzuki et al., 2007). Two lines of evidence indicate that Nanos3 may promote germline maintenance, at least in part, by repressing differentiation. First, ectopic expression of Nanos3 results in accumulation of cells in G1 phase (Lolicato et al. 2008). Second, Nanos3 is downregulated in response to retinoic acid, which is known to promote differentiation. Additionally, Nanos3 can also suppress Bax- dependent and independent apoptosis (Suzuki et al. 2008). Thus, Nanos3 appears to promote primordial germline stem cell maintenance via several different mechanisms. Unlike Nanos3, murine Nanos2 is expressed predominantly in the male germ line and upon its ablation, spermatogonia undergo apoptosis and male infertility occurs (Saga, 2008; Saga, 2010). The female germline instead does not express high levels of Nanos2 and its loss has no effect. Nanos2 is required in males to block meiosis by repression of key mRNA transcripts until the appropriate developmental timepoint and once again, retinoic acid antagonizes its expression (Saga, 2008). Intriguingly, in spite of these differences, Nanos2 can

27 rescue the loss of primordial germ cells that occurs following ablation of Nanos3, suggesting that these two proteins are functionally similar. The final member, Nanos1, is expressed in the adult murine germline, but a constitutive Nanos1 knockout mouse showed no developmental deficits (Haraguchi et al., 2003). However, human mutations in Nanos1 are associated with decreased fertility in males, suggesting that Nanos1 might indeed be important for mammalian germline stem cell maintenance (Kusz- Zamelczyk et al., 2013). Thus, all three mammalian Nanos family members appear to play somewhat overlapping roles in the germline. In this regard, one protein thought to interact with mammalian Nanos within the germline context is Pumilio2 (Murata and Wharton, 1995; Asaoka-Taguchi et al., 1999), which is also important for Nanos functions in the D. melanogaster germline (Kobayashi et al., 1996; Forbes and Lehmann, 1998). Despite their importance in germline stem cells, virtually nothing is known about mammalian Nanos proteins in other tissues, including the brain. However, the importance of nanos mRNA as a target for Smaug in early D. melanogaster development led me to hypothesize that it might be a similarly important target of Smaug in the context of murine cortical development. I have tested this hypothesis in Chapter 4.

1.4.3.2 4E-T:

In D. melanogaster, Smaug binds to and represses its target mRNAs as part of a complex with the Cup protein and eIF4E, as discussed above. However, Cup does not apparently have a mammalian homologue, implying that if Smaug forms repressive complexes that involve eIF4E in mammals, then it must do so in association with some other eIF4E-binding protein. In this regard, there is a mammalian eIF4E-binding protein called 4E transporter (4E-T), which shares a region of homology with D. melanogaster Cup (Nelson et al., 2004), and this protein is also involved in mRNA repression (Ferraiuolo et al., 2005). 4E-T binds to eIF4E to promote its import into the nucleus, where it is retained by 4E- BP1 (Rong et al., 2008). 4E-T also binds to the same site on eIF4E as eIF4G, and thus prevents the interaction between these two proteins and in so doing inhibits formation of the translation initiation complex (Dostie et al., 2000). Intriguingly, 4E-T associates with and is important for the formation of P-bodies, which are large cytoplasmic RNPs that are thought to be important for mRNA repression and decay (Ferraiuolo et al., 2005). Although 4E-T is not thought to be a direct mRNA binder, mRNA tethering assays have shown that it possesses intrinsic translational

28 repression activity (Kamenska et al., 2014). How does it do that? A recent report suggests that 4E-T, just like Smaug, has the ability to interact with the CCR4-NOT deadenylase complex and that it may do so together with the repressive RNA-helicase DDX6 (Ozgur et al., 2015). Furthermore, the interaction between 4E-T and CCR4-NOT enhances mRNA degradation (Nishimura et al., 2015), perhaps because 4E-T acts as a bridge between the 5’ and 3’ end of target mRNA and after binding of CCR4-NOT, it can promote the assembly of other degradation factors at the 5’ end, such as the decapping enzyme Dcp1. In agreement with a role for 4E-T in mRNA repression in mammals, our laboratory has recently described a repressive translational complex involving 4E-T and eIF4E that regulates mammalian neurogenesis by sequestering and silencing mRNAs encoding proneurogenic transcription factors (Yang et al., 2014). 4E-T is in this RNP with the P-body components Lsm1 and DDX6 and it binds to mRNAs with known roles in neurogenesis such as neurogenin1, neurogenin2 and neuroD. Upon 4E-T knockdown, radial precursors prematurely differentiated into neurons and ectopic expression of Neurogenin1, Neurogenin2 and NeuroD occurred, suggesting that a 4E-T-dependent RNP regulates neurogenesis by repressing the translation of neurogenic mRNAs. However, as mentioned above, 4E-T does not directly bind target mRNAs, and thus, there must be repressive RNA binding proteins associated with the complexes that were described in Yang et al. (2014). I therefore hypothesized that 4E-T in embryonic RPs functions in an analogous manner to Cup in early D. melanogaster embryos, and that one of the ways that it represses target mRNAs important for neurogenesis is by associating with Smaug. I have directly tested this hypothesis in Chapter 4 of this thesis.

Chapter 2 Experimental Procedures a) Animals. All animal use was approved by the Animal Care Committee of the Hospital for Sick Children in accordance with the Canadian Council of Animal Care policies. CD1 mouse embryos of either sex (Charles River Laboratories) were used for all experiments. b) Primary cultures and cell lines. Primary cortical precursor cultures were prepared as previously described (Gauthier et al., 2007). Briefly, dissociated E12.5 CD1 cortical cells were plated at 250,000 cells/well on glass coverslips precoated with 2% laminin (BD Biosciences) and 1% poly-D-lysine (Sigma) in Neurobasal medium (Invitrogen) supplemented with 2% B27 (Invitrogen), 0.5 mM L-glutamine (Invitrogen) and 40 ng/mL fibroblast growth factor 2 (FGF2) (BD Biosciences). For transfections, 1 μg of DNA in 100 μL of Opti-MEM was mixed with Lipofectamine LTX or to FuGene 6 (Roche) according to manufacturer’s instructions and added upon plating. In co-transfection experiments, an EGFP expression vector was mixed with shRNA or overexpression vectors at a 1:3 ratio; when three plasmids were used, the ratio was 1:1.5:1.5. Cells were immunostained three days later. Clonal analysis was performed as previously described (Gallagher et al., 2013). Briefly, cortical cultures were prepared as described above and then transfected with 1.5 μg of DNA at a 1:1:3 ratio of PCAG-PB-GFP, pCyL43-Pbase, and shRNA or control plasmids. HEK-293T cells were cultured in DMEM (Gibco) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Cells were transfected with Lipofectamine 2000 according to the manufacturer's instructions, and used 16- 24 hours later. c) In utero electroporation. In utero electroporations were performed as previously described (Gauthier et al., 2007; Yang et al., 2014). Briefly, a nuclear EGFP plasmid (pEF-EGFP) was co- electroporated with shRNA or overexpression vectors at a 1:3 ratio, or when two additional plasmids were co-electroporated, at a 1:2:2 ratio for a total final DNA concentration of 4.0 μg/μL. Prior to injection, plasmids were mixed with 0.5% trypan blue. Following injection into the lateral ventricles, the square electroporator CUY21 EDIT (TR Tech, Japan) was used to deliver five 50 ms pulses of 40-50 V with 950 ms intervals per embryo. For all experiments, we

30 analyzed at least 3 embryos per condition, each from an independent experiment with a different mother. d) Plasmids and shRNAs. The following plasmids used in this study have been previously published: pEF-EGFP (Paquin et al., 2005), pSUPERIOR shStaufen2, pSUPERIOR misStaufen2, pEYFP-N1 Staufen2R-EGFP, pEYFP-N1 dnStaufen2, pEYFP-N1 Staufen2-EGFP (all from Goetze et al., 2006), pSUPERIOR shPum2, pSUPERIOR misPum2 (all from Vessey et al., 2006). The prox1 EST used for generating the probes for FISH was obtained from Source BioScience (Nottingham, UK). The clone was sequence verified and located within the pCMV- SPORT6 vector. The Flag-tagged expression constructs for mouse and human Smaug2 and Smaug1 and mouse Nanos1, 2 and 3 and human Nanos1 were obtained from OriGene. Human pT7-EGFP-C1-HsNot7 and human -EGFP-C1-HsNot8 were a gift from Elisa Izaurralde (Braun et al. 2011) and were obtained from AddGene (CNOT7, #37325; CNOT8, # 37367). The pGPH1-GFP-NEO shRNAs and pGPH1-GFP-control were obtained from EZBiolabs (Carmel, IN, USA) and had the following sequences: Smaug2 shRNA-1 5’-GAG GAG AAC ATC ACC AGT TAC T-3’, Smaug2 shRNA-2 5’-GGG CTG GAA TGA GTG TGA ACAT-3’, Nanos1 shRNA-1 5’-GCACATACCATCAAGTATTGCT-3’, DDX1 shRNA 5’-GCA TTT AGT ATT CCC GTT ATC-3’ and the non-targeting sequence of the control shRNA was 5’-GTT CTC CGA ACG TGT CAC GT-3’. The 4E-T shRNA was previously described (Yang et al., 2014) and the sequence was 5'-CCG TTA TAC CAA AGA ACA A-3'. All clones were verified by sequencing. e) Immunostaining and quantitative analysis. Immunocytochemistry on cultured cells was performed as previously described (Vessey et al., 2012). Briefly, cells on glass coverslips were fixed for 15 minutes with 4% buffered paraformaldehyde (PFA), permeabilized with 0.2% NP40 in HBSS, blocked with 5% bovine serum albumin in HBSS (Jackson) for 1 hour and incubated with primary antibodies in HBSS for 2 hours at room temperature. Samples were washed 3x with HBSS, and secondary antibodies (1:1000) diluted in HBSS were added for an additional hour. Coverslips were mounted on glass slides with the ProLong Gold Antifade Reagent with DAPI (Invitrogen) or Hoechst 33258 (Sigma). For immunostaining of cortical sections, embryonic brains were dissected in ice-cold HBSS, fixed in 4% buffered paraformaldehyde at 4 ºC overnight, cryopreserved with 30% sucrose overnight, placed in OCT, and kept at -80 ºC for a few hours to overnight, and cryosectioned coronally with a thickness of 16 µm. Sections were 31 blocked at room temperature with 5% bovine serum albumin (Jackson Immunoresearch) and 0.3% TritonX in PBS, and incubated with primary antibodies in ½ blocking buffer overnight at 4 ºC, followed by incubation with appropriate secondary antibodies (1:1000) in PBS at room temperature for 1 hour. Nuclei were stained with Hoechst 33258 (Sigma). Quantification of immunostained cell cultures and brain sections was performed as previously described (Wang et al., 2010). Briefly, cells grown on glass coverslips were analyzed with a Zeiss Axioplan2 microscope. For cell identity analysis, at least 300 cells from different fields were counted per condition and results from at least three independent experiments were analyzed together. For in utero electroporation, 3-4 anatomically matched sections per brain from at least three embryos from different mothers were imaged with a 20X objective on an Olympus IX81 fluorescence microscope equipped with a Hamamatsu C9100-13 back-thinned EM-CCD camera and Okogawa CSU X1 spinning disk confocal scan head. Images were processed by using the Volocity software (Perkin Elmer). Pax6, Tbr2 and Hoechst staining were used to define the ventricular zone (VZ), subventricular zone (SVZ) and cortical plate (CP). f) Proximity ligation assays (PLA). Immunocytochemistry was performed as above up to primary antibody incubation, which was performed at 4°C overnight. Coverslips were then washed 3 times for 5 minutes with HBSS. Proximity ligation assays were performed as previously described (Gallagher et al., 2015) with a DuoLink in situ Red Starter Kit Mouse/Rabbit (Sigma, DUO92101) following manufacturer’s instructions. Briefly, the coverslips were incubated with the secondary antibodies provided in the kit for 1 hour, followed by ligation reaction for 30 minutes and signal amplification reaction for 1 hour and 40 minutes. All incubation steps were carried out in a humidified chamber at 37 °C. Washes were performed between each step as per manufacturer’s instructions. Following signal amplification and final washes, coverslips were mounted with the DAPI-containing mounting medium provided in the kit. g) Fluorescent in situ hybridization (FISH). The single molecule FISH was performed as previously described (Yang et al., 2014) using the RNAscope kit (Advanced Cell Diagnostics), according to the manufacturer's instructions. Briefly, freshly dissected embryonic brains were fixed overnight at 4 ºC in RNAse-free 4% PFA, cryopreserved overnight in 30% RNAse-free sucrose and placed in OCT at -80 ºC overnight. Brains were cryosectioned coronally at 14 μm. Sections were post-fixed with 4% PFA and washed with ethanol, followed by tissue 32 pretreatment, probe hybridization and signal amplification. Alternatively, E12.5 cortical precursors were cultured for three days before fixation, ethanol pretreatment, probe hybridization and signal amplification. In both cases, positive staining was identified as punctate dots. For simultaneous immunodetection of a protein of interest after FISH, sections or cultures were blocked and incubated with the relevant primary antibody overnight at 4 ºC, followed by 1 hour incubation with the appropriate Alexa secondary antibodies at room temperature before DAPI staining. Z-stacks of confocal images were taken with optical slice thickness of 0.1 μm. 200-800 mRNA granules in 80-300 Z-stacked images from random regions of the VZ/SVZ were used for 4E-T or Smaug2 colocalization quantification. Bright and clear mRNA granules that overlapped with immunostained 4E-T or Smaug2 granules were counted. Cortical precursor cultures were imaged with a 40X objective. Z-stacks of confocal images were taken with optical slice thickness of 0.1 μm and for each culture condition 20 to 30 random fields were imaged. For the current study the following probes were employed: nanos1 (NM_174421.3) cat. #431391, nanos2 (NM_194064.2) cat. # 450691, nanos3 (NM_194059.2) cat. #450721 and smaug2/samd4b (NM_175021.3) cat. #450731.

FISH for prox1 mRNA and subsequent IHC instead were performed according to the method in (Mikl et al., 2011). Sense and anti-sense probes for prox1 were generated using the SP6/T7 Transcription Kit (Roche) from a prox1-EST (accession BI732857.1). After fixation, sections underwent Proteinase K treatment (40 ng/mL) followed by blocking. Localization studies of prox1 mRNA were performed in VZ cells, defining apical versus basal enrichment with respect to the ventricular surface. For all experiments, at least three embryos from at least two separate mothers were analyzed. The number of cells analyzed per brain ranged between 100 and 200 cells. To quantify asymmetric distribution of Staufen2 protein and prox1 RNA in anaphase and telophase cells, the fluorescence intensities of Staufen2 immunoreactivity and prox1 mRNA in situ hybridization signal in prospective daughter cells were acquired on the confocal microscope, together with Hoechst 33258 DNA signal from the same cells (Bultje et al., 2009). ImageJ (NIH) was utilized to analyze confocal images of 30 to 40 dividing cells. Contours of both daughter cells were outlined with the cleavage furrow as the border of the two. Fluorescence intensities of DNA were measured and background was subtracted to obtain a ratio. Absolute equal values of intensity between daughter cells were considered to be 0.

33 h) Western Blot. Cortical tissue homogenates were prepared by dissecting embryonic cortices from CD1 mice at different developmental time-points or from HEK293T cell cultures. Both cortical tissues and HEK293 were rinsed twice in ice-cold phosphate-buffered saline (PBS) and lysed in RIPA lysis buffer supplemented with Complete Protease Inhibitor Tablets (Roche Applied Science) and 1 mM PMSF. 10-20 ug of total protein were loaded for HEK293T cell lysates, while 50ug total protein was loaded for cortical tissue homogenates. Proteins were separated by SDS-PAGE at 130V for 90 minutes, transferred to nitrocellulose membrane at 0.25 A for 90 minutes and blocked with 5% BSA-TBST for 1 hour while rocking at room temperature. Blots were incubated with primary antibodies diluted in 1%BSA-TBST overnight at 4° C and with secondary antibodies diluted in 1%BSA-TBST for 1 hour at room temperature while rocking. All washes were performed with TBST. Blots were developed by Enhanced Chemiluminescence (GE Healthcare). i) Antibodies. The primary antibodies used were mouse anti-4E-T (Abnova, H00056478-B01P, 1:2000), mouse anti-GFP (Invitrogen, GF28R, 1:1000), rabbit anti-GFP (Abcam, AB290, 1:2000), chicken anti-GFP (Abcam, AB13970, 1:2000), mouse anti-Ki67 (BD Pharmingen, 556003, 1:500), mouse anti-Satb2 (Abcam, AB51502, 1:400), mouse anti-βIII-tubulin (Covance, MMS-435P-250, 1:2000), rabbit anti-βIII-tubulin (Covance, PRB-435P, 1:2000), rabbit anti-Pax6 (Covance, PRB-278P, 1:1000), mouse anti-Pax6 (Millipore MAB5552, 1:1000), rabbit anti-Sox2 (Cell Signalling Technology, 3728, 1:1000), goat anti-Sox2 (Santa Cruz Biotechnology, sc17320, 1:500), rabbit anti-Tbr2 (Abcam, AB23345, 1:500), mouse anti- Dcp1 (Abnova, H00055802-M06, 1:2000), rabbit anti-ERK1/2 (Santa Cruz Biotechnology, SC93, 1:10,000), mouse anti-Flag (Origene, TA50011-100, 1:1000), rabbit anti-Nanos1 (Abcam, AB83417, 1:1000), rabbit anti-Smaug2 (Sigma, HPA016800-100UL, 1:1000), rabbit anti-Smaug1 (Sigma, HPA043061, 1:1000), rabbit anti-Staufen2 (Goetze et al., 2006, 1:5000 and Sigma, HPA019155, 1:250), rat anti-Nestin (Millipore, MAB353, 1:200), mouse anti-ZO-1 (Invitrogen, 339100, 1:1000), rabbit anti-DDX1 (Elvira et al., 2006, 1:1000), rabbit anti-Pum2 (Bethyl Laboratories A300-202A, 1:1000), mouse anti-γ-Tubulin (Abcam, AB27074,1:1000), mouse anti-Vimentin (phospho S55) (Abcam, AB22651, 1:1000), rabbit anti-Ki67 (Abcam, AB15580, 1:500, Abcam), rabbit anti-Prox1(1:1000; Reliatech, Germany), rabbit anti-Staufen1 (Abcam, AB50914, 1:500), rabbit anti-PABP (Abcam, AB21060, 1:1000), and anti-mouse Staufen2 (Abcam, AB60724, 1:1000), mouse CNOT7 (Abcam, AB57095, 1:1000) and rabbit CNOT7 (Abcam, AB103717, 1:1000). 34

The Alexa488, Alexa555, and Alexa647-conjugated secondary antibodies were obtained from Invitrogen. Nuclear staining was performed with Hoechst 33258 (Sigma). HRP-conjugated goat anti-mouse or anti-rabbit secondary antibodies (1:10000) were purchased from Boehringer Mannheim, and HRP-conjugated chicken anti-goat secondary antibody was obtained from Millipore (1:5000). j) Antibody Coupling. Cross-linking was performed by incubating the coupled antibodies with 40 mM dimethyl pimelimidate (DMP) for 30 minutes at RT followed by washing with 0.2 M triethanolamine (TEA). Incubation with DMP and washing was repeated 3 times. Unlinked antibodies were eluted with 0.2 M glycine (pH 2.5). k) Co-immunoprecipitations. Freshly dissected E12-13 murine cortices were lysed with Gentle Lysis Buffer (GLB) containing 25 mM Tris-HCl (pH 7.4), 2 mM EDTA, 1 mM EGTA, 10 mM NaCl, 0.5% Triton X-100 and 10% Glycerol supplemented with the Complete Protease Inhibitor Tablets (Roche Applied Science) and 1 mM PMSF. Total protein concentration was determined with the Protein Quantitation kit (Abcam). Lysates were first pre-cleared by incubating with Protein A/G magnetic beads (Millipore) for 1 to 2 hours at 4°C, followed by incubation for 3 hours with rabbit anti-Smaug2 or normal rabbit IgG (Santa Cruz Biotechnology) at 4 °C, and a 1-hour incubation with protein A/G magnetic beads at 4°C. Immunoprecipitates were washed three times with GLB buffer, boiled in 2x sample buffer with 1 mM dithiothreitol (DTT) for 3 minutes, and analyzed with SDS-PAGE as described previously (Yang et al., 2014). Alternatively, freshly dissected E12-13 murine cortices were dissociated and plated as described earlier and maintained in culture for three days, after which immunoprecipitation was performed as described above. The same procedure was followed for 4E-T immunoprecipitation; normal mouse IgG was used as control (Santa Cruz Biotechnology). l) RNA-immunoprecipitation (RIP). Freshly dissected E12-13 cortices were homogenized in Brain Extraction Buffer (BEB) (Vessey et al., 2012) containing 25 mM HEPES (pH 7.3), 150 mM KCl, 8% glycerol and 0.1% NP-40 supplemented with Complete Protease Inhibitor Tablets (Roche Applied Science), 1 mM PMSF and RiboLock RNAse Inhibitor (2 μL/mL, Thermo Scientific). One milligram of protein extract was precleared with 40 μL of 50% protein-A/G sepharose beads (Sigma) for 1 hr at 4 °C. 2 μg of Smaug2 antibody or 4E-T antibody or 2 μg of the relevant non-specific IgG were added to the precleared lysates and incubated at 4 °C with 35 rocking for 2-3 hours. Antibody-bound protein was isolated by adding 40 μL of 50% protein- A/G sepharose beads for 1 hour at 4 °C. Following 3 washes with BEB buffer, mRNA was isolated with Trizol reagent following the instructions of the manufacturer. m) RT-PCR. mRNA from E12-13 cortices or from RIP was isolated using the Trizol reagent (Invitrogen), cDNA was generated using the First Strand cDNA Synthesis Kit (Thermo Scientific) following the manufacturer’s protocol and used for PCR. The annealing temperature of all primers was approximately 60 °C and amplified products were between 100 and 300 nucleotides. All reactions were subjected to 35 cycles and products resolved on 2% agarose gels. Products were verified by sequence analysis. The following primers were used in this study: staufen2 forward (5’-GTG TTT GAG ATT GCG CTG AA-3’), staufen2 reverse (5’-TGC ATT ACG AAC TCT CGA CG-3’), gapdh forward (5’-CGA AGC TAA CGA CTA TCG CC-3’), gapdh reverse (5’-TTG TGA CTT TTT GGC CTT CC-3’), kif5c forward (5’GGC GGA TCC AGC CGA ATG CA-3’), kif5c reverse (5’-TGG GTT GTG TTG GGC GGC AG-3’), arnt forward (5’-AGG TCG GAT GAT GAG CAG AGCTC-3’) arnt reverse (5’-GGA CCC GCC CTG TTA GGG CAT-3’), β-actin forward (5’-GGA GAT GGC CAC TGC CGC AT-3’), β- actin reverse (5’-GCA GCT CAG TAA CAG TCC GCC TA-3’), prox1 forward (5’-TGG CAC CGA GCC CAG TTT CC-3’), prox1 reverse (5’-GTC GGC GCT CCT CTC GCT TC-3’), pum2 forward (5’-AGC TCT CAG TGG ATG TGG CT-3’), pum2 reverse (5’-ACG CCT TTT GAA TAA CAC GG-3’), ddx1 forward (5’-GGG GGA GTA TGC TGT CCG AGC A-3’), ddx1 reverse (5’-TCC CAC TCT GCC GAT CCG GTG-3’), smaug2 forward (5’CGA GGA GAA CAT CAC CAG TTA CC-3’) and smaug2 reverse 5’-CGG AGG AGT TTC AGC ACT TGCT- 3’ (OriGene), smaug1 forward (5’-TGC GCT CTT CTC GCA GAT GAC T-3’) and smaug1 reverse (5’-CCC TTT CCA AAG ACT TCA GGA GG-3’) (OriGene), gapdh forward (5’-GGG TGT GAA CCA CGA GAA ATA-3’) and gapdh reverse (5’-CTG TGG TCA TGA GCC CTT C-3’) (PrimerBank, Spandidos et al., 2010), nanos1 primer1 forward (5’-GTG TGT GTT TTG CCG GAA C-3’) and nanos1 primer1 reverse (5’-CTA GCG CAG CTT CTT GCT G-3’), nanos1 primer2 forward (5’-GGA GCT TCA GGT GTG TGT GTT-3’) and nanos1 primer 2 reverse (5’- CTA GCG CAG CTT CTT GCT G-3’), nanos2 forward (5’-ACC CTG GAT GTC TGC CTA CCA T-3’) and nanos2 reverse (5’-CAC ATA GTG CCT CAG GAT GGG A-3’) (Origene) and nanos3 forward (5’- TCT GCA GGC AAA AAG CTG ACC C-3’) and nanos3 reverse (5’- GGG CTT CCT GCC ACT TTT GGA A-3’) (Origene). For quantitative real-time PCR, 10 μL PCR reaction mixture containing FastStart DNA Master SYBR Green I (Roche 36

Molecular Biochemicals) was prepared according to the manufacturer's instructions, and loaded on to a 96 multiwell plate. The CFX96 thermocycler (Biorad) was used with a protocol involving an initial activation cycle (2 min, 95 °C), 45 cycles of denaturation (10 sec, 95 °C), annealing (20 sec, 60 °C) and elongation (20 sec, 72 °C). A single fluorescence reading was acquired at the end of each elongation step. A melting curve analysis cycle was performed after the PCR amplification. The primers used in RT-qPCR were nanos1 forward 5’-CTA CAC CAC ACA CAT CCT CAA GG-3’ and reverse 5’- GCA CTT TGG AGA GCG GGC AAT A-3’ (OriGene) and nanos2 and nanos3 (OriGene, same as above), ccnl2 forward 5’- GTG AGC GTA ATC AAC ACC TGG TC-3’ and reverse 5’- GCA GCA AGG TAG ATA CAG GCA C- 3’, mark4 forward 5’- CTG CGG AGA TTT CTG GTG CTG A-3’ and reverse 5’-TCC GTG TAT GGC TTC AGC TCC T-3’, xpnpep1 forward 5’-CAA AGC CGT GAA GAA CTC CGC T-3’ and reverse 5’-ATC TCC GTA ACG CCA CCT TTG G-3’, smaug2 was the same as above, huC forward 5’- TGC AGA CAA AGC CAT CAA CAC CC-3’ and reverse 5’- CCA CTG ACA TAC AGG TTG GCA TC-3’, vgll4 forward 5’- CTT GGA CTG TGA AAA CGA CCA CG-3’ and reverse 5’-ATC TCT GCG GCA GTC TCC GTT G-3’, mxd1 forward 5’- GCC TTC AAA CGG AGG AAC AAG C-3’ and reverse 5’- AGA GTG GTG TGT CGG CTT GAC T-3’, hes5 forward 5’-CCG TCA GCT ACC TGA AAC ACA G-3’ and reverse 5’- GGT CAG GAA CTG TAC CGC CTC-3’, ddx5 forward 5’-AGA GGT CAC AAC TGT CCA AAA CC-3’ and reverse 5’-CCA ATC CAC TGA GAG CAA CTG G-3’, btbd9 forward 5’- CTT GAT GCC ATC AAA GTG CGG TC-3’ and reverse 5’-CAG CTC TCC TTT CAC AAC CTG AG-3’, kdm6b forward 5’- AGA CCT CAC CAT CAG CCA CTG T-3’ and reverse 5’-TCT TGG GTT TCA CAG ACT GGG C-3’, mex3c forward 5’-AGT TCC GAG CAT GTC GCT GAG A-3’ and reverse 5’- CTT CAC CAC GAA CAG GTG TCT TG-3’, laptm4b forward 5’- CCG TGA ATC CTA CCT GTT TGG TC-3’ and reverse 5’-GAG GAG TTC CTG CCG TTG ATG T-3’, cggbp1 forward 5’-TGT TCG CAA GTC TGC CAT TAG TG-3’ and reverse 5’- GGA CTG TTG CAC TGA AGC GAT G-3’, rab39b forward 5’- TGG GAT ACA GCG GGT CAA GAG A-3’ and reverse 5’-GGC TGA ACG TGT ACT TTG GTC TC-3’, rab1 forward 5’-TGG TGT GGA TTT CAA GAT ACG AAC-3’ and reverse 5’-GCT CCT CTG TAA TAA CTG GAA GTG-3’, zfp781 forward 5’-ATG AGA GAA GTC ACA CTG GAG AG-3’ and reverse 5’- CAC GAC TAG CAA AGG CTT CA CC-3’, ppbp forward 5’-CTG ATC CTT GTT GCG CTG GCT C-3’ and reverse 5’-GCC TGT ACA CAT TCA CAA GGG AG-3’, vbp1 forward 5’-ATG GAA CTC AAC CTT GCT CAG AAA-3’ and reverse 5’- GCT TTG CAG TAC AGG TTA TCG GC-3’, efcab11 forward 5’-AGA AGG AAG CGC GAC TCT ATC G-3’ and reverse 5’- 37

TAG CTT GGG AGC CAC TCG ACT A-3’, rpp30 forward 5’-CTG TCC AGT GCT GCA GAA AGG C-3’ and reverse 5’-CAG CTC TGC AAT TTG TGG ACA CG-3’, gypa forward 5’- ACT CCT GTG GTG GCT TCA ACT G-3’ and reverse 5’-GTG TGG TGA GAC AGG CTG TTC T-3’. All primers for RT-qPCR were validated in accordance with MIQE guidelines (Bustin et al., 2009). n) RNA Immunoprecipitation and Microarray Analysis. E12/13 cortices were dissected as described above and immunoprecipitations were carried out with the Magna RNA-Binding Protein Immunoprecipitation Kit as per the manufacturer’s instructions (Millipore). Input lysates were precleared with protein A/G beads for 1 hour and incubated with 5 μg of rabbit Smaug2 (Sigma) or 5 μg of normal mouse IgG (Millipore) for 3 hr at 4 C. Total RNA was isolated, extracted with phenol/chloroform, and checked for quality on a BioAnalyzer (Agilent). RNA samples from 5 biological replicates for the total mRNA/Input and 4 biological replicates for IgG control and Smaug2 immunoprecipitates were amplified using the NuGEN Pico protocol (NuGEN) and analyzed on Mouse Gene 2.0 ST Arrays (Affymetrix). The raw microarray data from the Input was normalized together using robust multi-array normalization (RMA) using R- Studio and the Bioconductor Oligo Package. Probe sets of annotated transcripts with expression levels in the top 80% of total RNA input samples (approx. 20,000 probe sets) were considered as expressed in E12.5 cortices. IgG and Smaug2 immunoprecipitates raw data were normalized together, also with RMA in the Bioconductor Oligo Package. IgG and IP from each mouse were considered as a pair and each probe was compared with a paired t-test between IgG and Smaug2 IP. Pairs with an uncorrected p-value of < 0.01 and present in the 80% Input list were retained for further analysis. Transcripts showing at least a 1.5-fold change in Smaug2 immunoprecipitates versus IgG were used to create the high-confidence list of Smaug2 binders. Principal Component Analysis of the RIP-Chip datasets was performed on the Partek Genomic Suite with the default settings. o) Gene annotation enrichment analysis. The Smaug2 high confidence list was used to perform enrichment analysis for GO terms; to do so, the DAVID functional annotation GO server (Huang, 2009a; 2009b) and the PANTHER classification system were used (Muruganujan and Narechania, 2003). Genes that were identified as being Smaug2-associated were analyzed for enrichment against the background of expressed genes identified from the 5 input mRNA 38 samples, as previously described above. DAVID functional annotation clustering was performed using the default database settings. Clusters that had a score < 1.3, which corresponds to p < 0.05, where considered significant. For each cluster, one term or feature capturing the other terms in the cluster, and with a p-value < 0.05, was reported. Enrichment analysis was performed with PANTHER using the default settings. Terms with p-value < 0.05 were reported. Ingenuity Pathway Analysis (IPA-QIAGEN) was performed using the “Disease and Function Annotation” with the default settings. Because I was working with a small mRNA set that also had several unannotated mRNAs, potentially lowering the statistical power of the analysis, I chose to report terms with the uncorrected p-values < 0.05.

Venn Diagrams were drawn with BioVenn (Hulsen et al. 2008). Violin plots were drawn using the “easyGgplot2” library in RStudio. p) Heat map. To generate the heat map, Smaug2/IgG fold changes were log10-transformed and the positive and negative intervals were each divided in four bins of equal sizes. Smaug2 enrichment bins were then coloured in shades of red, the darker the shade the larger the enrichment. IgG enrichment bins instead were assigned shades of blue, and the darker the shade, the larger the IgG enrichment q) Electron Microscopy. E12.5 brains were fixed with 4% PFA in 0.1M phosphate buffer at RT for 2 hours and then overnight at 4°C. Following fixation, samples were washed in PBS and PFA was quenched with a 10 min wash in 0.15M glycine in PBS. Samples were then infiltrated with 12% gelatin at 37°C and hardened at 4°C. Gelatin-embedded samples were then put in 2.3M sucrose in PBS overnight at 4°C. Smaller pieces were cut by hand with a razor blade and placed on aluminum pins and flash frozen by dipping into liquid nitrogen. Frozen sections were cut at 80 nm using a Leica ultramicrotome with a cryochamber attachment. Sections were scooped with a loop using a 1:1 mixture of 2% methyl cellulose and 2.3M sucrose. Sections were then placed on nickel grids coated with formvar. Warm drops of PBS were placed on samples for 10 mins to remove the gelatin; samples were then placed on drops of 0.15M glycine in PBS for 10 mins. 1% BSA in PBS for 30 mins was used for blocking. The rabbit anti- Staufen2 antibody (1:100; Sigma) was diluted in 0.1% BSA and samples incubated with it for 30 to 45 minutes. Protein A-gold was used to detect the primary antibody. Sections were then incubated for 10 minutes in 2.5% glutaraldehyde in phosphate buffer and then incubated on 39 drops of 2% methylcellulose with 0.4% uranyl acetate for 10 min. on ice. The grids were picked up with loops and the excess methylcellulose, uranyl acetate solution was wiped off and the grids allowed to dry. Sections were imaged using a Tecnai 20 transmission electron microscope and micrographs taken with an AMT 16000 digital camera. The mean number of Staufen2 particles associating with putative RNA granules was determined by counting the number of gold particles overlaying an electron dense structure of an appropriate size for a granule. r) Statistical analysis. All data were expressed as the mean plus or minus the standard error of the mean (SEM). Statistic analysis was performed with a two-tailed Student's t-test, Mann- Whitney Rank Sum test. Where relevant, ANOVA with Tukey's or Student Neuman-Keuls multiple comparisons test or Bartlett’s test for equal variances post-hoc were used. To determine the correlation coefficient between microarray and RT-qPCR, each series of fold changes was tested for normality with the Shapiro-Wilk test. Since the test indicated that the values were not normally-distributed, the Spearman correlation ρ (rho) was utilized.

Chapter 3 An asymmetrically localized Staufen2-dependent RNA complex regulates maintenance of mammalian neural stem cells

The data presented in this chapter was published in the homonymous research paper in October 2012 in the journal “Cell Stem Cell”, 11: 517-528.

John P. Vessey1,2, Gianluca Amadei1,3, Sarah E. Burns2, Michael A. Kiebler 5, David R. Kaplan2,3, and Freda D. Miller1,3,4

Programs in Developmental and Stem Cell Biology1 and Cell Biology2, Hospital for Sick Children, Toronto, Canada M5G 1L7, Departments of Molecular Genetics3 and Physiology4, University of Toronto, Toronto, Canada M5G 1X5 and Department of Neuronal Cell Biology, Center for Brain Research, Medical University of Vienna,1090 Vienna, Austria5.

Author Contributions: J.P.V. performed the majority of the experiments presented in this paper, made the figures and wrote the draft of the paper. G.A. helped out with the majority of the experiments in this paper with the exception of prox1 in situ hybridization studies, immunoprecipitations and electron microscopy experiments. S.E.B. performed all in utero electroporations in this study and helped out with Pax6 quantification in vivo. M.A.K. generously provided some of the reagents utilized in this study, such as the Staufen2 antibody and the shStaufen2 constructs as well as intellectual input. F.D.M. and D.R.K. participated in the conception, design and analysis of experiments described here, and were involved with manuscript writing.

3.1 SUMMARY

The cellular mechanisms that regulate self-renewal versus differentiation of mammalian somatic tissue stem cells are still largely unknown. Here, we asked whether an RNA complex regulates this process in mammalian neural stem cells. We show that the RNA-binding protein Staufen2 is apically localized in radial glial precursors of the embryonic cortex, where it forms a complex with other RNA granule proteins including Pumilio2, and DDX1, and the mRNAs for β-actin and mammalian prospero, prox1. Perturbation of this complex by functional knockdown of Staufen2, Pumilio2, or DDX1 causes premature differentiation of radial glial precursors into neurons, and mislocalization and misexpression of prox1 mRNA. Thus, a Staufen2 and Pumilio2-dependent RNA complex directly regulates localization and, potentially, expression of target mRNAs like prox1 in mammalian neural stem cells, and in so doing regulates the balance of stem cell maintenance versus differentiation.

3.2 BRIEF INTRODUCTION AND RATIONALE

The mechanisms that regulate self-renewal of mammalian stem cells are still largely unknown. Insights into this issue come from model organisms like D. melanogaster, where pathways that regulate maintenance of germ and neural stem cells are well characterized. In this regard, one key D. melanogaster mechanism involves RNA: protein complexes that localize mRNAs encoding regulatory proteins (Chia et al., 2008). As examples of this, asymmetric localization and translational repression of prospero mRNA regulates maintenance versus differentiation of D. melanogaster neuroblasts (Knoblich, 2008), while asymmetric localization of oskar mRNA dictates appropriate oocyte development (Kuglar and Lasko, 2009). However, it is not yet clear whether similar asymmetrically localized RNA complexes exist in mammalian stem cells, although the RNA binding proteins Musashi-1and Pumilio-2 can bind and repress their target mRNAs in mammalian neural and germ line stem cells, thereby maintaining them in an undifferentiated state (Moore et al., 2003; Okano et al., 2005). Here, we have asked whether an asymmetrically localized RNA complex regulates mammalian stem cells, focusing upon Staufen, a double-stranded RNA-binding protein that regulates asymmetric RNA localization in D. melanogaster neuroblasts (Chia et al., 2008). In the fly, neuroblasts divide asymmetrically to produce a neuroblast and a ganglion mother cell, which then divides once to generate two neurons (Chia et al., 2008). These divisions, which require asymmetric localization of a Stau-containing RNA complex, have apparent similarities

42 to stem cell divisions in the mammalian central nervous system (CNS). For example, in the embryonic cortex many postmitotic neurons are generated by radial precursors (RPs) that undergo asymmetric divisions to produce both a copy of themselves and an intermediate progenitor that then divides to generate two neurons (Zhong and Chia, 2008). In spite of these similarities, it is not known whether a Staufen-containing RNA complex is present and/or important in mammalian neural stem cells. In mammalian neurons Staufen 1 and 2 bind their RNA targets and assemble into large, heterogeneous transport-competent RNA granules that undergo kinesin-dependent transport along microtubules into neuronal dendrites (Kiebler and Bassell, 2006). It is thought that mRNAs within these granules are repressed until synaptic activity induces dissociation of the granule, and that upon dissociation the mRNAs are locally translated. When Staufen1 or Staufen2 are depleted, this causes deficits in dendritic spines, presumably because this mechanism is disrupted (Goetze et al., 2006; Vessey et al., 2008). Here, we have asked whether a Staufen-dependent RNA complex is important in embryonic neural stem cells, and show that Staufen2 is apically enriched in RPs in vivo where it comprises a critical component of an RNA complex containing the translational repressor Pumilio2, the RNA helicase DDX1, and Staufen2 target mRNAs including prox1 and β-actin. Disruption of the complex by knockdown of any of these three protein components causes enhanced neurogenesis and depletion of RPs. Staufen2 knockdown also disrupts localization and expression of prox1 mRNA, and a Staufen2 mutant that cannot bind its target RNAs cannot promote RP self-renewal. Thus, a Staufen2, Pumilio2-dependent RNA complex directly regulates localization of proneurogenic mRNAs like prox1 in mammalian neural stem cells, and in so doing regulates the balance of stem cell maintenance versus differentiation.

3.3 RESULTS

3.3.1 Staufen2 is apically localized in embryonic radial glial precursors in the developing murine cortex

To ask about a Staufen-dependent RNA granule, we examined RPs of the embryonic murine cerebral cortex. These precursors generate neurons from around E12 until birth, and then switch to making glial cells. We focused upon Staufen2, which is enriched in the brain (Duchaîne et al., 2002). RT-PCR analysis showed that staufen2 mRNA was expressed in the E12.5 cortex, which is largely composed of precursors (Fig. 6A). Western blots confirmed that

Figure 6: (A-F) Staufen2 is expressed by RPs and newborn neurons during cortical development. (A) RT-PCR for staufen2 mRNA in E12.5 cortex (lane indicated with +; lane indicated with – is without reverse transcriptase). Molecular weight (MW) markers are shown to

44 the left. (B) Western blot analysis for three Staufen2 isoforms (as indicated) in E11 to P3 cortex. The blot was reprobed for Erk1/2. (C) Cultured E12.5 cortical precursors immunostained after 3 days for Staufen2 (red), and Nestin, Pax6 or ßIII-tubulin (all green). Scale bars represent 10 µm. (D,E) Immunoreactivity for Staufen2 (red) in coronal sections of E13 (D) or E15 (E) cortex. Nuclei were stained with Hoechst 33258 (blue). Arrows highlight Staufen2 in the apical VZ. Boundaries of the CP, VZ/SVZ and IZ are denoted with white lines. Scale bars represent 10 µm. (F) Confocal micrographs of the E13 VZ/SVZ co-stained for Staufen2 (red) and Nestin or ZO-1 (both green). Arrows denote the apical VZ. Also shown is co-staining for ßIII-tubulin (green) in the E15 CP (bottom panels). Scale bars represent 10 µm. (G-I) Staufen2 interacts with other RNA complex proteins. (G) Western blots of E12.5 cortical lysates immunoprecipitated with anti-Staufen2 (center lanes) or with control nonspecific rabbit IgG (right lanes), probed with antibodies for Pumilio2, DDX1, PABP or Staufen2. As a control, 10% of the input homogenate was loaded (left lanes). MW markers are shown to the left, in kDa. (H,I) Western blots of E12.5 cortical lysates immunoprecipitated with anti-Pumilio2 (H, center lanes) or anti-DDX1 (I, center lanes) or with a control nonspecific rabbit IgG (right lanes), probed with antibodies for Pumilio2, DDX1, or Staufen2, as indicated. As a control, 10% of the input homogenate was loaded (left lanes). MW markers are indicated to the left, in kDa.

45 all three Staufen2 splice variants were expressed (Fig. 6B), with the two smallest isoforms predominating at E11 and E12, and the largest 62 kDa isoform increasing in levels from E15 until postnatal day 3 (P3), coincident with increased neuron numbers. To define the cell types that express Staufen2, we immunostained E12.5 RP cultures (Gauthier et al., 2007). Staufen2 was detectable in almost all cells, including RPs that express Nestin and Pax6, and ßIII-tubulin-positive neurons that are born in these cultures (Fig. 6C). Immunostaining of the embryonic cortex confirmed a similar expression pattern in vivo. At E13 (Fig. 6D), Staufen2-positive cells were present in the ventricular and subventricular zones (VZ/SVZ) and the cortical plate (CP), regions that contain precursors and newborn neurons, respectively. Within the VZ, Staufen2 was enriched in the apical portion of Nestin-positive (Fig. 6F), where it colocalized with the apical marker ZO-1, a tight junction protein (Fig. 6F). At E15, Staufen2 was still enriched apically in the VZ, and was also robustly expressed in ßIII-tubulin- positive neurons in the CP (Fig. 6E,F). Thus, Staufen2 is asymmetrically localized in RPs in vivo.

3.3.2 Staufen2 is part of an apical RNA complex in radial glial precursors

We next asked whether Staufen2 in RPs was associated with three other RNA complex proteins, the RNA binding protein and translational repressor Pumilio2 (Moore et al., 2003), the RNA helicase DDX1 (Kanai et al., 2004), and Poly-A Binding protein (PABP), which binds polyadenylated mRNAs (Elvira et al., 2006). Staufen2 was immunoprecipitated from the E12.5 cortex, and Western blot analysis was performed on the immunoprecipitates with antibodies to Pumilio2, DDX1 and PABP. This analysis (Fig. 6G) demonstrated that Staufen2 co- immunoprecipitated with all three proteins. We confirmed these interactions by immunoprecipitating Pumilio2 (Fig. 6H) or DDX1 (Fig. 6I) and showing that Staufen2 co- immunoprecipitated with both proteins. To ask whether these proteins colocalized, we immunostained the embryonic cortex. Like Staufen2, at E13 Pumilio2 and DDX1 were enriched in the apical portion of the VZ, where they were expressed in Nestin-positive precursors (Fig.

7A,B) and colocalized with ZO-1 (Fig. 7C,D). To ask whether mRNAs were also associated with this apically-localized RP complex, we examined two mRNAs known to interact with Staufen2; β-actin mRNA, since it associates with Staufen2 in postmitotic neurons (Goetze et al., 2006), and prox1 mRNA, since it associates with Staufen2 in the mammalian brain (Furic et al., 2008), and since D. melanogaster Staufen

Figure 7: Staufen2 interacts and colocalizes with RNA complex proteins and target mRNAs. (A-D) Confocal micrographs of E13 cortex (A,B) or VZ/SVZ (C,D) immunostained for Nestin (green, A,B) or ZO-1 (green, C,D) and DDX1 (red, A,C) or Pumilio2 (red, B,D). Arrows denote

47 the apical VZ and boxed areas are enlarged in lower panels. Scale bars represent 10 µm. (E) RT- PCR for kif5c, arnt, β-actin, and prox1 mRNAs in E12.5 cortical lysates immunoprecipitated with anti-Staufen2 (+ lanes) or control nonspecific rabbit IgG (- lanes). As a control, RNA was isolated from input homogenates (In). MW markers are shown to the left. (F) Micrograph of FISH for prox1 mRNA (red) in the E15 VZ/SVZ. The boxed region is shown at higher magnification in the right panel. Arrows denote apical enrichment. As a control, adjacent sections were hybridized to a sense probe (left panel). Scale bars represent 10 µm. (G,H) FISH for prox1 mRNA (red) with immunocytochemistry for Staufen2 (G) or ZO-1 (H; both green) in E15 cortex (G) or the E15 VZ/SVZ (H). The boxed area in (G) is shown at higher magnification in the bottom panels. Arrows denote the apical VZ. Scale bars represent 10 µm. (I) Immunoelectron microscopy of E13 cortical sections probed with Staufen2 antibodies and visualized using gold particles. The left panel shows three immunogold particles over an electron dense structure below the nucleus and next to the plasma membrane lining the ventricles. The top right micrograph was obtained with a different primary antibody, and shows another immunogold-positive structure adjacent to the plasma membrane. Both particles are also shown at higher magnification. Scale bars represent 50 nm.

48 localizes prospero mRNA (Chia et al., 2008). Staufen2 was immunoprecipitated from the E12.5 cortex, and RNA was isolated from these immunoprecipitates. RT-PCR (Fig. 7E) demonstrated that both β-actin and prox1 mRNAs were present in the Staufen2 immunoprecipitates. These interactions were selective, since kif5c and arnt mRNAs, which encode the Kif5c motor protein and the aryl hydrocarbon receptor nuclear translocator protein, respectively, were not present (Fig. 7E). To ask whether these Staufen2-interacting mRNAs were apically localized, we performed in situ hybridization for prox1 mRNA in the E15 cortex. prox1 mRNA was expressed in cells in the VZ, and was apically enriched (Fig. 7F). Combined in situ hybridization and immunocytochemistry showed that prox1 mRNA colocalized apically with Staufen2 (Fig. 7G) and ZO-1, although it was not strictly limited to the apical-most extremity of the RPs, as was ZO-1 (Fig. 7H). Finally, we asked whether we could visualize Staufen2-containing complexes, since these have been seen in dendrites (Kiebler et al., 1999). Immunoelectron microscopy showed that Staufen2 immunoreactivity, as detected with immunogold, was localized to electron dense granule-like structures adjacent to the ventricle (Fig. 7I). Approximately 2 gold particles were associated with each positive granule. In contrast, very few immunogold particles were found on sections processed without the Staufen2 primary antibody, and virtually none of these were localized on granules. Similar results were obtained with two different Staufen2 antibodies (Fig. 7I).

3.3.3 Asymmetric localization of Staufen2 and prox1 mRNA in dividing radial glial precursors

In D. melanogaster, Staufen and its target mRNAs are asymmetrically segregated to daughter cells during differentiation (Chia et al., 2008). To ask if this was true for RPs, we immunostained E15 cortical sections for Staufen2 and for γ-tubulin, the latter to identify cleavage plane of dividing cells, and stained them with Hoechst 33258 to visualize condensing chromosomes. In many dividing ventricular precursors, Staufen2 levels were roughly similar on both sides of the cleavage plane, but in some, it was apparently enriched on one side versus the other (Fig. 8A,B). To quantify this effect, we used the confocal microscope to measure the level of Staufen2 immunoreactivity on both sides of the cleavage plane of ventricular anaphase or telophase cells (Fig. 8A,B), and compared these numbers to the Hoechst 33258 intensity, as a measure of DNA content (Fig. 8C) (Bultje et al., 2009). This analysis showed that Staufen2 was significantly asymmetrically segregated to the two presumptive daughter cells (Fig. 8C).

Figure 8: Staufen2 and its target mRNA prox1 are asymmetrically enriched in dividing ventricular precursors. E15 cortices were sectioned and processed by immunostaining for Staufen2 (red, A) or by FISH for prox1 mRNA (red, D). To locate cells undergoing mitosis, DNA was labelled with Hoechst (blue) and the mitotic spindle with γ-tubulin (green). The plane 50 of division was determined by the γ-tubulin staining and indicated with the long white arrow (left panels). The direction of the ventricular zone is indicated with the arrow head and the area of the dividing cell is traced with the dotted white line. The upper rows show symmetrically localized Staufen2 protein (A) or prox1 mRNA (D) and the lower rows show asymmetrically localized Staufen2 protein (A) or prox1 mRNA (D). Scale bars represent 5µm. (B,E) Confocal micrographs of ventricular cells in anaphase/telophase from E15 cortical sections analyzed by immunostaining for Staufen2 (red, B) or by FISH for prox1 mRNA (red, E), and stained with Hoechst 33258 (blue). Plane of division is indicated with the long white arrow (left panels), VZ is indicated by an arrowhead, and the dividing cell is traced with the dotted white lines. The upper two rows demonstrate symmetrically-localized and the bottom two asymmetrically- localized Staufen2 (B) or prox1 mRNA (D). Scale bars represent 5µm. (C,F) Quantification of the intensity of Staufen2 immunofluorescence (B) or prox1 mRNA in situ hybridization signal (E) on the two sides of the cleavage plane in anaphase/telophase cells. Data are normalized so that if fluorescence was the same on both sides, this would equal 0, and if it was only present on one side, this would equal 1. Data are represented as box plots with the box representing the data range between the 25th and 75th percentile, and the central line indicating the median value. Statistical outliers are indicated by black circles above and below whiskers which indicate the predicted maximum and minimum values. For comparison, a similar analysis was performed for Hoechst 33258 DNA, and is also shown as a box plot. (*p<0.05, **p <0.01; for Staufen2, Students t-test was used, while for prox1 mRNA, values did not represent a normal distribution, so the Mann-Whitney Rank Sum test was used).

We then performed a similar analysis for prox1 mRNA, combining fluorescence in situ hybridization with γ-tubulin immunostaining and/or visualization of DNA with Hoechst 33258. As seen for Staufen2, while prox1 mRNA levels were roughly similar on both sides of the cleavage plane in many dividing cells, in some cells it was clearly enriched on one side versus the other (Fig. 8D,E). Quantitative analysis demonstrated that prox1 mRNA was modestly but significantly asymmetrically segregated (Fig. 8F).

3.3.4 Knockdown of Staufen2 promotes neurogenesis and depletes radial glial precursors

To ask about the function of this apical RNA complex in RPs, we knocked down Staufen2 using a previously characterized shRNA (Goetze et al., 2006). We first confirmed that this shRNA knocked down Staufen2 in HEK293 cells (Fig. 9A, left) and that it did not affect Staufen1 expression (Fig. 9A, right), since staufen1 mRNA is also expressed in the embryonic cortex (Fig. 9B). We also confirmed that this shRNA was equally efficacious in RPs. Cultures were cotransfected with Staufen2 shRNA and an EGFP-expression plasmid; Staufen2- immunoreactivity was not detectable in EGFP-positive cells, but was readily detectable in EGFP-negative, non-transfected precursors (Fig. 9C). We then characterized the biology of similarly transfected RP cultures. Staufen2 knockdown had no effect on cell survival, as demonstrated by quantifying EGFP-positive condensed, apoptotic nuclei (Fig. 9D). However, it did affect neurogenesis; immunostaining for ßIII-tubulin three days post-transfection (Fig. 9E) demonstrated that, relative to a control shRNA, Staufen2 shRNA increased newborn neurons from approximately 30% to 80% (Fig. 9F). Staufen2 knockdown also affected proliferating precursors, causing a decrease in transfected cells expressing the proliferation marker Ki67 (Fig. 9G,H) and the RP marker Pax6 (Fig. 9I,J). In contrast, Staufen2 knockdown did not affect the relatively low proportion of Tbr2-positive basal progenitors (Fig. 9K,L), the neurogenic transit- amplifying cells in this system. These findings suggest that Staufen2 knockdown causes RPs to prematurely differentiate into neurons thereby depleting the stem cell pool. We performed two additional experiments that supported this conclusion. First, we performed clonal analysis and showed that the average number of cells generated from a single transfected precursor was decreased (Fig. 9M). Second, we triple labelled cultures for EGFP, Pax6 and Ki67 or ßIII-tubulin. Staufen2 knockdown caused a decrease in cycling Pax6-positive RPs (Fig. 9N) and an increase in RPs co-expressing

Figure 9: Knockdown of Staufen2 in culture causes increased neurogenesis and depletion of cycling precursors. (A, left) Western blot of HEK-293 cells transfected for 2 days with a control shRNA or a Staufen2 shRNA that targets the mouse and human proteins, probed with antibodies

53 for Staufen2 (top) and Erk1/2 (bottom). (A, right) Western blot of HEK-293 cells transfected for 2 days with a control shRNA or a Staufen2 shRNA that targets the mouse and human proteins, probed with antibodies for Staufen1 (top) and Erk1/2 (bottom). (B) RT-PCR analysis of E12.5 cortex RNA. staufen1 and staufen2 mRNA were detected at the predicted sizes. Molecular weight markers are shown in the left lane. (C-L) Cultured precursors were cotransfected with a nuclear EGFP plasmid plus Staufen2 or control shRNAs for 3 days. (C) Immunostaining of cultures transfected with Staufen2 shRNA for EGFP (green) and Staufen2 (red). Staufen2 was detectable in EGFP-negative, non-transfected cells (arrows) but not in EGFP-positive transfected cells (white stars). (D) Quantification of cells with condensed apoptotic nuclei as detected by staining with Hoechst 33258. (p>0.05; n = 3). (E-L) Cultures were immunostained for EGFP (green, E,G,I,K) and ßIII-tubulin (E), Ki67 (G), Pax6 (I) or Tbr2 (K; all red) and double-labelled cells (arrows) quantified (F,H,J,L). (**p<0.01, ***p <0.001; n = 3 in J,L, and 4 in F,H). (M) Quantification of the average number of cells in EGFP-positive clones deriving from precursors cotransfected with EGFP and Staufen2 or control shRNAs. (*p<0.05; n = 3). (N,O) Quantification of triple-labelled cells in cultures cotransfected with nuclear EGFP plus Staufen2 or control shRNAs, and 3 days later immunostained for EGFP, Pax6 and Ki67 (N) or ßIII-Tubulin (O). (***p<0.001; n = 3). (P,Q) Quantification of precursors cotransfected for 3 days with EGFP and Staufen2 or control shRNAs with or without an expression construct for shRNA-resistant Staufen2 (Staufen2R-EGFP), immunostained for EGFP and ßIII-tubulin (P) or Pax6 (Q). (*p<0.05; n = 3). All scale bars represent 10 µm. Statistical comparisons were performed using Student’s T-test, except for panels O and P, where we used a Student Neuman- Keuls post-hoc ANOVA. Error bars denote standard error of the mean (S.E.M.).

ßIII-tubulin (Fig. 9O) suggesting that Staufen2 knockdown promotes direct genesis of neurons from RPs. To ensure that these changes were due to Staufen2 knockdown, we performed rescue experiments using a previously described Staufen2-EGFP (Staufen2R ) that encodes the same Staufen2 protein but is resistant to Staufen2 shRNA (Goetze et al., 2006). Precursors were cotransfected with Staufen2 or control shRNA plus or minus Staufen2R, and immunostained three days later for ßIII-tubulin or Pax6. Staufen2R significantly rescued the Staufen2 knockdown phenotypes (Fig. 9P,Q), confirming the specificity of the shRNA. We asked whether these culture data were relevant in vivo by in utero electroporation of the Staufen2 shRNA together with an EGFP expression plasmid into E13/14 cortices (Gauthier- Fisher et al., 2009). This approach electroporates RPs that line the ventricles, many of which differentiate in the VZ/SVZ into neurons that migrate through the intermediate zone (IZ) into the CP. Analysis 3 days post-electroporation showed that Staufen2 knockdown caused mislocalization of EGFP-positive cells, with fewer transfected cells in the SVZ/VZ and CP, and more in the IZ (Fig. 10A,B). We asked about the cellular basis of these changes. Immunostaining for EGFP and ßIII-tubulin (Fig. 10C) showed that Staufen2 knockdown caused an increase in neurons (Fig. 10D), but that these neurons were mislocalized; in controls approximately 40% of EGFP-positive neurons were in the CP as compared to virtually none with Staufen2 knockdown (Fig. 10E). To confirm these findings, we analyzed another cortical neuron marker, SatB2, since the majority of RPs electroporated at E13/14 generate SatB2- positive neurons (Tsui et al. 2013). Staufen2 knockdown increased transfected SatB2-positive neurons from approximately 20% to 50% (Fig. 10F,G), but fewer of these were localized to the cortical plate (Fig. 10H), confirming the ßIII-tubulin data. We also asked if Staufen2 knockdown depleted RPs in vivo as it did in culture. Staufen2 knockdown decreased EGFP-positive, Pax6-positive RPs from approximately 40% to 20%, although virtually all of these were located within the VZ/SVZ in both conditions (Fig. 10I,J). As seen in culture, the proportion of EGFP-positive cells expressing both Pax6 and ßIII-tubulin increased two-fold to approximately 60% (Fig. 10K), suggesting that Staufen2 knockdown caused RPs to directly differentiate into neurons. Consistent with this, Staufen2 knockdown had no effect on Tbr2-positive intermediate progenitors at either 2 or 3 days post-electroporation (Fig. 10L,M). Thus, Staufen2 knockdown promotes genesis of neurons at the expense of RPs, but these neurons do not migrate appropriately. This latter phenotype might indicate a direct role

Figure 10: Depletion of Staufen2 in vivo increases genesis of neurons at the expense of RPs. (A,B) Micrographs (A) of E16/17 sections from cortices electroporated at E13/14 with EGFP

56 and Staufen2 or control shRNAs, that were immunostained for EGFP (green), stained with Hoechst 33258 (blue) and quantified (B) for the proportion of EGFP-positive cells in the CP, IZ and SVZ/VZ. (*p<0.05; n = 4 embryos each). (C-H) Cortical sections similar to those in (A) were immunostained for EGFP (green) and ßIII-tubulin (C) or SatB2 (F; both red) and transfected, marker-positive cells were quantified for their total numbers (D,G) and the proportion in the VZ/SVZ/IZ versus the CP (E,H). (*P<0.05, **P<0.01; n = 3 to 4 embryos each). (I-M) Sections as in (A) were immunostained and double- or triple-labelled cells quantified. (I,J) Quantification (J) of cells expressing EGFP (green, I) and Pax6 (red, I). (K) Quantification of cells expressing EGFP, Pax6 and ßIII-tubulin at 2 and 3 days (combined). (L,M) Quantification (M) of cells expressing EGFP (green, L) and Tbr2 (red, L) at 2 and 3 days. (*p<0.05, **p <0.01; n = 3 or 4 embryos each). In (C,F,I,L) the white line denotes the boundary between the IZ (above) and SVZ/VZ (below) and arrows indicate double-labelled cells. All scale bars represent 20 µm. Statistical comparisons used Student’s T-test, and error bars denote S.E.M.

57 for Staufen2 in neuronal migration, or it could be due to depletion of RPs, which provide essential migratory scaffolds.

3.3.5 Knockdown of either Pumilio2 or DDX1 phenocopies the effects of Staufen2 knockdown

To ask whether the Staufen2 knockdown phenotype was due to disruption of the apical RNA complex, we determined whether knockdown of Pumilio2 in cultured RPs caused the same phenotype. RT-PCR analysis (Fig. 11A) and immunostaining (Fig. 11C) showed that Pumilio2 was expressed in cultured Nestin-positive precursors and ßIII-tubulin-positive neurons, as seen in vivo (Fig. 7). To knockdown Pumilio2 in these cultures, we cotransfected them with a previously characterized shRNA and an EGFP plasmid (Vessey et al., 2006). Pumilio2 was readily detectable in EGFP-negative non-transfected cells, but not in EGFP- positive transfected cells (Fig. 11B), confirming Pumilio2 shRNA efficacy. Analysis of similarly-transfected cultures showed that Pumilio2 knockdown had no effect on cell survival, as indicated by EGFP-positive cells with condensed, apoptotic nuclei (Fig. 11D). In contrast, Pumilio2 knockdown almost doubled the percentage of transfected, ßIII-tubulin-positive neurons present 3 days post-transfection (Fig. 11E,F). Moreover, it caused a significant decrease in dividing RPs, as indicated by immunostaining for Ki67 and Pax6 (Fig. 11G-J), and by clonal analysis (Fig. 11K). Thus, Pumilio2 knockdown phenocopies Staufen2 knockdown. We performed similar studies for a third complex component, DDX1. RT-PCR (Fig. 11A) and immunostaining of cortical cultures (Fig. 11M) demonstrated that DDX1 was expressed in Nestin-positive and Pax6-positive precursors and in βIII-tubulin-positive neurons, consistent with the in vivo data (Fig. 7). We then confirmed the efficacy of a DDX1 shRNA by transfecting it into murine 3T3 cells; western blots 2 days post-transfection showed that the endogenously-expressed DDX1 was robustly knocked-down (Fig. 11L). We then cotransfected the DDX1 shRNA together with EGFP into precursor cultures; analysis at three days demonstrated that DDX1 knockdown doubled the proportion of βIII-tubulin-positive neurons and decreased Pax6-positive RPs (Fig. 11N,O). Thus, Staufen2, Pumilio2 and DDX1 are all necessary for maintaining RPs, a function they likely mediate by acting in a common RNA complex.

Figure 11: Knockdown of either Pumilio2 or DDX1, components of the apically localized Staufen2-containing complex, phenocopies Staufen2 knockdown. (A) RT-PCR for pumilio2 and ddx1 mRNAs in cultured precursors. MW markers are to the left. (B) Immunostaining for EGFP (green) and Pumilio2 (red) in cultured precursors transfected for 3 days with an EGFP plasmid and Pumilio2 shRNA. Pumilio2 was detectable in EGFP-negative, non-transfected cells (arrows) but not in EGFP-positive transfected cells (white stars). (C) Immunostaining of precursors cultured 3 days for Pumilio2 (red), and Nestin, Pax6, or βIII-tubulin (all green). Scale bars represent 10 µm. (D) Quantification of condensed apoptotic nuclei in precursors transfected as in (B). (P>0.05; n = 4). (E-J) Quantification of double-labelled cells (F,H,J) in cultures cotransfected as in (B) and immunostained for EGFP (green) and βIII-tubulin (E), Ki67 (G) and Pax6 (I; all red). Arrows denote double-labelled cells. (*p<0.05, **p <0.01, ***p <0.001; n = 3 or 4). (K) Quantification of the average number of cells in EGFP-positive clones deriving from precursors cotransfected with Pumilio2 or control shRNAs. (*P<0.05; n = 3). (L) Western blot of 3T3 cells transfected with control (lane 1) or DDX1 (lane 2) shRNAs, probed for DDX1 and Erk1/2. MW markers are indicated to the left, in kDa. (M) Immunostaining of precursors cultured 3 days for DDX1 (red), and Nestin, Pax6, or βIII-tubulin (all green). Scale bars represent 10 µm. (N,O) Quantification of double-labelled cells in cultures cotransfected with a nuclear EGFP plasmid plus DDX1 or control shRNAs for 3 days, and immunostained for EGFP and βIII-tubulin (N) or Pax6 (O). (**p<0.01, ***p <0.001; n = 3). Statistical comparisons used Student’s T-test, and error bars denote S.E.M.

3.3.6 Disruption of Staufen2-RNA interactions affects prox1 mRNA localization and expression and causes differentiation of radial glial precursors

To ask whether perturbation of the Staufen2/Pumilio2/DDX1 RNA complex might promote inappropriate neurogenesis by mislocalizing or derepressing target mRNAs, we examined prox1 mRNA localization. We electroporated E13/14 cortices with Staufen2 shRNA and EGFP, and 3 days later performed combined fluorescent in situ hybridization and immunocytochemistry. EGFP-positive cells within the VZ were classified into those where prox1 mRNA was apically-localized, those where it was localized on the other side of the cell nucleus from the ventricle (basal localization), and those where it was randomly localized (Fig. 12A). In control-electroporated cortices, approximately half of the transfected VZ cells showed apical prox1 mRNA localization (Fig. 12B). In contrast, in Staufen2 shRNA electroporated cortices, this number decreased to approximately 25%, and in most EGFP-positive VZ cells prox1 mRNA was randomly-localized (Fig. 12A,B). We next characterized Prox1 protein expression in control and Staufen2 knockdown cortices. Intriguingly, in contrast to prox1 mRNA, Prox1 protein was not detectable in the large majority of precursors lining the ventricles, suggesting that the apically localized prox1 mRNA in RPs was translationally repressed (Fig. 12C). In contrast, in cortices electroporated with Staufen2 shRNA, the proportion of EGFP-positive cells in the VZ/SVZ with nuclear Prox1 protein expression one day later was almost doubled (Fig. 12D), suggesting that translation of prox1 mRNA is repressed in a Staufen2-dependent fashion in cortical precursors. These findings indicate that the apical Staufen2-containing RNA granule localizes and represses prox1. To ask whether disruption of this RNA localization/repression activity could explain the aberrant differentiation seen when Staufen2 is knocked-down, we utilized a Staufen2 protein that lacks RNA-binding domain 5 and the putative tubulin-binding domain (Fig. 12E). This mutant Staufen2 can still bind its protein partners, but is unable to bind mRNA (Goetze et al., 2006). We confirmed that this mutant cannot bind target mRNAs by transfecting it into HEK293 cells. Immunoprecipitation and RT-PCR analysis of precipitates one day later showed that wildtype, but not mutant, Staufen2 co-immunoprecipitated with the endogenous β-actin and prox1 mRNAs (Fig. 12F, G). We then asked whether the Staufen2 mutant caused mislocalization of prox1 mRNA when electroporated into RPs in the embryonic cortex. As a control, we electroporated wildtype Staufen2. Analysis 3 days post-electroporation showed that mutant but not wildtype Staufen2 caused a significant decrease in transfected VZ cells with 61

Figure 12: Staufen2 mediates apical localization of prox1 mRNA by its RNA binding domain, and this RNA binding activity is essential for Staufen2 to promote RP maintenance. (A) High magnification micrographs through the VZ of coronal E16/17 cortical sections electroporated at

E13/14 with a nuclear EGFP plasmid and Staufen2 (center, right) or control (left) shRNAs, analyzed by FISH for prox1 mRNA (red) and immunostaining for EGFP (green; bottom panels). Dotted lines denote the ventricular surface (V), and panels show representative examples of apical, non-localized, and basal prox1 mRNA localization. (B) Quantification of sections as in (A). (*p<0.05; **p <0.01; n = 3 embryos each). (C,D) Quantification (D) of double-labelled cells (C, arrows) in cortices electroporated at E13/14 with EGFP and control (left) or Staufen2 (right) shRNAs and immunostained one day later for EGFP (green) and Prox1 (red) (C). The white line denotes the boundary between the IZ (above) and SVZ/VZ (below). (*p<0.05; n = 3 embryos each). (E) Schematic depicting the Staufen2 mutant (dnStaufen2) lacking RNA- binding domain (RBD) 5 and the putative tubulin-binding domain (TBD). (F) Western blot of HEK293 cells transfected with wildtype (wt) or mutant (dn) Staufen2 for 1 day, immunoprecipitated with anti-Staufen2 (center lanes) or control nonspecific rabbit IgG (right lanes) and probed with anti-Staufen2. As a control, 5% of the input homogenate was loaded (left lanes). MW markers are indicated to the left, in kDa. (G) RT-PCR for β-actin and prox1 mRNAs in RNA isolated from immunoprecipitates as in (F). In-wt and In-dn refer to wildtype and mutant Staufen2 input homogenates, wt-IP and dn-IP refer to wildtype and mutant anti- Staufen2 immunoprecipitates and wt-IgG and dn-IgG to the nonspecific rabbit IgG immunoprecipitates of the wildtype or mutant Staufen2-transfected cells. MW markers are shown to the left. (H) Quantification of EGFP-positive VZ cells showing apical, basal and non- localized prox1 mRNA in E16/17 cortical sections that were electroporated with EGFP and wildtype or mutant Staufen2 at E13/14, and analyzed as in (A,B). (*p<0.05; **p <0.01; n = 3 embryos each). (I, J) Quantification of double-labelled cells in precursor cultures cotransfected with plasmids encoding EGFP and wildtype (control) or mutant (dn) Staufen2, immunostained at 3 days for EGFP and βIII-tubulin (I) or Pax6 (J). (**p<0.01, ***p <0.001; n = 3). Statistical comparisons used Student’s T-test, except for panels B and H, where we used a Student Neuman-Keuls post-hoc ANOVA. Error bars denote S.E.M.

63 apical prox1 mRNA localization (Fig. 12H). Finally, we asked whether disruption of Staufen2 RNA binding perturbed RP maintenance. Transfection of mutant but not wildtype Staufen2 into cultured precursors for 3 days caused an increase in βIII-tubulin positive neurons and a decrease in Pax6 positive RPs (Fig. 12I, J). Thus, the Staufen2-dependent apical RNA complex apparently regulates stem cell maintenance by localizing and potentially repressing target mRNAs.

3.4 CONCLUSIONS

In summery, the data presented in this chapter supports four major conclusions. First, we demonstrate that Staufen2 is asymmetrically localized in RPs and is enriched in their apical endfeet at the ventricular surface. Second, our immunoprecipitation and colocalization studies demonstrate that Staufen2 forms part of an apically localized RNA complex that contains the RNA-binding proteins Pumilio2 and DDX1 and target mRNAs such as prox1 and β-actin. Our data also indicate that this complex is asymmetrically enriched in dividing ventricular precursors. Third, we show that disruption of this granule by genetic knockdown of Staufen2, Pumilio2, or DDX1 leads to genesis of neurons at the expense of precursors, indicating that this Staufen2-dependent RNA complex is essential for appropriate precursor maintenance. Fourth, we provide evidence that this RNA complex maintains precursors by binding, localizing and potentially repressing target mRNAs. In particular, we show that Staufen2 knockdown causes mislocalization and enhanced expression of prox1 mRNA, and that expression of a Staufen2 mutant that is unable to bind RNA also causes prox1 mRNA mislocalization and inappropriate neurogenesis. Together, these data support a model where Staufen2, Pumilio2 and DDX1 provide key components of an asymmetrically localized RNA granule that controls the translation of mRNAs that regulate the maintenance versus differentiation of RPs, and thereby development of the embryonic cortex. The impact of this work and the relationship to the companion paper published by Kusek et al. will be further discussed in Chapter 6.

Chapter 4 A Smaug2-based translational repression complex determines the balance between precursor maintenance versus differentiation during mammalian neurogenesis

The majority of the data presented in this chapter, with the exception of Figure 19, was published in November 2015 on The Journal of Neuroscience. 35: 15666-15681.

Gianluca Amadei1,3, Mark A. Zander1, Guang Yang1, Jason G. Dumelie4, John P. Vessey1, Howard D. Lipshitz2,3, Craig A. Smibert3,4, David R. Kaplan1,3 and Freda D. Miller1,3,5

Programs in Neurosciences and Mental Health1 and Development and Stem Cell Biology2, Hospital for Sick Children, Toronto, Canada M5G 1L7, Departments of Molecular Genetics3, Biochemistry4 and Physiology5, University of Toronto, Toronto, Canada M5G 1X5.

Author Contributions:

G.A. performed all aspects of the majority of the experiments presented in this chapter, with the exception of the following: M.Z. performed cell death quantification following Smaug2 knockdown in cultured cortical precursors, quantified cells expressing Nanos1 after Smaug2 and 4E-T knockdown in vivo and helped with initial design of the figures in the paper. G.Y. provided intellectual input and helped with the Smaug2 and 4E-T immunoprecipitations and also provided 4E-T knockdown brains. J.D. performed the computations to calculate the SRE scores of all transcripts expressed in mouse radial precursors at E12.5. J.P.V., H.D.L., and C.A.S. all contributed intellectually to the design and analysis of the experiments. D.R.K. and F.D.M. participated in the conception, design and analysis of the experiments described here and were involved in manuscript writing.

4.1 SUMMARY

Here, we have asked about post-transcriptional mechanisms regulating murine developmental neurogenesis, focusing upon the RNA-binding proteins Smaug2 and Nanos1. We identify, in embryonic neural precursors of the murine cortex, a Smaug2 protein/nanos1 mRNA complex that is present in cytoplasmic granules with the translational repression proteins Dcp1 and 4E-T. We show that Smaug2 inhibits and Nanos1 promotes neurogenesis, with Smaug2 knockdown enhancing neurogenesis and depleting precursors, and Nanos1 knockdown inhibiting neurogenesis and maintaining precursors. Moreover, we show that Smaug2 likely regulates neurogenesis by silencing nanos1 mRNA. Specifically, Smaug2 knockdown inappropriately increases Nanos1 protein, and the Smaug2 knockdown-mediated neurogenesis is rescued by preventing this increase. Thus, Smaug2 and Nanos1 function as a bimodal translational repression switch to control neurogenesis, with Smaug2 acting in transcriptionally-primed precursors to silence mRNAs important for neurogenesis, including nanos1 mRNA, and Nanos1 acting during the transition to neurons to repress the precursor state.

4.2 BRIEF INTRODUCTION AND RATIONALE

During mammalian brain development, embryonic neural precursors must generate the correct numbers and types of neurons and glia without exhausting themselves prematurely (Miller and Gauthier, 2007). The importance of this balance is exemplified by the observation that when neural cell genesis is perturbed genetically in humans, this can cause aberrant development and cognitive dysfunction (for example, see Gauthier et al., 2007; Wang et al., 2010). However, in spite of the fact that regulated cell genesis underlies the establishment of appropriate neural circuitry, the mechanisms controlling neural precursor differentiation are still not well understood. Studies in model organisms such as D. melanogaster have established post- transcriptional regulation as important for cellular differentiation. For example, during D. melanogaster oogenesis and early development, translational repression of unlocalized mRNAs together with active translation of these same transcripts after localization play crucial roles in pattern specification and germ plasm formation (Becalska and Gavis, 2009; Lasko, 2011; Lai and King, 2013). Indeed, in metazoan embryos, bulk degradation of maternal mRNAs may be essential for appropriate spatial and temporal regulation of gene expression (reviewed in Tadros and Lipshitz, 2009; Walser and Lipshitz, 2011). Is translational regulation similarly important in

66 mammalian stem cells? A few recent studies in neural stem cells suggest that it may be. In particular, the RNA binding protein Staufen2 asymmetrically segregates mRNAs associated with neural cell fate decisions (Kusek et al., 2012; Vessey et al., 2012). Moreover, we recently identified a repressive translational complex involving the eIF4E binding protein 4E-T that regulates mammalian neurogenesis by sequestering and silencing mRNAs encoding proneurogenic transcription factors (Yang et al., 2014). However, while these studies indicate that translational repression is important, the molecular players involved in this regulation are largely uncharacterized. For example, while Staufen2 and 4E-T are both part of translational regulatory complexes in mammalian neural precursors, neither protein is thought to be an mRNA-specific repressor protein; Staufen2 is a double-stranded RNA binding protein involved in mRNA transport, localization and stability (Miki et al., 2005; Tosar et al., 2012; Laver et al., 2013; Heraud-Farlow et al., 2013) and 4E-T does not itself bind to mRNAs (Kamenska et al., 2014). Thus, while translational repression is emerging as a key regulatory mechanism in neural precursors, many of the relevant molecular players have not been identified. We have addressed this issue here, focusing on the Smaug and Nanos families of RNA- binding proteins. In Drosophila, Smaug is a key translational repressor of mRNAs such as that encoding Nanos in the bulk cytoplasm of early embryos (Dahanukar et al., 1996; 1999; Smibert et al., 1996; 1999). Furthermore, Smaug functions during the maternal-to-zygotic transition in both the soma and the primordial germ cells of early embryos via translational repression and/or degradation of its target mRNAs (Tadros et al., 2007; Benoit et al., 2009; Siddiqui et al., 2012; Chen et al., 2014a). Smaug does this by binding target mRNAs via stem-loop structures known as Smaug Recognition Elements (SREs) (Aviv et al., 2003; Smibert et al., 1996; 1999; Dahanukar et al., 1999) and recruiting proteins involved in mRNA repression and/or degradation (Chen et al., 2014a). In spite of this broad and essential role for Smaug during D. melanogaster development, virtually nothing is known about Smaug during vertebrate development. Here, we show that murine Smaug2 is a key regulator of mammalian neural development, where it acts to maintain embryonic neural precursors in a stem cell state by binding to and repressing translation of the mRNA encoding a second repressive RNA-binding protein, Nanos1. Smaug2 silences nanos1 mRNA by recruiting it into a repressive Processing (P)-body-like granule in association with 4E-T. Moreover, we show that Nanos1, a known translational repressor, functions to promote the differentiation of precursors into neurons, likely by silencing translation of proteins associated with maintaining the stem cell state.

4.3 RESULTS

4.3.1 Smaug2 is expressed in embryonic cortical precursors during development

To ask whether Smaug is expressed in neural precursor cells, we focused upon embryonic murine cortical radial glial precursors during the neurogenic period from embryonic day 12 (E12) until birth. RT-PCR analysis showed that the mRNAs encoding the mammalian Smaug homologues Smaug1 and 2 were expressed in the embryonic cortex over this time-period (Fig. 13A). Western blots confirmed that Smaug2 protein was also readily detectable (Fig. 13B). However, Smaug1 protein was almost absent at these early stages, and was instead expressed in the adult cortex (Fig. 13C). To determine whether Smaug2 was expressed in embryonic neural precursors and/or neurons, we examined E12.5 cortical cultures, which are comprised of proliferating cortical precursors that generate neurons in vitro. Smaug2 immunoreactivity was present in cycling precursors that were positive for the proliferation marker Ki67, and in newborn neurons expressing βIII-tubulin (Fig. 13D). In both cell types, Smaug2 was expressed in a punctate cytoplasmic fashion, consistent with localization in granule-like structures. Immunostaining of E12.5 embryonic cortical sections showed a similar pattern; Smaug2 was detectable in most cells in the precursor regions of the cortex, the ventricular and subventricular zones (VZ/SVZ), and in the cortical plate (CP), which contains newborn neurons (Fig. 13E). Double labelling for the radial precursor marker Pax6 or the neural precursor markers Sox2 and Nestin confirmed the presence of punctate Smaug2-positive granule-like structures in precursors in vivo (Fig. 13F, G). Fluorescent in situ hybridization (FISH) confirmed that smaug2 mRNA was expressed in Smaug2-immunoreactive precursors in the VZ and SVZ (Fig. 13H).

4.3.2 Smaug2 regulates the genesis of cortical neurons

To understand the function of Smaug2 during cortical development, we performed acute knockdown experiments with shRNAs targeted to Smaug2. Initially we tested the efficacy and specificity of these Smaug2 shRNAs by co-transfecting them into HEK-293T cells together with an expression vector for either mouse Smaug2 or mouse Smaug1. Western blot analysis one day later showed that the Smaug2 shRNAs efficiently knocked down Smaug2 expression, but did not affect expression of Smaug1 (Fig. 13I). We showed that these shRNAs were equally effective in precursors by co-transfecting them into cultured E12.5 cortical precursors together

Figure 13: Smaug2 but not Smaug1 protein is expressed in apical precursors and newborn neurons during embryonic cortical neurogenesis. (A) RT-PCR for smaug1 and 2 mRNAs in the developing murine cortex from embryonic days 11 to 17 (E11-17) and at birth (P0). gapdh mRNA expression was monitored as a control. +ve indicates a sample with known expression of target mRNA and used as a positive control for the reaction and – ve the sample generated in the absence of reverse transcriptase. (B, C) Western blots for Smaug2 (B) and Smaug1 (C) in E11.5 to 2 month old (2mth) cortices. Blots were reprobed for ERK1/2 as a loading control. (D) Images of cortical precursors isolated at E12.5, cultured for 3 days and immunostained for Smaug2 (red) and the proliferation marker Ki67 or the early neuronal marker βIII-tubulin (both green). Arrows indicate a double-labelled cell. Scale bar = 10 μm. (E) Confocal image of a coronal E12.5 cortical section immunostained for Smaug2 (green). Boundaries of the ventricular/subventricular zones (VZ/SVZ) and cortical plate (CP) are denoted. The white v indicates the ventricle. Scale bar = 10 μm. (F, G) Confocal images of the E12.5 cortical VZ/SVZ immunostained for Smaug2 (green) and the radial precursor marker Pax6 (F, red) or the neural precursor markers Nestin (F, red) or Sox2 (G, red) at higher magnification. The lower panels in F were also counterstained with Hoechst to highlight nuclei. Arrows highlight Smaug2-positive granules. Scale bar in F = 10 μm, in G = 5 μm. (H) Confocal images of the E12.5 cortical VZ immunostained for Smaug2 (green) and subjected to FISH for smaug2 mRNA (red). The bottom panel shows the merge. v denotes ventricle. Scale bar = 10 μm. (I) Western blot analysis with anti-FLAG of HEK-293T cell lysates cotransfected with FLAG-tagged mouse Smaug2 or Smaug1 expression constructs and a control shRNA (con) or one of four different Smaug2 shRNAs (sh1, 2, 3 and 4). Cells were transfected with the expression construct alone as a positive control (oe). Blots were reprobed with ERK1/2. (J) Images of cultured precursors cotransfected with a nuclear EGFP construct and a control shRNA (top panel) or Smaug2 shRNA #2 (bottom panel, shSmaug2), and immunostained 3 days later for Smaug2 (red) and EGFP (green). Arrows and arrowheads denote EGFP-positive, Smaug2 positive cells and EGFP-positive, Smaug2-negative cells, respectively. Scale bar = 10 μm. (K) Quantification of transfected precursors as shown in (J) transfected with Smaug2 shRNA1 or shRNA2 (sh1 or sh2) and analyzed for their relative levels of immunodetectable Smaug2. ***p<0.001; n = 3 experiments. In K, statistics were performed with ANOVA and Tukey's post-hoc multiple comparisons test. Error bars indicate S.E.M.

70 with a nuclear EGFP plasmid. Immunostaining three days later showed that the Smaug2 shRNAs, but not a control shRNA caused a robust decrease in the proportion of EGFP-positive cells expressing detectable Smaug2 (Fig. 13J, K). We used these Smaug2 shRNAs to ask about a potential biological role for Smaug2, initially in cultured cortical precursors. E12.5 cortical cultures were co-transfected with one of the two Smaug2 shRNAs and a nuclear EGFP plasmid, and immunostained 3 days later for Pax6, Ki67 or βIII-tubulin (Fig. 14A). This analysis showed that the proportion of EGFP- positive and Pax6-positive proliferating precursors was significantly reduced by Smaug2 knockdown (Fig. 14B,C) while at the same time the proportion of EGFP-positive newborn neurons increased (Fig. 14D). These changes were not due to cell death since the proportion of cells with condensed, apoptotic nuclei was statistically similar between control and Smaug2 knockdown at days 1, 2 (p>0.05 for all comparisons) and 3 (Fig. 14E) post-transfection. These data indicate that in culture, Smaug2 maintains the neural precursor state and prevents neuronal differentiation. To ensure the specificity of these shRNA-dependent phenotypes, we performed rescue experiments using a human Smaug2 construct that did not contain the sequences targeted by the shRNAs. We transfected cultured precursors with EGFP and Smaug2 shRNA with or without the human Smaug2 expression vector. Quantification of Pax6-positive precursors and βIII-tubulin-positive newborn neurons 3 days later showed that coincident expression of human Smaug2 rescued both the decrease in precursors and the increase in neurons caused by Smaug2 knockdown (Fig. 14F, G). To ask if Smaug2 was important for embryonic cortical development in vivo, we next used in utero electroporation to co-transfect E13/14 cortical radial precursors that line the lateral ventricles with Smaug2 shRNA #2 and a nuclear EGFP expression plasmid. These transfected precursors generate neurons either directly, or indirectly (the latter via intermediate progenitor transit amplifying cells), and the newborn neurons then migrate from the VZ/SVZ through the intermediate zone (IZ) to the CP. Three days post-electroporation, coronal sections through the embryonic cortex were analyzed by immunostaining for EGFP (Fig. 14H). This analysis demonstrated that Smaug2 knockdown altered the location of transfected cells; fewer EGFP- positive cells were located in the VZ/SVZ and the CP, with coincidentally more in the IZ (Fig. 14I). We asked about the cellular basis of these alterations by immunostaining similar sections for EGFP and the radial precursor marker Pax6, the intermediate progenitor marker Tbr2, or the neuronal marker Satb2 (Fig. 14J). We chose Satb2 since our previous work showed that more than 90% of neurons born from cortical precursors over this timeframe express this marker (Tsui 71

Figure 14: Smaug2 knockdown in culture and in vivo increases neurogenesis and depletes cycling precursors. (A-E) Cultured cortical precursors were cotransfected with a nuclear EGFP construct and a control shRNA (con) or one of two Smaug2 shRNAs (sh1 and sh2), and analyzed by immunostaining 3 days later. (A) Images of precursors transfected with Smaug2 shRNA #1 and immunostained for EGFP (green) and Pax6, Ki67 or βIII-tubulin (all red). Arrows and arrowheads denote EGFP-positive, marker-positive cells and EGFP-positive, marker-negative cells, respectively. Scale bar = 10 μm. (B-D) Quantification of cultures as in (A) for the percentage of EGFP-positive cells expressing Pax6 (B), Ki67 (C) or βIII-tubulin (D). * p<0.05, ** p<0.01; n = 3 experiments. (E) Quantification of condensed nuclei to assess cell death in cultures as in (A). (F, G) Cultured precursors were cotransfected with a nuclear EGFP construct and a control shRNA (con) or Smaug2 shRNA #2 (shSmaug2) with or without an expression vector encoding a shRNA-resistant human Smaug2 (resc). Three days later cultures were immunostained and quantified for the percentage of EGFP-positive cells expressing Pax6 (F) or βIII-tubulin (G). * p<0.05, ** p<0.01; n = 3 experiments. (H-M) E13/14 murine cortices were co-electroporated with a nuclear EGFP construct together with control (con) or Smaug2 shRNA #2 (shSmaug2), and coronal sections were analyzed three days later at E16/17. (H) Confocal images of electroporated cortices immunostained for EGFP (green). VZ/SVZ = ventricular and subventricular zone, IZ = intermediate zone and CP = cortical plate. v denotes the ventricle. Scale bar = 30 μm. (I) Quantification of sections as in (H) for the percentage of EGFP-positive cells located in the different cortical regions. *p<0.05; n = 3 embryos each, at least 3 sections per embryo. (J) Confocal images of the VZ/SVZ of electroporated sections immunostained for EGFP (green) and Pax6, the intermediate progenitor marker Tbr2, or the neuronal marker Satb2 (all red). Arrows denote double-labelled cells, and arrowheads EGFP- positive, marker-negative cells. Scale bar = 10 μm. (K-M) Quantification of sections as in (J) for the percentage of EGFP-positive cells expressing Pax6 (K), Tbr2 (L) or Satb2 (M). *p<0.05; n = 3 embryos each, at least 3 sections per embryo. In panels B-G, statistics were performed with ANOVA and Tukey's post-hoc multiple comparisons test, and in panels I-M, with Student's t- test. Error bars indicate S.E.M.

et al., 2013). Quantification demonstrated a significant decrease in the proportion of EGFP- positive, Pax6-positive precursors following Smaug2 knockdown (Fig. 14J, K) with a coincident increase in Tbr2-positive intermediate progenitors and Satb2-positive neurons (Fig. 14J, L, M). Thus, as seen in culture, Smaug2 is required in vivo to maintain Pax6-positive precursors and to prevent their differentiation into neurons.

4.3.3 Smaug2 is sufficient to maintain cortical precursors

The preceding data indicate that Smaug2 is necessary to promote Pax6-positive cortical precursor maintenance. To ask if it is also sufficient, we overexpressed Smaug2. Initially, we performed this experiment in culture by cotransfecting E12.5 cortical precursors with a nuclear EGFP plasmid together with a murine Smaug2 expression construct. Immunostaining three days later demonstrated that Smaug2 overexpression increased the proportion of EGFP-positive, Ki67-positive proliferating precursors, and decreased the proportion of EGFP-positive, βIII- tubulin-positive newborn neurons (Fig. 15A-C). These data indicate that in culture Smaug2 is sufficient to promote precursor maintenance. To confirm this conclusion, we performed clonal analysis using the piggybac (PB) transposon system which indelibly marks precursors and their progeny (Gallagher et al., 2013; Tsui et al., 2013). E12.5 cortical precursors were cotransfected at low efficiency with Smaug2 or control expression vectors, together with plasmids encoding PB transposase and a PB EGFP reporter. Three days later, cultures were immunostained for EGFP. Smaug2 overexpression significantly increased the number of multicellular clones (Fig. 15D). To ask if Smaug2 was also sufficient to promote precursor maintenance in vivo, we electroporated E13/14 cortices with an EGFP plasmid and the Smaug2 expression construct, and analyzed coronal cortical sections three days later. Overexpression of Smaug2 caused a significant increase in the proportion of EGFP-positive cells in the VZ/SVZ (Fig. 15E, F). Immunostaining demonstrated that this increase in electroporated cells in the VZ/SVZ was due to an increase in EGFP-positive, Pax6-positive precursors (Fig. 15G, H). The increase in precursors occurred at the expense of differentiation, since there was a coincident decrease in EGFP-positive, Tbr2-positive intermediate progenitors and Satb2-positive neurons (Fig. 15G, I, J). Thus, Smaug2 is sufficient to maintain Pax6-positive precursors and to repress their differentiation into neurons.

Figure 15: Smaug2 overexpression in vitro and in vivo is sufficient to enhance cortical precursor self-renewal. (A-C) Cultured E12.5 precursors were cotransfected with a nuclear EGFP construct and control (con) or murine Smaug2 (Smaug2-OE) expression vectors, and analyzed by immunostaining 3 days later. (A) Images of transfected precursors immunostained for EGFP (green) and Ki67 or βIII-tubulin (both red). Arrows and arrowheads denote EGFP- positive, marker-positive cells and EGFP-positive, marker-negative cells, respectively. Scale bar = 10 μm. (B, C) Quantification of cultures as in (A) for the percentage of EGFP-positive cells expressing Ki67 (B) or βIII-tubulin (C). **p<0.01; n = 3 experiments. (D) Cortical precursor

75 cultures were cotransfected with the piggybac EGFP labelling system plus control (con) or murine Smaug2 (Smg-OE) expression vectors. Cultures were immunostained for EGFP 3 days later and clone size was scored. *p<0.05; n = 3 experiments. (E-J) E13/14 murine cortices were co-electroporated with a nuclear EGFP construct and control (con) or mouse Smaug2 (Smaug2- OE) expression vectors and coronal cortical sections were analyzed three days later at E16/17. (E) Confocal images of electroporated sections immunostained for EGFP (green). VZ/SVZ = ventricular and subventricular zone, IZ = intermediate zone and CP = cortical plate. v denotes the ventricle. Scale bar = 30 μm. (F) Quantification of sections similar to those in (E) for the percentage of EGFP-positive cells located in the different cortical regions. *p<0.05, ns=non- significant; n = 3 embryos each, at least 3 sections per embryo. (G) Confocal images of the VZ/SVZ (two top rows) or CP (bottom row) of electroporated sections immunostained for EGFP (green) and Pax6, Tbr2 or Satb2 (all red). Arrows denote double-labelled cells and arrowheads EGFP-positive, marker-negative cells. v denotes ventricle. Scale bar = 10 μm. (H-J) Quantification of sections as in (G) for the percentage of EGFP-positive cells that were positive for Pax6 (H), Tbr2 (I) or Satb2 (J). * p<0.05; **p<0.01; ***p<0.001; n = 3 embryos each, at least 3 sections per embryo. Statistics were performed with Student's t-test. Error bars indicate S.E.M.

4.3.4 The mRNA encoding nanos1, but not nanos2 or nanos3, is a target of Smaug2 in the embryonic cortex

In Drosophila, Smaug mediates its embryonic function, in part, by binding and repressing the translation of nanos mRNA, which itself encodes a repressive RNA-binding protein. In mammals there are three Nanos homologues, Nanos1, 2 and 3 (Jaruzelska et al., 2003; Tsuda et al., 2003; Lolicato et al., 2008), and RT-PCR analysis showed that mRNAs encoding all three of them were expressed in the embryonic cortex from E11 to birth (Fig. 16A). We asked whether any of these nanos mRNAs might be Smaug2 binding targets by searching for potential SREs using the consensus loop sequence of CNGGN(0-3) flanked by nucleotides that have the potential to form a loop on a 4-nucleotide non-specific stem (Aviv et al., 2003; 2006). These analyses showed that nanos1 mRNA contained five potential SREs (Fig. 16B) with the following sequences: #1) cguccgggggcg starting at base 71; #2) gggcccgggccc starting at base 472; #3) acggccgggugcugu starting at base 911; #4) gcuccaggucugagu starting at base 3029; and #5) ggcacagguugugcu starting at base 3755. In contrast, neither nanos2 nor nanos3 mRNAs contained SREs. A similar result was obtained when we applied a computational approach that includes a thermodynamic-based calculation of the likelihood that an SRE will fold properly, to generate an SRE score for a transcript (Chen et al., 2014a). nanos1 mRNA had an SRE score of 88, similar to the scores for Drosophila Smaug’s two best-characterized target mRNAs Hsp83 and nanos which have SRE scores of 70 and 80, respectively. In contrast, nanos2 and nanos3 had SRE scores of 1.9 and 22, respectively. We therefore focused on nanos1 mRNA as a potential Smaug2 binding target. Western blot analysis showed that Nanos1 protein, like its mRNA, was expressed in the embryonic cortex from E11.5 through to adulthood (Fig. 16C). We asked whether nanos1 mRNA co- purified with Smaug2 in the embryonic cortex. Initially, we confirmed that the Smaug2 antibody was able to immunoprecipitate FLAG-tagged overexpressed Smaug2 from transfected HEK- 293T cells (Fig. 16D). We then used this antibody to immunoprecipitate Smaug2 from the embryonic cortex at E12.5 (Fig. 16E). RT-PCR analysis showed that nanos1 mRNA was present in the Smaug2 immunoprecipitates, but not in control immunoprecipitates (Fig.16E). In addition, as predicted, neither nanos2 nor nanos3 mRNAs were present in the Smaug2 immunoprecipitates (Fig. 16E). These data indicate that nanos1 mRNA, but not nanos2 or nanos3 mRNAs, is associated with Smaug2 in the embryonic cortex. To determine whether this occurs in precursors, we

Figure 16: nanos1 mRNA is a Smaug2 target in embryonic cortical precursors. (A) RT-PCR for nanos1, nanos2 and nanos3 mRNAs in murine cortices from E11 to birth (P0). nanos1 mRNA expression was detected using two different primer sets. PCR products were sequenced to confirm specificity. +ve indicates a sample with known expression of target mRNA and used as a positive control for the reaction and –ve the sample generated in the absence of reverse transcriptase. (B) Schematic of Smaug Recognition Elements (SREs) in the nanos1 mRNA transcript. The yellow arrow labelled CDS denotes the protein-coding region. (C) Western blot analysis for Nanos1 in E11.5 to 2 month old cortices. The blot was reprobed for ERK1/2 as a loading control. (D) Western blot of HEK-293T cells transfected with a Flag-tagged mouse Smaug2 construct and immunoprecipitated with anti-Smaug2 or with control non-specific rabbit IgG, probed with antibodies for Smaug2. As a control, 10% of the input homogenate was loaded. (E) Western blot (top panel) of E12.5 cortical lysates immunoprecipitated with the same Smaug2 antibody as in (D) or with control, non-specific rabbit IgG and probed with anti- Smaug2. As a positive control, 10% of the input homogenate was loaded. Similar immunoprecipitates were generated in parallel, mRNA was extracted, and the samples were analyzed for nanos1, nanos2 and nanos3 mRNAs using RT-PCR (second to bottom panels). (F) Confocal images of FISH for nanos1 (left), nanos2 (center) and nanos3 (right) mRNAs (black granules) in coronal sections of the E12.5 cortex. VZ/SVZ = ventricular and subventricular zones, CP = cortical plate, v = ventricle. Scale bar = 10 μm. (G) Higher magnification confocal images of the VZ/SVZ of an E13.5 cortical section showing FISH for nanos1 mRNA (red) and immunostaining for Smaug2 (green; merge is shown in the top panels). The boxed regions are shown at higher magnification in the right panels, which also show colocalization of Smaug2 and nanos1 mRNA on the Z axis (XZ and YZ), as indicated by the hatched white lines. Scale bar = 10 μm. (H) Confocal images of the E12.5 cortex showing FISH for nanos1 mRNA (magenta) and immunostaining for Smaug2 (green). The VZ/SVZ is divided into five bins of identical width, as denoted by the hatched white lines, and boxed regions within some of these bins are shown at higher magnification in the right panels. The arrows denote foci with colocalized nanos1 mRNA and Smaug2, and the arrowheads foci with only nanos1 mRNA. v denotes the ventricle. Scale bar = 10 μm. (I, J) Quantification of sections similar to that shown in (H) for the distribution of total nanos1 mRNA-positive foci (I) and the relative proportion of nanos1 mRNA-positive foci that colocalize with Smaug2 in each bin (J). * p<0.05; **p<0.01; ***p<0.001; n = 3. (K, L) Quantification of sections similar to those shown in (H) for the proportion of nanos1, nanos2 or nanos3 mRNA foci that colocalize with Smaug2 across the 79 entire E12.5 VZ/SVZ (K) or only in Bin1 (L), the apical-most region of the VZ. * p<0.05; ***p<0.001; n = 3. Statistics were performed with ANOVA and Tukey's post-hoc multiple comparisons test. Error bars indicate S.E.M.

80 performed histological analyses, comparing nanos1, nanos2, and nanos3 mRNAs. Fluorescence in situ hybridization (FISH) demonstrated that all three mRNAs were expressed throughout the E12.5 cortex (Fig. 16F). To ask whether nanos1 mRNA and Smaug2 colocalized during this period of development, we combined FISH with immunocytochemistry for Smaug2. This analysis showed that nanos1 mRNA partially colocalized with Smaug2-positive foci (Fig. 16G). This localization was not exclusive; Smaug2 was present in many granules that did not contain nanos1 mRNA, and the converse was also true, with nanos1 mRNA present in granules that did not contain Smaug2. We confirmed this colocalization with Z-stack analysis (Fig. 16G). Next, we quantified the nanos1 mRNA/Smaug2 foci in the VZ/SVZ of E12.5 cortical sections (Fig. 16H-J). 34 ± 2% of nanos1 mRNA foci colocalized with Smaug2. These complexes were not, however, evenly distributed across the VZ/SVZ. Of all of the nanos1 mRNA-positive foci, the highest proportion was found in the basal-most region of the SVZ, at the border of the newly-formed cortical plate (Fig. 16H, I; Bin 5). In contrast, the highest percentage of nanos1 mRNA-positive foci that were colocalized with Smaug2 was in the apical- most part of the VZ (Fig. 16H, J; Bin 1). In this region, which is predominantly composed of radial precursors, almost 60% of nanos1 mRNA foci were colocalized with Smaug2 (Fig. 16J). Thus, nanos1 mRNA is physically associated with and colocalizes with Smaug2, particularly in apical precursor cells. As a control for the specificity of this analysis, we performed similar studies for nanos2 and nanos3 mRNAs, quantifying the proportion of mRNA-positive foci that colocalized with Smaug2 in E12.5 cortical sections. While approximately 35% of total nanos1 mRNA foci were colocalized with Smaug2, this was decreased to approximately 15% for nanos2 and nanos3 mRNAs (Fig. 16K). Moreover, when only the apical-most radial precursor region of the VZ was considered (Bin 1), approximately 60% of nanos1 mRNA foci versus 15-20% of nanos2 and nanos3 mRNA foci were colocalized with Smaug2 (Fig. 16L).

4.3.5 Nanos1 promotes the genesis of neurons from cortical precursors

These findings define a nanos1 mRNA/Smaug2 complex in embryonic cortical precursors. We therefore next asked about a potential biological role for Nanos1 in cortical development. To do this we generated a Nanos1 shRNA, and verified that it knocked down Nanos1 but not Nanos2 or Nanos3 when co-transfected with the relevant expression constructs into HEK-293T cells (Fig. 17A). We then used this shRNA to assess a functional role for

Figure 17: Nanos1 is necessary and sufficient to promote neurogenesis in vivo. (A) Western blots of HEK-293T cell lysates cotransfected with murine Nanos1 or Flag-tagged murine Nanos2 or Nanos3 expression constructs and a control shRNA (Con) or a Nanos1 shRNA (shNos1) and probed with anti-Nanos1 or anti-Flag, as indicated. The blots were reprobed with ERK1/2 as a loading control. (B-H) E13/14 murine cortices were co-electroporated with a nuclear EGFP construct and either a control (con) or Nanos1 shRNA (shNos1) and coronal sections were analyzed three days later at E16/17. (B) Images of electroporated sections immunostained for EGFP (green). VZ/SVZ = ventricular and subventricular zone, IZ = intermediate zone and CP = cortical plate. v denotes ventricle. Scale bar = 10 μm. (C) Quantification of sections similar to those in (B) for the percentage of EGFP-positive cells located in the different cortical regions. **p<0.01; n = 3 embryos each, at least 3 sections per embryo. (D) Confocal micrographs of the VZ/SVZ (three top rows) or CP (bottom row) of electroporated sections immunostained for EGFP (green) and Pax6, Ki67, Tbr2 or Satb2 (all red). Arrows denote double-labelled cells. v denotes ventricle. Scale bar = 10 μm. (E-H) Quantification of sections similar to those in (D) for the percentage of EGFP-positive cells that expressed Pax6 (E), Ki67 (F), Tbr2 (G) or Satb2 (H). **p<0.01, ***p<0.001; n = 3 embryos each, at least 3 sections per embryo. (I-K) E13/14 cortices were co-electroporated with a nuclear EGFP construct and a control (con) or Nanos1 shRNA (shNos1) plus or minus a shRNA- resistant human Nanos1 expression vector (resc) and coronal sections were analyzed three days later at E16/17. (I) Images of electroporated sections immunostained for EGFP (green). v denotes ventricle. Scale bar = 10 μm (J, K) Sections similar to those in (I) were immunostained for EGFP and Pax6 or Satb2 and the proportion of EGFP-positive cells that were also positive for the marker was quantified. **p<0.01, ***p<0.001; n = 3 embryos each, at least 3 sections per embryo. (L-R) E13/14 cortices were co-electroporated with a nuclear EGFP construct and either a control (con) or murine Nanos1 (Nos1-OE) expression vector and coronal sections were analyzed three days later at E16/17. (L) Images of electroporated sections immunostained for EGFP (green). v denotes ventricle. Scale bar = 10 μm. (M) Quantification of sections as in (L) for the percentage of EGFP-positive cells located in the different cortical regions. *p<0.05, ns = non-significant; n = 3 embryos each, at least 3 sections per embryo. (N) Confocal images of the VZ/SVZ (top three rows) or CP (bottom row) of electroporated sections similar to those in (L) immunostained for EGFP (green) and Pax6, Ki67, Tbr2 or Satb2 (all red). Arrows denote double-labelled cells and arrowheads EGFP-positive, marker-negative cells. v denotes ventricle. Scale bar = 10 μm. (O-R) Quantification of sections as in (N) for the percentage of EGFP- 83 positive cells that were also positive for Pax6 (O), Ki67 (P), Tbr2 (Q) or Satb2 (R). *p<0.05, **p<0.01, ***p<0.001; n = 3 embryos each, at least 3 sections per embryo. In panels J and K, statistics were performed with ANOVA and Tukey's post-hoc multiple comparisons test, and in all other panels with Student's t-test. Error bars indicate S.E.M.

Nanos1 by electroporating it into the E13/14 embryonic cortex. Analysis of coronal cortical sections three days post-electroporation demonstrated that Nanos1 knockdown resulted in an accumulation of transfected EGFP-positive cells in the VZ/SVZ, and fewer cells in the CP (Fig. 17B, C). Consistent with this increase in EGFP-positive cells in the precursor zones, immunostaining demonstrated an increase in transfected Pax6-positive and Ki67-positive cortical precursors following Nanos1 knockdown (Fig. 17D-F). At the same time, EGFP- positive, Tbr2-positive intermediate progenitors and Satb2-positive neurons were decreased by Nanos1 knockdown (Fig. 17D, G, H). To verify the specificity of these phenotypes, we performed rescue experiments with a human Nanos1 expression vector that was resistant to the shRNA. Analysis 3 days later showed that coincident expression of Nanos1 rescued the perturbations in location of EGFP-positive cells observed following Nanos1 knockdown (Fig. 17I), as well as the increase in Pax6-positive precursors and decrease in Satb2-positive neurons (Fig. 17J, K), thereby confirming the specificity of the knockdown phenotypes. These data indicate that Nanos1 normally functions to promote the genesis of neurons from precursors. We therefore asked whether ectopic expression of Nanos1 was sufficient to increase neurogenesis. We co-electroporated plasmids encoding murine Nanos1 and nuclear- localized EGFP into the E13/14 cortex and analyzed coronal sections three days later. Immunostaining showed that EGFP-positive cells were mislocalized following Nanos1 overexpression, with a significant decrease in transfected cells in the VZ/SVZ (Fig. 17L, M). This loss of transfected cells in the precursor zones was due to a decrease in Ki67-positive, Pax6-positive precursors and a coincident increase in Tbr2-positive intermediate progenitors and Satb2-positive transfected neurons (Fig. 17N-R). Thus, Nanos1 depletion and overexpression result in reciprocal phenotypes: loss of Nanos1 results in increased apical precursors and fewer neurons whereas overexpression of Nanos1 results in fewer apical precursors and more neurons.

4.3.6 Smaug2 and nanos1 mRNA are present in RNP granules containing the repressors Dcp1 and 4E-T

These data are consistent with the hypothesis that Smaug2 and Nanos1 mediate opposing functions during the precursor to neuron transition, and that Smaug2 maintains cortical precursors in a stem cell state, perhaps by associating with nanos1 mRNA. In this regard, we recently (Yang et al., 2014) identified a repressive P-body-like RNA granule in embryonic cortical precursors involving the eIF4E binding protein 4E-T which is distantly related to Drosophila Cup, an eIF4E-binding protein that interacts with Smaug and contributes to Smaug-

85 mediated translational repression (Nelson et al., 2004). We therefore asked whether the Smaug2/nanos1 mRNA complex might be associated with the 4E-T repression complex. Initially we asked whether Smaug2 was associated with 4E-T in the embryonic cortex. We immunoprecipitated endogenous Smaug2 from cultured cortical precursors and found that 4E-T co-immunoprecipitated with Smaug2 (Fig. 18A). This association was confirmed by immunoprecipitating 4E-T and showing co-immunoprecipitation of Smaug2 (Fig. 18B). We further assessed association between these two proteins using two additional approaches. First, we immunostained cultured cortical precursors. As we previously reported (Yang et al., 2014), much of the 4E-T immunoreactivity was present in fairly large, punctate cytoplasmic granules (Fig. 18C). Many of these particles were also positive for Smaug2, although much of the Smaug2 was also present in smaller cytoplasmic puncta (Fig. 18C). Second, we used the proximity ligation assay (PLA) which is based upon antibodies binding to protein targets that are within 30-40 nm of each other (Weibrecht et al., 2010). This analysis showed that Smaug2 and 4E-T interacted in multiple bright cytoplasmic puncta in cortical precursors (Fig. 18D). We previously showed that these large 4E-T-positive granules in cortical precursors contained other P-body proteins (Yang et al., 2014). Double-label immunostaining showed that Smaug2 was indeed colocalized with another P-body protein, Dcp1, in large foci (Fig. 18E). Moreover, PLA with antibodies for Dcp1 and Smaug2 showed multiple bright cytoplasmic foci (Fig. 18F), consistent with this colocalization. These data are consistent with the idea that Smaug2 is associated, in part, with 4E-T- containing, P-body-like repressive granules in cortical precursors. We therefore asked whether nanos1 mRNA was also present in these granules. Immunoprecipitation of 4E-T from the E12.5 cortex followed by RT-qPCR demonstrated that nanos1 mRNA was enriched in 4E-T immunoprecipitates relative to controls (Fig. 18G, H). We then asked whether nanos1 mRNA, Smaug2 and 4E-T were all colocalized, performing FISH and immunostaining on cultured cortical precursors. This analysis showed that almost 70% of nanos1 mRNA was colocalized with 4E-T in large, P-body-like granules (Fig. 18I, J). Approximately 45% of these also contained Smaug2 protein, and almost all of the Smaug2/nanos1 mRNA complexes were also associated with 4E-T (Fig. 18I, J). In contrast, only approximately 20% of nanos1 mRNA was not associated with either protein. We also asked whether nanos1 mRNA was colocalized with 4E-T in vivo, performing FISH and immunostaining on E12.5 cortical sections (Fig. 18K). For comparison, we performed similar studies with nanos2 and nanos3 mRNAs. Approximately 40% of nanos1 mRNA foci 86

Figure 18: Smaug2 and nanos1 mRNA are associated with 4E-T in a P-Body-like granule in Pax6-positive apical precursors. (A) Western blot analysis for Smaug2 (Smaug2) and 4E-T in lysates of E12.5 cortical precursors cultured for three days and immunoprecipitated with anti- Smaug2 or with control, non-specific rabbit IgG. As a positive control, 10% of the input homogenate was loaded. (B) Western blot analysis for Smaug2 and 4E-T in lysates of E12.5 cortical precursors cultured for three days and immunoprecipitated with anti-4E-T or with control, non-specific mouse IgG. As a positive control, 10% of the input homogenate was loaded. (C) Confocal images of E12.5 cortical precursors cultured for 3 days and immunostained for Smaug2 (green) and 4E-T (magenta). Cultures were also counterstained with Hoechst (blue). The boxed areas in the top panels are shown at higher magnification in the bottom panels. Arrows denote granules that are positive for both Smaug2 and 4E-T. Scale bar = 5 μm. (D) Confocal images of E12.5 3 day cortical precursor cultures after the Proximity Ligation Assay (PLA) with Smaug2 and 4E-T antibodies. Cultures were also counterstained with Hoechst (blue). The boxed areas in the left panel are shown at higher magnification to the right. Scale bar = 10 μm. (E) Confocal images of E12.5 cortical precursors cultured for 3 days and immunostained for Smaug2 (red) and Dcp1 (green). Cultures were also counterstained with Hoechst (blue). The boxed areas in the top panels are shown at higher magnification in the bottom panels. Arrows denote granules that are double labelled for Smaug2 and Dcp1. Scale bar = 10 μm. (F) Confocal images of cortical precursor cultures after PLA with Smaug2 and Dcp1 antibodies. Cultures were also counterstained with Hoechst (blue). The boxed areas in the left panels are shown at higher magnification on the right. Scale bar = 10 μm. (G) RT-PCR analysis for nanos1 mRNA in 4E-T immunoprecipitates (4E-T IP) from the E12.5 cortex. As a control, similar lysates were immunoprecipitated with a control, non-specific mouse IgG (IgG). (H) RT- qPCR analysis for nanos1 mRNA enrichment in multiple independent 4E-T immunoprecipitates from the E12.5 cortex, in comparison with control IgG immunoprecipitates. (I) Confocal images of E12.5 cortical precursors cultured for 3 days and analyzed by FISH for nanos1 mRNA (red or magenta) and immunostaining for 4E-T or Smaug2 (both green). Cultures were also counterstained with Hoechst (blue). Arrows and arrowheads denote nanos1 mRNA-positive foci that are or are not positive for the relevant protein, respectively. Scale bar = 5 μm. (J) Quantification of cultures as in (I) for the percentage of total nanos1 mRNA-positive foci that also colocalized with Smaug2 (Smaug2) or 4E-T alone, or with both together. *p<0.05, **p<0.01, ***p<0.01; n = 3. (K) Confocal images of the E12.5 cortical VZ immunostained with 4E-T (green) and subjected to FISH (magenta) with a nanos1 mRNA probe shown at low 88 magnification (left) and high magnification (right). Cell nuclei were counterstained with Hoechst (blue), and the merge is shown on the bottom. The boxed regions in the left panels are shown at high magnification to the right. Arrows denote foci positive for both nanos1 mRNA and 4E-T and the arrowheads nanos1 mRNA foci that are negative for 4E-T. v denotes ventricle. Scale bar = 10 μm. (L) Quantification of sections similar to that shown in (K) for the relative proportion of nanos1 mRNA-positive foci that colocalized with 4E-T in each bin of the VZ/SVZ, as defined in Figure 16H. **p<0.01; n = 3. (M, N) Quantification of sections similar to those shown in (K) for the proportion of nanos1, nanos2 or nanos3 mRNA foci that colocalized with Smaug2 across the entire E12.5 VZ/SVZ (M) or only in Bin1 (N), the apical- most region of the VZ. * p<0.05; n = 3. Statistics were performed with ANOVA and Tukey's post-hoc multiple comparisons test. Error bars indicate S.E.M.

89 colocalized with 4E-T in the VZ/SVZ (Fig. 18L, M). These complexes were not, however, evenly distributed, and the highest percentage, approximately 55%, were in the apical-most part of the VZ (Fig. 18L, N; Bin 1). In contrast, many fewer nanos2 and nanos3 mRNA foci were colocalized with 4E-T in the entire VZ/SVZ (Fig. 18M) or in the apical-most region of the VZ (Bin 1; Fig. 18N).

4.3.7 Smaug2 and 4E-T may repress nanos1 with CNOT7, a conserved mammalian deadenylase

In D. melanogaster and S. cerevisiae Smaug has been shown to repress its targets by recruiting the CCR4-NOT deadenylase complex, which performs mRNA deadenylation, thus targeting mRNA for translational repression and decay (Semotok et al., 2005; 2008). The interaction between Smaug and CCR4-NOT deadenylase is mediated by the subunit Caf1 (Zaessinger et al., 2006), which in mammals has two homologues, CNOT7 and CNOT8. To test whether this interaction was conserved, I overexpressed human Smaug2-Flag and human CNOT7-GFP, or human Smaug2-flag and human CNOT8-GFP and immunoprecipitated Smaug2-Flag using a Flag antibody (Fig. 19A, B). Upon Smaug2 immunoprecipitation I observed that human CNOT7-GFP (Fig. 19A) and human CNOT8-GFP (Fig. 19B) co- immunoprecipitated, suggesting that there could be a conserved interaction between mammalian Smaug2 and CNOT7/8. For the subsequent analysis I focused on one of these two proteins, CNOT7. To test the association between Smaug2 and CNOT7 further, I used two additional approaches. First, I immunostained cultured cortical precursors for CNOT7 and Smaug2. Both proteins were present in cytoplasmic puncta of variable sizes and they co-localized in several foci, although this co-localization was not exclusive (Fig. 19C). Second, I used the proximity ligation assay (PLA) which is based upon antibodies binding to protein targets that are within 30-40 nm of each other (Weibrecht et al., 2010). This analysis showed that Smaug2 and CNOT7 interacted in multiple bright cytoplasmic puncta in cortical precursors (Fig. 19D). I then asked whether nanos1 mRNA, Smaug2 and CNOT7 were all colocalized, performing FISH and immunostaining on cultured cortical precursors (Fig. 19E). Quantification showed that approximately 50% of nanos1 mRNA FISH grains were colocalized with CNOT7 (Fig. 19F) and that more than half of these also contained Smaug2 protein. In contrast 10% of the nanos1 mRNA foci associated only with Smaug2 but not CNOT7 and 40% of nanos1

Figure 19: Smaug2, 4E-T and nanos1 mRNA are associated with CNOT7, a conserved mammalian mRNA deadenylase protein. (A) Western blot analysis for human Smaug2-Flag and human CNOT7-GFP expressed in HEK-293T cells for one day and immunoprecipitated

91 with anti-Flag or with control, non-specific mouse IgG. Immunoprecipitation of Smaug2-Flag was verified by probing with mouse anti-Flag and successful co-immunoprecipitation of CNOT7 was verified using rabbit anti-GFP. As a positive control, 10% of the input homogenate was loaded. (B) Western blot analysis for human Smaug2-Flag and human CNOT8-GFP expressed in HEK-293T cells for one day and immunoprecipitated with anti-Flag or with control, non-specific mouse IgG. Immunoprecipitation of Smaug2-Flag was verified by probing with mouse anti-Flag and successful co-immunoprecipitation of CNOT8 was verified using rabbit anti-GFP. As a positive control, 10% of the input homogenate was loaded. (C) Confocal images of E12.5 cortical precursors cultured for 3 days and immunostained for Smaug2 (magenta) and CNOT7 (green). Cultures were also counterstained with Hoechst (blue). The boxed areas in the top panels are shown at higher magnification in the bottom panels. Arrows denote granules that are positive for both Smaug2 and CNOT7. Scale bar = 5 μm. (D) Confocal images of E12.5 3 day cortical precursor cultures after the Proximity Ligation Assay (PLA) with Smaug2 and CNOT7 antibodies. Cultures were also counterstained with Hoechst (blue). The boxed areas in the top panel are shown at higher magnification in the bottom panels. Scale bar = 10 μm. (E) Confocal images of E12.5 cortical precursors cultured for 3 days and analyzed by FISH for nanos1 mRNA (magenta) and immunostaining for CNOT7 or Smaug2 (both green). Cultures were also counterstained with Hoechst (blue). Arrows denote nanos1 mRNA-positive foci that are positive for the relevant protein, respectively. Scale bar = 10 μm. (F) Quantification of cultures as in (E) for the percentage of total nanos1 mRNA-positive foci that also colocalized with Smaug2 (Smaug2) or CNOT7 alone, or with both together. *p<0.05, **p<0.01, ***p<0.01; n = 3. (G) Confocal images of E12.5 3 day cortical precursor cultures after the Proximity Ligation Assay (PLA) with 4E-T and CNOT7 antibodies. Cultures were also counterstained with Hoechst (blue). The boxed areas in the left panel are shown at higher magnification in the right panels. Scale bar = 10 μm. Statistics were performed with ANOVA and Tukey's post-hoc multiple comparisons test. Error bars indicate S.E.M.

92 mRNA foci were not associated with either protein. These data suggest that some of the previously described granules containing 4E-T and Smaug2 may also contain CNOT7. To test whether 4E-T also associates with CNOT7, I performed PLA with 4E-T and CNOT7 antibodies and I observed that also those two proteins interacted in multiple bright cytoplasmic foci (Fig. 19G). Taken together, these results further indicate that the Smaug2/4E-T granule is a repressive one and it contains a well-characterized member of the RNA degradation pathway, CNOT7. Furthermore, this suggests that RNA deadenylation may be an important step in the repression of nanos1 mRNA.

4.3.8 Enhanced neurogenesis following Smaug2 knockdown is caused by derepression of nanos1 mRNA translation

Together, these data suggest that Smaug2 binds and represses nanos1 mRNA in cortical apical precursors, and that disruption of this complex might promote neurogenesis by causing aberrant Nanos1 translation. To test this hypothesis, we in utero electroporated E13/14 cortices with Smaug2 shRNA and nuclear EGFP, and immunostained cortices 3 days later for Nanos1 protein. In control cortices, Nanos1 immunoreactivity was most robust in the CP in newborn neurons (Fig. 20A) although some scattered positive cells with cytoplasmic staining were also observed in the VZ/SVZ. Quantification showed that only approximately 10% of EGFP-positive control cells expressed detectable Nanos1 three days post-electroporation (Fig. 20B, C). In contrast, knockdown of Smaug2 increased the proportion of EGFP-positive, Nanos1-positive cells to almost 30% (Fig. 20B, C), consistent with derepression of nanos1 mRNA. We performed similar experiments knocking down 4E-T in the E13/14 cortex with a shRNA that we previously characterized (Yang et al., 2014), and then analyzing Nanos1 protein-positive cells two days later. This analysis showed that the proportion of EGFP-positive cells expressing detectable Nanos1 doubled following 4E-T knockdown (Fig. 20D, E). These findings are consistent with the idea that a Smaug2/4E-T complex represses nanos1 mRNA translation in apical precursors. Finally, we asked whether the enhanced neurogenesis observed following Smaug2 knockdown might be due to this aberrant derepression of Nanos1 translation. To do this, we electroporated E13/14 cortices with nuclear EGFP and Smaug2 shRNA, either with or without Nanos1 shRNA. Analysis three days later demonstrated that knockdown of Nanos1 substantially rescued the aberrant distribution of EGFP-positive cells that occurred following Smaug2

Figure 20: Knockdown of Smaug2 or 4E-T causes aberrant Nanos1 expression, and this is responsible for the Smaug2 knockdown-mediated increase in neurogenesis. (A) Nanos1 immunoreactivity in a coronal section of the E16/17 cortex. VZ/SVZ = ventricular and subventricular zone, IZ = intermediate zone and CP = cortical plate. v denotes ventricle. Scale bar = 10 μm. (B) Confocal images of cells at the border of the SVZ and the IZ of E16/17 murine cortices that were co-electroporated 3 days earlier with a nuclear EGFP construct and control (con) or Smaug2 (shSmaug2) shRNAs. Sections were immunostained for EGFP (green) and Nanos1 (red). Arrows and arrowheads denote EGFP-positive cells that do or do not express Nanos1, respectively. Scale bar = 10 μm. (C) Quantification of the proportion of EGFP-positive cells expressing detectable Nanos1 in sections similar to those in (B). ***p<0.001; n = 3 embryos each, at least 3 sections per embryo. (D) Confocal images of cells at the border of the SVZ and the IZ of E15/16 murine cortices that were co-electroporated 2 days earlier with a nuclear EGFP construct and control (con) or 4E-T (sh4ET) shRNAs. Sections were immunostained for EGFP (green) and Nanos1 (red). Arrows and arrowheads denote EGFP- positive cells that do or do not express Nanos1, respectively. Scale bar = 10 μm. (E) Quantification of the proportion of EGFP-positive cells expressing detectable Nanos1 in sections similar to those in (D). *p<0.05; n = 3 embryos each, at least 3 sections per embryo. (F- I) E13/14 cortices were co-electroporated with a nuclear EGFP construct and control (con) or Smaug2 shRNA (shSmaug2) plus or minus Nanos1 shRNA (shNos1), and coronal cortical sections were analyzed three days later at E16/17. (F) Images of electroporated sections immunostained for EGFP (green). v denotes the ventricle. Scale bar = 10 μm. (G) Quantification of sections similar to those in (F) for the percentage of EGFP-positive cells located in the different cortical regions. *p<0.05, ** p<0.01, ns=non-significant; n = 3 embryos each, at least 3 sections per embryo. (H, I) Quantification of EGFP-positive, marker-positive cells in sections as in (F) immunostained for EGFP and either Pax6 (H) or Satb2 (I). *p<0.05, **p<0.01; n = 3 embryos each, at least 3 sections per embryo. (J) Schematic showing the proposed repressive complex involving Smaug2, 4E-T, CNOT7, Dcp1 and nanos1 mRNA (top panel). When the complex is disrupted, either by environmental signals, or by knockdown of complex components like Smaug2, this causes aberrant translation of Nanos1, thereby promoting neurogenesis (bottom panel). In panels G-I, statistics were performed with ANOVA and Tukey's post-hoc multiple comparisons test, and in other panels with Student's t-test. Error bars indicate S.E.M.

95 knockdown, with the exception of cell in the cortical plate (Fig. 20F, G), while also rescuing the decrease in Pax6-positive precursors and the increase in Satb2-positive neurons (Fig. 20H, I). These results are consistent with a rescue of the Smaug2 knockdown phenotype by knocking down Nanos1, although we cannot rule out the possibility that nanos1 knockdown is genetically dominant with regard to cellular phenotype. Together, these findings support a model where Smaug2-mediated translational repression regulates expression of Nanos1 (Fig. 20J) and in doing so, ensures appropriate generation of neurons during cortical development.

4.4 CONCLUSIONS

In summary, in the work presented in this chapter I identify a key role for translational repression, showing that two RNA-binding proteins, Smaug2 and Nanos1, regulate the balance of self-renewal versus differentiation of embryonic neural precursors, and in doing so determine the timing and numbers of neurons that are generated. We show that these two repressive proteins do this by functioning in opposition, with Smaug2 maintaining the precursor state and inhibiting differentiation, and Nanos1 promoting differentiation and depleting precursors. Moreover, we show that these antagonistic activities are coordinated, with Smaug2 associating with and silencing nanos1 mRNA, potentially by localizing it to P-body-like granules in association with the translational repressor 4E-T and with the deadenylase CNOT7. Based upon these findings, we propose a translational repression "switch" model for the precursor to neuron transition. In this model, developing neural precursors are transcriptionally primed to generate neurons, but Smaug2 and 4E-T-dependent mRNA repression/silencing maintains them in a stem cell state. In this model, extrinsic proneurogenic cues would disrupt these repressive complexes, thereby releasing the relevant mRNAs and allowing for rapid, precise, and coordinated translation of proteins that promote neurogenesis. This environmentally-driven dissociation of Smaug2 repressive complexes would also derepress nanos1 mRNA, and the newly-translated Nanos1 would then act to repress mRNAs associated with and necessary for the stem cell state. Thus, the RNA-binding proteins Smaug2 and Nanos1 act as a "switch" - when Smaug2 is active or "turned on", this keeps Nanos1 "turned off", thereby maintaining translation of precursor proteins and silencing neuronal translation. However, when Smaug2 is turned off, this turns Nanos1 on, allowing it to silence translation of precursor proteins. Coincident with this Nanos1-dependent silencing of the precursor state, other derepressed proteins, like the neurogenic bHLHs (Yang et al., 2014) would promote

96 establishment of a neuronal phenotype. The impact of this work will be further discussed in Chapter 6.

Chapter 5: Smaug2 RIP-Chip identifies approximately 200 mRNAs associating with Smaug2 in cortical radial precursors in vivo.

5.1 SUMMARY:

Post-transcriptional regulation has been shown to be important for mammalian neural precursor development but much work is needed to identify the mRNAs and RNA-binding proteins that are present in these repressive complexes. Here, I identified mRNAs associating with Smaug2 to investigate additional Smaug2 functions that are independent from nanos1 regulation and to identify additional mRNAs that are jointly regulated by Smaug2 and 4E-T. I show that Smaug2 associates with several mRNAs that may explain the nanos1-independent phenotype observed upon Smaug2 knockdown and that Smaug2 and 4E-T largely associate with different sets of mRNAs, suggesting that multiple repressive complexes may exist in the developing mammalian cortex.

5.2 BRIEF INTRODUCTION AND RATIONALE

Several recent studies have shown that post-transcriptional regulation is an important mechanism for regulating mammalian radial precursor development (Okano et al., 2005; Kusek et al., 2012; Vessey et al., 2012; Yang et al., 2014; Amadei et al., 2015). On the basis of the current evidence, I have proposed a model whereby radial precursors are primed to generate neurons by expression of proneurogenic mRNAs whose translation is prevented by binding to translational repressors. In this model, when these repressive complexes are inherited by differentiating neurons, translational repression of the proneurogenic mRNAs is lifted, allowing differentiation. This model of regulation is appealing, because it would explain how radial precursor cells generate neurons in response to developmental cues without exhausting themselves prematurely, but it also raises many questions. What mRNAs are in these repressive complexes, and what RBPs do they associate with? Do different RBPs cooperate together to regulate neurogenesis? How are the different repressive complexes regulated, and how are they targeted to regulate different aspects of cortical development? Here, I propose to start to answer some of these questions, focusing on the potential repressive complexes that include Smaug2. 98

In my previous work I showed that the RNA-binding protein Smaug2 maintains radial precursors and prevents their early differentiation by repressing nanos1 mRNA in a P-body-like granule with 4E-T (Amadei et al., 2015). Knockdown of Smaug2 caused premature radial precursor differentiation into neurons, mislocalization of these newborn neurons and ectopic expression of Nanos1 protein. Prevention of ectopic expression of Nanos1 was sufficient to rescue the cell fate defect observed upon Smaug2 knockdown, but not the newborn neuron localization defect, suggesting that Smaug2 may regulate additional aspects of cortical development independently from nanos1, possibly by regulating additional mRNAs. What could these mRNAs be? Most of our knowledge of Smaug mRNA targets comes from extensive studies in D. melanogaster, where Smaug has been shown to destabilize a large fraction of unstable maternal transcripts in mature eggs, bulk cytoplasm of the early embryo and primordial germ cells (PGCs) (Tadros et al., 2007; Siddiqui et al., 2012; Chen et al., 2014a). Isolation of Smaug- associated mRNAs in mature eggs by RIP-Chip and their analysis by Gene Ontology showed that they regulate important processes such as the cell cycle, proliferation, DNA replication and protein catabolism (Tadros et al., 2007). Similar studies from D. melanogaster early embryos reported enrichment of Smaug-associated mRNAs for similar GO terms but also additional ones such as protein folding, ubiquitination, lipid droplets and glucose metabolic process (Chen et al., 2014a). In contrast, in D. melanogaster PGCs, the Smaug mRNA targets were enriched for terms including post-transcriptional regulation of germ plasm assembly, control of stem cell division, transcriptional regulation and developmental proteins (Siddiqui et al., 2012). Thus, these studies indicate that in D. melanogaster Smaug does not simply repress nanos mRNA but that it also represses other mRNAs with a wide range of functions, therefore raising the possibility that mammalian Smaug2 may also regulate additional aspects of radial precursor development independently from nanos1 mRNA. In my previous study, I also showed that another important factor involved in the Smaug2-dependent repression of nanos1 is the repressor 4E-T, but whether or not they cooperate to repress multiple mRNA transcripts beside nanos1 is currently an open question. Our laboratory has recently characterized 4E-T function and the mRNA transcripts associating with it in the early radial precursors of the mammalian cortex (Yang et al., 2014). 4E-T regulates many mRNAs important for different aspects of cortical development, such as radial precursor differentiation, proliferation and neuronal development and morphology. However, it is currently not clear how 4E-T recognizes its target mRNAs since it is not thought to interact 99 with mRNA directly (Kamenska et al., 2014). The interaction between Smaug2 and 4E-T on nanos1 thus suggests that 4E-T may utilize Smaug2 to recognize and repress additional mRNA targets. To address these questions, I isolated mRNA transcripts associated with Smaug2 from murine cortical radial precursors at embryonic day 12-13 by performing RIP-Chip with a rabbit antibody specific for Smaug2. Analysis of the high-confidence Smaug2 binders via Gene Ontology (GO) allowed me to identify several mRNAs with known roles in neural cell migration that might be responsible for the nanos1-independent defect in neuronal localization I observe upon Smaug2 knockdown. I also compared the Smaug2-associated mRNAs with those associated with 4E-T and showed that there is a small but significant overlap between the two, suggesting that although 4E-T and Smaug2 may repress some transcripts together, they largely regulate different subsets of mRNAs, possibly by forming different repressive complexes.

5.3 RESULTS

5.3.1 Smaug2 RIP-Chip identifies mRNAs associating with Smaug2 in vivo

To identify mRNAs that associate with Smaug2, I isolated mRNAs that associate with Smaug2 from the E12.5 developing cortex by immunoprecipitation of Smaug2 and analysis of Smaug2-associated mRNAs using an Affymetrix GeneChip Mouse Gene 2.0 ST Array (RIP- Chip). To control for non-specific binding, I performed similar immunoprecipitations using a rabbit control immunoglobulin (IgG). As a quality control measure, I performed Principal Component Analysis (PCA) (Fig. 21A) on all transcripts from each sample studied. PCA indicated that each series of samples that were derived from an individual mouse (hereafter referred to as biological replicates) were different from one another. Such inter-animal differences could result from a variety of confounding variables. For example the variability could be due to small differences in the age of the embryos that were used for the tissue dissections or from slight variability in the tissue dissections themselves. Therefore, given this inter-animal variability, I analyzed these datasets using a pairwise analysis of the Smaug2 versus control IgG immunoprecipitates for each biological replicate. To do so, the IgG and Smaug2 samples for each biological replicate were analyzed as pairs and each gene in the array was analyzed by paired t-test. This analysis allows one to account for inter animal differences in the samples, thus increasing the chance of detecting transcripts which are significantly associating with Smaug2. Following statistical testing, I focused on transcripts that were at least 1.5 fold

100

101

Figure 21: RIP-Chip of endogenous Smaug2 isolated from E12.5 mouse cortices identifies 200 high confidence Smaug2 targets. (A) Principal Component Analysis (PCA) of the mRNA samples generated from total mRNA extraction from the E12.5 mouse cortex, immunoprecipitation with a rabbit Smaug2-specific antibody, or with a non-specific rabbit IgG control analyzed on microarray (RIP-Chip). Each series of colours represents a different, independent biological replicate (blue, green, brown, pink and orange). Each shade of colour represents a sample type. The darkest shade is the total mRNA, the intermediate one is the IgG IP control, and the lightest shade is the Smg2 IP sample. The fifth biological replicate (orange) did not have a matching IgG. The PCA graph is presented from two different perspectives to provide a more faithful representation of the data points in a three-dimensional space. (B) RT- qPCR quantification of selected mRNAs that were significantly associated with Smaug2 (red, orange, pink bars) or with control IgG (blue and green bars) from the RIP-Chip analysis, following paired analysis of the microarray samples. The dashed line indicates a value of 1, where an mRNA was present at equal levels in the Smaug2 and control IgG immunoprecipitates. Values above 1 indicate enrichment for an mRNA in the Smaug2 immunoprecipitates, while values below 1 indicate enrichment in the IgG immunoprecipitate. Error bars indicate S.E.M. *p<0.05, **p<0.01, ***p<0.005; n = 4 experiments. Statistics were performed with the Student's t-test. (C) “Pseudo” heat map of the RT-qPCR data presented in panel (B) for each of the mRNAs in the individual biological replicates. Shades of red indicate that a transcript was enriched in the Smaug2 immunoprecipitate while shades of blue indicate that a transcript was enriched in the IgG immunoprecipitate. The darker the shade, the larger the fold change, in either direction. Green indicates no successful RT-qPCR amplification. (D) Correlation of the fold change (Smaug2/IgG) for each of the candidate mRNAs as measured by paired analysis of the microarray and by the RT-qPCR analysis shown in (B). Data was tested for normality with the Shapiro Wilk test and correlation was tested with the Spearman correlation ρ (rho) (E) Box-plot of the Smaug2 high confidence binders (fold change ≥ 1.5, p- value < 0.01) that also had a calculated SRE score (172 genes) (right box) compared to the SRE score of a matched number of transcripts that were most significantly associated with the IgG (middle box) and to the SRE score of all E12.5 mouse cortex mRNAs for which an SRE score was available (15,558 mRNAs). The top and bottom of the boxes indicate the first and third quartiles, respectively. The horizontal bars in the boxes represent the median. Top whiskers represent the maximum value. ***p<0.005, as determined by a one-way ANOVA with a post- hoc Bartlett’s test for equal variances. (F) Violin plots of the data presented in E to show the 102 probability density of the genes in the set at different SRE values. The black bar indicates the standard deviation and the purple dot indicates the mean.

103 enriched in Smaug2 versus control IgG immunoprecipitates with a p-value < 0.01. This approach identified 197 genes (Table 1- see Appendix).

5.3.2 Smaug2 RIP-Chip was validated by two independent methods

To validate this analysis, I used two different approaches. In the first approach, I performed RT-qPCR for a subset of the mRNAs that fulfilled the criteria for significant enrichment in the Smaug2 immunoprecipitates, analyzing 4 independent RNA- immunoprecipitation experiments. I chose these mRNAs based on the fact that their predicted SRE scores were higher then 90 and based on their roles as either transcription factors or transcriptional cofactors (kdm6b, ccnl2, hes5, xpnpep1, mxd1, btbd9, vgll4), or nucleic acid binders (mex3c, ddx5, huC and samd4b). I decided to focus on these categories of transcripts for two reasons. First of all, a large fraction of mRNAs associating with 4E-T encode transcription factors (Yang et al., 2014) and thus I hypothesized that also in this case transcription factors would be an important group of mRNAs. Secondly, several of these mRNAs have known roles in nervous system development, thus making them interesting candidates to explore nanos1- independent Smaug2 functions. I also chose mark4, which encodes a microtubule binding protein, because it has a high SRE score and because out of the 197 RNAs in the Smaug2 RIP- Chip, it was the one with the largest Smaug2 enrichment. Of the 12 transcripts that were analyzed, 9 were significantly enriched in the Smaug2 RIP (Fig. 21B). One mRNA, mxd, was enriched in 3 out of 4 Smaug2 RIP samples, but was not enriched in the fourth. Finally, 2 out of the 12 genes (kdm6b and mex3c) did not show enrichment in the Smaug2 or IgG RIP (Fig. 21B, C). For a negative control comparison, I performed a similar analysis with 10 mRNAs that were identified as being significantly enriched in the IgG control immunoprecipitation. In this case, the hypothesis is that these 10 mRNAs should not be enriched in the Smaug2 RIP. To choose these 10 mRNA transcripts I relied upon enrichment in the control IgG as described above, as well as upon the fact that they had low SRE scores. This analysis demonstrated that 2 out of 10 mRNAs showed significant enrichment in the IgG, while 8 out of 10 were present at statistically similar levels in the control IgG and Smaug2 immunoprecipitates. 2 out of these 8 (efcab11 and gypa) have a fold change larger than 1 because in one of the RT-qPCR replicates they were enriched in the Smaug2 RIP (Fig. 21B,C). A comparison of the relative fold enrichment as observed by microarray and by RT-qPCR indicated a significant correlation

104 between these two approaches (Spearman correlation coefficient: 0.78, p = 0.0001) (Fig. 21D) thus giving me confidence that the paired analysis of the IgG and Smaug2 RIP-Chip samples allowed for faithful prediction of mRNA transcripts that associated with Smaug2. As a second approach to validate my microarray analysis, I utilized a computational approach that takes advantage of the well-defined Smaug binding preference. As discussed in the introduction, Smaug binds to and regulates its target mRNAs through SRE stem loop structures that are, like Smaug itself, conserved from yeast to humans (Aviv et al., 2003; 2006). I therefore predicted that mRNAs bound by Smaug2 should be enriched for SRE sequences. To calculate this enrichment, we applied a computational approach that includes a thermodynamic- based calculation of the likelihood that an SRE will fold properly (Chen et al., 2014a), to generate an SRE score for nearly every transcript expressed in E12.5 cortical radial precursors (15558 transcripts). I then assigned the calculated SRE for each mRNA transcript that was significantly enriched in the Smaug2 RIP (Table 2). If a given transcript on the Smaug2 list had no calculated SRE score, it was removed from the list. Of the 197 mRNA transcripts that were significantly enriched in the Smaug2 immunoprecipitates, 172 of them had calculated SRE scores. Of the remaining 25 mRNAs, only 5 encoded known proteins, and the remainder were miRNAs (4), long non-coding RNAs (2), small nucleolar RNAs (2) and sequences for which a definitive annotation was not available (11). As a comparison, I selected an equal sized control list of mRNA transcripts that were enriched in the IgG RIP. To select these, I ranked the transcripts based on fold-enrichment with respect to the IgG, and selected the 172 that were most enriched in the IgG and also had a calculated SRE score (Table 2). I then compared the average SRE scores for the Smaug2-enriched list, the IgG-enriched list and for all mRNA transcripts that were expressed in the E12.5 cortex that were annotated with an SRE score (15558 transcripts, Input column) (Fig. 21E). This comparison demonstrated that the average SRE scores were statistically significantly different between these different groups, with the Smaug2-associated mRNAs having the highest score (Smaug2, 74.6; Input, 37; IgG, 19.2), indicating that the Smaug2 immunoprecipitates were enriched for mRNA transcripts with higher SRE scores. These data were also represented using violin plots, which indicate the probability of finding transcripts at a given SRE value (Fig. 21F). This analysis showed that the mRNA transcripts enriched in the Smaug2 RIP were consistently distributed towards higher SRE scores, unlike those from the IgG RIP or the input samples. Thus, the computational approach

105

Table 2: SRE score of mRNA transcripts expressed at E12.5 in radial precursors, of mRNAs significantly associated in the control IgG RIP and of mRNAs significantly associated in the Smaug2 RIP (Full table in the attached file) SRE scores were calculated by applying a computational approach that includes a thermodynamic-based calculation of the likelihood that an SRE will fold properly (Chen et al., 2014a). In this table, mRNAs expressed in E12.5 cortical radial precursors and for which an SRE score is also calculated are reported with the corresponding SRE score (15558 transcripts). Similarly, mRNA transcripts that were significantly enriched in the Smaug2 RIP and that had an SRE score are reported. If a given transcript on the Smaug2 list had no calculated SRE score, it was removed from the list. Of the 197 mRNA transcripts that were significantly enriched in the Smaug2 immunoprecipitates, 172 of them had calculated SRE scores. As a comparison, an equal sized control list of mRNA transcripts that were enriched in the IgG RIP was selected by ranking the transcripts based on fold-enrichment with respect to the IgG, and selecting the 172 that were most enriched in the IgG and also had a calculated SRE score. For each mRNA in the table, the SRE score is reported.

Sm2 Input SRE scores IgG transcripts SRE scores transcripts SRE scores Macf1 489.9980952 1110058L19Rik 0.143809524 0610011F06Rik 23.91809524 Tbc1d16 454.5695238 1110059G10Rik 0 2410015M20Rik 0.444761905 Ubr4 435.752381 2610524H06Rik 0 2610507B11Rik 154.4685714 Bsn 409.0028571 Acot13 0.022222222 A430005L14Rik 4.492380952 Zfp651 380.1409524 Adam10 3.526666667 Abhd17a 69.77428571 Elavl3 371.9628571 Ammecr1 147.5895671 Abhd4 28.0747619 Acacb 371.9542857 Anapc10 116.2142857 Acox1 47.62952381 Lrp4 369.0971429 Anapc16 21.14857143 Ahsa1 11.78857143 Dstyk 365.9709347 Ankrd49 28.12095238 Ambra1 169.76 Dst 361.372381 Ap3s1 0.08 Ap1ar 4.48 Cux2 360.5142857 Apoa1bp 0.001904762 Arhgap1 15.56380952 Xirp1 356.9223214 Arl5a 22.90952381 Arhgap11a 42.76380952 Plec 355.8409524 Atp5e 0 Arhgap33 105.3304762 Kcnq2 351.0314286 Atp6v1d 2.963809524 Arhgap42 102.5161905 Kif3c 349.2857143 Atp6v1e1 0.244761905 Arpc5 3.347619048 Col7a1 345.4057143 Brk1 0.112380952 Ash2l 69.25428571 Cep250 344.584 Bzw2 6.988181818 Asphd2 3.726666667 Gucy2e 341.6340548 Calr 2.104761905 Atf7 90.07770723 Plekhm2 341.2697619 Casc4 16.94285714 Atxn7l2 167.8866667 Samd4b 340.1266667 Casp3 0.091279762 BC037034 147.9761905 Sspo 336.3694931 Ccdc124 0 Bet1l 13.23809524 Acan 333.6504762 Ccdc127 66.12190476 Braf 81.54857143

106

Agap2 332.9152381 Ccdc18 9.443809524 Cacfd1 19.71619048 Myh10 332.26 Ccp110 17.72666667 Camkk2 162.567619 Grik5 325.7485714 Cdh2 8.38 Camsap3 132.6085714 Akap13 324.0533333 Cdk6 0.286666667 Cbfb 127.5138961 Myo10 323.5009524 Cdo1 2.115238095 Ccnl2 148.5190476 Bptf 322.2422969 Cenpk 0.085714286 Cdk5 0.681904762 Mga 322.1066667 Cenpl 20.29619048 Cds2 263.1914286 Tbc1d8 321.94 Cetn2 0 Cfdp1 6.755238095 Sfxn5 321.2238095 Cggbp1 5.778095238 Cnih2 48.9352381 Otog 314.2022222 Chic1 146.4295238 Cnnm3 26.69809524 Adora1 314.0992857 Chmp2a 97.75714286 Cntfr 79.38190476 Obscn 311.4714286 Chmp6 79.72095238 Copb2 56.79010582 Mef2c 310.7580952 Cln8 94.32571429 Cryz 97.82857143 Pitpnm2 306.5209524 Clvs1 22.28190476 Cst3 41.37692641 Herc2 306.1469841 Cmc2 3.488571429 Ctage5 11.90761905 Dlec1 305.9314286 Commd9 22.67428571 Ctps 4.676190476 Bcl6b 301.5266667 Cops4 3.812380952 Dclk2 121.1209524 Ptpn14 301.0542857 Cox6b2 0.011428571 Ddx19a 13.46380952 Shank3 298.7857143 Cpa2 0 Dhps 7.726666667 Kndc1 293.0466667 Creb5 7.019047619 Dnpep 0.326666667 Prpf40b 292.7771726 Crls1 3.488571429 Dusp8 122.7 Fras1 290.2447619 Csnk1g3 6.576190476 Dvl3 87.67333333 Lrp1b 288.4380952 Cstb 0 Ehmt2 29.17904762 Unc5b 287.2590476 Cyb5b 0.045714286 Elavl3 371.9628571 Ttl 285.9029228 Dck 79.41238095 Elp3 34.17428571 Celsr1 285.1657143 Ddx3y 31.82285714 Enc1 115.3790476 Maml1 283.0857143 Ddx6 2.26952381 Epha4 220.8761905 Neurl1a 282.0047619 Derl2 2.025714286 Fam102b 78.61047619 Dnah11 281.9933333 Ebf1 14.14380952 Fasn 210.5304762 Evpl 281.2514286 Efcab11 44.92285714 Fbxw5 46.36285714 C77080 281.1228571 Elavl1 2.834285714 Fchsd2 111.44 Nos3 280.2533333 Elovl5 72.52666667 Fndc4 28.8352381 Pcdh12 277.1819048 Exoc4 15.8447619 Fryl 156.8371429 Dnah5 275.147619 Fabp5 0 Fzd8 60.42190476 Sh3pxd2b 274.7914286 Fbxl5 15.91809524 Fzr1 222.1102381 Scarf2 273.5190476 Fign 47.47428571 Gde1 49.72571429 Lpgat1 272.832381 Flywch2 0 Gid4 80.19809524 Pip4k2c 272.7514286 Gm15127 0.419047619 Git1 49.76 Dlc1 269.9 Gm561 0 Glul 83.54095238 Kmt2b 266.5704762 Gng5 16.64666667 Gm5617 3.148571429 Dusp13 266.3285714 Gpx4 2.401904762 Gmeb2 172.8742857 Ltn1 265.94 Gypa 0 Gpbp1l1 12.03333333 Lphn1 263.4304762 H2afx 10.17333333 Gprin1 132.0114286 Btbd9 263.3580952 Has1 132.8093333 Gramd1a 153.4209524 Cds2 263.1914286 Haus1 0 Grk6 121.0542424 Mapkbp1 263.1142857 Hcfc1r1 0 Grtp1 1.372380952 Lrrn2 262.9247619 Hnrnpll 82.2 Gsto1 0.176190476

107

Ppm1h 262.8230252 Hsf2 7.800952381 Hes5 148.02 Morc2a 262.767619 Isca1 1.894444444 Ing2 1.057142857 Slc30a1 261.6514286 Isg20l2 53.14857143 Ipmk 149.4324466 Zfp407 260.1847619 Itfg1 0.393333333 Kdm6b 167.7552381 Chrm1 259.5695238 Klhl13 0.421904762 Kpna1 7.003809524 Cdh23 259.2304762 Kras 4.285714286 Kpna6 95.36761905 Ppp1r18 258.8529004 Krr1 9.194285714 Lfng 152.9485714 Sema6b 257.9961905 Laptm4b 28.38571429 Lrrc1 206.387619 Ap4e1 257.9580952 Larp4 84.1952381 Lyrm4 16.88333333 Dscam 256.7666667 Lrrn1 0.397142857 Mark4 132.3438095 Arhgap27 256.5685714 Lsm1 15.9152381 Mcm3 124.5333333 Ep300 255.5514286 Lsm5 0.095238095 Mcmbp 0.232380952 Camta1 254.9171429 Ly96 0 Mcrs1 78.31619048 Gata4 252.9628571 Lyplal1 36.54285714 Med24 56.17428571 Mdn1 252.9133333 Lztfl1 94.24666667 Mex3b 33.16952381 Dnaaf2 251.9266667 Magohb 3.298095238 Mex3c 123.2604365 Pvrl4 251.5657143 Med9 15.24952381 Mgat4b 52.3152381 Smg1 251.0952381 Mmgt1 0.100952381 Mlx 68.72190476 Ephb4 250.7771429 Mpc1 0.04 Mrpl45 46.15238095 Il2rb 250.712381 Mrpl22 0.000952381 Msh2 59.8184127 Nlgn2 250.5657143 Naa40 64.02 Mthfd2 30.79142857 Usp35 249.1895238 Nap1l2 1.098095238 Mxd1 109.6520238 Pnpla1 248.9819048 Ndufa2 0 Mxi1 2.510476191 Olfr1380 248.4095238 Ndufa6 0.010476191 Mxra8 10.94860029 Ccdc88b 246.8171429 Ndufaf3 5.513333333 Myh10 332.26 Wipf1 244.2022711 Ndufb6 20.55619048 Ncan 162.3895238 4932438A13Rik 244.1571429 Ndufb7 0.000952381 Nelfe 12.57619048 Zfp697 244.1238095 Ndufs4 0 Nhlh1 159.34 Cct2 243.32 Nhlh2 89.4447619 Nhp2 9.454285714 Ndst3 242.2571429 Nop56 0.002857143 Nol11 168.3638095 Asxl2 242.223208 Nudcd3 77.89238095 Nt5dc2 8.135238095 Adcy5 242.1542857 Olfr543 0.666666667 Ormdl1 66.81047619 Zfp191 241.8 Pdzd8 15.40380952 Osbpl2 1.734285714 Rnf40 241.3609524 Peli1 0.443809524 Paxip1 54.14952381 Plin4 240.3866667 Pgls 3.766666667 Pddc1 64.95238095 Dnah10 238.6047619 Pkia 7.282857143 Pgrmc2 7.250476191 Apc2 237.4647619 Ppbp 0 Phf11a 7.996190476 Hic2 237.3942857 Prkaa2 5.712380952 Phtf1 10.78095238 Klhdc7a 237.312381 Prkacb 4.586666667 Pola2 33.42571429 Kifc2 236.232381 Prkag1 0.01070028 Ppap2b 33.97710637 Ptprg 234.9380952 Ptn 22.34571429 Ppil2 3.693333333 Ppl 234.1904762 Pygo1 69.16571429 Ppp1r26 90.78095238 Rfx1 233.852381 Rab1 0.015238095 Ppp6r2 22.62380952 Sec14l1 233.8285714 Rab11b 132.9733333 Prdm8 79.16190476 Muc5b 233.34 Rab39b 0.155238095 Prkcd 75.45809524 Pkd1 233.3342243 Rab6b 0.86 Prpf6 84.4047619 Gmip 232.6865256 Rb1 2.178095238 Ptov1 3.925714286

108

Atp2a3 232.2704762 Rbpj 0.253333333 Rab3d 86.74 Aars 232.0333333 Rbpms 5.763809524 Rcan3 228.8514286 Chst3 231.7304762 Reep3 99.75904762 Rfc2 14.68761905 Zfp335 231.1695238 Rnf2 60.74190476 Rhot2 208.3152381 AI317395 230.1628571 Rpl14 0.279047619 Rnf115 11.5952381 Rcan3 228.8514286 Rpl35 8.397142857 Rnf19b 63.83142857 Rexo1 228.8190476 Rpl6 31.9847619 Rnf31 29.84952381 Bbc3 228.6809524 Rpp30 0.596190476 Rps6ka2 95.90779762 Dctn1 228.6209524 Rufy3 13.12857143 Samd4b 340.1266667 C2cd2l 228.3771429 Ryk 12.14761905 Samm50 0.234285714 Usp31 227.6121429 S100a6 11.59809524 Sh3bp1 149.1371429 Vars 227.5257143 Saysd1 78.13829365 Sh3gl1 2.248571429 Ush2a 227.4380952 Scd1 50.51714286 Slain1 122.1114286 1700037C18Rik 226.6714286 Selt 12.75714286 Slc25a22 44.14857143 Lzts3 226.1107937 Serinc3 34.63142857 Slc35e3 81.51142857 Dnah17 225.6180952 Shoc2 0.231428571 Slc43a1 17.33333333 Anapc1 225.0333333 Siah3 0.410476191 Smarcd2 51.55809524 Rsad2 224.9838095 Slain2 1.635036075 Snapin 41.84857143 Celsr3 224.827619 Slc35f1 15.38571429 Snf8 54.53047619 Agtpbp1 224.7752381 Snrnp35 0.010476191 Snn 21.91047619 Spint2 224.7152381 Snx24 17.32095238 Soat1 114.3980952 Tmem2 224.4564583 Spata24 1.010476191 Socs7 91.6152381 Atm 224.3819048 Srm 18.24571429 Sphk2 140.5847619 Sorbs2 224.2638095 Sst 10.03809524 Spop 23.64190476 Mon1b 222.9721612 Stau1 41.84190476 Ssh1 182.2428571 Klf13 222.3742857 Stx7 1.966451613 St5 143.4266667 Fzr1 222.1102381 Taf7 3.170476191 St8sia2 78.70190476 Lca5l 221.8667725 Taok1 82.68095238 Suds3 2.600952381 Fbxo41 221.8419048 Tax1bp1 23.02857143 Sult4a1 22.22666667 Zfp518b 221.7828571 Tbca 0.104761905 Supt6 105.0228571 Kif1c 221.5285714 Tcf7l2 12.97142857 Taf1d 43.78694444 Kctd15 221.432381 Tm2d2 3.136190476 Tbc1d22b 33.10952381 Epha4 220.8761905 Tmem184c 4.722640693 Tbcb 83.31047619 Mthfr 220.212381 Tmem192 3.744761905 Tgfbrap1 36.60380952 Emc1 219.987619 Tmem30a 1.22952381 Tle4 24.50095238 Ttc39c 219.9771429 Trappc6b 8.766666667 Tmem50b 22.36666667 Sh3rf1 219.8190476 Tspyl1 36.40857143 Tmem70 73.1047619 Stxbp5l 219.73337 Ttc9b 3.192592593 Tomm34 83.13714286 Ky 218.212381 Twsg1 22.5552381 Traf4 36.82215303 Lemd2 218.1895238 Ube2j1 31.40695971 Trappc10 23.55142857 Mlec 218.1571429 Vbp1 0 Tsn 126.0460504 Prkdc 217.3714286 Vps4a 9.978095238 Tsnax 41.2152381 Gse1 217.2866667 Vta1 0.007619048 Tstd1 1.386666667 Zfp704 216.9133333 Wapal 16.4247619 Tubb6 3.552380952 Cwh43 216.6378571 Xrcc6bp1 17.09619048 Uaca 53.18952381 Slc43a3 216.4942857 Ybx3 1.044761905 Ube4b 143.08 Scube3 215.8609524 Zc3h7a 22.16571429 Unc119b 57.71428571

109

Abcg4 215.7942857 Zcchc17 1.178095238 Unc13b 8.816190476 Cenpf 215.792381 Zfand6 0.000952381 Vgll4 27.4047619 Golga3 215.7914286 Zfp266 4.273333333 Vmn1r119 5.78 Gfod2 215.4295238 Zfp275 3.646666667 Wdr59 19.49619048 Engase 215.0371429 Zfp467 8.586666667 Xpnpep1 127.468381 Rin1 214.3495238 Zfp781 0.030476191 Xpot 83.82095238 Pvrl1 214.067619 Zfp846 0.024761905 Zbtb41 62.96082102 Dync1h1 213.5685714 Zfp937 12.36857143 Zfp36l1 122.1542857 Plxdc2 213.4419048 Zic1 3.734285714 Zswim5 55.44285714

110 and the RT-qPCR analysis validated the enrichment for Smaug2-associated mRNAs in the Smaug2 RIP-Chip.

5.3.3 Ingenuity Pathway Analysis highlights several Smaug2-associated mRNAs with known roles in migration

To look for candidate mRNAs that could be involved in the neuronal localization defect I observed upon Smaug2 knockdown, I performed Ingenuity Pathway Analysis (IPA) on the Smaug2 list (Table 3). I chose to report terms related to nervous system development, cell cycle, gene expression and cellular processes with a p-value < 0.05. Categories which were excluded were the ones with very few genes (5 or less), terms related to infection, disease or abnormal development, and categories that were redundant or fully overlapping with other categories already displayed. For the selected categories, p-value, number of genes and identity of the genes are reported. Interestingly, one of the terms identified was “migration of brain cells” and it contains several mRNAs with known roles for migration in the nervous system and could thus be involved in the neuronal localization defect observed following knockdown of Smaug2. However, this is not the only interesting category in this list because several others are very relevant for radial precursor development and neuronal development and could provide interesting targets to investigate additional Smaug2 roles during, for instance, neuronal maturation.

5.3.4 PANTHER and DAVID analysis of the Smaug2-associated transcripts hints at additional Smaug2 roles in cortical development

To look for additional terms of enrichment that may have been missed by IPA, I analyzed the Smaug2 high confidence list with the “PANTHER Classification System”. (Protein ANalysis THrough Evolutionary Relationships). PANTHER classifies proteins and their genes according to groups of evolutionarily conserved functions (Families), or according to Molecular Function, Biological Process, or Pathway (Muruganujan et al., 2003). Of the 197 Smaug2- associated mRNAs, 172 mapped to known genes in the database but only 100 of these were annotated in the PANTHER database. I therefore restricted my analysis to these 100 mRNAs. Analysis of the list via “protein class” demonstrated significant enrichment for 5 terms: nucleic acid binding (34 genes), transcription factors (17 genes), kinases (11 genes), ligases (7 genes) and enzyme modulators (17 genes) (Table 4). An interesting subset of enzyme modulators

111

Table 3: Ingenuity Pathway Analysis (IPA) of the Smaug2 associated transcripts. Smaug2- associated transcripts were analyzed with the “Disease and Functional Annotation” tool of IPA with the default settings. The categories reported are those involving nervous system development, cell cycle, gene expression and cellular processes, with a p-value < 0.05. Categories which were excluded were the ones with very few genes (5 or less), terms related to infection, disease or abnormal development, and categories which were redundant or fully overlapping with other categories already displayed. For the reported categories, p-value, number of genes and identity of the genes are reported.

Categories Functions Annotation p-value Molecules # Cell Morphology, BRAF,CAMKK2,CDK5, Nervous System CNTFR,GIT1,HES5,MARK4, Development and morphology of neurons 9,89E-04 Mir124a- 13 Function, Tissue 1hg,MYH10,SH3GL1, Morphology ST8SIA2,UBE4B,UNC13B ARHGAP1,BRAF,HES5, Embryonic size of embryo 9,69E-03 IPMK,LFNG,MTHFD2,Paxip1 9 Development ,TRAF4,ZFP36L1 Cell Cycle, DNA Replication, FBXW5,KPNA1,MSH2, DNA recombination 9,30E-04 7 Recombination, and PRPF6,RFC2,SUPT6H,TSN Repair Embryonic Development, Nervous System Development and BRAF,CDK5,EPHA4,HES5, Function, Organ formation of forebrain 9,30E-03 7 MYH10,PRDM8,ST8SIA2 Development, Organismal Development, Tissue Development Cellular Function and CDK5,MYH10,PRKCD, Maintenance, exocytosis 8,97E-03 6 RAB3D,SNAPIN,UNC13B Molecular Transport CBFB,MCRS1,MXD1, Gene Expression repression of RNA 8,24E-03 6 MXI1,SPOP,SUDS3 AHSA1,AMBRA1,ARHGAP1, ARHGAP33,ASH2L,BRAF, CACFD1,CAMKK2,CBFB, CCNL2,CDK5,CFDP1,CNTF R, COPB2,CST3,CTPS1,DHPS, DUSP8,DVL3,EHMT2,ENC1, Cellular Growth and proliferation of cells 5,57E-03 EPHA4,FASN,FZD8,FZR1,GI 58 Proliferation T1, GLUL,GPRIN1,HES5,ING2,I PMK, MCM3,MCRS1,MGAT4B,mir- 467, MSH2,MXD1,MXI1,MYH10,N CAN,

112

NELFE,OSBPL2,Paxip1,PRK

CBFB,CDK5,ENC1,FBXW5, Post-Translational MED24,Paxip1,PPIL2,RNF11 ubiquitination of protein 4,25E-04 11 Modification 5, RNF31,SOCS7,UBE4B BRAF,CDK5,CST3,DCLK2, Nervous System ELP3,EPHA4,HES5,IPMK, development of central Development and 4,21E-03 KDM6B,MYH10,NHLH1, 15 nervous system Function PRDM8,SH3GL1,SPHK2, ST8SIA2 ASH2L,ATF7,BRAF,CAMKK2 CBFB,CDK5,DCLK2,DVL3,E HMT2, ELP3,FZD8,GMEB2,HES5,IN G2, KDM6B,KPNA6,MED24,mir- Gene Expression transcription 3,60E-02 467, 32 MLX,MXD1,MXI1,NELFE,NH LH1, PRDM8,PRKCD,PRPF6,RPS 6KA2, SNF8,SPOP,SUDS3, TGFBRAP1,TLE4 Cellular ARHGAP33,BRAF,CAMKK2, Development, CDK5, Cellular Growth and FZD8,FZR1,GPRIN1,MYH10, proliferation of neuronal Proliferation, Nervous 3,47E-02 NCAN, 11 cells System Development PRKCD,UBE4B and Function, Tissue Development DNA Replication, AP1AR,CST3,EHMT2, Recombination, and DNA replication 3,45E-02 MCM3,Paxip1,POLA2 6 Repair FZR1,MCM3,MXD1,MXI1, Cell Cycle S phase 1,44E-02 PRKCD,PTOV1,RNF31 7 CDK5,MYH10,PRKCD, Cellular Movement migration of brain cells 1,14E-04 SOCS7,ST8SIA2,TSNAX 6

113

Table 4: PANTHER GO enrichment analysis of the Smaug2-associated mRNAs. Enrichment analysis for GO terms of the Smaug2 list was performed with the PANTHER “Protein Class” online tool with the default settings. Genes were analyzed for enrichment against the background of expressed genes in radial precursors at E12.5 identified from the 5 input/total mRNA samples. Terms with p-value < 0.05, were considered significant and reported. For each term, the p-value and the associated genes are reported. In “Nucleic Acid binding”, genes highlighted in red are transcription factors, genes highlighted in blue are RNA-binding proteins and genes highlighted in purple have both of those annotations. Under “transcription factors”, genes highlighted in green are the ones with an RNA-binding protein annotation also. Under “enzyme modulators”, genes highlighted in yellow are G-protein modulators.

Panther Protein Class Enzyme Nucleic Acid Transcription Kinase Ligase Modulator (p = Binding (p = 0.013) Factor (p = 0.047) (p = 0.0027) (p = 0.011) 0.047) Ash2l Atf7 Braf Ctps1 Ahsa1 Atf7 Ccnl2 Camkk2 Fasn Arhgap1 Ccnl2 Hes5 Cdk5 Gde1 Arhgap11a Ddx19a Ing2 Dclk2 Glul Arhgap42 Ehmt2 Mlx Grk6 Mex3b Ccnl2 Git1 Mxd1 Ipmk Mthfd2 Cst3 Gsto1 Mxi1 Mark4 Rnf19b Dvl3 Hes5 Nhlh1 Prkcd Fzr1 Ing2 Pddc1 Rps6ka2 Git1 Mcm3 Phtf1 Smok4a Grtp1 Mex3b Prdm8 Sphk2 Myh10 Mlx Ptov1 Ppp6r2 Msh2 Spop Rhot2 Mxd1 Supt6 Sh3gl1 Mxi1 Tle4 Socs7 Nhlh1 Xpnpep1 Tbc1d22b Nhp2 Zbtb41 Ube4b Pddc1 Phf11a Phtf1 Pola2 Prdm8 Prpf6 Ptov1 Rfc2 Slc25a22 Smarcd2 Spop Supt6

114

Tle4 Xpnpep1 Xpot Zbtb41 Zfp36l1

115 comprised G-protein modulator (8 genes, p-value = 0.023, genes highlighted in yellow in Table 4). The category “nucleic acid binding”, besides the transcription factors, included 6 RNA binding proteins (Table 4, green highlights). Two genes in the transcription factors class (Ccnl2; Pddc1) also had an RNA-binding function. The remaining 11 genes encoded 2 histone methyltransferases, Ash2l and Ehmt2, 1 GTPase activating protein, Git1, 1 DNA helicase, Mcm3, a DNA repair protein, Msh2, 2 ribosomal proteins, Nhp2 and Slc25a22, a DNA polymerase subunit, Pola2, 1 replication factor Rfc2 and a chromatin-binding protein, Smarcd2. Phf11a, despite being listed as a nucleic acid binder, is not very well-characterized. In contrast to the nucleic acid binding protein category, the ligase, kinase and enzyme modulator categories all contained distinct mRNAs (Table 4). Analysis of the list under the PANTHER "molecular function" category yielded essentially the same terms, with the addition of RNA binding (p- value = 0.011) and protein binding (p-value = 0.0038). To confirm these enrichment terms, I performed a similar analysis of the Smaug2- associated mRNAs using the “DAVID” database. Like PANTHER, DAVID looks for statistically overrepresented terms but it does so by searching multiple databases and by grouping similar terms in clusters. Therefore, one would expect that a significant term of enrichment should be returned by more than one database and combined in a cluster. On the other hand, a significant term which is returned only once is not placed in any cluster. Clusters are then given a relative score by combining individual p-values for the terms within a cluster and ranking the clusters according to this score (Huang et al., 2009a). As a first step for DAVID analysis, I wanted to check how well DAVID identified transcripts on the Smaug2 list and how this compared to the entries identified by PANTHER. DAVID identified 177 out of the 197 Smaug2-associated mRNAs; 159 genes were identified by both PANTHER and DAVID. PANTHER was not able to identify 19 transcripts, which included long-non coding RNAs, miRNAs and transcripts that were not well-characterized. DAVID instead did not recognize 13 transcripts which corresponded to protein coding genes. A small subset (6 transcripts), comprising miRNAs (mir669h, mir5125), lncRNA (pla2g10os, chd3os), a ribosomal RNA (rn45s), an unannotated transcript (loc102635985), was not recognized by either tool (Table 5- see Appendix). DAVID analysis of the Smaug2 list identified 4 clusters with enrichment scores corresponding to p values < 0.05 (Table 6). The first of these was the transcription cluster, and the main terms within were “nucleus” and “transcription regulation”. Out of the 177 Smaug2- associated genes analyzed with DAVID, 51 were under “nucleus” and 24 in “transcription 116

Table 6: DAVID GO enrichment analysis of the Smaug2-associated mRNAs. Enrichment analysis for GO terms of the Smaug2 list was performed with the DAVID functional annotation GO server. Genes were analyzed for enrichment against the background of expressed genes in radial precursors at E12.5 identified from the 5 input/total mRNA samples. DAVID functional annotation clustering was performed using the default database settings. Clusters that had a score < 1.3, which corresponds to p < 0.05, where considered significant. For each cluster, one or two terms or features capturing the other terms in the cluster, and with a p-value < 0.05, were reported. For each term, the number of genes, what percentage of the input list they represent, p- value, gene names and fold of enrichment over the input population, are reported.

Annotation Enrichment Cluster 1 Score: 1.65 Category Term Count % p-value Genes Fold Enrichment SP_PIR_KE nucleus 51 28.8 8.68E-04 PTOV1, ING2, 1.539574965 YWORDS GPBP1L1, TAF1D, TSNAX, PPIL2, MED24, POLA2, MXI1, CBFB, ZFP36L1, ASH2L, SMARCD2, DDX19A, SNF8, PHTF1, IPMK, TRAF4, SPOP, XPOT, ELP3, MCRS1, MSH2, GMEB2, MEX3B, ZBTB41, TLE4, SOCS7, TSN, MCM3, EHMT2, MXD1, PRPF6, CCNL2, PRDM8, PAXIP1, UACA, HES5, RPS6KA2, RFC2, MLX, ATF7, MEX3C, NOL11, VGLL4, NHP2, DUSP8, KPNA1, KDM6B, NHLH1, TXNL4A SP_PIR_KE transcription 24 13.5 0.007729 ELP3, PTOV1, ING2, 1.771662231 YWORDS regulation GPBP1L1, GMEB2, TAF1D, ZBTB41, MED24, TLE4, MCM3, MXI1, MXD1, CCNL2, PRDM8, SMARCD2, HES5, ASH2L, MLX, SNF8, ATF7, PHTF1, VGLL4, NHLH1, SUDS3 Annotation Enrichment Cluster 2 Score: 1.49 Category Term Count % p-value Genes Fold Enrichment UP_SEQ_F domain:Heli 6 3.4 0.004786 HES5, MLX, PHTF1, 5.461075725

117

EATURE x-loop-helix MXD1, MXI1, NHLH1 motif Annotation Enrichment Cluster 3 Score: 1.37 Category Term Count % p-value Genes Fold Enrichment GOTERM_ GO:0008047 8 4.5 0.005103 GIT1, ARHGAP33, 3.772357724 MF_FAT ~enzyme GRTP1, ARHGAP1, activator ARHGAP11A, activity SH3BP1, TBC1D22B, AHSA1 Annotation Enrichment Cluster 4 Score: 1.31 Category Term Count % p-value Genes Fold Enrichment SP_PIR_KE transferase 23 13.0 0.004456 MGAT4B, SOAT1, 1.890577947 YWORDS ELP3, SMOK4A, SPHK2, BRAF, ST8SIA2, EHMT2, MARK4, PRKCD, CDK5, CAMKK2, CDS2, SULT4A1, EPHA4, RPS6KA2, FASN, GRK6, DCLK2, DHPS, GSTO1, IPMK, LFNG INTERPRO IPR002290:S 8 4.5 0.007524 SMOK4A, RPS6KA2, 3.532653276 erine/threoni GRK6, DCLK2, CDK5, ne protein PRKCD, MARK4, kinase CAMKK2

118 regulation”. When compared to the input population, which comprises mRNAs expressed at E12.5 in cortical radial precursors, these two categories showed a modest enrichment of 1.5 and 1.77, respectively, in the Smaug2 list. The second cluster included 6 transcription factors of the bHLH family, including Hes5, Mlx, Phtf1, Mxd1, Mxi1 and Nhlh1 and it had an enrichment of 5.5 fold compared to the input population. The third cluster included 8 genes involved in GTPase activator activity and included Git1, which is a G protein-coupled receptor kinase- interactor 1 and 2 Rho GTPase activating proteins, Arhgap1, Arhgap11a and it was enriched by 3.8 fold. The fourth cluster instead comprised as main term “transferase”; an interesting subset of this category involved “Serine/threonine protein kinases” and included genes such as Dclk2, Cdk5, Mark4, and Camkk2. The two terms reported from cluster 4 had an enrichment of 1.9 and 3.5, respectively. Importantly, these broad categories were similar for both DAVID and PANTHER, including transcription factors, which were also identified by IPA. All the identified genes from these different analyses with a known function in the context of brain development are summarized (Table 7 – see Appendix). Genes that have a known function in brain development and that were identified by manual curation of the Smaug2 list are also included (Table 8).

5.3.5 Comparison of Smaug2 and 4E-T associated transcripts show a small but significant overlap

Finally, I wanted to ask about the potential overlap between Smaug2 and 4E-T- associating mRNAs in the embryonic cortex, given the data that is presented in Chapter 4 showing that Smaug2 represses nanos1 mRNA in association with 4E-T. To do this, I made a direct comparison of the list of mRNAs bound by Smaug2 and 4E-T in the E12.5 cortex, as identified by RIP (Fig. 22A). This analysis demonstrated a small but significant overlap between these two groups of mRNAs (Fig. 22A). Of 19,513 mRNAs in the E12.5 cortex, 1582 were associated with 4E-T, 197 with Smaug2, and 28 of these were associated with both proteins. These shared target mRNAs included many with known functions in the developing cortex, as shown in Fig. 22B (Gene function description in Table 7 and 8). In agreement with this, analysis of this overlap with DAVID returned only one significantly enriched cluster with the terms developmental protein (p-value = 0.00093; elavl3, Lfng, ambra1, enc1, hes5, nhlh1, ppap2b) and neurogenesis (p-value = 0.018; elavl3, hes5 and ambra1). This small overlap suggests that Smaug2 and 4E-T could regulate these mRNAs together but they also regulate distinct mRNA subsets, possibly by forming different repressive complexes.

119

Table 8: Smaug2-associated mRNAs that were NOT highlighted by IPA, DAVID or PANTHER but with a documented role in the CNS. All Smaug2-associated mRNAs that have a known functions in the CNS as identified by manual curation only. For each of them, gene symbol, description of the name and a brief description of the function are reported. Gene Description Documented CNS function Symbol Alpha/beta hydrolase It is a regulator of N-acyl-phospholipid Abhd4 domain-containing metabolism in the mammalian nervous protein 4 system (Lee et al. 2015) It is involved in lipid metabolism in the brain Peroxisomal acyl- Acox1 (Miyazaki et al., 2013) and this gene is coenzyme A oxidase 1 decreased in aged rats (Yang et al. 2014) Deficient mice have altered circadian BTB/POZ domain- Btbd9 rhythm and motor control (DeAndrade et al. containing protein 9 2012) HuC, together with HuB and HuD, is a neuron specific RNA binding protein (Fornaro et al 2007) They are mRNA stabilizers and they bind to AU rich Huc Elav-like protein 3 / HuC sequences (AREs). Huc and Hub are sufficient to induce neuronal differentiation and markers when overexpressed in vivo via in utero electroporation (Akamatsu et al. 1999, PNAS) Matrix remodelling It is expressed in astrocytes, it might be Mxra8 associated protein 8 / involved in maintenance of the blood brain limitrin barrier (Yonezawa et al. 2003) Membrane-associated It increases proliferation of rat neural Pgrmc2 progesterone receptor precursors in response to progesterone component 2, (Liu et al., 2009) Knockdown impairs neuronal differentiation Phospholipid Ppap2b of ES cells and neurite outgrowth phosphatise 3 (Sanchez-Sanchez et al., 2012) Microtubule plus-end tracking protein, it regulates microtubule dynamics (Van deer SLAIN motif-containing Slain1 Baart et al., 2012). After E11, its expression protein 1 is restricted mostly to the CNS (Hirst et al., 2010)

120

It is involved in actin assembly and disassembly, by dephosphorylating the factor Cofilin, that becomes active and can stimulate actin disassembly. Cofilin is Phosphatase Slingshot Ssh1 phosphorylated by Limk1. Ssh1 can homolog 1 dephosphorylate and deactivate also Limk1. This process is definitely important for the regulation of dendritic spines (Yuen et al., 2010) It is a guanine nucleotide exchange factor, Suppression of not very studied but its alteration may result St5 tumorigenicity 5 in intellectual disability and CNS problems (Goring et al., 2010) It is expressed predominantly in the brain Sult4a1 Sulfotransferase 4A1 and it has been suggested that it is linked to schizophrenia (Meltzer et al. 2008) Tubulin-folding cofactor It regulates neuronal growth cones (lopez- Tbcb b Fenarraga et al. 2007) It is overexpressed in Down syndrome and expressed in mouse cortical precursors and Transmembrane protein Tmem50b glia during development (Moldrich et al., 50b 2008) but its function has not been characterized yet. Its function is not characterized but it is Zinc finger SWIM expressed in progenitors of the Lateral or Zswim5 domain-containing Median Ganglionic Eminence in the mouse protein 5 brain (Tucker et al. 2008), it is also highly expressed in gliomas (Meyer, 2014)

121

Figure 22: Comparison of mRNAs associated with Smaug2 and 4E-T reveals a small but significant shared subset. (A) The left panel shows a Venn diagram of all mRNAs expressed in the E12.5 cortex (blue circle), those enriched in the 4E-T immunoprecipitates (yellow circle) (from Yang et al., 2014) and those enriched in the Smaug2 immunoprecipitates (pink circle). The Venn diagram of the latter two populations is shown enlarged in the right panel. (B) Table showing the mRNAs that significantly associated with both 4E-T and Smaug2, as determined by RIP-Chip. The genes highlighted in yellow have known function in the nervous system. Description of the genes highlighted in yellow are in Tables 7 and 8.

122

5.4 CONCLUSIONS

In summary, experiments presented in this chapter used RIP-Chip to identify potential Smaug2 target mRNAs in early developing radial precursors. Gene Ontology (GO) analysis of these mRNAs allowed me to identify several candidates that could be involved in the nanos1- independent neuronal mislocalization phenotype I observed upon Smaug2 knockdown in vivo. Furthermore, GO analysis highlighted other interesting targets whose function in the context of Smaug2 regulation should be investigated in the future, since it suggested that Smaug2 may have additional roles in cortical development subsequent to neurogenesis. Finally, these studies showed that some mRNAs associated with both Smaug2 and the repressor 4E-T overlap. However, since most of the Smaug2 mRNA targets were not also identified targets of 4E-T, then this suggests that while there is a common complex between these proteins, it is also likely that Smaug2 and 4E-T are part of different, independent, repressive complexes in radial precursors. The impact of this work will be further discussed in chapter 6.

123

Chapter 6 Discussion The mechanisms regulating genesis of appropriate numbers and types of neurons are a key question in mammalian development. While transcriptional mechanisms determining neurogenesis have been well-studied (Miller and Gauthier, 2007; Imayoshi and Kageyama, 2014), little is known about post-transcriptional regulatory mechanisms. In this thesis, I focused on Staufen2, a double-stranded RNA-binding protein with known roles in mRNA transport and localization, and Smaug2, which is the mammalian homologue of an important translational repressor in D. melanogaster, and I showed that both of these proteins perform important functions in regulating maintenance and differentiation of mammalian cortical radial precursors. These studies, together with additional work from our laboratory and others, add another layer of complexity to neural precursor development and are likely relevant for other types of mammalian stem cells.

6.1 An asymmetrically localized, Staufen2-dependent RNA granule controls radial precursor maintenance and differentiation

In this study, we demonstrated that a Staufen2-dependent RNA granule plays an important role in regulating self-renewal and differentiation of mammalian stem cells. We focused on radial glial precursor cells of the embryonic mammalian cortex at a time point when they undergo both symmetric self-renewing and asymmetric neurogenic divisions. Our studies in this system supported four major conclusions. First, we demonstrated that Staufen2 is asymmetrically localized in RPs and is enriched in their apical end-feet at the ventricular surface. Second, our immunoprecipitation and colocalization studies demonstrated that Staufen2 forms part of an apically-localized RNA complex that contains the RNA-binding proteins Pumilio2 and DDX1 and target mRNAs such as prox1 and β-actin. Our data also indicated that this complex is asymmetrically-enriched in dividing ventricular precursors. Third, we showed that disruption of this granule by genetic knockdown of Staufen2, Pumilio2, or DDX1 leads to genesis of neurons at the expense of precursors, indicating that this Staufen2-dependent RNA complex is essential for appropriate precursor maintenance. Fourth, we provided evidence that this RNA complex maintains precursors by binding, localizing and potentially repressing target mRNAs. In particular, we showed that Staufen2 knockdown causes mislocalization and enhanced expression of prox1 mRNA, and that expression of a Staufen2 mutant that is unable to bind RNA also causes prox1 mRNA mislocalization and inappropriate neurogenesis. Together, these data support a model where Staufen2, Pumilio2 and DDX1 provide key components of an 124 asymmetrically-localized RNA granule that controls the translation of mRNAs that regulate the maintenance versus differentiation of RPs, and thereby development of the embryonic cortex. While asymmetrically-localized RNA granules have not been previously shown to regulate mammalian stem cell biology, they are clearly important in model organisms. In D. melanogaster, Staufen itself plays a key role in asymmetric RNA localization in oocytes and in neuroblasts. In oocytes, a Stau-containing RNP localizes oskar mRNA to the posterior pole of the developing oocyte, and disruption of this complex perturbs development (Micklem et al., 2000). In neuroblasts, Stau binds prospero mRNA and comprises part of an RNA complex that also contains the translational repressor Brat and the adaptor protein Miranda. This complex is basally localized, and segregates into the neurogenic ganglion mother cell where Miranda degradation allows the complex to release its associated mRNAs, and the prodifferentiation factors Prospero, Brat and Numb (Chia et al., 2008). In this regard, the Brat homologue Trim32 was recently shown to play a key role in regulating cortical precursor maintenance versus differentiation (Schwamborn et al., 2009), and mammalian prox1 has been implicated in regulating the maintenance and differentiation of mammalian neural transit-amplifying cells (Dyer et al., 2003; Kaltezioti et al, 2010). The two mammalian Staufen have been the focus of significant interest in postmitotic neurons, where they are thought to play a key role in regulating local activity-dependent translation in dendrites. In particular, Staufen2 is known to bind to its target mRNAs in neuronal cell bodies and to assemble them into a transport-capable RNA granule (Goetze et al., 2006). This RNA granule transports these mRNAs and localizes them to distal dendrites in a repressed state until activity-dependent local signals promote their translation, an event thought to be essential for synaptic plasticity (Sossin and DesGroseillers, 2006). Studies investigating the molecular composition of these Staufen2-dependent RNA complexes have shown them to be heterogeneous, both with regard to the mRNAs they bind (Maher-Laporte and DesGroseillers, 2010) and their protein partners, which include proteins involved in transport, translational repression and translational activation (Kanai et al., 2004; Elvira et al., 2006; Maher-Laporte et al., 2010). Intriguingly, our data showing that Staufen2 knockdown causes mislocalization of newly-born neurons suggests that these Staufen2-dependent RNA granules may also play a role in neuronal migration. Alternatively, this mislocalization may simply be the result of depletion of RPs, since these cells provide essential migratory scaffolds for newborn neurons. Our data suggest that the Staufen2-dependent RNA complexes in cortical precursors share many similarities with the previously-characterized RNA granules in neurons. In 125 particular, we show that they contain the DEAD-box RNA helicase, DDX1, which was previously found as part of a neuronal Staufen2-containing RNA transport granule (Elvira et al., 2006). Moreover, we show that Staufen2 forms a complex with the translational repressor Pumilio2, which is also localized in dendrites, where it is thought to repress the translation of localized RNAs until the appropriate stimulus is encountered (Vessey et al., 2006). The functional importance of these proteins within RPs is illustrated by our data showing that knockdown of Pumilio2 or DDX1 phenocopies the Staufen2 knockdown phenotype, causing inappropriate neurogenesis and depletion of the stem cell pool. How then would such an RNA granule regulate RP maintenance versus differentiation? By analogy to the RNA complexes in neuronal dendrites and in model organisms, we propose that it does so by binding to, localizing, and repressing target mRNAs that are important for promoting neurogenesis until a given RP receives the appropriate extracellular signals. This model is supported by our data showing that (i) prox1 mRNA, which is normally apically- localized in RPs, becomes randomly-localized when Staufen2 is depleted, (ii) Prox1 protein expression appears to be repressed in RPs and when Staufen2 is knocked-down, this enhances Prox1 expression, (iii) expression of a Staufen2 mutant that cannot bind RNA in RPs has the same effect as depleting Staufen2 and (iv) knockdown of Pumilio2, which is a known translational repressor, phenocopies Staufen2 knockdown. Such a model is somewhat analogous to what is seen in D. melanogaster neuroblasts, where Stau participates in asymmetric localization and repression of prospero mRNA, which is then released from repression once it is segregated into the transit amplifying neurogenic progenitor in this system, the ganglion mother cell (Chia et al., 2008). However, while there are similarities between RPs and D. melanogaster neuroblasts, there are also a number of key differences. First, while the Staufen-containing RNA complex in D. melanogaster neuroblasts is clearly asymmetrically-segregated into the ganglion mother cell, our data indicate that in RPs this is not an absolute segregation, but is instead an enrichment. Nonetheless, it is plausible and even likely that enrichment of cell-fate determinants in one daughter cell versus another is sufficient to promote different cell fates. Alternatively, detachment of the neurogenic daughter cell from the ventricular surface and the resultant disruption of apical localization may itself provide a signal for derepression of the mRNAs within the Staufen2 complex. Second, in D. melanogaster neuroblasts, prospero alone is sufficient to regulate stem cell maintenance versus neurogenesis, but we feel this is unlikely to be the case in mammals. Instead, mammalian Staufens are thought to bind to many unique RNAs (Maher-Laporte and DesGroseillers, 2010), and we propose that the RP Staufen2 126 complex contains multiple target mRNAs that are important for mammalian neurogenesis, many of which collaborate to regulate genesis of intermediate progenitors and newborn neurons. Some of these mRNAs have been identified in the Staufen2 companion paper (Kusek et al., 2012) and together these findings support the idea that Staufen2 may regulate radial precursor development on multiple levels, as discussed in the following section.

6.1.1 The companion paper published with this study (Kusek et al., 2012) supports our results

When we were working on Staufen2 functional characterization in radial precursors, Dr. Sally Temple’s laboratory was also studying Staufen2 function in the same model system. Our papers were published together and importantly, our results are largely the same. In particular, they also reported (i) apical enrichment of Staufen2 in radial precursors, (ii) depletion of radial precursors and an increase in neurons upon Staufen2 knockdown, and (iii) Staufen2 enrichment in one of two daughter cells during cell division in vitro and in vivo. Kusek et al. also performed Staufen2 RIP-Chip and the mRNAs that they identified suggested that Staufen2 may control neurogenesis through multiple mechanisms. Most notably, they identified trim32 mRNA as a Staufen2-associated mRNA. As mentioned above, Trim32 is the mammalian homologue of D. melanogaster Brat and it has previously been shown to segregate asymmetrically in dividing radial precursor cells (Schwamborn et al., 2009). Furthermore, Trim32 knockdown was shown to promote radial precursor cell self-renewal and its overexpression was shown to promote neurogenesis (Schwamborn et al., 2009). Trim32 is thought to mediate these activities by (i) ubiquitinating c-Myc and promoting its degradation and (ii) binding Argonaute1 and thereby stimulating the activity of pro-neurogenic miRNAs, such as let-7 (Schwamborn et al., 2009). Although Kusek et al. could not show that Staufen2 and trim32 mRNA co-localized because of technical limitations, they did show that trim32 mRNA was asymmetrically localized during the neurogenic period and that upon Staufen2 knockdown, asymmetric segregation was lost. Hence, from their cumulative evidence they concluded that the daughter cell with more Staufen2 becomes an intermediate progenitor because of all the pro-neurogenic mRNAs that were bound to, and segregated with, Staufen2. Upon Staufen2 knockdown, these determinants were symmetrically distributed between daughter cells, thereby causing the observed increase in Tbr2-positive intermediate progenitors. Together with our findings, this suggests that in addition to prox1 mRNA, Staufen2 asymmetrically segregates trim32 mRNA to promote differentiation.

127

Their RIP-Chip study also identified negative regulators of the cell cycle and factors that promote cell cycle exit, consistent with the observed depletion of radial precursors following Staufen2 knockdown. Interestingly enough, they also identified mRNAs involved in cilium and centrosome assembly and they report that this is important because mutation is some of these genes cause ciliopathies and neurological disorders (Zaghloul and Katsanis, 2009). Furthermore, the primary cilium is also thought to be a cell “antenna” that is responsible for transducing signals from important pathways such as Shh (Fuccillo et al., 2006), Wnt (Goetz and Anderson, 2010; Quinlan et al., 2008) and PDGF (Schneider et al., 2005; Gazea et al., 2016). The cilium has also been shown to translocate from the apical side to the basolateral side of radial precursors before inducing the expression of Tbr2, which marks their transition to intermediate progenitors, and delamination from the ventricular surface (Wilsch-Bräuninger et al., 2012). It would be interesting to test whether the localization defects observed upon Staufen2 knockdown are caused by misregulation of cilium or centrosome components. Although the RIP-Chip experiment identified many different mRNAs that are associated with Staufen2, there are a number of caveats with this analysis. One caveat is that it was performed at E13.5 and therefore the mRNAs isolated do not represent a pure radial precursor population but are likely to include contributions from intermediate progenitors and deep layer neurons as well. Secondly, prox1 mRNA was not among the Staufen2 associated mRNAs identified in this analysis. However, it is possible that prox1 did not reach statistical significance due to experimental variability. It would therefore be informative to repeat this experiment at an earlier timepoint or with a sorted radial precursor cell population to distinguish Staufen2- associated mRNAs in radial precursors versus neurons. Although the majority of our findings coincide, one major and interesting difference between our work and that reported by Kusek et al. is that they reported that the increase in neurogenesis was due to an expansion of the intermediate progenitor pool, while we did not observe an increase in Tbr2-positive intermediate progenitors in any of our experiments. Instead, we saw an increase only in neurons, leading us to suggest that Staufen2 knockdown in our experiments was increasing the direct differentiation of radial precursor into neurons. This discrepancy might be explained by our different experimental conditions. For instance, for in vitro experiments they dissected cortices at E13.5 and plated at clonal density of 40-80 cells per well, while we dissected cortices at E12.5 and plated 150,000 cells per well. Thus, our cultures differed both in the developmental age and in density. Our laboratory has recently shown that the density of plating of radial precursor cells profoundly affects proliferation and differentiation 128 and that at lower densities there is increased generation of intermediate progenitor cells (Yuzwa and Miller, unpublished observations). Thus, their in vitro conditions were quite different from ours. Furthermore, we performed in utero electroporation at E13.5 while they performed it as late as E15. Since direct genesis of neurons from radial precursors is favoured at earlier ages and the intermediate progenitor neurogenic pathway is favoured at later stages, then this might also explain our differences in this regard.

6.1.2 Conserved developmental roles for the Staufen proteins?

A final key question with regard to our Staufen2 study is whether similar RNA complexes play important roles in other mammalian stem cells. While this is still an open question, RNA-binding proteins have been implicated in other types of stem cells. For example, Musashi1 binds its target RNA, m-numb, and represses translation, thereby keeping epithelial and neural stem cells in an undifferentiated state (Okano et al., 2005). Likewise, Pumilio2 binds its target mRNA deleted in azoospermia-like (dazl) and represses its translation in germ line stem cells, an interaction that is important for maintaining these stem cells in an undifferentiated state (Moore et al., 2003). This latter finding, together with the work reported here, raises the intriguing possibilities that perhaps a related Staufen2 and Pumilio2-dependent RNA granule regulates the maintenance versus differentiation of multiple mammalian stem cell populations. In support of this idea, Staufen2 knockdown in the developing chicken eye has been shown to cause microphthalmia because of decreased proliferation of eye precursors (Cockburn et al., 2012). This is similar to the phenotype we observe following Staufen2 knockdown, and suggests that perhaps Staufen2 may regulate different types of precursors or stem cells. On the other hand, a role for Staufen1 in developing radial precursors has not been described but studies in other developmental systems suggest that it would also be important. For instance, early mouse embryonic stem cell differentiation is impaired in pre-implantation embryos when Staufen1 is knocked down (Gautrey et al., 2008). In contrast, in skeletal muscle, Staufen1 appears to regulate myogenic differentiation by regulating c-Myc translation (Ravel-Chapuis et al., 2014). If Staufen1 were to promote c-Myc translation also in radial precursors, it would likely be involved in the maintenance of radial precursor self-renewal. In conclusion, the work presented in this chapter shows that the double stranded RNA binding protein Staufen2 plays a critical role in the regulation of radial precursor development. Of course this begs the question of whether the Staufen proteins play important roles in other

129 types of stem cells. Based on the studies that have already been published, this is likely to be the case. Hopefully with the development of more sophisticated tools, we will be able to clarify the unanswered aspects of Staufen function, such as the role of the different Staufen isoforms. The Staufen2-dependent granule, however, is unlikely to be the only type of RNP that is important for brain development. In this regard, studies in neurons have shown that many heterogeneous RNPs exist in dendrites and respond to different cues to regulate synaptic plasticity and neuronal remodelling. Similarly, one might expect that fine-tuning of early brain development may also require many RNPs acting in concert or independently as needed to give rise to the diversity of cell types and circuitry that are required for brain function. So far this seems to be the case because several RBPs have been shown to have a role in radial precursor development and in the following section I will discuss my work on the function of translational repressor Smaug2.

6.2 A Smaug2-based translational repression complex determines the balance between precursor maintenance versus differentiation during mammalian neurogenesis

Data presented in Chapter 4 showed that two RNA-binding proteins, Smaug2 and Nanos1, regulate the balance of self-renewal versus differentiation of embryonic neural precursors, and in doing so determine the timing and numbers of neurons that are generated. We showed that these two repressive proteins did this by functioning in opposition, with Smaug2 maintaining the precursor state and inhibiting differentiation, and Nanos1 promoting differentiation and depleting precursors. Moreover, we showed that these antagonistic activities were coordinated, with Smaug2 associating with and silencing nanos1 mRNA, potentially by localizing it to P-body-like granules in association with the translational repressor 4E-T and with the deadenylase CNOT7. Based upon these findings, we propose a translational repression "switch" model for the precursor to neuron transition. In this model, developing neural precursors are transcriptionally primed to generate neurons, but Smaug2 and 4E-T-dependent mRNA repression/silencing maintain them in a stem cell state. In this model, extrinsic proneurogenic cues would disrupt these repressive complexes, thereby releasing the relevant mRNAs and allowing for rapid, precise, and coordinated translation of proteins that promote neurogenesis. This environmentally-driven dissociation of Smaug2 repressive complexes would also derepress nanos1 mRNA, and the newly-translated Nanos1 would then act to repress mRNAs associated

130 with and necessary for the stem cell state. Thus, the RNA-binding proteins Smaug2 and Nanos1 act as a "switch" - when Smaug2 is active or "turned on", this keeps Nanos1 "turned off", thereby maintaining translation of precursor proteins and silencing neuronal translation. However, when Smaug2 is turned off, this turns Nanos1 on, allowing it to silence translation of precursor proteins. Coincident with this Nanos1-dependent silencing of the precursor state, other derepressed proteins, like the neurogenic bHLHs (Yang et al., 2014) would promote establishment of a neuronal phenotype. One attractive feature of this model is that it would ensure that cells did not become "confused" by adopting a neuronal phenotype while at the same time maintaining features of cycling precursors. Thus, adoption of one phenotype would coincide with repression of the other. A second key feature of this model is that it involves transcriptional "priming" of precursors, which would allow for phenotypic preprogramming of different neuronal phenotypes since these mRNAs could be sequestered by Smaug2/4E-T/CNOT7 complexes until the appropriate developmental timepoint. Support for this translational repression switch model comes from our data showing that i) Smaug2 was necessary and sufficient to maintain neural precursors in a stem cell state, ii) Nanos1 was necessary and sufficient to promote neurogenesis and deplete precursors, iii) nanos1 mRNA was highly colocalized with Smaug2, 4E-T and CNOT7, iv) knockdown of Smaug2 or 4E-T causes aberrant Nanos1 expression, presumably because of complex disruption, and v) coincident Nanos1 knockdown largely rescued the increase in neurogenesis and loss of Pax6-positive precursors caused by Smaug2 knockdown. Further support for this model comes from our recent work showing that 4E-T, which does not itself bind mRNAs, is necessary to maintain radial precursors in a stem cell state, and that when it is knocked-down, this derepresses translation of the proneurogenic bHLHs Neurogenin1, Neurogenin2, and NeuroD1 (Yang et al., 2014). How does Smaug2 silence nanos1 mRNA and its other potential target mRNAs? In D. melanogaster, Smaug silences its bound mRNAs in several ways. First, it associates with the eIF4E-binding protein Cup, which prevents eIF4E from interacting with its binding partner eIF4G and in so doing inhibits recruitment of the 40S subunit of the ribosome to the bound mRNA (Nelson et al., 2004). Second, Smaug can recruit Argonaute 1 to a target mRNA in a miRNA-independent fashion to repress translation (Pinder and Smibert, 2013a and b). Finally, Smaug can recruit the CCR4-NOT deadenylase complex, leading to mRNA deadenylation and destabilization (Semotok et al., 2005; 2008; Zaessinger et al., 2006). Our data do not conclusively distinguish these different potential mechanisms with regard to developing radial 131 precursors. We have, however, shown that almost all of the nanos1 mRNA that is associated with Smaug2 in Pax6-positive precursors is also colocalized with 4E-T, which is distantly related to the D. melanogaster Cup. These large 4E-T-containing granules also contain Dcp1, eIF4E, DDX6 and Lsm1 (shown in Yang et al., 2014), and thus resemble the P-bodies that are seen in other cells (Dostie et al., 2000; Ferraiuolo et al., 2005). Thus, P-body-like, 4E-T- containing granules are a major site for the Smaug2/nanos1 mRNA complex and likely, for many other repressive complexes in radial precursors. Consistently with this, our results show that at least a subset of these repressive complexes also contains the conserved deadenylase CNOT7, which is a subunit of the multimeric and evolutionarily conserved CCR4-NOT deadenylase complex. We suggest that potentially 4E-T and Smaug2 may recruit CCR4-NOT to their targets and trigger their deadenylation to repress them. The interaction between Smaug2 and CNOT7 is in agreement with studies on D. melanogaster Smaug and S. cerevisiae Vts1p, which have been shown to associate with CNOT7 homologue Caf1 (Semotok et al. 2005; 2008, Zaessinger et al. 2006). It is interesting to note that also 4E-T has been shown to interact with this complex (Ozgur et al. 2015; Nishimura et al. 2015) and that this interaction increases the efficiency of the repression. Hence, several RNA-binding proteins may act independently or cooperatively to move mRNAs towards this conserved repression pathway. The localization does not, however, allow us to distinguish Smaug2-mediated mRNA repression versus degradation; deadenylation is generally the first, rate-determining step for both processes by decreasing translational efficiency of mRNA and by promoting decapping, and both processes are thought to occur in P-bodies (Balagopal and Parker, 2009). Furthermore, our work also suggests that not all nanos1 complexes all contain 4E-T, Smaug2 and CNOT7 at the same time, perhaps suggesting that target mRNA may be differentially regulated depending on what RBPs associate with it. Interestingly, Smaug2 localization is different from that seen in D. melanogaster embryos, where Smaug is found in foci that appear to be distinct from P-bodies (Zaessinger et al., 2006). Moreover, one of the few studies on Smaug in mammals shows that in adult hippocampal neurons Smaug1 is localized in foci that lack Dcp1 (Baez and Boccaccio, 2005; Baez et al., 2012). All together, it appears that Smaug proteins can associate with different granules in different cell types. While we have focused here on a function for Smaug2 in apical precursors, our data have shown that it is also expressed in newborn cortical neurons. Moreover, Smaug2 knockdown caused enhanced genesis of neurons, but these neurons did not reach the cortical plate. While there are a number of potential explanations for this neuronal mislocalization, these 132 data might indicate that Smaug2 plays an important role in migration of newborn neurons. In this regard, while a function for Smaug2 in the adult brain has not been described, previous work has shown that the related family member Smaug1 forms mRNA-silencing foci that reversibly release their associated mRNAs upon specific synaptic activation by N-methyl-D- aspartic acid (NMDA) in dendrites of hippocampal neurons (Baez and Boccaccio, 2005; Baez et al., 2012). In addition, mutation of Smaug1 causes a lean phenotype in mice, likely due to deregulation of the mTOR pathway (Chen et al., 2014b) consistent with the recent finding that D. melanogaster Smaug regulates a large set of metabolic transcripts (Chen et al., 2014a). Thus, mammalian Smaug 1 and 2 are likely to play diverse roles in multiple cell types, in part dependent upon the constellation of target mRNAs that are expressed in each case. One key remaining issue involves the molecular mechanism(s) responsible for the proneurogenic effects of Nanos1. In D. melanogaster, Nanos associates with two other RNA- binding proteins, Pumilio, and Brain tumour (Brat) (Loedige et al., 2014). This complex binds to and represses target mRNAs, a function that is essential for germ cell development (Hayashi et al., 2004; Lai and King, 2013). This role as a translational repressor in germ cells is conserved in vertebrates, including mice, where there are three family members, Nanos1, 2, and 3 (Jaruzelska et al., 2003; Tsuda et al., 2003; Lolicato et al., 2008). We therefore propose that Nanos1 regulates neurogenesis by associating with and repressing mRNAs associated with the neural precursor state. Interestingly enough, both Nanos and its binding partner Pumilio are known to also interact with the CCR4-NOT deadenylase complex (Van Etten et al., 2012; Suzuki et al., 2014; Raisch et al., 2016). What then are the relevant Nanos1 target mRNAs in newborn neurons? In D. melanogaster and Xenopus Nanos associates with Pumilio to repress cyclin B mRNA and to thereby lock cells out of the cell cycle (Nakahata et al., 2001; Kadyrova et al., 2007). This repression might be equally essential for newborn neurons, which, unlike their precursor parents, are postmitotic. Another potential target is Sox2. In murine germ cells Nanos2 has been shown to associate with sox2 mRNA (Saba et al., 2014), and Sox2 is a key stem cell gene in cortical radial precursors (Julian et al., 2013). Thus, we propose that Nanos1- mediated translational repression of mRNAs encoding proteins like Cyclin B and Sox2 represses their previous precursor identity in newborn neurons, thereby allowing them to fully establish a neuronal phenotype. In summary, our work identified an essential role for translational repression in the neural precursor to neuron transition. The bimodal translational regulation we identify here may provide a widespread mechanism for ensuring that newborn progeny completely adopt their 133 differentiated phenotype during development. As mentioned previously, however, Smaug2 may regulate radial precursor development through targets other than nanos1 mRNA. To address this possibility, I isolated mRNAs associating with Smaug2 in the early embryonic cortex, and I will discuss those findings in the following section.

6.3 Smaug2 RIP-Chip identifies approximately 200 mRNAs associating with Smaug2 in cortical radial precursors in vivo.

Studies published in recent years have shown that mechanisms of post-transcriptional regulation play a critical role in the development of the mammalian cerebral cortex (Okano et al., 2005; Kusek et al., 2012; Vessey et al., 2012; Yang et al., 2014). In order to elucidate the mode of action of these RNA-binding proteins and whether or not they interact with one another, it is critical to identify all the different mRNA transcripts associated with them. Here, I identified approximately 200 RNA transcripts associating with Smaug2 in the radial precursors of the early developing cerebral cortex to investigate additional, nanos1-independent, Smaug2 roles, as suggested by the data I presented in Chapter 4. I performed Gene Ontology (GO) analysis on these identified transcripts, to gain insight into their known roles and functions and to seek candidates to explain the additional phenotype observed upon Smaug2 knockdown, as I discuss in the following section.

6.3.1 Smaug2 functions that are independent of nanos1 regulation.

With the data presented in Chapter 4, I showed that Smaug2 promotes radial precursor maintenance and prevents early neuronal differentiation by repressing nanos1 mRNA with 4E-T in a P-body like granule. I also showed that upon knockdown of Smaug2 or 4E-T, ectopic expression of Nanos1 protein occurred, and that increased neurogenesis observed upon Smaug2 knockdown could be rescued by coincidentally knocking down nanos1, which argues that the ectopic increase in Nanos1 protein was responsible for these phenotypes. Interestingly, upon Smaug2 knockdown I also observed mislocalization of the new-born neurons, which were “stuck” in the intermediate zone and did not reach the cortical plate as they do in control brains. Nanos1 knockdown, however, did not rescue this defect. One potential explanation for this is that these new neurons do not the reach the cortical plate because the premature differentiation of their radial precursor parents alters the basally-projecting fibers that they utilize as neuronal guides during migration. Premature radial precursor differentiation, however, does not necessarily cause impaired neuronal localization. For instance, with 4E-T knockdown, which 134 also resulted in increased neurogenesis and radial precursor depletion, newborn neurons reached the cortical plate (Yang et al., 2014). Therefore, it is possible that Smaug2 has additional roles in regulating newborn neuron migration and that these roles are independent of nanos1 repression. I therefore analyzed the Smaug2-interacting mRNAs using Gene Ontology analysis to identify potential regulators of neuronal migration. One caveat to keep in mind for GO analysis is that several genes, despite being recognized by the tools I used, were not functionally characterized in the databases and hence it is possible to miss interesting categories and/or gene candidates. The first candidates that stand out in the context of neuronal migration are the ones clustered in the category “migration of brain cells”, which were identified with IPA, all of which have known functions in this process. Cdk5 is a cyclin-dependent kinase that is involved in neuronal migration (Kawauchi, 2015) and dendritic spines formation (Mita et al., 2016), Myh10 is required for centriole remodelling during cilia formation and cell migration (Hong et al., 2015), Prkcd, is a protein kinase which is also involved in different phases of migration (Nishimura et al. 2010), and Socs7 regulates formation of the cortical layers by modulating Reelin signalling (Sekine et al., 2014; Lawrenson et al., 2015). Smaug2-dependent misregulation of these mRNA transcripts could potentially be involved in the deficits in neuronal migration that I observed. The remainder of the GO analysis, however, hints at the idea that Smaug2 may also regulate other aspects of neuronal maturation and circuitry formation, as indicated by several more categories that stand out. One of these comprises G-protein modulators, such as RhoGTPase proteins, which have well-known roles in the context of cytoskeletal remodelling (Gauthier-Fisher et al., 2009). During differentiation, intermediate progenitors and newborn neurons undergo extensive cytoskeletal remodelling to acquire the appropriate morphology for each developmental stage and Smaug2 may be involved in regulating the expression of key mRNA transcripts for this process. Potential RhoGTPase mRNAs with known roles in the nervous system include Arhgap1, which promotes the epithelial to mesenchymal transition in neural crest cells (Clay and Halloran, 2013), Arhgap33, which is involved in synapse maturation (Schuster et al., 2015) and intracellular trafficking of the neurotrophin receptor TrkB (Nakazawa et al., 2016), and Git1, which is GTPase activating protein whose alteration causes microcephaly in a genetic mouse model (Hong et al., 2015). Another category that is overrepresented in the Smaug2-associated mRNA transcript list is the kinase category. This includes proteins that have already been characterized as important 135 for appropriate neurogenesis including Braf, which is a kinase required for precursor survival, proliferation, neurogenesis and migration (Camarero et al., 2006), and Mark4, which is upregulated in differentiating neurons (Moroni et al., 2006). Finally, the transcription factor category is another intriguing category represented in the Smaug2-associated mRNA list. Several of these mRNAs encode proteins with known roles in brain development, including Tle4, which in the mouse is expressed by specific cortical layers (Fauser et al., 2013) and in the chick neuroepithelium is required for the regional organization of the brain (Agoston and Schulte 2009), Prdm8, which is required for the specification of cortical upper layers (Inoue et al., 2015) and for the transition from intermediate progenitors to neurons (Inoue et al., 2014), and Nhlh1, which is downstream or other well-known differentiation factors such as NeuroD1 (Kim, 2012). In the work presented in Chapter 4 I examined early neuronal differentiation but I did not investigate subsequent neuronal development or maturation. On the basis of these candidate Smaug2-interacting mRNAs, it will be an interesting question for future studies to examine these candidates and investigate whether Smaug2 has additional roles in neuronal maturation or even during later stages of cortical development. That Smaug2 may play a role in regulating aspects of neuronal development past early neuronal differentiation would not be surprising, since Smaug1 in cultured hippocampal neurons is required for proper development of dendritic spines and synapses (Baez et al., 2005). Taking all this together, I think that GO analysis allowed me to identify several candidates that could explain the mislocalization of new-born neurons following Smaug2 knockdown, and has suggested that Smaug2 may play further, additional roles during neuronal maturation. Having said that, there are also potential caveats associated with this analysis, such as the fact that in order to identify those enrichment categories I had to use less stringent statistical testing. This was not completely unexpected because my Smaug2 high-confidence list was relatively small and towards the lower end of what is considered optimal range for this type of analysis, between 100 and 4000 genes (Huang et al., 2009a). Furthermore, many of the mRNA transcripts on the list, while they were recognized by the tools I utilized, were not fully annotated, and were therefore excluded from the analysis. It is possible that with a larger list and with more complete annotation, more categories would have been identified and/or a stronger enrichment would have been observed. Because of these limitations, the results of the Smaug2 list enrichment analysis should be taken with caution, and largely used simply as a guide for

136 future experiments, such as for identifying candidates that could be studied further within the context of additional Smaug2 roles in cortical development. One final thing to note is that while Smaug2 was associated with mRNA transcripts involved in neuronal differentiation, as described above, it also was bound to mRNAs known to regulate precursor proliferation, such as Hes5 (Kobayashi et al. 2014) and Lfng (Kato et al., 2010; Nikolaou et al., 2009). However, in spite of this, upon Smaug2 knockdown, I observed increased neurogenesis and decreased radial precursor proliferation. Similarly, a subset of the mRNAs associated with both Smaug2 and 4E-T encoded proteins implicated in radial precursor proliferation. Possible explanations for this are discussed in later sections after I talk about the relationship between Smaug2 and 4E-T targets, which will be discussed in the following section.

6.3.2 The relationship between Smaug2 and 4E-T

In Chapter 4, I presented data supporting the conclusion that 4E-T and Smaug2 cooperated together to repress nanos1 mRNA in cortical radial precursors. In light of the notion that 4E-T is a repressor that does not contact mRNA directly but possesses translational repressor activity when artificially tethered to transcripts (Kamenska et al., 2014), this finding was interesting because it suggested that 4E-T employed a sequence-specific RNA binding protein, Smaug2, to interact with target RNAs. This simple model would predict that a large subset of mRNAs might be shared between the two. To test this, I compared transcripts associating with Smaug2 to those previously found to be associated with 4E-T in the embryonic cortex at the same timepoint (Yang et al., 2014). My comparison did not, however, demonstrate a large overlap between Smaug2 and 4E- T-associated mRNAs. Only approximately 15% of Smaug2 transcripts were shared with 4E-T and 2% of 4E-T target mRNAs were associated with Smaug2, indicating that their interaction was not as straightforward as predicted by the model. One possible explanation for this is that Smaug2 is only one of several sequence specific RNA-binding proteins functioning with 4E-T, each of them associating with a subset of all 4E-T mRNAs. A precedent for this possibility already exists, since a recent study showed that TTP (Tristetraprolin), which is an RBP that recognizes AU-rich sequences in the 3’ UTR of its target mRNA, recruits the CCR4-NOT deadenylase complex, which in turn recruits the decapping and degradation machinery containing 4E-T (Nishimura et al., 2015). In this regard, Smaug2 might display a similar mode

137 of regulation since, like TTP, it binds mRNA and it associates with CNOT7, one of the deadenylases of the CCR4-NOT deadenylase complex, as well as with Dcp1 and 4E-T. One attractive feature of this model is that many different RNA targets could be globally regulated by one common repression pathway involving 4E-T but smaller scale regulation could also be possible by altering the binding of the sequence specific RBPs. One good candidate RBP for testing this model further would be Pumilio2, which is highly expressed in radial precursors and is known to regulate cortical neurogenesis (Vessey et al., 2012), has a well-characterized sequence specificity, and is known to interact with CCR4-NOT (Van Etten et al., 2012). My data indicate that about 85% of Smaug2 mRNAs do not interact with 4E-T within the embryonic cortex, and therefore, their regulation is currently an open question. One possible explanation for this observation is that different Smaug2-associated repression complexes exist in radial precursors. For instance, it is well known that while the CCR4-NOT deadenylase complex is conserved from yeast to humans, gene duplications events have occurred in mammals. Hence, the yeast deadenylase CCR4 has two equivalents in humans, CNOT6 and CNOT6L, and the deadenylase CAF1 has two equivalents, CNOT8 and CNOT7. Because these paralogues have been reported to be mutually exclusive (Shirai et al., 2014), one could imagine that even by simply considering these four proteins, there could be four slightly different versions of this complex. Consistent with the idea that heterogeneous repressive complexes exist in radial precursors, my data in chapter 4 showed that there is only partial co-localization between Smaug2 and 4E-T and between Smaug2 and CNOT7. If this is the case, then, how is a particular mRNA targeted to a particular repressive complex? One possibility is that, in addition to the sequence motifs for a particular RBD, additional features on any given mRNA may promote the assembly of other factors which in turn dictate the assembly of the correct repressive complex. These different factors may do so by favouring the stabilization and thus subunit assembly of one versus another. What would be the functional significance of having different repressive complexes? Perhaps, depending on what subunits are present in the complex, the mRNA is subjected to different fates. Since the decapping complex with 4E-T has been shown to promote transcript degradation, it is possible that the 4E-T/Smaug2 subset is immediately pegged for translational repression and degradation while the subset with Smaug2 alone is more transiently repressed, due to polyA tail shortening, but the mRNA is not degraded immediately and it may resume translation more quickly. This would be compatible with the observation that only a fraction of the total Smaug2 actually co-localizes with P-Body markers. Also in agreement with this 138 interpretation are the observations that (i) Smaug1 represses the translation of SRE-containing reporter mRNA but does not alter the mRNA stability (Baez et al., 2005) and (ii) Smaug1 represses mRNA in hippocampal neurons without being part of canonical P-bodies (Baez et al., 2011). If this were the case, mammalian Smaug2 and 4E-T function would be reversed in comparison to D. melanogaster Smaug and Cup because the former can both repress and degrade mRNA (Chen et al., 2014a) while the latter represses translation by deadenylation but then protects the mRNA from decapping and further decay (Igreja and Izaurralde, 2011).

6.3.2.1 Significance of the mRNAs bound by both 4E-T and Smaug2

Is there anything special about the transcripts regulated together by Smaug2 and 4E-T? Interestingly enough, analysis of those shared transcripts with DAVID reports only one significantly enriched cluster, which contains terms such as neurogenesis, developmental proteins and differentiation. Even though this cluster only contains 7 genes (elavl3, Lfng, ambra1, enc1, hes5, nhlh1, ppap2b) several others in this shared subset have known roles in the context of brain development. As previously mentioned, it is interesting to note that while some of these shared mRNAs have known roles in promoting differentiation, which would be in line with the phenotype observed upon Smaug2 and 4E-T knockdown, some of these mRNAs have well-characterized roles in promoting radial precursor proliferation. Why is this the case? One possibility would be that P-bodies act as a “storage room” for mRNAs, ensuring that the cellular levels of mRNAs with short half-lives like hes5 mRNA would not drop below a critical level. In this model, hes5 mRNA that is not immediately translated would be shuttled to P-bodies for storage and decay, but could be moved back to ribosomes if the cytosolic supply of hes5 should falter. When would such a scenario occur? One instance could be during cell division, since the splitting of the cytoplasm in two could lead to the depletion of needed mRNAs. Alternatively, reports have shown that mammalian transcription occurs in “bursts” (Halpern et al., 2015) and cells have developed systems to prevent these bursts of transcription from dramatically altering the mRNA content of the cytoplasm. Another possibility is that rapid shuttling of mRNAs from polysomes to P-bodies allows a tighter control of translation. This would be important for hes5 because its expression has been shown to oscillate in cortical radial precursors (Imayoshi and Kageyama, 2014b) and P-bodies-mediated repression, together with oscillatory transcription, may be needed to achieve the correct oscillation period.

139

Such a model implies that P-body function is more nuanced than what we currently think. In support of this view, a very recent study in D. melanogaster showed that two very important mRNAs for D. melanogaster development, bicoid and gurken, relied on P-bodies for both translational repression and activation. This study showed that bicoid mRNA was repressed by being stored at the core of the P-body, which was ribosome free, and that gurken mRNA was instead docked at the edge of P-bodies, which were decorated with ribosomes and translational activators such as Orb (CPEB in mammals), and that it was translated, not repressed (Davidson et al., 2016). If this mechanism is conserved in mammals, one could easily imagine that two mRNA populations may exist in P-bodies, one that is repressed at the core and one that instead is being translated at the edge. One attractive feature of this model is that the close association between P-bodies and ribosomes would allow for quick adjustments in the translation status of different mRNAs upon changing cellular conditions. Having said that, however, yet another possibility is that additional factors may bind Smaug2 and 4E-T thus further influencing the fate of their transcripts in different ways, such as altering transcript stability. Finally, it is also possible that expression of the prodifferentiation factors “outcompetes” the self-renewal ones and promotes differentiation. Further work will be required to really characterize the fate of these targets following knockdown or overexpression of 4E-T and Smaug2 to distinguish among all these possibilities.

6.3.3 Pairwise analysis of Smaug2 RIP-chip: pro and cons

One issue with my RIP-Chip dataset was that the biological replicates were quite variable, as shown by Principal Component Analysis. In order to minimize the effects of this variability and to be able to identify transcripts that were significantly associated with Smaug2 I utilized a pairwise comparison and uncorrected p-values < 0.01. Even though this analysis may not be as stringent as it is for other published studies, RT-qPCR and SRE score validations of the Smaug2 list supported the validity of my analysis because I was able to confirm the enrichment of several genes. Comparison of the RT-qPCR results with the microarray results showed good agreement between the two methods and the correlation coefficient I observed is in line with other published studies (Morey et al., 2006). It has been reported that generally a higher correlation between RT-qPCR and microarray is observed for fold changes larger than 1.4 in upregulated genes (Beckman et al., 2004). Reasons to explain this include that low expressing genes or genes downregulated to a low level of expression are more susceptible to

140 false calls due to background noise and normalization procedures at the microarray level. Also, low expressing genes are going to be amplified at later cycles by RT-qPCR and those reads may be affected by decreased efficiency of the polymerase. My data is in line with these limitations, since for the upregulated genes there was a higher correlation coefficient and the majority of the Smg2/IgG fold changes detected by RT-qPCR are in the same direction predicted by the microarray, while the agreement was not as close for genes predicted not to be enriched in the Smaug2 RIP-Chip. Of the 200 Smaug2-associated mRNAs that I identified by this analysis, there were several interesting candidates for further investigation that will hopefully provide insights into Smaug2 functions that are independent of nanos1 mRNA, and that might shed additional light on the role of different repressive complexes in radial precursors. Nonetheless, nanos1 mRNA was not identified as a Smaug2 target using this approach, suggesting that the variability in the individual samples and the necessary pairwise validation only identified a subset of bona fide Smaug2-associated mRNAs. In agreement with this conclusion, in earlier experiments to identify Smaug2 binders via an SRE-based candidate approach, I identified and validated huc/elavl3, smaug2, aspm and mef2c mRNAs as Smaug2-associated mRNAs in the embryonic cortex, but only two of these, huc and smaug2 mRNAs were also identified in the microarray. Hence, my list is likely to be incomplete and in order to identify targets that may be missing, I would have to repeat the RIP-Chip with new samples and ensure that the age of the replicates matches throughout.

6.3.3.1 Limitations of the SRE score

Successful identification of mRNA transcripts associated with Smaug2 was not only tested by RT-qPCR but also by SRE score enrichment. Utilization of SRE scores relies on the high degree of conservation of the SAM domain across species and thus, high degree of conservation of Smaug binding preference (Aviv et al. 2003; 2006). The very good SRE score enrichment I observed between IgG and Smaug2 binders is in line with a similar analysis done in D. melanogaster embryos, showing that also in that model system mRNAs binding to Smaug are strongly enriched for SREs compared to IgG (Chen et al., 2014). Hence, the SRE enrichment analysis suggests that I identified several bona fide Smaug2-associated mRNAs. However, one aspect of the Smaug2-associated mRNA list that stands out is that some mRNAs that are associated with Smaug2 have low SRE scores, while others that are present in

141 the IgG have high SRE scores. There are several explanations for these observations. First and foremost, taking into account that 2 out of 12 transcripts that were predicted to be enriched, did not show any enrichment by RT-qPCR, it is possible that 17% of mRNAs on the Smaug2 list are incorrect and possibly some of these incorrect ones would be the ones with a low SRE score. Conversely, it is possible that some mRNAs on the IgG list with a high SRE score have been wrongly assigned to the list, although it should be noted that only 10 mRNAs on the IgG list, out of 172, have a score of 88 or higher, which is the score of nanos1. Alternatively, another explanation for transcripts in the Smaug2 list with low SRE score could be that these transcripts may be associating with Smaug2 indirectly via another RBP bound to Smaug. This could be plausible in light of the recent work showing that Smaug can associate with Argonaute1 and Argonaute2 in D. melanogaster (Pinder and Smibert, 2013a; 2013b). Another possibility is that mRNAs with SREs that fold poorly, and therefore have low SRE scores, could still bind to Smaug2 if they are concentrated by other factors localizing the mRNA to Smaug2. On the other hand, a transcript with a very high SRE will likely not bind to Smaug2 if they do not come in close proximity, such as being in different compartments, or if the binding of other factors masks the SRE inaccessible for binding. Another factor that may bias SRE scores in both directions also lies in the limitations of the folding model itself. In fact, the folding environment in which Smaug2 binds in vivo may be very different from the folding environment simulated in silico or from the environment of in vivo folding experiments (Zemora and Waldsich, 2010). It may be interesting to scan the 3’UTR of the Smaug2 binders for other RBP motifs, which may suggest additional factors interacting with these mRNAs. Still, even with these limitations, the availability of the SRE scores provided a great tool to validate the Smaug2 RIP-Chip and it also allowed the identification of Smaug2 high confidence binders when the initial candidate approach was undertaken.

6.4 General Discussion and Future Directions

The studies presented in this thesis indicate that mechanisms of post-transcriptional regulation, such as mRNA localization and mRNA repression play an important role in regulating radial precursor development. The presence of repressive RNPs is advantageous because it primes redial precursor to quickly generate their differentiated progeny in response to the appropriate environmental stimuli. In this model, radial precursors could transcribe and have

142 at the ready as many pro-differentiation mRNAs as needed but mechanisms of translational repression would prevent early translation of these differentiation factors within radial precursors, which would thus retain their precursor potential. However, translational repression would be lifted in response to environmental signals in the differentiating cell, driving subsequent development. Furthermore, the presence of these repressive RNPs would provide yet another means for fine-tuning development since different RNPs could respond to different stimuli. Finally, repressive RNPs could contribute to radial precursor diversity by targeting different subsets of mRNA for repression versus degradation, and thereby personalizing the radial precursors and perhaps conferring different developmental biases. Despite the appeal of this model, however, much work is needed to answer all the potential questions that it raises. Characterization of the different protein-mRNA complexes in radial precursors will be required to fully understand their heterogeneity and to understand if and to what extent they interact. In order to elucidate the protein composition of these RNPs, immunoprecipitation experiments for these proteins should be carried out and the isolated proteins should be analyzed by mass spectrometry, as has been done in rodent hippocampal neurons (see for example Fritzsche et al., 2013). An alternative approach would be to utilize a technique such as BioID to identify potential RNP-interacting proteins. In this approach, the protein of interest is fused to an E.coli biotin ligase domain. When other proteins interact with this fusion protein, even transiently, they are biotinylated, and can thus be readily purified and identified by mass spectrometry (Roux KJ et al. 2012). Putative interactors could then be validated endogenously in vitro and in vivo by, for instance, Proximity Ligation Assay and colocalization studies, or biochemically by immunoprecipitation. Furthermore, potential RNP complexes should be characterized in radial precursors and in neurons to ask whether they remain the same in both cell types or if they are remodelled. If remodelling occurred, then it would be very interesting because it may provide clues as to how these complexes are regulated. In addition to the identification of potential protein components in these RNP complexes, it would also be very informative to identify the mRNAs in these RNPs, and to compare these mRNAs in radial precursors and neurons, using approaches similar to those I have described in this thesis. For example, RIP-Chip to identify mRNAs associated with Nanos1 would allow me to ask whether or not it actually represses cell-cycle or self-renewal genes, as we have proposed, while identification of Pumilio2-interacting mRNAs would help me determine if all Pumilio2-containing RNPs in radial precursors also contain Staufen2, or whether Pumilio2 binds to a set of mRNAs independent of Stau2. 143

An additional open question involves the physiologically relevant mRNA targets of the different RNA binding proteins. Both Staufen2 and 4E-T associate with approximately 1500 mRNAs (Kusek et al., 2012; Yang et al., 2014) and Smaug2 associates with 200, although as discussed before, this number could be an underestimate. As previously discussed, it is likely that there are different subpopulations of mRNAs present in these repressive complexes and that, in addition to preventing early translation of mRNAs required for differentiation, these complexes also serve “housekeeping” functions. For example, they might ensure that steady- state translation continues as needed, prevent depletion of limiting translation factors, or recycle components of the translational machinery at the end of a given mRNAs life. To start testing all of this, it is paramount to elucidate how RBPs such as Staufen2, Smaug2 and 4E-T regulate their target mRNAs within the context of cortical development, since work to date has suggested that RBPs can promote different fates for their target mRNAs, and that these different regulatory mechanisms are cell type or context dependent (Kim et al., 2005; Sugimoto et al., 2015). In this regard, Smaug2, 4E-T and Staufen2 mRNA lists are valuable tools because one could pick candidate genes and then test in vitro or in vivo whether their expression at the mRNA and protein levels changes in response to RBP knockdown or overexpression. As one example of this type of question, I previously suggested that perhaps mRNAs that associate with Smaug2 but not 4E-T are translationally repressed but not degraded while perhaps those that associate with both Smaug2 and 4E-T might be targeted for degradation. One prediction of this model is that for mRNAs associated only with Smaug2, Smaug2 knockdown would cause an increase in the proteins that they encode without a corresponding change in mRNA expression. Conversely, for mRNAs associated with both Smaug2 and 4E-T (and predicted to be targeted for degradation), Smaug2 knockdown would cause an increase in both mRNA and protein levels. Of course, this model is likely to be simplistic, because for instance also 4E-T is known to have a role in translational repression independent of any role in degradation, but experiments like this would be starting points for dissecting these mechanisms Kamenska et al., 2014). Several good candidates for this kind of analysis would be the mRNAs associating with both Smaug2 and 4E-T. Also, as I showed in Chapter 4, the CCR4-NOT deadenylase complex is likely to be involved with Smaug2 and 4E-T regulation and this finding would suggest that Smaug2 may be involved with mRNA deadenylation. To test this further, one would have to assess polyA tail length of Smaug2 targets to determine whether the length increases upon Smaug2 or deadenylase complex downregulation. Also more generally, the CCR4-NOT complex is an 144 important regulator of mRNA turnover in mammalian cells but a role in the cortex has not been described, possibly due to the fact that several knockout models for the various subunits of the complex are embryonically lethal (Shirai et al., 2014). Hence, conditional knockouts or acute knockdown studies are needed. As previously suggested, identification of Smaug2 binding partners via mass spectrometry or BioID would be informative to generate a list of potential protein interactor candidates that could be tested further to fully determine how Smaug2 regulates radial precursor cell biology. Another important question that will ultimately need to be answered is how these RNPs are regulated. One obvious possibility would be phosphorylation and there are well- characterized examples of this, such as Zipcode-Binding Protein (ZBP). ZBP binds and represses β-actin mRNA in neurons but when ZBP is phosphorylated by Src, it releases β-actin mRNA, which is then translated (Hüttelmaier et al., 2005). Another important RBP whose translational repression is controlled by phosphorylation is FMRP, but unlike ZBP, it represses translation only when it is phosphorylated (Coffee Jr et al., 2012). With respect to Smaug regulation, phosphorylation may be involved, since Smaug1 has been shown to be phosphorylated in vitro by Akt (Chen et al., 2014b) and this site in conserved in Smaug2 (Johnson et al., 2011). 4E-T instead is phosphorylated by JNK (Cargnello et al. 2012) but interestingly 4E-T phosphorylation promotes its inclusion to P-bodies. Staufen instead has not been shown to be phosphorylated, but it does interact with protein phosphatase 1 in the rat brain (Monshausen et al. 2002). Hence, Smaug2, 4E-T and Staufen2 could all be regulated by phosphorylation although it will have to be determined whether phosphorylation promotes their repressive activity or not. Of course, the repressive nature of the Staufen2 granule could also be controlled by regulation of Pumilio2, which is known to function as a translational repressor. Phosphorylation, however, is unlikely to be the only way to regulate these repressive complexes and perhaps in some instances, translation occurs when the repressor is degraded or when a different isoform is expressed. Interestingly, Staufen1 has been recently described to bind to the CDS of poorly translated mRNAs in HEK-293T cells but their translation is promoted by Staufen1 degradation (Sugimoto et al. 2015), which occurs at mitosis in some cell types (Boulay et al., 2014). On the other hand, a study in mouse lung endothelial cells showed that two 4E-T isoforms exist and that while the short one promotes mRNA degradation in P- bodies, the long one does not (Chang et al., 2014). Interestingly, the genesis of one form versus the other occurs by alternative splicing and is regulated by HuR, an RBP that regulates neurogenesis in the mouse cortex (Kraushar et al., 2014). Our laboratory has shown that 4E-T is 145 a critical regulator of cortical development but we still don't know what regulates 4E-T, and HuR would be an interesting candidate in this regard. It will be interesting to see whether naturally-occurring variants of Smaug2 also occur in vivo and whether or not they regulate mRNA repression. Another challenge for future studies will be to identify the signalling pathways upstream of these RNPs that regulate their assembly, disassembly and/or remodelling. As previously mentioned, in this thesis I mainly focused on the role of these RBPs in the context of radial precursor development; however, work on Stau2 and Smaug1 in other systems and the localization defect I observe upon Staufen2 and Smaug2 knockdown, hint at the fact that they may also play additional roles in neurons. To date, conditional or cell-type specific knockout models for these two proteins are not available but they would be useful to address these questions. In the interim, acute knockdown of Smaug2 or Stau2 could be performed selectively in neurons by expressing the shRNA under the control of a neuronal promoter, such as Talpha1 alpha-tubulin (Gloster et al., 1999). Since GO analysis of the Stau2 and Smaug2 mRNA targets has already highlighted potential candidates involved in migration, it would be interesting to knockdown Staufen2 and Smaug2 in culture and test how the expression of these candidate genes changes at the mRNA and protein levels. For candidates whose expression was altered, it would be interesting to try to rescue the neuronal migration deficits that occur following Smaug2 knockdown by coincidently knocking down these putative targets, as I did for Smaug2 and Nanos1 with regard to neurogenesis. Also, cortical precursor or cortical neuron cultures could be utilized to test whether Smaug2 or Stau2 knockdown in maturing or mature neurons impacts neuronal maturation and morphology. Are RBPs also relevant for human brain development as they are in mouse development? As previously mentioned, the mouse and human brains share several important developmental similarities, such as conserved cell types and cortical layer organization. However, they are also different in a number of important ways. First and foremost, the human and mouse brains differ with respect to their size. The larger human brain is thought to arise as a consequence of a larger human founder population of neuroepithelial stem cells that undergoes extended proliferation and then extended neurogenesis and gliogenesis (Sidman & Rakic, 1973; Rakic, 1995). Second, the human brain has a highly folded structure, whereas the mouse brain is smooth. This difference likely arose to increase the overall surface area of the human brain without massively increasing the overall cranial size or cortical thickness (it is noteworthy that the human cortex has one thousand times more surface area than the mouse brain but it is only ten times thicker) (Fernández et al., 2016). This lateral expansion and gyrification of the human 146 cortex is thought to be the result of a more heterogeneous population of human neural precursors and an additional germinal zone called the outer SVZ (Smart et al., 2002; Fietz et al., 2010; Hansen et al., 2010; Reillo and Borrell, 2012). This increased complexity is thought to allow for much greater amplification of the progeny of single neuroepithelial stem cells, thereby allowing for the development of a larger, gyrified cortex. Support for this idea comes from experiments where mouse neural precursors were manipulated so that the population of radial precursor cells proliferated and expanded laterally, a manipulation that led to in some cases led to gyrification of the mouse cortex (Chenn and Walsh, 2002). Several explanations have been put forth to explain these differences at the molecular level. One potential explanation is that the human brain has specific proteins conferring these differences and in support of this view, approximately 50 proteins specific to the human brain have recently been identified (Florio et al., 2015). Another interesting observation is that several human genes encoding important developmental and morphogenetic proteins are flanked by non-coding regulatory regions with a very large rate of mutation, which might have modified the activity of their neighbouring coding sequences over time (Pollard et al., 2006a, b). Last but not least, however, a role in this increased complexity could also have been played by RBPs. For instance, it has been observed that there is an increased number of non-coding RNAs at points of evolutionary divergence (Heimberg et al., 2008; Aprea and Calegari, 2015; Liu and Sun, 2015). Work in macaque embryos has shown that hundreds of miRNAs are expressed in the macaque developing cortex but not in the mouse, and that those miRNAs regulate cell cycle and neurogenesis genes (Arcila et al., 2014). Therefore, since all of the RBPs studied in this thesis have clear homologues in humans, it is possible to speculate that i) they also play important roles in human cortical development and ii) we are actually underestimating the developmental importance of RBPs in human brain development because the mouse model may only capture certain aspects. In conclusion, the work presented here shows that in addition to the known roles that post-transcriptional regulation plays in the mature nervous system, it is also plays an important role in ensuring correct development of the mammalian cortex. With Staufen2, Smaug2, 4E-T and Nanos1, we are just beginning to discover an additional layer of radial precursor regulation involving RNA binding proteins and their associated repressive complexes. Moreover, emerging evidence indicates that dysregulation of these proteins in the humans might be important for neurodevelopmental disorders (Yang et al., 2014; Zhou et al., 2014; Sander, 2015; Bryant and Yazdani, 2016; Yano et al., 2016). Hopefully, by gaining a better understanding of how these 147

RNA-protein complexes work, we will be able to develop novel tools to restore physiological function and to promote repair.

148

REFERENCES

Aaku-Saraste E, Hellwig A and Huttner WB (1996) Loss of occludin and functional tight junctions, but not ZO-1, during neural tube closure-remodelling of the neuroepithelium prior to neurogenesis. Dev Biol 180: 664-679.

Abe T and Fujimori T (2013) Reporter mouse lines for fluorescence imaging. Dev Growth Differ 55: 390-405.

Agoston Z and Schulte D (2009) Meis2 competes with the Groucho co-repressor Tle4 for binding to Otx2 and specifies tectal fate without induction of a secondary midbrain-hindbrain boundary organizer. Development 136: 3311-3322.

Ahmed I, Sbodio JI, Harraz MM, Tyagi R, Grima JC, Albacarys LK, Hubbi ME, Xu R, Kim S, Paul BD and Snyder SH (2015) Huntington's disease: Neural dysfunction linked to inositol polyphosphate multikinase. Proc Natl Acad Sci U S A 112: 9751-9756.

Akamatsu W, Okano HJ, Osumi N, Inoue T, Nakamura S, Sakakibara S, Miura M, Matsuo N, Darnell RB and Okano H (1999) Mammalian ELAV-like neuronal RNA-binding proteins HuB and HuC promote neuronal development in both the central and the peripheral nervous systems. Proc Natl Acad Sci U S A 96: 9885-9890.

Akimoto J, Itoh H, Miwa T and Ikeda K (1993) Immunohistochemical study of glutamine synthetase expression in early glial development. Brain Res Dev Brain Res 72: 9-14.

Alvarez-Buylla A and Lim DA (2004) For the long run: maintaining germinal niches in the adult brain. Neuron 41:683-686.

Anderson SA, Kaznowski CE, Horn C, Rubenstein JL and McConnell SK (2002) Distinct origins of neocortical projection neurons and interneurons in vivo. Cereb Cortex 12: 702-709.

149

Anthony TE, Klein C, Fishell G and Heintz N (2004) Radial glia serve as neuronal progenitors in all regions of the central nervous system. Neuron 41: 881-890.

Aprea J and Calegari F (2015) Long non-coding RNAs in corticogenesis: deciphering the non- coding code of the brain. EMBO J 34: 2865–2884.

Arcila ML, Betizeau M, Cambronne XA, Guzman E, Doerflinger N, Bouhallier F, Zhou H, Wu B, Rani N, Bassett DS, Borello U, Huissoud C, Goodman RH, Dehay C, Kosik KS (2014) Novel primate miRNAs coevolved with ancient target genes in germinal zone-specific expression patterns. Neuron 81: 1255–1262.

Arnold SJ, Huang GJ, Cheung AF, Era T, Nishikawa S, Bikoff EK, Molnár Z, Robertson EJ and Groszer M (2008) The T-box transcription factor Eomes/Tbr2 regulates neurogenesis in the cortical subventricular zone. Genes Dev 22: 2479-2484.

Asaoka-Taguchi M, Yamada M, Nakamura A, Hanyu K and Kobayashi S (1999) Maternal Pumilio acts together with Nanos in germline development in Drosophila embryos. Nat Cell Biol 1: 431-437.

Aviv T, Lin Z, Lau S, Rendl LM, Sicheri F and Smibert CA (2003) The RNA-binding SAM domain of Smaug defines a new family of post-transcriptional regulators. Nat Struct Biol 10:614–621.

Aviv T, Lin Z, Ben-Ari G, Smibert CA and Sicheri F (2006) Sequence-specific recognition of RNA hairpins by the SAM domain of Vts1p. Nat Struct Mol Biol 13:168-176.

Baas AF, Gabbett M, Rimac M, Kansikas M, Raphael M, Nievelstein RA, Nicholls W, Offerhaus J, Bodmer D, Wernstedt A, Krabichler B, Strasser U, Nyström M, Zschocke J, Robertson SP, van Haelst MM and Wimmer K (2013) Agenesis of the corpus callosum and gray matter heterotopia in three patients with constitutional mismatch repair deficiency syndrome. Eur J Hum Genet 21: 55-61.

150

Badura M, Braunstein S, Zavadil J and Schneider RJ (2012) DNA damage and eIF4G1 in breast cancer cells reprogram translation for survival and DNA repair mRNAs. Proc Natl Acad Sci U S A 109: 18767-72.

Baez MV and Boccaccio GL (2005) Mammalian Smaug is a translational repressor that forms cytoplasmic foci similar to stress granules. J Biol Chem 280:43131-43140.

Baez MV, Luchelli L, Maschi D, Habif M, Pascual M, Thomas ME and Boccaccio GL (2011) Smaug1 mRNA-silencing foci respond to NMDA and modulate synapse formation. J Cell Biol 195:1141-1157.

Bahar Halpern K, Caspi I, Lemze D, Levy M, Landen S, Elinav E, Ulitsky I and Itzkovitz S (2015) Nuclear Retention of mRNA in Mammalian Tissues. Cell Rep 13: 2653-2662.

Balagopal V and Parker R (2009) Polysomes, P bodies and stress granules: states and fates of eukaryotic mRNAs. Curr Opin Cell Biol 21: 403-408.

Battelli C, Nikopoulos GN, Mitchell JG and Verdi JM (2006) The RNA-binding protein Musashi-1 regulates neural development through the translational repression of p21WAF-1.

Battiste J, Helms AW, Kim EJ, Savage TK, Lagace DC, Mandyam CD, Eisch AJ, Miyoshi G and Johnson JE (2007) Ascl1 defines sequentially generated lineage-restricted neuronal and oligodendrocyte precursor cells in the spinal cord. Development 134: 285-293.

Baxter PA, Lin Q, Mao H, Kogiso M, Zhao X, Liu Z, Huang Y, Voicu H, Gurusiddappa S, Su JM, Adesina AM, Perlaky L, Dauser RC, Leung HC, Muraszko KM, Heth JA, Fan X, Lau CC, Man TK, Chintagumpala M and Li XN (2014) Silencing BMI1 eliminates tumour formation of paediatric glioma CD133+ cells not by affecting known targets but by down-regulating a novel set of core genes. Acta Neuropathol Commun 2:160.

Bayatti N, Moss JA, Sun L, Ambrose P, Ward JFH, Lindsay S, Clowry GJ (2008) A molecular neuroanatomical study of the developing human neocortex from 8 to 17 postconceptional weeks

151 revealing the early differentiation of the subplate and subventricular zone. Cereb Cortex 18: 1536–1548.

Bayod S, Felice P, Andrés P, Rosa P, Camins A, Pallàs M and Canudas AM (2015) Downregulation of canonical Wnt signalling in hippocampus of SAMP8 mice. Neurobiol Aging. 36: 720-9.

Becalska AN and Gavis ER (2009) Lighting up mRNA localization in Drosophila oogenesis. Development 136: 2493-2503.

Becker AJ, Blümcke I, Urbach H, Hans V and Majores M (2006) Molecular neuropathology of epilepsy-associated glioneuronal malformations. J Neuropathol Exp Neurol 65: 99-108.

Becker L, Mito T, Takashima S and Onodera K (1991) Growth and development of the brain in Down syndrome. Prog Clin Biol Res 373: 133-152.

Beckman KB, Lee KY, Golden T and Melov S (2004) Gene expression profiling in mitochondrial disease: assessment of microarray accuracy by highthroughput Q-PCR. Mitochondrion 4: 453-470.

Benoit B, He CH, Zhang F, Votruba SM, Tadros W, Westwood JT, Smibert CA, Lipshitz HD and Theurkauf WE (2009) An essential role for the RNA-binding protein Smaug during the Drosophila maternal-to-zygotic transition. Development 136:923-932.

Blaise S, Kneib M, Rousseau A, Gambino F, Chenard MP, Messadeq N, Muckenstrum M, Alpy F, Tomasetto C, Humeau Y and Rio MC (2012) In vivo evidence that TRAF4 is required for central nervous system myelin homeostasis. PLoS One 7: e30917.

Bonfanti L and Peretto P (2007) Radial glial origin of the adult neural stem cells in the subventricular zone. Prog Neurobiol 83: 24-36.

152

Boron WF and Boulpaep EL (2009) Medical Physiology (2nd ed.). Philadelphia, PA: Saunder Elsevier.

Boulay K, Ghram M, Viranaicken W, Trépanier V, Mollet S, Fréchina C and DesGroseillers L (2014) Cell cycle-dependent regulation of the RNA-binding protein Staufen1. Nucleic Acids Res 42: 7867-7883.

Braun JE, Huntzinger E, Fauser M and Izaurralde E (2011) GW182 proteins directly recruit cytoplasmic deadenylase complexes to miRNA targets. Mol Cell 44: 120-133.

Briscoe J and Small S (2015) Morphogen rules: design principles of gradient-mediated embryo patterning. Development 142: 3996-4009.

Bryant CD and Yazdani N (2016) RNA-binding proteins, neural development and the addictions. Genes Brain Behav 15: 169-186.

Bultje RS, Castaneda-Castellanos DR, Jan LY, Jan YN, Kriegstein AR, Shi SH (2009) Mammalian Par3 regulates progenitor cell asymmetric division via notch signaling in the developing neocortex. Neuron 63: 189-202.

Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, Mueller R, Nolan T, Pfaffl MW, Shipley GL, Vandesompele J and Wittwer CT (2009) The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem 55:611-622.

Calegari F and Huttner WB (2003) An inhibition of cyclin-dependent kinases that lengthens, but does not arrest, neuroepithelial cell cycle induces premature neurogenesis. J Cell Sci 116: 4947-4955.

Calegari F, Haubensak W, Haffner C and Huttner WB (2005) Selective lengthening of the cell cycle in the neurogenic subpopulation of neural progenitor cells during mouse brain development. J Neurosci 25: 6533-6538.

Camarero G, Tyrsin OY, Xiang C, Pfeiffer V, Pleiser S, Wiese S, Götz R and Rapp 153

UR (2006) Cortical migration defects in mice expressing A-RAF from the B-RAF locus. Mol Cell Biol 26: 7103-7115.

Campbell K (2003) Dorso-ventral patterning in the mammalian telencephalon. Curr Opin Neurobiol 13: 50-56.

Cao W, Sohail M, Liu G, Koumbadinga GA, Lobo VG and Xie J (2011) Differential effects of PKA-controlled CaMKK2 variants on neuronal differentiation. RNA Biol 8: 1061-72.

Cappello S, Attardo A, Wu X, Iwasato T, Itohara S, Wilsch-Bräuninger M, Eilken HM, Rieger MA, Schroeder TT, Huttner WB, Brakebusch C and Götz M (2006) The Rho-GTPase cdc42 regulates neural progenitor fate at the apical surface. Nat Neurosci 9: 1099-1107.

Cargnello M, Tcherkezian J, Dorn JF, Huttlin EL, Maddox PS, Gygi SP, Roux PP (2012) Phosphorylation of the eukaryotic translation initiation factor 4E-transporter (4E-T) by c-Jun N- terminal kinase promotes stress-dependent P-body assembly. Mol Cell Biol 32: 4572-4584.

Chang SH, Elemento O, Zhang J, Zhuang ZW, Simons M and Hla T (2014) ELAVL1 regulates alternative splicing of eIF4E transporter to promote postnatal angiogenesis. Proc Natl Acad Sci U S A 111: 18309-14.

Chen JA, Wang Q, Davis-Turak J, Li Y, Karydas AM, Hsu SC, Sears RL, Chatzopoulou D, Huang AY, Wojta KJ, Klein E, Lee J, Beekly DL, Boxer A, Faber KM, Haase CM, Miller J, Poon WW, Rosen A, Rosen H, Sapozhnikova A, Shapira J, Varpetian A, Foroud TM, Levenson RW, Levey AI, Kukull WA, Mendez MF, Ringman J, Chui H, Cotman C, DeCarli C, Miller BL, Geschwind DH and Coppola G (2015) A multiancestral genome-wide exome array study of Alzheimer disease, frontotemporal dementia, and progressive supranuclear palsy. JAMA Neurol 72:414-22.

Chen L, Dumelie JG, Li X, Cheng MHK, Yang Z, Laver JD, Siddiqui NU, Westwood JT, Morris Q, Lipshitz HD and Smibert CA (2014a) Global regulation of mRNA translation and stability in the early Drosophila embryo by the Smaug RNA-binding protein. Genome Biol 15:1-21. 154

Chen Z, Holland W, Shelton JM, Ali A, Zhan X, Won S, Tomisato W, Liu C, Li X, Moresco EM and Beutler B (2014b) Mutation of mouse Samd4 causes leanness, myopathy, uncoupled mitochondrial respiration, and dysregulated mTORC1 signaling. PNAS 111:7367-7372.

Chenn A and McConnell SK (1995) Cleavage orientation and the asymmetric inheritance of Notch1 immunoreactivity in mammalian neurogenesis. Cell 82: 631-641.

Chenn A, Zhang YA, Chang BT and McConnell SK (1998) Intrinsic polarity of mammalian neuroepithelial cells. Mol Cell Neurosci 11: 183-193.

Chenn A and Walsh CA (2002) Regulation of cerebral cortical size by control of cell cycle exit in neural precursors. Science 297: 365 – 369.

Chia W, Somers WG and Wang H (2008) Drosophila neuroblast asymmetric divisions: cell cycle regulators, asymmetric protein localization, and tumorigenesis. J Cell Biol 180: 267-272.

Cho EA, Prindle MJ and Dressler GR (2003) BRCT domain-containing protein PTIP is essential for progression through mitosis. Mol Cell Biol 23: 1666-16673.

Choi BH (1981) Radial glia of the developing human fetal spinal cord: Golgi, immunohistochemical and electron microscopic study. Brain Res 227: 249-267.

Clay MR and Halloran MC (2013) Rho activation is apically restricted by Arhgap1 in neural crest cells and drives epithelial-to-mesenchymal transition. Development 140: 3198-3209.

Cockburn DM, Charish J, Tassew NG, Eubanks J, Bremner R, Macchi P and Monnier PP (2012) The double-stranded RNA-binding protein Staufen 2 regulates eye size. Mol Cell Neurosci 51: 101-111.

Coffee RL Jr, Williamson AJ, Adkins CM, Gray MC, Page TL, Broadie K (2012) In vivo neuronal function of the fragile X mental retardation protein is regulated by phosphorylation. Hum Mol Genet 21: 900-915. 155

Cohen R, Basel-Vanagaite L, Goldberg-Stern H, Halevy A, Shuper A, Feingold-Zadok M, Behar DM and Straussberg R (2014) Two siblings with early infantile myoclonic encephalopathy due to mutation in the gene encoding mitochondrial glutamate/H+ symporter SLC25A22. Eur J Paediatr Neurol 18: 801-805.

Conti L, Pollard SM, Gorba T, Reitano E, Toselli M, Biella G, Sun Y, Sanzone S, Ying QL, Cattaneo E and Smith A (2005) Niche independent symmetrical self-renewal of a mammalian tissue stem cell. PLoS Biol 3:e283.

Corbeil D, Fargeas CA and Huttner WB (2001) Rat prominin, like its mouse and human orthologues, is a pentaspan membrane glycoprotein. Biochem Biophys Res Commun 285: 939- 944.

Cremisi F (2013) MicroRNAs and cell fate in cortical and retinal development. Front Cell Neurosci 7:141-149.

Creppe C, Malinouskaya L, Volvert ML, Gillard M, Close P, Malaise O, Laguesse S, Cornez I, Rahmouni S, Ormenese S, Belachew S, Malgrange B, Chapelle JP, Siebenlist U, Moonen G, Chariot A and Nguyen L (2009) Elongator controls the migration and differentiation of cortical neurons through acetylation of alpha-tubulin. Cell 136: 551-64

Culurgioni S and Mapelli M (2013) Going vertical: functional role and working principles of the protein Inscuteable in asymmetric cell divisions. Cell Mol Life Sci 70: 4039-4046.

Dahanukar A and Wharton RP (1996) The Nanos gradient in Drosophila embryos is generated by translational regulation. Genes & Dev 10:2610-2620.

Dahanukar A, Walker JA and Wharton RP (1999) Smaug, a novel RNA-binding protein that operates a translational switch in Drosophila. Mol Cell 4: 209–218.

156

Datla US, Scovill NC, Brokamp AJ, Kim E, Asch AS and Lee MH (2014) Role of PUF-8/PUF protein in stem cell control, sperm-oocyte decision and cell fate reprogramming. J Cell Physiol 229: 1306-1311.

Davidovic L, Bechara E, Gravel M, Jaglin XH, Tremblay S, Sik A, Bardoni B and Khandjian EW (2006) The nuclear microspherule protein 58 is a novel RNA-binding protein that interacts with fragile X mental retardation protein in polyribosomal mRNPs from neurons. Hum Mol Genet 15:1525-38.

Davidson A, Parton RM, Rabouille C, Weil TT, Davis I (2016) Localized Translation of gurken/TGF-α mRNA during Axis Specification Is Controlled by Access to Orb/CPEB on Processing Bodies. Cell Rep 14: 2451-2462.

Davis AA and Temple S (1994) A self-renewing multipotential stem cell in embryonic rat cerebral cortex. Nature 372: 263-266.

Dawson MR, Polito A, Levine JM and Reynolds R (2003) NG2-expressing glial progenitor cells: an abundant and widespread population of cycling cells in the adult rat CNS. Mol Cell Neurosci 24: 476-488.

DeAndrade MP, Johnson RL Jr, Unger EL, Zhang L, van Groen T, Gamble KL and Li Y (2012) Motor restlessness, sleep disturbances, thermal sensory alterations and elevated serum iron levels in Btbd9 mutant mice. Hum Mol Genet 21: 3984-3992.

Delgado-Esteban M, Garcia-Higuera I, Moreno S and Almeida A (2014) The E3 ubiquitin ligase APC/C-Cdh1 coordinates neurogenesis and cortical size during development. Free Radic Biol Med 75 doi: 10.1016

DiCicco-Bloom E, Lord C, Zwaigenbaum L, Courchesne E, Dager SR, Schmitz C, Schultz RT, Crawley J and Young LJ (2006) The developmental neurobiology of autism spectrum disorder. J Neurosci 26: 6897-6906.

157

Ding D, Song T, Jun W, Tan Z and Fang J (2015) Decreased expression of the SPOP gene is associated with poor prognosis in glioma. Int J Oncol 46: 333-341.

Doe CQ and Skeath JB (1996) Neurogenesis in the insect central nervous system. Curr Opin Neurobiol 6: 18-24.

Doetsch F, Caillé I, Lim DA, García-Verdugo JM and Alvarez-Buylla A (1999) Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell 97: 703-716.

Dostie J, Ferraiuolo M, Pause A, Adam SA and Sonenberg N (2000) A novel shuttling protein, 4E-T, mediates the nuclear import of the mRNA 5' cap-binding protein, eIF4E. EMBO J 19:3142-3156.

Doyle M and Kiebler MA (2011) Mechanisms of dendritic mRNA transport and its role in synaptic tagging. EMBO J 30: 3540-3552.

Du C, Wu H and Leng RP (2016) UBE4B targets phosphorylated p53 at serines 15 and 392 for degradation. Oncotarget 7: 2823-2836.

Dubnau J, Chiang AS, Grady L, Barditch J, Gossweiler S, McNeil J, Smith P, Buldoc F, Scott R, Certa U, Broger C and Tully T (2003) The staufen/Pumilio pathway is involved in Drosophila long-term memory. Curr Biol 13: 286-296.

Duchaȋne TF, Hemraj I, Furic L, Deitinghoff A, Kiebler MA and DesGroseillers L (2002) Staufen2 isoforms localize to the somatodendritic domain of neurons and interact with different organelles. J Cell Sci 115: 3285-3295.

Dyer MA, Livesey FJ, Cepko CL and Oliver G (2003) Prox1 function controls progenitor cell proliferation and horizontal cell genesis in the mammalian retina. Nat Genet 34: 53-58.

Edri R, Yaffe Y, Ziller MJ, Mutukula N, Volkman R, David E, Jacob-Hirsch J, Malcov H, Levy C, Rechavi G, Gat-Viks I, Meissner A and Elkabetz Y (2015) Analysing human neural stem cell

158 ontogeny by consecutive isolation of Notch active neural progenitors. Nat Commun 6: 6500- 6515.

Elbediwy A, Zihni C, Terry SJ, Clark P, Matter K and Balda MS (2012) Epithelial junction formation requires confinement of Cdc42 activity by a novel SH3BP1 complex. J Cell Biol 198: 677-93.

Elvira G, Wasiak S, Blandford V, Tong XK, Serrano A, Fan X, del Rayo Sánchez-Carbente M, Servant F, Bell AW, Boismenu D (2006) Characterization of an RNA granule from developing brain. Mol Cell Proteomics 5: 635-651.

Englund C, Fink A, Lau C, Pham D, Daza RA, Bulfone A, Kowalczyk T and Hevner RF (2005) Pax6, Tbr2 and Tbr1 are expressed sequentially by radial glia, intermediate progenitor cells and postmitotic neurons in developing neocortex. J Neurosci 25: 247-251.

Erfani P, Tome-Garcia J, Canoll P, Doetsch F and Tsankova NM (2015) EGFR promoter exhibits dynamic histone modifications and binding of ASH2L and P300 in human germinal matrix and gliomas. Epigenetics 10: 496-507.

Fauser S, Häussler U, Donkels C, Huber S, Nakagawa J, Prinz M, Schulze-Bonhage A, Zentner J and Haas CA (2013) Disorganization of neocortical lamination in focal cortical dysplasia is brain-region dependent: evidence from layer-specific marker expression. Acta Neuropathol Commun 1: 47.

Feng L, Hatter ME and Heintz N (1994) Brain lipid-binding protein (BLBP): A novel signaling system in the developing mammalian CNS. Neuron 12: 895-908.

Fernandez-Alvarez AJ, Pascual ML, Boccaccio GL and Thomas MG (2016) Smaug variants in neural and non-neuronal cells. Communicative & Integrative Biology DOI: 10.1080/19420889.2016.1139252

Fernandez-Moya SM, Bauer KE and Kiebler MA (2014) Meet the players: local translation at the synapse. Front Mol Neurosci 7: 1-6. 159

Fernández V, Llinares-Benadero C and Borrell V (2016) Cerebral cortex expansion and folding: what have we learned? EMBO J 35: 1021-1044.

Ferrari F, Mercaldo V, Piccoli G, Sala C, Cannata S, Achsel T and Bagni C (2007) The fragile X mental retardation protein-RNP granules show an mGluR-dependent localization in the post- synaptic spines. Mol Cell Neurosci 34: 343-354.

Ferraiuolo MA, Basak S, Dostie J, Murray EL, Schoenberg DR and Sonenberg N (2005) A role for the eIF4E-binding protein 4E-T in P-body formation and mRNA decay. J Cell Biol 170:913- 924.

Fietz SA, Kelava I, Vogt J, Wilsch-Bräuninger M, Stenzel D, Fish JL, Corbeil D, Riehn A, Distler W, Nitsch R and Huttner WB (2010) OSVZ progenitors of human and ferret neocortex are epithelial-like and expand by integrin signalling. Nat Neurosci 13: 690-699.

Fish JL, Dehay C, Kennedy H, and Huttner WB (2008) Making bigger brains-the evolution of neural-progenitor-cell division. J Cell Sci 121: 2783–2793.

Florio M, Albert M, Taverna E, Namba T, Brandl H, Lewitus E, Haffner C, Sykes A, Wong FK, Peters J, Guhr E, Klemroth S, Prufer K, Kelso J, Naumann R, Nusslein I, Dahl A, Lachmann R, Paabo S and Huttner WB (2015) Human specific gene ARHGAP11B promotes basal progenitor amplification and neocortex expansion. Science 347: 1465–1470.

Forbes A and Lehmann R (1998) Nanos and Pumilio have critical roles in the development and function of Drosophila germline stem cells. Development 125: 679-690.

Fornaro M, Raimondo S, Lee JM and Giacobini-Robecchi MG (2007) Neuron-specific Hu proteins sub-cellular localization in primary sensory neurons. Ann Anat 189: 223-228.

Fransson S, Ruusala A and Aspenström P (2006) The atypical Rho GTPases Miro-1 and Miro-2 have essential roles in mitochondrial trafficking. Biochem Biophys Res Commun 344: 500-510.

160

Fritzsche R, Karra D, Bennett KL, Ang FY, Heraud-Farlow JE, Tolino M, Doyle M, Bauer KE, Thomas S, Planyavsky M, Arn E, Bakosova A, Jungwirth K, Hörmann A, Palfi Z, Sandholzer J, Schwarz M, Macchi P, Colinge J, Superti-Furga G and Kiebler MA (2013) Interactome of two diverse RNA granules links mRNA localization to translational repression in neurons. Cell Rep 5: 1749-1762.

Fuccillo M, Rutlin M and Fishell G (2006) Removal of Pax6 partially rescues the loss of ventral structures in Shh null mice. Cereb Cortex 16: 96-102.

Fuerstenberg S, Peng CY, Alvarez-Ortiz P, Hor T and Doe CQ (1998) Identification of Miranda protein domains regulating asymmetric cortical localization, cargo binding, and cortical release. Mol Cell Neurosci 12: 325-339.

Furic L, Maher-Laporte M and DesGroseillers L (2008) A genome-wide approach identifies distinct but overlapping subsets of cellular mRNAs associated with Staufen1- and Staufen2- containing ribonucleoprotein complexes. RNA 14: 324-335.

Gage FH (2002) Neurogenesis in the adult brain. J Neurosci 22: 612-613.

Gallagher D, Norman AA, Woodard CL, Yang G, Gauthier-Fisher A, Fujitani M, Vessey JP, Cancino GI, Sachewsky N, Woltjen K, Fatt MP, Morshead CM, Kaplan DR, Miller FD (2013) Transient maternal IL-6 mediates long-lasting changes in neural stem cell pools by deregulating an endogenous self-renewal pathway. Cell Stem Cell 13: 564-576.

Gallagher D, Voronova A, Zander MA, Cancino GI, Bramall A, Krause MP, Abad C, Tekin M, Neilsen PM, Callen DF, Scherer SW, Keller GM, Kaplan DR, Walz K and Miller FD (2015) Ankrd11 is a chromatin regulator involved in autism that is essential for neural development. Dev Cell 32:31-42.

Gauthier AS, Furstoss O, Araki T, Chan R, Neel BG, Kaplan DR, and Miller FD (2007) Control of CNS cell-fate decisions by SHP-2 and its dysregulation in Noonan syndrome. Neuron 54: 245-262.

161

Gauthier-Fisher A, Lin DC, Greeve M, Kaplan DR, Rottapel R and Miller FD (2009) Lfc and Tctex-1 regulate the genesis of neurons from cortical precursor cells. Nat Neurosci 12: 735-744.

Gautrey H, McConnell J, Lako M, Hall J and Hesketh J (2008) Staufen1 is expressed in preimplantation mouse embryos and is required for embryonic stem cell differentiation. Biochim Biophys Acta 1783: 1935-1942.

Gavis ER and Lehmann R (1994) Translational regulation of nanos by RNA localization. Nature 369: 315-318.

Gazea M, Tasouri E, Tolve M, Bosch V, Kabanova A, Gojak C, Kurtulmus B, Novikov O, Spatz J, Pereira G, Hübner W, Brodski C, Tucker KL and Blaess S (2016) Primary cilia are critical for Sonic hedgehog-mediated dopaminergic neurogenesis in the embryonic midbrain. Dev Biol 409: 55-71.

Ginter-Matuszewska B, Spik A, Rembiszewska A, Koyias C, Kupryjanczyk J, Jaruzelska J (2009) The SNARE-associated component Snapin Pumilio and Nanos1 proteins in human male germ cells. Mol Hum Reprod 15: 173-179.

Ginter-Matuszewska B, Kusz K, Spik A, Grzeszkowiak D, Rembiszewska A, Kupryjanczyk J, Jaruzelska J (2011) Nanos1 and Pumilio2 bind microRNAs biogenesis factor Gemin3, within chromatoid body in human germ cells. Histochem Cell Biol 136: 279-287.

Gloster A, El-Bizri H, Bamji SX, Rogers D and Miller FD (1999) Early induction of Talpha1 alpha-tubulin transcription in neurons of the developing nervous system. J Comp Neurol 405: 45-60.

Goetz SC and Anderson KV (2010) The primary cilium: a signalling centre during vertebrate development. Nat Rev Genet 11: 331-344.

162

Goetze B, Tuebing F, Xie Y, Dorostkar MM, Thomas S, Pehl U, Boehm S, Macchi P and Kiebler MA (2006) The brain-specific double-stranded RNA-binding protein Staufen2 is required for dendritic spine morphogenesis. J Cell Biol 172: 221-231.

Göhring I, Tagariello A, Endele S, Stolt CC, Ghassibé M, Fisher M, Thiel CT, Trautmann U, Vikkula M, Winterpacht A, FitzPatrick DR and Rauch A (2010) Disruption of ST5 is associated with mental retardation and multiple congenital anomalies. J Med Genet 47: 91-98.

Götz M, Hartfuss E and Malatesta P (2002) Radial glial cells as neuronal precursors: a new perspective on the correlation of morphology and lineage restriction in the developing cerebral cortex of mice. Brain Res Bull 57: 777-788.

Götz M and Huttner WB (2005) The cell biology of neurogenesis. Nat Rev Mol Cell Biol 6: 777-788.

Green JB, Gardner CD, Wharton RP and Aggarwal AK (2003) RNA recognition via the SAM domain of smaug. Molecular Cell 11: 1537-1548.

Griffin JN, Sondalle SB, Del Viso F, Baserga SJ and Khokha MK (2015) The ribosome biogenesis factor Nol11 is required for optimal rDNA transcription and craniofacial development in Xenopus. PLoS Genet 11:e1005018.

Grindley JC, Davidson DR and Hill RE (1995) The role of Pax-6 in eye and nasal development. Development 121: 1433-1442.

Grove EA, Williams BP, Li DQ, Hajihosseini M, Friedrich A and Price J (1993) Multiple restricted lineages in the embryonic rat cerebral cortex. Development 117: 553-561.

Grove EA and Fukuchi-Shimogori T (2003) Generating the cerebral cortical area map. Annu Rev Neurosci 26: 355-380.

Guérout N, Li X and Barnabé-Heider F (2014) Cell fate control in the developing central nervous system. Exp Cell Res 321: 77-83. 163

Ha JY, Kim JS, Kang YH, Bok E, Kim YS and Son JH (2014) Tnfaip8 l1/Oxi-β binds to FBXW5, increasing autophagy through activation of TSC2 in a Parkinson's disease model. J Neurochem 129: 527-38.

Hansen DV, Lui JH, Parker PR and Kriegstein AR (2010) Neurogenic radial glia in the outer subventricular zone of human neocortex. Nature 464: 554-561.

Haraguchi S, Tsuda M, Kitajima S, Sasaoka Y, Nomura-Kitabayashid A, Kurokawa K and Saga Y (2003) Nanos1: a mouse nanos gene expressed in the central nervous system is dispensable for normal development. Mech Dev 120: 721-731.

Hartfuss E, Galli R, Heins N and Götz M (2001) Characterization of CNS precursor subtypes and radial glia. Dev Biol 229: 15-30.

Hartsock A and Nelson WJ (2008) Adherens and tight junctions: structure, function and connections to the actin cytoskeleton. Biochim Biophys Acta 1778: 660-669.

Haubensak W, Attardo A, Denk W and Huttner B (2004) Neurons arise in the basal neuroepithelium of the early mammalian telecephalon: a major site of neurogenesis. Proc Natl Acad Sci USA 101: 3196-3201.

Hayashi Y, Hayashi M and Kobayashi S (2004) Nanos suppresses somatic cell fate in Drosophila germ line. PNAS 101:10338-10342.

Heimberg AM, Sempere LF, Moy VN, Donoghue PC and Peterson KJ (2008) MicroRNAs and the advent of vertebrate morphological complexity. Proc Natl Acad Sci USA 105: 2946–2950.

Heller S, Ernsberger U and Rohrer H (1996) Extrinsic signals in the developing nervous system: the role of neurokines during neurogenesis. Perspect Dev Neurobiol 4: 19-34.

164

Heraud-Farlow JE, Sharangdhar T, Li X, Pfeifer P, Tauber S, Orozco D, Hörmann A, Thomas S, Bakosova A, Farlow AR, Edbauer D, Lipshitz HD, Morris QD, Bilban M, Doyle M and Kiebler MA (2013) Staufen2 regulates neuronal target RNAs. Cell Rep 5:1115-1118.

Hernandez MC, Andres-Barquin PJ, Martinez S, Bulfone A, Rubenstein JL and Israel MA (1997) ENC-1: a novel mammalian kelch-related gene specifically expressed in the nervous system encodes an actin-binding protein. J Neurosci 17: 3038-51.

Hippenmeyer S (2014) Molecular pathways controlling the sequential steps of cortical projection neuron migration. Adv Exp Med Biol 800: 1-24.

Hirst CE, Lim SM, Pereira LA, Mayberry RA, Stanley EG and Elefanty AG (2010) Expression from a betageo gene trap in the Slain1 gene locus is predominantly associated with the developing nervous system. Int J Dev Biol 54: 1383-1388.

Hoeft F, Dai L, Haas BW, Sheau K, Mimura M, Mills D, Galaburda A, Bellugi U, Korenberg JR and Reiss AL (2014) Mapping genetically controlled neural circuits of social behavior and visuo-motor integration by a preliminary examination of atypical deletions with Williams syndrome. PLoS One 9:e104088.

Holmberg J and Frisén J (2002) Ephrins are not only unattractive. Trends Neurosci 25: 239-243.

Holt CE and Schuman EM (2013) The central dogma decentralized: new perspectives on RNA function and local translation in neurons. Neuron 80: 648-657.

Hong H, Kim J and Kim J (2015) Myosin heavy chain 10 (Myh10) is required for centriole migration during the biogenesis of primary cilia. Biochem Biophys Res Commun 461: 180-185.

Hong ST and Mah W (2015) A Critical Role of GIT1 in Vertebrate and Invertebrate Brain Development. Exp Neurobiol 24: 8-16.

Horisawa K, Imai T, Okano H and Yanagawa H (2009) 3’-Untranslated region of doublecortin mRNA is a binding target of the Musashi1 RNA-binding protein. FEBS Lett 583: 2429-2434. 165

Huang CJ, Chen HY, Lin WY and Choo KB (2014) Differential expression of speckled POZ protein, SPOP: putative regulation by miR-145. J Biosci 39: 401-413.

Huang DW, Sherman BT, Lempicki RA (2009a) Systematic and integrative analysis of large gene lists using DAVID Bioinformatics Resources. Nature Protoc 4:44-57.

Huang DW, Sherman BT, Lempicki RA (2009b) Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 37: 1-13. Proc Natl Acad Sci U S A 104: 4194-4199.

Huang Y1, Higginson DS, Hester L, Park MH and Snyder SH (2007) Neuronal growth and survival mediated by eIF5A, a polyamine-modified translation initiation factor.

Hulbert SW and Jiang YH (2016) Monogenic mouse models of autism spectrum disorders: Common mechanisms and missing links. Neuroscience 321: 3-23.

Hulsen T, de Vlieg J and Alkema W (2008) BioVenn - a web application for the comparison and visualization of biological lists using area-proportional Venn diagrams BMC Genomics 9: 488.

Hüttelmaier S, Zenklusen D, Lederer M, Dictenberg J, Lorenz M, Meng X, Bassell GJ, Condeelis J and Singer RH (2005) Spatial regulation of beta-actin translation by Src-dependent phosphorylation of ZBP1. Nature: 438: 512-515.

Huttner WB and Brand M (1997) Asymmetric division and polarity of neuroepithelial cells. Curr Opin Neurobiol 7: 29-39.

Hwangpo TA, Jordan JD, Premsrirut PK, Jayamaran G, Licht JD, Iyengar R and Neves SR (2012) G Protein-regulated inducer of neurite outgrowth (GRIN) modulates Sprouty protein repression of mitogen-activated protein kinase (MAPK) activation by growth factor stimulation. J Biol Chem 287: 13674-13685.

166

Igreja C and Izaurralde E (2011) Cup promotes deadenylation and inhibits decapping of mRNA targets. Genes Dev 25: 1955-1967.

Imai T, Tokunaga A, Yoshida T, Hashimoto M, Mikoshiba K, Weinmaster G, Nakafuku M, Okano H (2001) The neural RNA-binding protein Musashi1 translationally regulates mammalian numb gene expression by interacting with its mRNA. Mol Cell Biol 21: 3888-3900.

Imayoshi I and Kageyama R (2014a) bHLH Factors in self-renewal, multipotency, and fate choice of neural progenitor cells. Neuron 82:9-23.

Imayoshi I and Kageyama R (2014b) Oscillatory control of bHLH factors in neural progenitors. Trends in Neurosci 37: 531-538.

Inglis-Broadgate SL, Ocaka L, Banerjee R, Gaasenbeek M, Chapple JP, Cheetham ME, Clark BJ, Hunt DM and Halford S (2005) Isolation and characterization of murine Cds (CDP- diacylglycerol synthase) 1 and 2. Gene 356: 19-31.

Inoue M, Kuroda T, Honda A, Komabayashi-Suzuki M, Komai T, Shinkai Y and Mizutani K (2014) Prdm8 regulates the morphological transition at multipolar phase during neocortical development. PLoS One 9:e86356.

Inoue M, Iwai R, Yamanishi E, Yamagata K, Komabayashi-Suzuki M, Honda A, Komai T, Miyachi H, Kitano S, Watanabe C, Teshima W, Mizutani K (2015) Deletion of Prdm8 impairs development of upper-layer neocortical neurons. Genes Cells 20: 758-770.

Jaruzelska J, Kotecki M, Kusz K, Spik A, Firpo M and Reijo Pera RA (2003) Conservation of a Pumilio-Nanos complex from Drosophila germ plasm to human germ cells. Dev Genes Evol 213:120-126.

Javaherian A and Kriegstein A (2009) A stem cell niche for intermediate progenitor cells of the embryonic cortex. Cereb Cortex 19: 70-77.

167

Johnson C, Tinti M, Wood NT, Campbell DG, Toth R, Dubois F, Geraghty KM, Wong BH, Brown LJ, Tyler J, Gernez A, Chen S, Synowsky S and MacKintosh C (2011) Visualization and biochemical analyses of the emerging mammalian 14-3-3-phosphoproteome. Mol Cell Proteomics 10: M110.005751.

Joly W, Chartier A, Rojas-Rios P, Busseau I and Simonelig M (2013) The CCR4 deadenylase acts with Nanos and Pumilio in the fine-tuning of Mei-P26 expression to promote germline stem cell self-renwal. Stem Cell Reports 1: 411-424.

Julian LM, Vandenbosch R, Pakenham CA, Andrusiak MG, Nguyen AP, McClellan KA, Svoboda DS, Lagace DC, Park DS, Leone G, Blais A, and Slack RS (2013) Opposing regulation of Sox2 by cell-cycle effectors E2f3a and E2f3b in neural stem cells. Cell Stem Cell 12:440- 452.

Jung H, Gkogkas CG, Sonenberg N and Holt CE (2014) Remote control of gene function by local translation. Cell 157: 26-40.

Jung YS, Lee BK, Park HS, Shim JK, Kim SU, Lee SH, Baik EJ and Moon CH (2005) Activation of protein kinase C-delta attenuates kainate-induced cell death of cortical neurons. Neuroreport 16: 741-744.

Kadyrova LY, Habara Y, Lee TH and Wharton RP (2007) Translational control of maternal Cyclin B mRNA by Nanos in the Drosophila germline. Development 134:1519-1527.

Kafka A, Tomas D, Beroš V, Pećina HI, Zeljko M, Pećina-Šlaus N (2014) Brain metastases from lung cancer show increased expression of DVL1, DVL3 and beta-catenin and downregulation of E-cadherin. Int J Mol Sci 15: 10635-51.

Kaltezioti V, Kouroupi G, Oikonomaki M, Mantouvalou E, Stergiopoulos A, Charonis A, Rohrer H, Matsas R and Politis PK (2010) Prox1 regulates the notch1-mediated inhibition of neurogenesis. PLoS Biol 8: e1000565.

168

Kamenska A, Lu WT, Kubacka D, Broomhead H, Minshall N, Bushell M, and Standart N (2014) Human 4E-T represses translation of bound mRNAs and enhances microRNA-mediated silencing. Nucl Acids Res 42:3298-3313.

Kanai Y, Dohmae N and Hirokawa N (2004) Kinesin transports RNA: isolation and characterization of an RNA-transporting granule. Neuron 43: 513-525.

Kaneko-Oshikawa C, Nakagawa T, Yamada M, Yoshikawa H, Matsumoto M, Yada M, Hatakeyama S, Nakayama K and Nakayama KI (2005) Mammalian E4 is required for cardiac development and maintenance of the nervous system. Mol Cell Biol 25 :10953-10964.

Kato TM, Kawaguchi A, Kosodo Y, Niwa H and Matsuzaki F (2010) Lunatic fringe potentiates Notch signaling in the developing brain. Mol Cell Neurosci 45: 12-25.

Kawauchi T (2015) Cellular insights into cerebral cortical development: focusing on the locomotion mode of neuronal migration. Front Cell Neurosci 9: 394.

Kerjan G, Koizumi H, Han EB, Dubé CM, Djakovic SN, Patrick GN, Baram TZ, Heinemann SF and Gleeson JG (2009) Mice lacking doublecortin and doublecortin-like kinase 2 display altered hippocampal neuronal maturation and spontaneous seizures. Proc Natl Acad Sci U S A 106: 6766-71.

Kessaris N, Fogarty M, Iannarelli P, Grist M, Wegner M and Richardson WD (2006) Competing waves of oligodendrocytes in the forebrain and postnatal elimination of an embryonic lineage. Nat Neurosci 9: 173-179.

Kiebler MA, Hemraj I, Verkade P, Köhrmann M, Fortes P, Marión RM, Ortín J and Dotti CG (1999) The mammalian staufen protein localizes to the somatodendritic domain of cultured hippocampal neurons: implications for its involvement in mRNA transport. J Neurosci 19: 288– 297.

Kiebler MA and Bassell GJ (2006) Neuronal RNA granules: movers and makers. Neuron 51: 685-690. 169

Kim DW and Hirth F (2009) Genetic mechanisms regulating stem cell self-renewal and differentiation in the central nervous system of Drosophila. Cell Adh Migr 3: 402-411.

Kim WY (2012) NeuroD1 is an upstream regulator of NSCL1. Biochem Biophys Res Commun. 419: 27-31.

Kim YK, Furic L, DesGroseillers L and Maquat LE (2005) Mammalian Staufen1 recruits Upf1 to specific mRNA 3’ UTRs so as to elicit mRNA decay. Cell 120: 195-208.

Kimble J and Crittenden SL (2007) Controls of germline stem cell, entry into meiosis, and the sperm/oocyte decision in Caenorhabditis elegans. Annu Rev Cell Dev Biol 23: 405-433.

Klisch TJ, Souopgui J, Juergens K, Rust B, Pieler T and Henningfeld KA (2006) Mxi1 is essential for neurogenesis in Xenopus and acts by bridging the pan-neural and proneural genes. Dev Biol 292: 470-485.

Knoblich JA (1997) Mechanisms of asymmetric cell division during animal development. Curr Opin Cell Biol 9: 833-841.

Knoblich JA (2008) Mechanisms of asymmetric stem cell division. Cell 132: 583-597.

Knobloch M, Braun SM, Zurkirchen L, von Schoultz C, Zamboni N, Araúzo-Bravo MJ, Kovacs WJ, Karalay O, Suter U, Machado RA, Roccio M, Lutolf MP, Semenkovich CF and Jessberger S (2013) Metabolic control of adult neural stem cell activity by Fasn-dependent lipogenesis. Nature 493: 226-230.

Kobayashi S, Yamada M, Asaoka M and Kitamura T (1996) Essential role of the posterior morphogen nanos for germline development in Drosophila. Nature 380: 708-711.

Kobayashi T, Kageyama R (2014) Expression dynamics and functions of Hes factors in development and diseases. 110: 263-283.

170

Köhrmann M, Luo M, Kaether C, DesGroseillers L, Dotti CG and Kiebler MA (1999) Microtubule-dependent recruitment of Staufen-green fluorescent protein into large RNA- containing granules and subsequent dendritic transport in living hippocampal neurons. Mol Biol Cell 10:2945-2953.

Konno D, Shioi G, Shitamukai A, Mori A, Kiyonari H, Miyata T and Matsuzaki F (2008) Neuroepithelial progenitors undergo LGN-dependent planar divisions to maintain self- renewability during mammalian neurogenesis. Nat Cell Biol 10: 93-101.

Kosodo Y, Röper K, Haubensak W, Marzesco AM, Corbeil D and Huttner WB (2004) Asymmetric distribution of the apical plasma membrane during neurogenic divisions of mammalian neuroepithelial cells. EMBO 23: 2314-2324.

Kotani T, Yasuda K, Ota R and Yamashita M (2013) Cyclin B1 mRNA translation is temporally controlled through formation and disassembly of RNA granules. J Cell Biol 202: 1041-1055.

Kowalczyk T, Pontius A, Englund C, Daza RA, Bedogni F, Hodge R, Attardo A, Bell C, Huttner WB, Hevner RF (2009) Intermediate neuronal progenitors (basal progenitors) produce pyramidal-projection neurons for all layers if cerebral cortex. Cereb Cortex 19: 2439-2450.

Krause SA, Loupart ML, Vass S, Schoenfelder S, Harrison S and Heck MM (2001) Loss of cell cycle checkpoint control in Drosophila Rfc4 mutants. Mol Cell Biol 21: 5156-5168.

Kraushar ML, Thompson K, Wijeratne HR, Viljetic B, Sakers K, Marson JW, Kontoyiannis DL, Buyske S, Hart RP and Rasin MR (2014) Temporally defined neocortical translation and polysome assembly are determined by the RNA-binding protein Hu antigen R. 111: E3815-24.

Krebs DL, Metcalf D, Merson TD, Voss AK, Thomas T, Zhang JG, Rakar S, O'bryan MK, Willson TA, Viney EM, Mielke LA, Nicola NA, Hilton DJ and Alexander WS (2004) Development of hydrocephalus in mice lacking SOCS7. Proc Natl Acad Sci U S A 101: 15446- 15451.

171

Krichevsky AM and Kosik KS (2001) Neuronal RNA granules: a link between RNA localization and stimulation-dependent translation. Neuron 32: 683-696.

Kriegstein A and Alvarez-Buylla A (2009) The glial nature of embryonic and adult neural stem cells. Annu Rev Neurosci 32: 149-184.

Kriegstein AR and Götz M (2003) Radial glia diversity: a matter of cell fate. Glia 43: 37-43.

Kröcher T, Röckle I, Diederichs U, Weinhold B, Burkhardt H, Yanagawa Y, Gerardy-Schahn R, and Hildebrandt H (2014) A crucial role for polysialic acid in developmental interneuron migration and the establishment of interneuron densities in the mouse prefrontal cortex. Development 141: 3022-3032.

Krushel LA, Johnston JG, Fishell G, Tibshirani R and Van der Kooy D (1993) Spatially localized neuronal cell lineages in the developing mammalian forebrain. Neuroscience 53: 1035- 1047.

Kugler J and Lasko P (2009) Localization, anchoring and translational control of oskar, gurken, bicoid and nanos mRNA during Drosophila oogenesis. Fly (Austin) 3: 15-28.

Kusek G, Campbell M, Doyle F, Tenenbaum SA, Kiebler M and Temple S (2012) Asymmetric segregation of the double-stranded RNA binding protein Staufen2 during mammalian neural stem cell divisions promotes lineage progression. Cell Stem Cell 11:505-516.

Kusz-Zamelczyk K, Sajek M, Spik A, Glazar R, Jędrzejczak P, Latos-Bieleńska A, Kotecki M, Pawelczyk L, Jaruzelska J (2013) Mutations of Nanos1, a human homologue of the Drosophila morphogens, are associated with a lack of germ cells in testes or severe oligo-astheno- teratozoospermia. J Med Genet 50: 187-193.

Iacopetti P, Michelini, M, Stuckmann, I, Oback B, Aaku-Saraste E and Huttner WB (1999). Expression of the antiproliferative gene TIS21 at the onset of neurogenesis identifies single neuroepithelial cells that switch from proliferative to neuron-generating division. Proc Natl Acad Sci USA 96: 4639-4644. 172

LaMonica BE, Lui JH, Wang X and Kriegstein AR (2012) OSVZ progenitors in the human cortex: an updated perspective on neurodevelopmental disease. Curr Opin Neurobiol 22: 747- 753.

Lancaster MA and Knoblich JA (2012) Spindle orientation in mammalian cerebral cortical development. Curr Opin Neurobiol 22: 737-746.

Langley AR, Smith JC, Stemple DL and Harvey SA (2014) New insights into the maternal to zygotic transition. Development 141: 3834-3841.

Lasko P (2009) Translational control during early development. Progress in Molecular Biology and Translational Science 90: 211-254.

Lasko P (2011) Posttranscriptional regulation in Drosophila oocytes and early embryos. RNA 2:408-416.

Lasko P (2012) mRNA localization and translational control in Drosophila oogenesis. Cold Spring Harb Perspect Biol 4: 1-15.

Lavado A and Oliver G (2007) Prox1 expression patterns in the developing and adult murine brain. Dev Dyn 236: 518-524.

Laver JD, Li X, Ancevicius K, Westwood JT, Smibert CA, Morris QD and Lipshitz HD (2013) Genome-wide analysis of Staufen-associated mRNAs identifies secondary structures that confer target specificity. Nucl Acids Res 41: 9438-9460.

Lai F and King ML (2013) Repressive translational control in germ cells. Mol Repro Dev 80: 665-676.

Larrieu D, Ythier D, Brambilla C Pedeux R (2010) ING2 controls the G1 to S-phase transition by regulating p21 expression. Cell Cycle 9: 3984-3990.

Lawrenson ID, Krebs DL, Linossi EM, Zhang JG, McLennan TJ, Collin C, McRae HM, Kolesnik TB, Koh K, Britto JM, Kueh AJ, Sheikh BN, El-Saafin F, Nicola NA, Tan SS, Babon 173

JJ, Nicholson SE, Alexander WS, Thomas T and Voss AK (2015) Cortical Layer Inversion and Deregulation of Reelin Signaling in the Absence of SOCS6 and SOCS7. Cereb Cortex pii: bhv253. [Epub ahead of print].

Lebeau G, DesGroseillers L, Sossin W and Lacaille JC (2011) mRNA binding protein Staufen 1-dependent regulation of pyramidal cell spine morphology via NMDA receptor-mediated synaptic plasticity. Mol Brain 4: 22-29.

Lebeau G, Miller LC, Tartas M, McAdam R, Laplante I, Badeaux F, DesGroseillers L, Sossin WS and Lacaille JC (2011) Staufen 2 regulates mGluR long-term depression and Map1b mRNA distribution in hippocampal neurons. Learn Mem 18: 314-326.

Lee HC, Simon GM and Cravatt BF (2015) ABHD4 regulates multiple classes of N-acyl phospholipids in the mammalian central nervous system. Biochemistry 54: 2539-2549.

Lesage B, Gutierrez I, Martí E and Gonzalez C (2010) Neural stem cells: the need for a proper orientation. Curr Opin Genet Dev 20: 438-442.

Lee KJ and Jessell TM (1999) The specification of dorsal cell fates in the vertebrate central nervous system. Annu Rev Neurosci 22: 261-294.

Lee SY, Lim IA, Kang GU, Cha SJ, Altanbyek V, Kim HJ, Lee S, Kim K and Yim J (2015) Protective effect of Drosophila glutathione transferase omega 1 against hydrogen peroxide- induced neuronal toxicity. Gene 568: 203-210.

Levitt P and Rakic P (1980) Immunoperoxidase localization of glial fibrillary acidic protein in radial glial cells and astrocytes of the developing rhesus monkey brain. J Comp Neurol 193: 815-840.

Li P, Yang X, Wasser M, Cai Y and Chia W (1997) Inscuteable and Staufen mediate asymmetric localization and segregation of prospero RNA during Drosophila neuroblast cell divisions. Cell 90: 437-447.

174

Li Z, Wu Y and Baraban JM (2008) The Translin/Trax RNA binding complex: clues to function in the nervous system. Biochim Biophys Acta 1779: 479-85.

Lipshitz HD and Smibert CA (2000) Mechanisms of RNA localization and translational regulation. Curr Opin Genet Dev 10: 476-488.

Liu L, Wang J, Zhao L, Nilsen J, McClure K, Wong K and Brinton RD (2009) Progesterone increases rat neural progenitor cell cycle gene expression and proliferation via extracellularly regulated kinase and progesterone receptor membrane components 1 and 2. Endocrinology 150: 3186-3196.

Liu X and Sun T (2015) microRNAs and molecular pathogenesis of microcephaly. Curr Mol Pharmacol doi:10.2174/1874467208666150928153949.

Loedige I, Stotz M, Qamar S, Kramer K, Hennig J, Schubert T, Löffler P, Längst G, Merkl R, Urlaub H and Meister G (2014) The NHL domain of BRAT is an RNA-binding domain that directly contacts the hunchback mRNA for regulation. Genes & Dev 28:749-764.

Lolicato F, Marino R, Paronetto MP, Pellegrini M, Dolci S, Geremia R and Grimaldi P (2008) Potential role of Nanos3 in maintaining the undifferentiated spermatogonia population. Dev Biol 313:725-738.

Lopez-Fanarraga M, Carranza G, Bellido J, Kortazar D, Villegas JC and Zabala JC (2007) Tubulin cofactor B plays a role in the neuronal growth cone. J Neurochem 100:1680-7.

Lui JH, Hansen DV and Kriegstein AR (2011) Development and evolution of the human neocortex. Cell 146: 18-36.

Lukaszewicz A, Savatier P, Cortay V, Giroud P, Huissoud C, Berland M, Kennedy H, Dehay C (2005) G1 Phase Regulation, Area-Specific Cell Cycle Control, and Cytoarchitectonics in the Primate Cortex. Neuron 47: 353–364.

175

Luskin MB, Pearlman AL and Sanes JR (1988) Cell lineage in the cerebral cortex of the mouse studied in vivo and in vitro with a recombinant retrovirus. Neuron 1: 635-647.

Luskin MB, Parnavelas JG and Barfield JA (1993) Neurons, astrocytes and oligodendrocytes of the rat cerebral cortex originate from separate progenitor cell: an ultrastructural analysis of clonally related cells. J Neurosci 13: 1730–1750.

MacNicol AM, Hardy LL, Spencer HJ, MacNicol MC (2015) Neural stem and progenitor cell fate transition requires regulation of Musashi1 function. BMC Dev Biol 15: 1-9.

Maher-Laporte M, Berthiaume F, Moreau M, Julien LA, Lapointe G, Mourez M and DesGroseillers L (2010) Molecular composition of Staufen2-containing ribonucleoproteins in embryonic rat brain. PLoS ONE 5: e11350.

Maher-Laporte M and DesGroseillers L (2010) Genome wide identification of Staufen2-bound mRNAs in embryonic rat brains. BMB Rep 43: 344-348.

Makeyev AV and Bayarsaihan D (2011) Molecular basis of Williams-Beuren syndrome: TFII-I regulated targets involved in craniofacial development. Cleft Palate Craniofac J 48: 109-16.

Malatesta P, Hartfuss E and Götz M (2000) Isolation of radial glial cells by fluorescent activated cell sorting reveals a neuronal lineage. Development 127: 5253-5263.

Malatesta P, Hack MA, Hartfuss E, Kettenmann H, Klinkert W, Kirchhoff F and Götz M (2003) Neuronal or glial progeny: regional differences in radial glial fate. Neuron 37: 751-764.

Malatesta P, Appolloni I and Calzolari F (2008) Radial glia and neural stem cells. Cell Tissue Res 331: 165-178.

Managò F, Espinoza S, Salahpour A, Sotnikova TD, Caron MG, Premont RT and Gainetdinov RR (2012) The role of GRK6 in animal models of Parkinson's disease and L-DOPA treatment. Sci Rep 2:301. doi: 10.1038/srep00301.

176

Martell KJ, Seasholtz AF, Kwak SP, Clemens KK and Dixon JE (1995) hVH-5: a protein tyrosine phosphatase abundant in brain that inactivates mitogen-activated protein kinase. J Neurochem 65: 1823-1833.

Masson R, Régnier CH, Chenard MP, Wendling C, Mattei MG, Tomasetto C and Rio MC (1998) Tumor necrosis factor receptor associated factor 4 (TRAF4) expression pattern during mouse development. Mech Dev 71: 187-91.

Mathiens V and ffrench-Constant C (2009) Adherens junction domains are split by asymmetric division of embryonic neural stem cells. EMBO Rep 10: 515-520.

Matsuzaki M, Honkura N, Ellis-Davies GC and Kasai H (2004) Structural basis of long-term potentiation in single dendritic spines. Nature 429: 761-766.

Matynia A1, Anagnostaras SG, Wiltgen BJ, Lacuesta M, Fanselow MS (2008) A high through- put reverse genetic screen identifies two genes involved in remote memory in mice. PLoS One 3:e2121.

McCarthy M, Turnbull DH, Walsh CA and Fishell G (2001) Telencephalic neural progenitors appear to be restricted to regional and glial fates before the onset of neurogenesis. 21: 6772- 6781.

Mei L and Xiong WC (2008) Neuregulin 1 in neural development, synaptic plasticity and schizophrenia. Nat Rev Neurosci 9: 437-452.

Meltzer HY, Brennan MD, Woodward ND and Jayathilake K (2008) Association of Sult4A1 SNPs with psychopathology and cognition in patients with schizophrenia or schizoaffective disorder. Schizophr Res 106: 258-264.

Merkle FT, Tramontin AD, García-Verdugo JM and Alvarez-Buylla A (2004) Radial glia give rise to adult neural stem cells in the subventricular zone. Proc Natl Acad Sci U S A 101: 17528 – 17532.

177

Meyer MA (2014) Highly expressed genes in human high grade gliomas: immunohistochemical analysis of data from the human protein atlas. Neurol Int 6: 5348.

Micklem DR, Adams J, Grünert S and St Johnston D (2000) Distinct roles of two conserved Staufen domains in oskar mRNA localization and translation. EMBO J 15: 1366-1377.

Miki T, Takano K and Yoneda Y (2005) The role of mammalian Staufen on mRNA traffic: a view from its nucleocytoplasmic shuttling function. Cell Struc Func 30: 51-56.

Mikl M, Vendra G and Kiebler MA (2011) Independent localization of MAP2, CaMKIIα and β- actin RNAs in low copy numbers. EMBO Rep 12: 1077-1084.

Miller FD and Gauthier AS (2007) Timing is everything: making neurons versus glial in the developing cortex. Neuron 54: 357–369.

Miller MA and Olivas WM (2011) Roles of Puf proteins in mRNA degradation and translation. Wiley Interdiscip Rev RNA 2: 471-492.

Ming GL and Song H (2005) Adult neurogenesis in the mammalian central nervous system. Annu Rev Neurosci 28: 223-250.

Mione MC, Cavanagh JF, Harris B and Parnavelas JG (1997) Cell fate specification and symmetrical/asymmetrical divisions in the developing cortex. J Neurosci 17: 2018-2029.

Mission JP, Takahashi T and Caviness VS Jr (1991) Ontogeny of radial and other astroglial cells in murine cerebral cortex. Glia 4:138-148.

Mita N, He X, Sasamoto K, Mishiba T and Ohshima T (2016) Cyclin-Dependent Kinase 5 regulates dendritic spine formation and maintenance of cortical neuron in the mouse brain. Cereb Cortex 26: 967-976.

Miyata T, Kawaguchi A, Okano H and Ogawa M (2001) Asymmetric inheritance of radial glial fibers by cortical neurons. Neuron 31: 727-741.

178

Miyata T, Kawaguchi A, Saito K, Kawano M, Muto T and Ogawa M (2004) Asymmetric production of surface-dividing and non-surface dividing cortical progenitor cells. Development 131: 3133-3145.

Miyata T (2007) Asymmetric cell division during brain morphogenesis. Prog Mol Subcell Biol. 45: 121-142.

Miyazaki C, Saitoh M, Itoh M, Yamashita S, Miyagishi M, Takashima S, Moser AB, Iwamori M and Mizuguchi M (2013) Altered phospholipid molecular species and glycolipid composition in brain, liver and fibroblasts of Zellweger syndrome. Neurosci Lett 552: 71-75.

Mo Z and Zecevic N (2008) Is Pax6 critical for neurogenesis in the human fetal brain? Cereb Cortex 18: 1455–1465.

Moldrich RX, Lainé J, Visel A, Beart PM, Laffaire J, Rossier J and Potier MC (2008) Transmembrane protein 50b (C21orf4), a candidate for Down syndrome neurophenotypes, encodes an intracellular membrane protein expressed in the rodent brain. Neuroscience 154: 1255-1266.

Molyneaux BJ, Arlotta P, Menezes JR and Macklis JD (2007) Neuronal subtype specification in the cerebral cortex. Nat Rev Neurosci 8: 427-437.

Monshausen M, Rehbein M, Richter D and Kindler S (2002) The RNA-binding protein Staufen from rat brain interacts with protein phosphatase-1. J Neurochem 81: 557-564.

Moore FL, Jaruzelska J, Fox MS, Urano J, Firpo MT, Turek PJ, Dorfman DM and Reijo Pera RA (2003) Human Pumilio-2 is expressed in embryonic stem cells and germ cells and interacts with DAZ (Deleted in AZoospermia) and DAZ-Like proteins. Proc. Natl. Acad. Sci. USA 100: 538–543.

179

Morey JS, Ryan JC and Van Dolah FM (2006) Microarray validation: factors influencing correlation between oligonucleotide microarrays and real-time PCR. Biol Proced Online 8: 175- 193.

Mori Y, Matsui T, Omote D and Fukuda M (2013) Small GTPase Rab39A interacts with UACA and regulates the retinoic acid-induced neurite morphology of Neuro2A cells. Biochem Biophys Res Commun 435: 113-119.

Moroni RF, De Biasi S, Colapietro P, Larizza L, Beghini A (2006) Distinct expression pattern of microtubule-associated protein/microtubule affinity-regulating kinase 4 in differentiated neurons. Neuroscience 143: 83-94.

Murata Y and Wharton RP (1995) Binding of Pumilio to maternal hunchback mRNA is required for posterior patterning in Drosophila embryos. Cell 80: 747-756.

Nagy A, Mar L and Watts G (2009) Creation and use of a cre recombinase transgenic database. Methods Mol Biol 530: 365-378.

Nakahata S, Katsu Y, Mita K, Inoue K, Nagahama Y and Yamashita M (2001) Biochemical identification of Xenopus Pumilio as a sequence-specific cyclin B1 mRNA-binding protein that physically interacts with a Nanos homolog, Xcat-2, and a cytoplasmic polyadenylation element- binding protein. J Biol Chem 276:20945-209453.

Nakazawa T, Hashimoto R, Sakoori K, Sugaya Y, Tanimura A, Hashimotodani Y, Ohi K, Yamamori H, Yasuda Y, Umeda-Yano S, Kiyama Y, Konno K, Inoue T, Yokoyama K, Inoue T, Numata S, Ohnuma T, Iwata N, Ozaki N, Hashimoto H, Watanabe M, Manabe T, Yamamoto T, Takeda M and Kano M (2016) Emerging roles of ARHGAP33 in intracellular trafficking of TrkB and pathophysiology of neuropsychiatric disorders. Nat Commun 7: 10594.

Nawabi H, Belin S, Cartoni R, Williams PR, Wang C, Latremolière A, Wang X, Zhu J, Taub DG, Fu X, Yu B, Gu X, Woolf CJ, Liu JS, Gabel CV, Steen JA and He Z (2015) Doublecortin- Like Kinases Promote Neuronal Survival and Induce Growth Cone Reformation via Distinct Mechanisms. Neuron 88: 704-719. 180

Nelson MR, Leidal AM and Smibert CA (2004) Drosophila Cup is an eIF4E-binding protein that functions in Smaug-mediated translational repression. EMBO J 23:150–159.

Neumüller RA and Knoblich JA (2009) Dividing cellular asymmetry: asymmetric cell division and its implications for stem cells and cancer. Genes Dev 23: 2675-2699.

Nikolaou N, Watanabe-Asaka T, Gerety S, Distel M, Köster RW and Wilkinson DG (2009) Lunatic fringe promotes the lateral inhibition of neurogenesis. Development 136: 2523-2533.

Ninkovic J and Götz M (2013) Fate specification in the adult brain—lessons for eliciting neurogenesis from glial cells. Bioessays 35: 242-252.

Nishimura T, Padamsi Z, Fakim H, Milette S, Dunham WH, Gingras AC and Fabian MR (2015) The eIF4E-binding protein 4E-T is a component of the mRNA decay machinery that bridges the 5’ and 3’ termini of target mRNAs. Cell Rep 11: 1425-1436.

Nishimura YV, Sekine K, Chihama K, Nakajima K, Hoshino M, Nabeshima Y and Kawauchi T (2010) Dissecting the factors involved in the locomotion mode of neuronal migration in the developing cerebral cortex. J Biol Chem 285: 5878-5887.

Noctor SC, Flint AC, Weissman TA, Dammerman RS and Kriegstein AR (2001) Neurons derived from radial glial cells establish radial units in neocortex. Nature 409: 714-720.

Noctor SC, Flint AC, Weissman TA, Wong WS, Clinton BK and Kriegstein AR (2002) Dividing precursor cells of the embryonic cortical ventricular zone have morphological and molecular characteristics of radial glia. J Neurosci 22: 3161-3173.

Noctor SC, Martínez-Cerdeño V, Ivic L and Kriegstein AR (2004) Cortical neurons arise in symmetric and asymmetric division zones and migrate through specific phases. Nat Neurosci 7: 136-144.

181

Noctor SC, Martínez-Cerdeño V and Kriegstein AR (2004) Contribution of intermediate progenitor cells to cortical histogenesis. Arch Neurol 64: 639-642.

Noctor SC, Martínez-Cerdeño V and Kriegstein AR (2008) Distinct behaviours of neural stem and progenitor cells underlie cortical neurogenesis. J Comp Neurol 508: 28-44.

Nottebohm F (2004) The road we travelled: discovery, choreography, and significance of brain replaceable neurons. Ann N Y Acad Sci 1016: 628-658.

Nousch M and Eckmann CR (2013) Translational control in the Caenorhabditis elegans germ line. Adv Exp Med Biol 757: 205-247.

Ohba C, Shiina M, Tohyama J, Haginoya K, Lerman-Sagie T, Okamoto N, Blumkin L, Lev D, Mukaida S, Nozaki F, Uematsu M, Onuma A, Kodera H, Nakashima M,Tsurusaki Y, Miyake N, Tanaka F, Kato M, Ogata K, Saitsu H and Matsumoto N (2015) GRIN1 mutations cause encephalopathy with infantile-onset epilepsy, and hyperkinetic and stereotyped movement disorders. Epilepsia 56: 841-848.

Okabe M, Sawamoto K, Imai T, Sakakibara S, Yoshikawa S and Okano H (1997) Intrinsic and extrinsic determinants regulating cell fate decision in developing nervous system. Dev Neurosci 19: 9-16.

Okano H, Kawahara H, Toriya M, Nakao K, Shibata S and Imai T (2005) Function of RNA- binding protein Musashi-1 in stem cells. Exp Cell Res 306: 349-356.

Ortuño-Sahagún D, Rivera-Cervantes MC, Gudiño-Cabrera G, Junyent F, Verdaguer E, Auladell C, Pallàs M, Camins A and Beas-Zárate C (2012) Microarray analysis of rat hippocampus exposed to excitotoxicity: reversal Na(+)/Ca(2+) exchanger NCX3 is overexpressed in glial cells. Hippocampus 22: 128-40.

Ozgur S, Basquin J, Kamenska A, Filipowicz W, Standart N and Conti E (2015) Structure of a Human 4E-T/DDX6/CNOT1 complex reveals the different interplay of DDX6-binding proteins with the CCR4-NOT complex. Cell Rep 13: 703-711. 182

Paquin A, Barnabé-Heider F, Kageyama R and Miller FD (2005) CCAAT/enhancer-binding protein phosphorylation biases cortical precursors to generate neurons rather than astrocytes in vivo. J Neurosci 25: 10747-10758.

Paridaen JT and Huttner WB (2014) Neurogenesis during development of the vertebrate central nervous system. EMBO Rep 15: 351-364.

Park E, Gleghorn ML and Maquat LE (2013) Staufen2 functions in Staufen1-mediated mRNA decay by binding to itself and its paralog and promoting Upf1 helicase but not ATPase activity. Proc Natl Acad Sci U S A 110: 405-412.

Park E and Maquat LE (2013) Staufen-mediated mRNA decay. Wiley Interdiscip Rev RNA 4:423-435.

Parker R (2012) RNA degradation in Saccharomyces cerevisiae. Genetics 191: 671-702.

Parnavelas JG, Barfield JA, Franke E and Luskin MB (1991) Separate progenitor cells give rise to pyramidal and nonpyramidal neurons in the rat telencephalon. 1: 463-468.

Pavlos NJ, Cheng TS, Qin A, Ng PY, Feng HT, Ang ES, Carrello A, Sung CH, Jahn R, Zheng MH and Xu J (2011) Tctex-1, a novel interaction partner of Rab3D, is required for osteoclastic bone resorption. Mol Cell Biol 31: 1551-64.

Peredo J, Villacé P, Ortín J and de Lucas S (2014) Human Staufen1 associates to miRNAs involved in neuronal cell differentiation and is required for correct dendritic formation. PLoS One 9: 113704-113725.

Pfeiffer V, Götz R, Xiang C, Camarero G, Braun A, Zhang Y, Blum R, Heinsen H, Nieswandt B and Rapp UR (2013) Ablation of BRaf impairs neuronal differentiation in the postnatal hippocampus and cerebellum. 8: e58259.

183

Pfeilschifter W, Czech-Zechmeister B, Sujak M, Mirceska A, Koch A, Rami A, Steinmetz H, Foerch C, Huwiler A and Pfeilschifter J (2011) Activation of sphingosine kinase 2 is an endogenous protective mechanism in cerebral ischemia Biochem Biophys Res Commun 413:212-7.

Pilz GA, Shitamukai A, Reillo I, Pacary E, Schwausch J, Stahl R, Ninkovic J, Snippert HJ, Clevers H, Godinho L, Guillemot F, Borrell V, Matsuzaki F, Götz M (2013) Amplification of progenitors on the mammalian telencephalon includes a new radial glial cell type. Nat Commun 4: 2125-2136.

Pinder BD and Smibert CA (2013a) Smaug: an unexpected journey into the mechanisms of post-transcriptional regulation. Fly 3:142-145.

Pinder BD and Smibert CA (2013b) microRNA-independent recruitment of Argonaute 1 to nanos mRNA through the Smaug RNA-binding protein. EMBO Rep 14:80-86.

Pinto L and Götz M (2007) Radial glial cell heterogeneity--the source of diverse progeny in the CNS. Prog Neurobiol 83: 2-23.

Poduri A, Heinzen EL, Chitsazzadeh V, Lasorsa FM, Elhosary PC, LaCoursiere CM, Martin E, Yuskaitis CJ, Hill RS, Atabay KD, Barry B, Partlow JN, Bashiri FA, Zeidan RM, Elmalik SA, Kabiraj MM, Kothare S, Stödberg T, McTague A, Kurian MA, Scheffer IE, Barkovich AJ, Palmieri F, Salih MA and Walsh CA (2013) SLC25A22 is a novel gene for migrating partial seizures in infancy. Ann Neurol 74: 873-82.

Pollard KS, Salama SR, King B, Kern AD, Dreszer T, Katzman S, Siepel A, Pedersen JS, Bejerano G, Baertsch R, Rosenbloom KR, Kent J and Haussler D (2006a) Forces shaping the fastest evolving regions in the human genome. PLoS Genet 2: e168.

Pollard KS, Salama SR, Lambert N, Lambot MA, Coppens S, Pedersen JS, Katzman S, King B, Onodera C, Siepel A, Kern AD, Dehay C, Igel H, Ares M Jr, Vanderhaeghen P and Haussler D (2006b) An RNA gene expressed during cortical development evolved rapidly in humans. Nature 443:167–172. 184

Postiglione MP, Jüschke C, Xie Y, Haas GA, Charalambous C and Knoblich JA (2011) Mouse inscutable induces apical-basal spindle orientation to facilitate intermediate progenitor generation in the developing neocortex. Neuron 72: 269-284.

Price J and Thurlow L (1988) Cell lineage in the rat cerebral cortex: a study using retroviral- mediated gene transfer. Development 104: 473-482.

Price J, Williams BP and Götz M (1995) The generation of cellular diversity in the cerebral cortex. Ciba Found Symp 193: 117-126.

Puklowski A, Homsi Y, Keller D, May M, Chauhan S, Kossatz U, Grünwald V, Kubicka S, Pich A, Manns MP, Hoffmann I, Gönczy P and Malek NP (2011) Nat Cell Biol 13: 1004-9.

Purger D, Gibson EM and Monje M (2015) Myelin plasticity in the central nervous system. Neuropharmacology 15: 30047-30052.

Qian X, Shen Q, Goderie SK, He W, Capela A, Davis AA and Temple S (2000) Timing of CNS cell generation: a programmed sequence of neuron and glial cell production from isolated murine cortical stem cells. Neuron 28: 69-80.

Quinlan RJ, Tobin JL and Beales PL (2008) Modeling ciliopathies: Primary cilia in development and disease. Curr Top Dev Biol 84: 249-310.

Rakic P (1974) Neurons in rhesus monkey visual cortex: systematic relation between time of origin and eventual disposition. Science 183: 425-427.

Rakic P (1995) Radial versus tangential migration of neuronal clones in the developing cerebral cortex. Proc Natl Acad Sci USA 92: 11323-11327.

Rakic P (2003) Developmental and evolutionary adaptations of cortical radial glia. Cereb Cortex 13: 541-549.

185

Raine A (2006) Schizotypal personality: neurodevelopmental and psychosocial trajectories. Annu Rev Clin Psychol 2: 291-326.

Raisch T, Bhandari D, Sabath K, Helms S, Valkov E, Weichenrieder O and Izaurralde E (2016) Distinct modes of recruitment of the CCR4-NOT complex by Drosophila and vertebrate Nanos. EMBO J 35: 974-990.

Ratié L, Ware M, Jagline H, David V and Dupé V (2014) Dynamic expression of Notch- dependent neurogenic markers in the chick embryonic nervous system. Front Neuroanat 18:158.

Raum H, Dietsche B, Nagels A, Witt SH, Rietschel M, Kircher T and Krug A (2015) A genome- wide supported psychiatric risk variant in NCAN influences brain function and cognitive performance in healthy subjects. Hum Brain Mapp 36: 378-390.

Ravel-Chapuis A, Crawford TE, Blais-Crépeau ML, Bélanger G, Richer CT and Jasmin BJ (2014) The RNA-binding protein Staufen1 impairs myogenic differentiation via a c-myc- dependent mechanism. Mol Biol Cell 25: 3765-3778.

Reichert H (2011) Drosophila neural stem cells: cell cycle control of self-renewal, differentiation, and termination in brain development. Results Probl Cell Differ 53: 529-546.

Reid CB, Liang I and Walsh C (1995) Systematic widespread clonal organization in cerebral cortex. Neuron 15: 299-310.

Reillo I and Borrell V (2012) Germinal zones in the developing cerebral cortex of ferret: ontogeny, cell cycle kinetics, and diversity of progenitors. Cereb Cortex 22: 2039-2054.

Rong L, Livingstone M, Sukarieh R, Petroulakis E, Gingras AC, Crosby K, Smith B, Polakiewicz RD, Pelletier J, Ferraiuolo MA and Sonenberg N (2008) Control of eIF4E cellular localization by eIF4E-binding proteins, 4E-BPs. RNA 14: 1318-1327.

186

Ross SE, McCord AE, Jung C, Atan D, Mok SI, Hemberg M, Kim TK, Salogiannis J, Hu L, Cohen S, Lin Y, Harrar D, McInnes RR and Greenberg ME (2012) Bhlhb5 and Prdm8 form a repressor complex involved in neuronal circuit assembly. Neuron 73: 292-303.

Rouget C, Papin C, Boureux A, Meunier AC, Franco B, Robine N, Lai EC, Pelisson A and Simonelig M (2010). Maternal mRNA deadenylation and decay by the piRNA pathway in the early D. melanogaster embryo. Nature 467: 1128-1132.

Roux KJ, Kim DI, Raida M and Burke B (2012) A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells. J Cell Biol 196: 801-810.

Rowitch DH and Kriegstein AR (2010) Developmental genetics of vertebrate glial-cell specification. Nature 468: 214-222.

Saba R, Kato Y and Saga Y (2014) NANOS2 promotes male germ cell development independent of meiosis suppression. Dev Biol 385:32-40.

Saga Y (2008) Sexual development of mouse germ cells: Nanos2 promotes the male germ cell fate by suppressing the female pathway. Dev Growth Differ 50: 141-147.

Saga Y (2010) Function of Nanos2 in the male germ cell lineage in mice. Cell Mol Life Sci 67: 3815-3822.

Sakakibara S, Nakamura Y, Satoh H and Okano H (2001) RNA-binding protein Musashi2: developmentally regulated expression in neural precursor cells and subpopulations of neurons in mammalian CNS. J Neurosci 21: 8091-8107.

Sakikibara S, Nakamura Y, Yoshida T, Shibata S, Koike M, Takano H, Ueda S, Uchiyama Y, Noda T, Okano H (2002) Proc Natl Acad Sci U S A 99: 15194-15199.

Sánchez-Sánchez R, Morales-Lázaro SL, Baizabal JM, Sunkara M, Morris AJ and Escalante- Alcalde D (2012) Lack of lipid phosphate phosphatase-3 in embryonic stem cells compromises neuronal differentiation and neurite outgrowth. Dev Dyn 241: 953-964. 187

Sancho-Tello M, Vallés S, Montoliu C , Renau-Piqueras J and Guerri C (1995) Developmental pattern of GFAP and vimentin gene expression in rat brain and in radial glial cultures. Glia 15: 157-166.

Sander SJ (2015) First glimpses of the neurobiology of autism spectrum disorder. Curr Opin Genet Dev 33: 80-92.

Schlüter OM, Schmitz F, Jahn R, Rosenmund C and Südhof TC (2004) A complete genetic analysis of neuronal Rab3 function. J Neurosci 24: 6629-37.

Schneider L, Clement CA, Teilmann SC, Pazour GJ, Hoffmann EK, Satir P and Christensen ST (2005) PDGFRalpha signalling is regulated through the primary cilium in fibroblasts. Curr Biol 15: 1861-1866.

Schnitzer J, Frank WW and Schachner M (1981) Immunocytochemical demonstration of vimentin in astrocytes and ependymal cells of developing and adult mouse nervous system. J Cell Biol 90: 435-447.

Schober M, Schaefer M and Knoblich JA (1999) Bazooka recruits inscuteable to orient asymmetric cell divisions in Drosophila neuroblasts. Nature 402: 548-551.

Schuster S, Rivalan M, Strauss U, Stoenica L, Trimbuch T, Rademacher N, Parthasarathy S, Lajkó D, Rosenmund C, Shoichet SA, Winter Y, Tarabykin V and Rosário M (2015) NOMA- GAP/ARHGAP33 regulates synapse development and autistic-like behavior in the mouse. Mol Psychiatry 20: 1120-1131.

Schuurmans C, Armant O, Nieto M, Stenman JM, Britz O, Klenin N, Brown C, Langevin LM, Seibt J, Tang H, Cunningham JM, Dyck R, Walsh C, Campbell K, Polleux F and Guillemot F (2004) Sequential phases of cortical specification involve Neurogenin-dependent and -independent pathways. EMBO J 23: 2892-2902.

188

Schwamborn JC, Berezikov E and Knoblich JA (2009) The TRIM-NHL protein TRIM32 activates microRNAs and prevents self-renewal in mouse neural progenitors. Cell 136: 913-925.

Sekine K, Kubo K and Nakajima K (2014) How does Reelin control neuronal migration and layer formation in the developing mammalian neocortex?. Neurosci Res 86: 50-58.

Selkoe DJ and Hardy J (2016) The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol Med doi: 10.15252/emmm.201606210.

Semotok JL, Cooperstock RL, Pinder BD, Vari HK, Lipshitz HD and Smibert CA (2005) Smaug recruits the CCR4/POP2/NOT deadenylase complex to trigger maternal transcript localization in the early Drosophila embryo. Curr Biol 15:284–294.

Semotok JL, Luo H, Cooperstock RL, Karaiskakis A, Vari HK, Smibert CA and Lipshitz HD (2008) Drosophila maternal Hsp83 mRNA destabilization is directed by multiple SMAUG recognition elements in the open reading frame. Mol Cell Biol 28:6757– 6772.

Sessa A, Mao CA, Hadjantonakis AK, Klein WH and Broccoli V (2008) Tbr2 directs conversion of radial glia into basal precursors and guides neuronal amplification by indirect neurogenesis in the developing neocortex. 60: 59-69.

Sheenan-Rooney K, Swartz ME, Zhao F, Liu D and Eberhart JK (2013) Ahsa1 and Hsp90 activity confers more severe craniofacial phenotypes in a zebrafish model of hypoparathyroidism, sensorineural deafness and renal dysplasia (HDR). Dis Model Mech 6: 1285-91.

Shibata T, Yamada K, Watanabe M, Ikenaka K, Wada K, Tanaka K and Inoue Y (1997) Glutamate transporter GLAST is expressed in the radial glia-astrocyte lineage of developing mouse spinal cord. J Neurosci 17: 9212-9219.

Shirai YT, Suzuki T, Morita M, Takahashi A and Yamamoto T (2014) Multifunctional roles of the mammalian CCR4-NOT complex in physiological phenomena. Front Genet 5: 286.

189

Shitamukai A, Konno D, Matsuzaki F (2011) Oblique radial glial divisions in the developing mouse neocortex induce self-renewing progenitors outside the germinal zone that resemble primate outer subventricular zone progenitors. J Neurosci 31: 3683–3695.

Siddiqui NU, Li X, Luo H, Karaiskakis A, Hou H, Kislinger T, Westwood JT, Morris Q and Lipshitz HD (2012) Genome-wide analysis of the maternal-to-zygotic transition in Drosophila primordial germ cells. Genome Biol 13:R11.

Sidman RL and Rakic P (1973) Neuronal migration, with special reference to developing human brain: a review. Brain Res 62: 1–35.

Smart IH (1973) Proliferative characteristics of the ependymal layer during the early development of the mouse neocortex: a pilot study based on recording the number, location and plane of cleavage of mitotic figures. J Anat 116: 67–91.

Smart IHM, Dehay C, Giroud P, Berland M, Kennedy H (2002) Unique morphological features of the proliferative zones and postmitotic compartments of the neural epithelium giving rise to striate and extrastriate cortex in the monkey. Cereb Cortex 12: 37–53.

Smibert CA, Wilson JE, Kerr K and Macdonald PM (1996) Smaug protein represses translation of unlocalized nanos mRNA in the Drosophila embryo. Genes & Dev 10:2600-2609.

Smibert CA, Lie YS, Shillinglaw W, Henzel WJ and Macdonald PM (1999) Smaug, a novel and conserved protein, contributes to repression of nanos mRNA translation in vitro. RNA 5:1535– 1547.

Sofroniew MV and Vinters HV (2010) Astrocytes: biology and pathology. Acta Neuropathol 119: 7-35.

Sossin WS and DesGroseillers L (2006) Intracellular trafficking of RNA in neurons. Traffic 7: 1581-1589.

190

Spandidos A, Wang X, Wang H and Seed B (2010) PrimerBank: a resource of human and mouse PCR primer pairs for gene expression detection and quantification. Nucl Acids Res 38:792-799.

Spassky N, Olivier C, Cobos I, LeBras B, Goujet-Zalc C, Martínez S, Zalc B and Thomas JL (2001) The early steps of oligodendrogenesis: insights from the study of the plp lineage in the brain of chicks and rodents. 23: 318-326.

Spassky N, Merkle FT, Flames N, Tramontin AD, García-Verdugo JM and Alvarez-Buylla A (2005) Adult ependymal cells are postmitotic and are derived from radial glial cells during embryogenesis. J Neurosci 25: 10-18.

Spassov DS and Jurecic R (2003) Mouse Pum1 and Pum2 genes, members of the Pumilio family of RNA-binding proteins, show differential expression in fetal and adult hematopoietic stem cells and progenitors. Blood Cells Mol Dis 30: 55-69.

Spik A, Oczkowski S, Olszak A, Kotecki M, Formanowicz P, Blazewicz J and Jaruzelska J (2006) Candidate mRNAs interacting with fertility protein Pumilio 2 in the human germ line. Reprod Biol 6: 37-42.

Stergiopoulos A, Elkouris M and Politis PK (2015) Prospero-related homeobox (Prox1) at the crossroads of diverse pathways during adult neural fate specification. Front Cell Neurosci 8:454- 462.

Stubbs D, DeProto J, Nie K, Englund C, Mahmud I, Hevner R and Molnár Z (2009) Neurovascular congruence during cerebral cortical development. Cereb Cortex 19: 32-41.

Sugimoto Y, Vigilante A, Darbo E, Zirra A, Militti C, D’Ambrogio A, Luscombe NM and Ule J (2015) hiCLIP reveals the in vivo atlas of mRNA secondary structures recognized by Staufen1. Nature 519: 491-494.

Suzuki A and Tsuda M and Saga Y (2007) Functional redundancy among Nanos proteins and a distinct role of Nanos2 during male germ cell development. Development 134: 77-83. 191

Suzuki H, Tsuda M, Kiso M and Saga Y (2008) Nanos3 maintains the germ cell lineage in the mouse by suppressing both Bax-dependent and –independent apoptotic pathways. Dev Biol 318: 133-142.

Suzuki A, Niimi Y and Saga Y (2014) Interaction of Nanos2 and Nanos3 with different components of the CNOT complex may contribute to the functional differences in mouse male germ cells. Biol Open 3: 1207-1216.

Tadros W, Goldman AL, Babak T, Menzies F, Vardy L, Orr-Weaver T, Hughes TR, Westwood JT, Smibert CA and Lipshitz HD (2007) SMAUG is a major regulator of maternal mRNA destabilization in Drosophila and its translation is activated by the PAN GU kinase. Dev Cell 12:143-155.

Tadros W and Lipshitz HD (2009) The maternal-to-zygotic transition: a play in two acts. Development 136:3033-3042.

Takahashi T, Nowakowski RS and Caviness VS Jr (1996) Interkinetic and migratory behavior of a cohort of neocortical neurons arising in the early embryonic murine cerebral wall. J Neurosci 16: 5762-5776.

Takahashi T, Goto T, Miyama S, Nowakowski RS and Caviness VS Jr (1999) Sequence of neuron origin and neocortical laminar fate: relation to cell cycle of origin in the developing murine cerebral wall. J Neurosci 19: 10357-10371.

Takahashi T, Caviness VS Jr and Bhide PG (2002) Analysis of cell generation in the telencephalic neuroepithelium. Methods Mol Biol 198: 101-113.

Tamamaki N, Nakamura K, Okamoto K and Kaneko T (2001) Radial glia is a progenitor of neocortical neurons in the developing cerebral cortex. Neurosci Res 41: 51-60.

192

Tanackovic G, Ransijn A, Ayuso C, Harper S, Berson EL and Rivolta C (2011) A missense mutation in PRPF6 causes impairment of pre-mRNA splicing and autosomal-dominant retinitis pigmentosa. Am J Hum Genet 88: 643-649.

Tao H, Li H, Su Y, Feng D, Wang X, Zhang C, Ma H and Hu Q (2014) Histone methyltransferase G9a and H3K9 dimethylation inhibit the self-renewal of glioma cancer stem cells. Mol Cell Biochem 394: 23-30.

Tarabykin V, Stoykova A, Usman N and Gruss P (2001) Cortical upper layer neurons derive from the subventricular zone as indicated by Svet1 gene expression. Development: 128: 1983-1993.

Tata A, Stoppel DC, Hong S, Ben-Zvi A, Xie T and Gu C (2014) An image-based RNAi screen identifies SH3BP1 as a key effector of Semaphorin 3E-PlexinD1 signaling. J Cell Biol 205: 573-90.

Taverna E, Götz M and Huttner WB (2014) The cell biology of neurogenesis: toward an understanding of the development and evolution of the neoxortex. Annu Rev Cell Dev Biol 30:465-502.

Tekki-Kessaris N, Woodruff R, Hall AC, Gaffield W, Kimura S, Stiles CD, Rowitch DH and Richardson WD (2001) Hedgehog-dependent oligodendrocyte lineage specification in the telencephalon. 128: 2545-2554.

Thomas MG, Pascual ML, Maschi D, Luchelli L and Boccaccio GL (2014) Synaptic control of local translation: the plot thickens with new characters. Cell Mol Life Sci 71: 2219-2239.

Thomas PD, Campbell MJ, Kejariwal A, Mi H, Karlak B, Daverman R, Diemer K, Muruganujan A, Narechania A (2003) PANTHER: a library of protein families and subfamilies indexed by function. Genome Res 13: 2129-2141.

Timmerman C, Suppiah S, Gurudatta BV, Yang J, Banerjee C, Sandstrom DJ, Corces VG and Sanyal S (2013) The Drosophila transcription factor Adf-1 (nalyot) regulates dendrite growth by 193 controlling FasII and Staufen expression downstream of CamkII and neural activity. J Neurosci 33: 11916-11931.

Tizon B, Ribe EM, Mi W, Troy CM and Levy E (2010) Cystatin C protects neuronal cells from amyloid-beta-induced toxicity. J Alzheimers Dis 19: 885-94.

Tosar LJ, Thomas MG, Baez MV, Ibanez I, Chernomoretz A and Boccaccio GL (2012) Staufen: from embryo polarity to cellular stress and neurodegeneration. Frontier Biosci 4:432-452.

Tripathi RB, Clarke LE, Burzonato V, Kessaris N, Anderson PN, Attwell D and Richardson WD (2011) Dorsally and ventrally derived oligodendrocytes have similar electrical properties but myelinate preferred tracts. J Neurosci 31: 6809-6819.

Tsuda M, Sasaoka Y, Kiso M, Abe K, Haraguchi S, Kobayashi S and Saga Y (2003) Conserved role of nanos proteins in germ cell development. Science 301:1239-1241.

Tsui D, Vessey JP, Tomita H, Kaplan DR and Miller FD (2013) FoxP2 regulates neurogenesis during embryonic cortical development. J Neurosci 33:244-258.

Tucker ES, Segall S, Gopalakrishna D, Wu Y, Vernon M, Polleux F and Lamantia AS (2008) Molecular specification and patterning of progenitor cells in the lateral and medial ganglionic eminences. J Neurosci 28: 9504-9518.

Tuoc TC, Pavlakis E, Tylkowski MA and Stoykova A (2014) Control of cerebral size and thickness. Cell Mol Life Sci 71: 3199-3218.

Tuzovic L, Yu L, Zeng W, Li X, Lu H, Lu HM, Gonzalez KD and Chung WK (2013) A human de novo mutation in MYH10 phenocopies the loss of function mutation in mice. Rare Dis 14:1 e26144. van de Bospoort R, Farina M, Schmitz SK, de Jong A, de Wit H, Verhage M and Toonen RF (2012) Munc13 controls the location and efficiency of dense-core vesicle release in neurons. J Cell Biol 199: 883-891. 194

van der Vaart B, Franker MA, Kuijpers M, Hua S, Bouchet BP, Jiang K, Grigoriev I, Hoogenraad CC and Akhmanova A (2012) Microtubule plus-end tracking proteins SLAIN1/2 and ch-TOG promote axonal development. J Neurosci 32: 14722-14728.

Van Etten J, Schagat TL, Hrit J, Weidmann CA, Brumbaugh J, Coon JJ and Goldstrohm AC (2012) Human Pumilio proteins recruit multiple deadenylases to efficiently repress messenger RNAs. J Biol Chem 287: 36370-36383.

Vardy L and Orr-Weaver TL (2007) Regulating translation of maternal messages: multiple repression mechanism. Trends Cell Biol 17: 547-554.

Vasistha NA, García-Moreno F, Arora S, Cheung AF, Arnold SJ, Robertson EJ and Molnár Z (2015) Cortical and clonal contribution of Tbr2 expressing progenitors in the developing mouse brain. Cereb Cortex 25: 3290-3302.

Vessey JP, Vaccani A, Xie Y, Dahm R, Karra D, Kiebler MA and Macchi P (2006) Dendritic localization of the translational repressor Pumilio 2 and its contribution to dendritic stress granules. J Neurosci 26: 6496-6508.

Vessey JP, Macchi P, Stein JM, Miki M, Hawker KN, Vogelsang P, Wieczorek K, Vendra G, Riefler J, Tübing F, Aparicio SA, Abel T and Kiebler MA (2008) A loss of function allele for murine Staufen1 leads to impairment of dendritic Staufen1-RNP delivery and dendritic spine morphogenesis. Proc Natl Acad Sci U S A 105: 16374-16379.

Vessey JP, Amadei G, Burns SE, Kiebler MA, Kaplan DR and Miller FD (2012) An asymmetrically localized Staufen2-dependent RNA complex regulates maintenance of mammalian neural stem cells. Cell Stem Cell 11:517-528.

Villacé P, Marión RM and Ortín J (2004) The composition of Staufen-containing RNA granules from human cells indicates their role in the regulated transport and translation of messenger RNAs.

195

Vives V, Alonso G, Solal AC, Joubert D and Legraverend C (2003) Visualization of s100β- positive neurons and glia in the central nervous system of EGFP transgenic mice. J Comp Neurol 457: 404-419.

Volta M, Bulfone A, Gattuso C, Rossi E, Mariani M, Consalez GG, Zuffardi O, Ballabio A, Banfi S and Franco B (1999) Identification and characterization of CDS2, a mammalian homolog of the Drosophila CDP-diacylglycerol synthase gene. Genomics 55: 68-77.

Walser CB and Lipshitz HD (2011) Transcript clearance during the maternal-to-zygotic transition. Curr Opin Genet Dev 4:431- 434.

Walsh C and Cepko CL (1988) Clonally related cortical cells show several migration patterns. Science 241: 1342-1345.

Walsh C and Cepko CL (1992) Widespread dispersion of neuronal clones across functional regions of the cerebral cortex. Science 255: 434-440.

Walther C and Gruss P (1991) Pax-6, a murine paired box gene, is expressed in the developing CNS. Development 113: 1435:1349.

Wang C and Lehmann R (1991) Nanos is the localized posterior determinant in Drosophila. Cell 66: 637-647.

Wang H and Chia W (2005) Drosophila neural progenitor polarity and asymmetric division. Biol Cell 97: 63-74.

Wang J, Weaver IC, Gauthier-Fisher A, Wang H, He L, Yeomans J, Wondisford F, Kaplan DR and Miller FD (2010) CBP histone acetyltransferase activity regulates embryonic neural differentiation in the normal and Rubinstein-Taybi syndrome brain. Dev Cell 18:114-125.

Wang J, Gallagher D, DeVito LM, Cancino GI, Tsui D, He L, Keller GM, Frankland PW, Kaplan DR and Miller FD (2012) Metformin activates an atypical PKC-CBP pathway to promote neurogenesis and enhance spatial memory formation. Cell Stem Cell 11: 23-35. 196

Wang X, Takenaka K and Takeda S (2010) PTIP promotes DNA double-strand break repair through homologous recombination. Genes Cells 15: 243-254.

Wang X, Tsai JW, LaMonica B, Kriegstein AR (2011) A new subtype of progenitor cell in the mouse embryonic neocortex. Nat Neurosci 14: 555–561.

Weibrecht I, Leuchowius KJ, Clausson CM, Conze T, Jarvius M, Howell WM, Kamali- Moghaddam M and Söderberg O (2010) Proximity ligation assays: a recent addition to the proteomics toolbox. Exp Rev Proteom 7:401-409.

Weigmann A, Corbeil D, Hellwig A and Huttner WB (1997) Prominin, a novel microvilli- specific polytopic membrane protein of the apical surface of epithelial cells, is targeted to plasmalemmal protrusions of non-epithelial cells. Proc Natl Acad Sci USA 94: 12425-12430.

Wharton RP and Struhl G (1991) RNA regulatory elements mediate control of D.melanogaster body pattern by the posterior morphogen nanos. Cell 67: 955-967.

Wieczorek D, Newman WG, Wieland T, Berulava T, Kaffe M, Falkenstein D, Beetz C, Graf E, Schwarzmayr T, Douzgou S, Clayton-Smith J, Daly SB, Williams SG, Bhaskar SS, Urquhart JE, Anderson B, O'Sullivan J, Boute O, Gundlach J, Czeschik JC, van Essen AJ, Hazan F, Park S, Hing A, Kuechler A, Lohmann DR, Ludwig KU, Mangold E, Steenpaß L, Zeschnigk M, Lemke JR, Lourenco CM, Hehr U, Prott EC, Waldenberger M, Böhmer AC, Horsthemke B, O'Keefe RT, Meitinger T, Burn J, Lüdecke HJ and Strom TM (2014) Compound heterozygosity of low-frequency promoter deletions and rare loss-of-function mutations in TXNL4A causes Burn-McKeown syndrome. Am J Hum Genet 95: 698-707.

Wilde JJ, Petersen JR and Niswander L (2014) Genetic, epigenetic and environmental contributions to neural tube closure. Annu Rev Genet 48: 583-611.

Williams BP, Read J and Price J (1991) The generation of neurons and oligodendrocytes from a common precursor cell. Neuron 7: 685-693.

197

Wilsch-Bräuninger M, Peters J, Paridaen JT and Huttner WB (2012) Basolateral rather apical primary cilia on neuroepithelial cells committed to delamination. Development 139: 95-105.

Wodarz A, Ramrath A, Kuchinke U and Knust E (1999) Bazooka provides an apical cue for Inscuteable localization in Drosophila neuroblasts. Nature 402: 544-547.

Wodarz A and Huttner WB (2003) Asymmetric cell division during neurogenesis in Drosophila and vertebrates. Mech Dev 120: 1297-1309.

Wu S, Wu Y and Capecchi MR (2006) Motoneurons and oligodendrocytes are sequentially generated from neural stem cells but do not appear to share common lineage-restricted progenitors in vivo. Development 133: 581-590.

Xie Z and Chin LS (2008) Molecular and cell biology of brain tumor stem cells: lessons from neural progenitor/stem cells. Neurosurg Focus 24: 3-25.

Yang G, Smibert CA, Kaplan DR and Miller FD (2014) An eIF4E1/4E-T complex determines the genesis of neurons from precursors by translationally repressing a proneurogenic transcription program. Neuron 84:723-739.

Yang J, Liu W, Lu X, Fu Y, Li L and Luo Y (2015) High expression of small GTPase Rab3D promotes cancer progression and metastasis. Oncotarget 6: 11125-11138.

Yang L, Zhang Y, Wang S, Zhang W and Shi R (2014) Decreased liver peroxisomal β-oxidation accompanied by changes in brain fatty acid composition in aged rats. Neurol Sci 35: 289-293.

Yano M, Hayakawa-Yano Y and Okano H (2016) RNA regulation went wrong in neurodevelopmental disorders: The example of Msi/Elavl RNA binding proteins. Int J Dev Neurosci 5748: 30163-30165.

Yazdankhah M, Farioli-Vecchioli S, Tonchev AB, Stoykova A and Cecconi F (2014) The autophagy regulators Ambra1 and Beclin 1 are required for adult neurogenesis in the brain subventricular zone. 4: e1403. 198

Yonezawa T, Ohtsuka A, Yoshitaka T, Hirano S, Nomoto H, Yamamoto K and Ninomiya Y (2003) Limitrin, a novel immunoglobulin superfamily protein localized to glia limitans formed by astrocyte endfeet. Glia 44: 190-204.

Yoon SH, Bae YS, Mun MS, Park KY, Ye SK, Kim E and Kim MH (2012) Developmental retardation, microcephaly, and peptiduria in mice without aminopeptidase P1. Biochem Biophys Res Commun 429: 204-209.

Yuen EY, Liu W, Kafri T, van Praag H and Yan Z (2010) Regulation of AMPA receptor channels and synaptic plasticity by cofilin phosphatase Slingshot in cortical neurons. J Physiol 588: 2361-2371.

Zaessinger S, Busseau I and Simonelig M (2006) Oskar allows nanos mRNA translation in Drosophila embryos by preventing its deadenylation by Smaug/CCR4. Development 133:4573– 4583.

Zaghloul NA and Katsanis N (2009) Mechanistic insights into Bardet-Biedl syndrome, a model ciliopathy. J Clin Invest 119: 428-437.

Zamenhof S (1985) Differential effects of antifolate on the development of brain parts in chick embryos. Growth 49: 28-33.

Zecevic N, Chen Y, Filipovic R (2005) Contributions of cortical subventricular zone to the development of the human cerebral cortex. J Comp Neurol 491: 109–122.

Zemora G and Waldsich C (2010) RNA folding in living cells. RNA Biology 7: 634-641.

Zhadanov AB, Provance DW Jr, Speer CA, Coffin JD, Goss D, Blixt JA, Reichert CM and Mercer JA (1999) Absence of the tight junctional protein AF-6 disrupts epithelial cell-cell junctions and cell polarity during mouse development. Curr Biol 9: 880-888.

Zhang C, Zhu T, Chen Y and Xu Y (2015) Loss of preimplantation embryo resulting from a Pum1 gene trap mutation. Biochem Biophys Res Commun 462: 8-13. 199

Zhao S, Yang XF, Gou WF, Lu H, Li H, Zhu ZT, Sun HZ and Zheng HC (2016) Expression profiles of inhibitor of growth protein 2 in normal and cancer tissues: An immunohistochemical screening analysis. Mol Med Rep 13:1881-7.

Zhong W and Chia W (2008) Neurogenesis and asymmetric cell division. Curr Opin Neurobiol 18: 4-11.

Zhou B, Zhu YB, Lin L, Cai Q and Sheng ZH (2011) Snapin deficiency is associated with developmental defects of the central nervous system. Biosci Rep 31:151-8.

Zhou B, Cai Q, Xie Y and Sheng ZH (2012) Snapin recruits dynein to BDNF-TrkB signaling endosomes for retrograde axonal transport and is essential for dendrite growth of cortical neurons. Cell Rep 2: 42-51.

Zhou H, Mangelsdorf M, Liu J, Zhu L and Wu JY (2014) RNA-binding proteins in neurological diseases. Sci China Life Sci 57: 432-444.

200

APPENDIX

Table 1: mRNAs significantly associated with Smaug2, as identified by Smaug2 RIP-Chip. Smaug2 RIP was performed from E12.5 cortical radial precursors with a Smaug2-specific antibody. As a control, similar RIPs were performed with a non-specific IgG. Isolated RNAs from four biological replicates were analysed on Affymetrix Mouse Gene 2.0 ST arrays. Raw probe intensity values were background corrected, normalized with quantile normalization, transformed into the log2 scale, and summarized into probe sets with the RMA algorithm as described in the Experimental Methods section. Statistically enriched mRNAs with the Smaug2 IP or IgG IP were identified by a pairwise analysis. The IgG and Smaug2 samples for each biological replicate were analyzed as pairs and each gene in the array was analyzed by paired t- test. mRNA transcripts with an uncorrected p-value < 0.01 and a Smaug2/IgG fold change ≥ 1.5 were considered to be significantly associated with Smaug2. For each mRNA, fold-change, unadjusted p-value, gene symbol, gene description and ENTREZ gene identifier are reported.

Fold- Unadjusted p- ENTREZ Gene Symbol Description change value Gene ID MAP/microtubule affinity- 12.1717769 0.006996052 Mark4 232944 regulating kinase 4 sterile alpha motif domain 10.3809744 0.003652185 Samd4b 233033 containing 4B glycerophosphodiester 8.67244992 0.000278373 Gde1 56209 phosphodiesterase 1 7.887075199 0.006246418 Vgll4 vestigial like 4 (Drosophila) 232334 7.824091645 0.008782828 Gm5617 predicted gene 5617 434402 zinc finger protein 36, C3H 5.865827356 0.003308548 Zfp36l1 12192 type-like 1 dishevelled 3, dsh homolog 5.641663029 0.000151807 Dvl3 13544 (Drosophila) G protein-coupled receptor 5.161774083 0.000117958 Git1 216963 kinase-interactor 1 ras homolog gene family, 4.948677845 0.008570271 Rhot2 214952 member T2 Rho GTPase activating 4.845862706 0.001328985 Arhgap11a 228482 protein 11A matrix-remodelling associated 4.615629999 0.003014735 Mxra8 74761 8 thiosulfate sulfurtransferase 4.599107219 0.006358557 Tstd1 (rhodanese)-like domain 226654 containing 1 4.556116509 0.000496783 Snapin SNAP-associated protein 20615 4.016605815 0.000630487 BC037034 cDNA sequence BC037034 231807 3.94495827 0.009796043 Ccnl2 cyclin L2 56036

201

craniofacial development 3.825863695 0.000664777 Cfdp1 23837 protein 1 3.814683149 0.000174473 Kpna6 karyopherin (importin) alpha 6 16650 frizzled homolog 8 3.806852102 0.002602653 Fzd8 14370 (Drosophila) hairy and enhancer of split 5 3.783579575 0.006063348 Hes5 15208 (Drosophila) phosphatidic acid 3.726708602 0.006419707 Ppap2b 67916 phosphatase type 2B aspartate beta-hydroxylase 3.604500349 0.003628896 Asphd2 72898 domain containing 2 Rho GTPase activating 3.418164999 0.009933398 Arhgap33 233071 protein 33 3.35080209 0.00585075 Cbfb core binding factor beta 12400 3.224870665 0.006695799 Sh3gl1 SH3-domain GRB2-like 1 20405 leucine rich repeat containing 3.127445185 0.002893152 Lrrc1 214345 1 fibronectin type III domain 3.12526377 0.008868552 Fndc4 64339 containing 4 methylenetetrahydrofolate dehydrogenase (NAD+ 3.089784471 0.003111292 Mthfd2 dependent), 17768 methenyltetrahydrofolate cyclohydrolase 3.016860406 0.007323291 Mlx MAX-like protein X 21428 G protein-coupled receptor 2.978680019 0.005439047 Grk6 26385 kinase 6 ELAV (embryonic lethal, 2.921032827 0.005856099 Elavl3 abnormal vision, Drosophila)- 15571 like 3 (Hu antigen C) 2.911775873 0.007432606 Soat1 sterol O-acyltransferase 1 20652 2.907293855 0.004475111 Supt6 suppressor of Ty 6 20926 cornichon homolog 2 2.903818773 0.00724218 Cnih2 12794 (Drosophila) furry homolog-like 2.850920697 0.00337117 Fryl 72313 (Drosophila) TATA box binding protein 2.835734533 0.002244823 Taf1d (Tbp)-associated factor, RNA 75316 polymerase I, D 2.822268857 0.000737168 Ube4b ubiquitination factor E4B 63958 2.792756375 0.005136466 Cnnm3 cyclin M3 94218 2.791685963 0.004924387 Prkcd protein kinase C, delta 18753 LFNG O-fucosylpeptide 3- 2.784963499 0.000772594 Lfng beta-N- 16848 acetylglucosaminyltransferase GC-rich promoter binding 2.731490433 0.002117129 Gpbp1l1 77110 protein 1-like 1 GID complex subunit 4, 2.720031544 0.000247734 Gid4 VID24 homolog (S. 66771 cerevisiae) 2.70302319 0.008840149 Atxn7l2 ataxin 7-like 2 72522 2.688982133 0.00643791 Fasn fatty acid synthase 14104 2.665845404 0.004916132 Gmeb2 glucocorticoid modulatory 229004

202

element binding protein 2 glutamate-ammonia ligase 2.641314908 0.001912 Glul 14645 (glutamine synthetase) 2.637367499 0.004894393 Cst3 cystatin C 13010 X-prolyl aminopeptidase 2.6316997 0.002998336 Xpnpep1 170750 (aminopeptidase P) 1, soluble inositol polyphosphate 2.602027344 0.000218042 Ipmk 69718 multikinase TNF receptor associated 2.599777357 0.001228101 Traf4 22032 factor 4 2.586269485 0.00835953 Tmem50b transmembrane protein 50B 77975 inhibitor of growth family, 2.562837953 0.00420758 Ing2 69260 member 2 actin related protein 2/3 2.512969203 0.0088913 Arpc5 67771 complex, subunit 5 zinc finger and BTB domain 2.507830312 0.009907278 Zbtb41 226470 containing 41 homolog 2.485849614 0.000291947 Slain1 SLAIN motif family, member 1 105439 2.433939462 0.004714468 Cryz crystallin, zeta 12972 phospholipase A2, group X, 2.419698479 0.001775431 Pla2g10os 76684 opposite strand 2.408444761 0.001050582 Dusp8 dual specificity phosphatase 8 18218 2.374516392 0.007829107 Dhps deoxyhypusine synthase 330817 uveal autoantigen with coiled- 2.353994601 0.008832756 Uaca coil domains and ankyrin 72565 repeats calcium/calmodulin- 2.312516573 0.004065347 Camkk2 dependent protein kinase 207565 kinase 2, beta autophagy/beclin 1 regulator 2.293245998 0.006924366 Ambra1 228361 1 myosin, heavy polypeptide 2.263572988 0.008860325 Myh10 77579 10, non-muscle 2.256643902 0.006884698 Ncan neurocan 13004 SWI/SNF related, matrix associated, actin dependent 2.250753663 0.008789103 Smarcd2 83796 regulator of chromatin, subfamily d, member 2 2.247751963 0.00441268 Gramd1a GRAM domain containing 1A 52857 2.21232215 0.000948522 Mir5125 microRNA 5125 100628593 2.206647059 0.000436151 Rnf115 ring finger protein 115 67845 slingshot homolog 1 2.141021062 0.008462469 Ssh1 231637 (Drosophila) ash2 (absent, small, or 2.140583317 0.001143754 Ash2l 23808 homeotic)-like (Drosophila) suppressor of cytokine 2.137323891 0.000659618 Socs7 192157 signaling 7 family with sequence 2.132520601 0.004785807 Fam102b 329739 similarity 102, member B small nucleolar RNA host 2.121910815 0.00566322 Snhg6 73824 gene 6 solute carrier family 43, 2.121306923 0.004921491 Slc43a1 72401 member 1

203

suppression of tumorigenicity 2.120742275 0.008533135 St5 76954 5 2.112586296 0.001996627 Enc1 ectodermal-neural cortex 1 13803 fizzy/cell division cycle 20 2.106851848 0.00676837 Fzr1 56371 related 1 (Drosophila) 2.100396695 0.001937563 Snn stannin 20621 2.0746825 0.008118288 Sh3bp1 SH3-domain binding protein 1 20401 2.047486077 0.001611908 Tsn translin 22099 2.035985333 0.006594828 Lyrm4 LYR motif containing 4 380840 ST8 alpha-N-acetyl- 2.014416131 0.006425868 St8sia2 neuraminide alpha-2,8- 20450 sialyltransferase 2 2.013690907 0.000762948 Tsnax translin-associated factor X 53424 solute carrier family 25 2.002020646 0.009793969 Slc25a22 (mitochondrial carrier, 68267 glutamate), member 22 1.993677972 0.001545757 Mir669h microRNA 669h 100316831 RAB3D, member RAS 1.961495066 0.003884489 Rab3d 19340 oncogene family G protein-regulated inducer of 1.955915424 0.009289908 Gprin1 26913 neurite outgrowth 1 CDP-diacylglycerol synthase 1.942803217 0.003586602 Cds2 (phosphatidate 110911 cytidylyltransferase) 2 sulfotransferase family 4A, 1.941693209 0.00992638 Sult4a1 29859 member 1 1.927304319 0.003024233 Mex3b mex3 homolog B (C. elegans) 108797 ciliary neurotrophic factor 1.925542273 0.000426614 Cntfr 12804 receptor 1.893559669 0.000400525 Mir297a-4 microRNA 297a-4 100124435 1.875804618 0.009129538 Ctage5 CTAGE family, member 5 217615 ribosomal protein S6 kinase, 1.864100737 0.008938523 Rps6ka2 20112 polypeptide 2 TBC1 domain family, member 1.852082549 0.006927566 Tbc1d22b 381085 22B 1.85056389 0.002574093 Mxd1 MAX dimerization protein 1 17119 progesterone receptor 1.845922428 0.002089593 Pgrmc2 70804 membrane component 2 1.831750341 0.000370206 Rcan3 regulator of calcineurin 3 53902 1.829351657 0.008376777 Mxi1 Max interacting protein 1 17859 PAX interacting (with 1.823345713 0.005670401 Paxip1 transcription-activation 55982 domain) protein 1 RIKEN cDNA 2810047C21 1.8209024 0.001560102 2810047C21Rik1 72716 gene 1 1.819424343 0.008812541 Cdk5 cyclin-dependent kinase 5 12568 1.796340209 0.000806148 Spaca6 sperm acrosome associated 6 75202 1.789384184 0.001740497 Tbcb tubulin folding cofactor B 66411 1.787230266 0.002912533 Nol11 nucleolar protein 11 68979 suppressor of defective 1.78607195 0.001754093 Suds3 silencing 3 homolog (S. 71954 cerevisiae)

204

blocked early in transport 1 1.771501029 0.003388415 Bet1l 54399 homolog (S. cerevisiae)-like putative homeodomain 1.770353646 0.00996394 Phtf1 18685 transcription factor 1 translocase of outer 1.767896797 0.000674579 Tomm34 67145 mitochondrial membrane 34 FCH and double SH3 1.763824915 0.001311061 Fchsd2 207278 domains 2 1.763804953 0.008985085 Rnf19b ring finger protein 19B 75234 1.759974027 0.00342428 Dnpep aspartyl aminopeptidase 13437 solute carrier family 35, 1.749196915 0.00253378 Slc35e3 215436 member E3 1.747909277 0.002726796 Smok4a sperm motility kinase 4A 272667 Parkinson disease 7 domain 1.747226771 0.00276064 Pddc1 213350 containing 1 5'-nucleotidase domain 1.740743662 0.002514389 Nt5dc2 70021 containing 2 1.739750063 0.003264206 Epha4 Eph receptor A4 13838 adaptor-related protein 1.737380749 0.003940076 Ap1ar complex 1 associated 211556 regulatory protein mannoside 1.734754134 0.002145413 Mgat4b acetylglucosaminyltransferase 103534 4, isoenzyme B prostate tumor over 1.733255733 0.002000005 Ptov1 84113 expressed gene 1 euchromatic histone lysine N- 1.732424374 0.001230682 Ehmt2 110147 methyltransferase 2 glutathione S-transferase 1.730674888 0.000342535 Gsto1 14873 omega 1 RIKEN cDNA A430104N18 1.726981642 0.003036828 A430104N18Rik 78591 gene 1.717094511 0.00756329 Gm5665 predicted gene 5665 435366 expressed sequence 1.703211088 0.003136255 AA474331 213332 AA474331 activating transcription factor 1.702848594 0.009913352 Atf7 223922 7 AHA1, activator of heat shock 1.701829773 0.005458347 Ahsa1 217737 protein ATPase 1 transducin-like enhancer of 1.697549281 0.005680503 Tle4 split 4, homolog of Drosophila 21888 E(spl) 1.694861973 0.007376743 Tubb6 tubulin, beta 6 class V 67951 calcium channel flower 1.687272307 0.005941816 Cacfd1 381356 domain containing 1 RIKEN cDNA A930011O12 1.687234382 0.005041646 A930011O12Rik 268755 gene oxysterol binding protein-like 1.682798638 0.000563298 Osbpl2 228983 2 unc-119 homolog B (C. 1.680776833 0.002409471 Unc119b 106840 elegans) DEAD (Asp-Glu-Ala-Asp) box 1.670797621 0.008378744 Ddx19a 13680 polypeptide 19a 1.665936542 0.002147938 Nhp2 NHP2 ribonucleoprotein 52530

205

RIKEN cDNA 2610507B11 1.665336038 0.002054411 2610507B11Rik 72503 gene 1.661322136 0.008349523 Mcrs1 microspherule protein 1 51812 1.658998825 0.002255445 Tmem70 transmembrane protein 70 70397 1.657459324 0.005384255 Phf11a PHD finger protein 11A 219131 SNF8, ESCRT-II complex 1.653300881 0.0046656 Snf8 subunit, homolog (S. 27681 cerevisiae) abhydrolase domain 1.653173328 0.003571788 Abhd4 105501 containing 4 negative elongation factor 1.652019427 0.001793945 Nelfe 27632 complex member E, Rdbp abhydrolase domain 1.650601373 0.000372433 Abhd17a 216169 containing 17A 1.649442316 0.001500746 Vmn1r119 vomeronasal 1 receptor 119 384696 1.647966168 0.006133594 Msh2 mutS homolog 2 (E. coli) 17685 RIKEN cDNA 5430416N02 1.646777325 0.001175154 5430416N02Rik 100503199 gene MCM (minichromosome 1.643870897 0.003301296 Mcmbp maintenance deficient) 210711 binding protein 1.643066616 0.000817912 Zfp934 zinc finger protein 934 77117 PRP6 pre-mRNA splicing 1.640956368 0.000701131 Prpf6 68879 factor 6 homolog (yeast) 1.63003822 0.00677829 Rnf31 ring finger protein 31 268749 1.625377698 0.008155468 Kpna1 karyopherin (importin) alpha 1 16646 F-box and WD-40 domain 1.625359568 0.001973982 Fbxw5 30839 protein 5 1.618497332 0.007485789 Spop speckle-type POZ protein 20747 1.618110984 0.007401256 Gm3453 predicted gene 3453 100041651 1.616110282 0.001920935 Prdm8 PR domain containing 8 77630 trafficking protein particle 1.61428904 0.000728499 Trappc10 216131 complex 10 1.613012047 0.008328006 Mex3c mex3 homolog C (C. elegans) 240396 1.611891612 0.007090348 Wdr59 WD repeat domain 59 319481 1.610693542 0.003844187 Braf Braf transforming gene 109880 1.606945122 0.00166705 Rn45s 45S pre-ribosomal RNA 100861531 1.604887593 0.006341156 Sphk2 sphingosine kinase 2 56632 mitochondrial ribosomal 1.600572092 0.008539053 Mrpl45 67036 protein L45 uncharacterized 1.598712418 0.005827198 LOC102635985 102635985 LOC102635985 RIKEN cDNA 2610203C20 1.592935772 0.006715345 2610203C20Rik 100042464 gene replication factor C (activator 1.586447675 0.000377571 Rfc2 19718 1) 2 RIKEN cDNA 4930578G10 1.585635285 0.00141483 4930578G10Rik 75952 gene peptidylprolyl isomerase 1.581750543 0.001593646 Ppil2 66053 (cyclophilin)-like 2 Rho GTPase activating 1.5816041 0.00599437 Arhgap1 228359 protein 1

206

store-operated calcium entry- 1.581089978 0.004413837 Saraf 67887 associated regulatory factor polymerase (DNA directed), 1.580214267 0.003288512 Pola2 18969 alpha 2 coatomer protein complex, 1.577402657 0.002888361 Copb2 50797 subunit beta 2 (beta prime) acyl-Coenzyme A oxidase 1, 1.576020531 0.00375365 Acox1 11430 palmitoyl zinc finger SWIM-type 1.575105657 0.007153035 Zswim5 74464 containing 5 RIKEN cDNA A430005L14 1.570108126 0.006933031 A430005L14Rik 97159 gene minichromosome 1.569325672 0.004564865 Mcm3 maintenance deficient 3 (S. 17215 cerevisiae) 1.567996221 0.008728907 Txnl4a thioredoxin-like 4A 27366 protein phosphatase 1, 1.563122159 0.002810524 Ppp1r26 241289 regulatory subunit 26 RIKEN cDNA 0610011F06 1.559266943 0.002126318 0610011F06Rik 68347 gene protein phosphatase 6, 1.559231009 0.003380353 Ppp6r2 71474 regulatory subunit 2 calmodulin regulated spectrin- 1.556925237 0.008561784 Camsap3 associated protein family, 69697 member 3 RIKEN cDNA 2410015M20 1.556002308 0.003622226 2410015M20Rik 224904 gene chromodomain helicase DNA 1.55219723 0.001180689 Chd3os binding protein 3, opposite 80515 strand unc-13 homolog B (C. 1.550855769 0.003101731 Unc13b 22249 elegans) elongator acetyltransferase 1.550334666 0.00664429 Elp3 74195 complex subunit 3 1.548192014 0.001823714 Dclk2 doublecortin-like kinase 2 70762 transforming growth factor, 1.544934888 0.009870556 Tgfbrap1 beta receptor associated 73122 protein 1 1.544068942 0.008566695 Grtp1 GH regulated TBC protein 1 66790 Rho GTPase activating 1.539445868 0.00044117 Arhgap42 71544 protein 42 1.536294894 0.004447679 Gm5553 predicted gene 5553 433874 1.531106121 0.003227514 Med24 mediator complex subunit 24 23989 exportin, tRNA (nuclear 1.527219787 0.002744602 Xpot 73192 export receptor for tRNAs) cytidine 5'-triphosphate 1.518317386 0.004347347 Ctps 51797 synthase small nucleolar RNA, C/D box 1.517695079 0.001050025 Snord49b 100217426 49B 1.516255139 0.000361591 Mir466d microRNA 466d 100124465 1.509359022 0.001423028 Ormdl1 ORM1-like 1 (S. cerevisiae) 227102 KDM1 lysine (K)-specific 1.509111892 0.002078319 Kdm6b 216850 demethylase 6B

207

sorting and assembly 1.507725524 0.000396737 Samm50 machinery component 50 68653 homolog (S. cerevisiae) 1.504284725 0.003407551 Nhlh1 nescient helix loop helix 1 18071

208

Table 5: Comparison of mRNAs on the Smaug2 list identified by DAVID and PANTHER. mRNAs associated with Smaug2 were analyzed with DAVID and PANTHER and the mRNA sets identified with each tool were compared. The reported mRNAs are grouped according to whether they were correctly identified by both tools, by DAVID alone, by PANTHER alone or by neither tool.

Mapped by both DAVID and NOT Mapped by NOT Mapped NOT Mapped PANTHER (159) PANTHER(19) by DAVID (13) by EITHER (6) Tsn BC037034 Arhgap42 Mir669h Camkk2 Gm3453 Mcmbp Mir5125 Tstd1 Snord49b Gid4 Rn45s Ahsa1 Gm5665 Saraf Pla2g10os Xpnpep1 5430416N02Rik Supt6 Chd3os Fam102b A430005L14Rik Ppp1r26 LOC102635985 Samm50 Gm5617 Zfp934 Rnf19b 4930578G10Rik Cacfd1 Rhot2 2610507B11Rik Camsap3 Grtp1 A430104N18Rik Phf11a Snf8 Snhg6 Abhd17a Trappc10 2810047C21Rik1 Ppp6r2 Arhgap33 Mir297a-4 Spaca6 Kdm6b Mir466d Ipmk AA474331 Tomm34 A930011O12Rik Mcrs1 Gm5553 Gramd1a 0610011F06Rik Med24 2610203C20Rik Mxd1 Atf7 Tbcb Paxip1 Lrrc1 Smok4a Abhd4 Fndc4 St5 Xpot Acox1 Epha4 Tbc1d22b Unc119b Taf1d Mxra8

209

Cryz Smarcd2 Mcm3 Snn Fzd8 Kpna1 Txnl4a Dnpep Ctps Ormdl1 Suds3 Pgrmc2 Nol11 Nelfe Git1 Mxi1 Cfdp1 Fchsd2 Gsto1 Rfc2 Ube4b Tgfbrap1 Rnf115 Slc25a22 Cnih2 Glul Fbxw5 Msh2 Mex3b Zfp36l1 Mthfd2 Hes5 Braf Sh3gl1 Elavl3 Bet1l Slc35e3 Tmem70 Ambra1 Ccnl2 Osbpl2 Prkcd Sh3bp1 Mlx Arhgap1 Dusp8 Prdm8

210

Samd4b Cdk5 Kpna6 Ehmt2 Tubb6 Prpf6 Rnf31 Cst3 Mex3c Spop Atxn7l2 Elp3 Cds2 Mrpl45 Slain1 Rps6ka2 Grk6 Fzr1 Cbfb Traf4 Sphk2 Gmeb2 Copb2 Dclk2 Unc13b Asphd2 Gde1 St8sia2 Pddc1 Enc1 Soat1 Rab3d Nhlh1 2410015M20Rik Lfng Phtf1 Cnnm3 Ctage5 Ppap2b Fryl Ppil2 Ap1ar Wdr59 Gpbp1l1 Ing2 Sult4a1 Lyrm4

211

Nhp2 Cntfr Slc43a1 Arhgap11a Dvl3 Rcan3 Ncan Vgll4 Pola2 Mark4 Mgat4b Ptov1 Nt5dc2 Gprin1 Tmem50b Ddx19a Snapin Myh10 Zbtb41 Vmn1r119 Zswim5 Arpc5 Ssh1 Socs7 Tsnax Uaca Tle4 Fasn Dhps Ash2l

212

Table 7: Smaug2-associated mRNAs highlighted by IPA, DAVID or PANTHER with a documented role in the CNS. All mRNAs in the statistically overrepresented categories previously identified by IPA, DAVID or PANTHER were manually curated for known functions in the CNS and the ones with a documented function were included. For each of them, gene symbol, description of the name and a brief description of the function are reported.

Gene Description Symbol Documented CNS function Activator of 90 kDa With Hsp90, it is involved in the occurrence Ahsa1 heat shock protein of craniofacial abnormalities in a zebrafish ATPase homolog 1 model (Sheenan-Rooney et al., 2013) It is an autophagy regulator and its activity Activating molecule in is involved with adult neurogenesis in the Ambra1 BECN1-regulated subventricular zone (Yazdankhah et al. autophagy protein 1 2014) It restricts Rho activation to the apical side Rho GTPase-activating in neural crest cells and promotes Arhgap1 protein 1 epithelial-to-mesenchymal transition (Clay and Halloran, 2013) It is involved in synapse development and Rho GTPase-activating autistic-like behaviour in mice (Schuster et Arhgap33 protein 33 al. 2015), and in intracellular trafficking of TrkB (Nakazawa et al. 2016) It promotes activation of the Egfr promoter Set1/Ash2 histone in neural precursors during development Ash2l methyltransferase and also in glioma samples (Erfani et al. complex subunit ASH2 2015)

Serine/threonine- It is required in the postnatal hippocampus Braf protein kinase B-raf and cerebellum for proper neurogenesis and differentiation (Pfeiffer et al. 2013) It is not characterized at all in mouse but Calcium channel flower the D. melanogaster homolog is a calcium Cacfd1 homolog channel, regulating endo and exocytosis at the synapse (Yao et al. 2009) Calcium/calmodulin- It is a modulator of neuronal differentiation, Camkk2 dependent protein with different isoforms triggering different kinase kinase 2 responses (Cao et al. 2011) It is involved in neuronal migration Cyclin-dependent (Kawauchi, 2015) as well as dendritic spine Cdk5 kinase 5 formation and cortical neuron maintenance (Mita et al. 2016)

213

It is associated with the endoplasmic reticulum (Inglis-Broadgate et al. 2005). CDP-diacylglycerol This homologue is highly expressed in Cds2 synthase differentiating neuroblasts in the mouse retina; it is involved in phototransduction (Volta et al. 1999) It is involved in craniofacial development Craniofacial Cfdp1 and Williams-Beuren syndrome (Makeyev development protein 1 et al. 2011) Ciliary neurotrophic It is involved in neurogenesis and Cntfr factor receptor subunit gliogenesis in the developing nervous alpha system (Heller et al.1996) It is studied in the context of Alzheimer’s Cst3 Cystatin-C disease, it protects neurons (Tizon et al. 2010) It promotes neuronal survival in the retina Serine/threonine- (Nawabi et al. 2015). Mice without this Dclk2 protein kinase double- gene display altered hippocampal neuronal cortin like kinase 2 maturation and have seizures (Kerjan et al. 2009) It is indirectly involved in extension of Deoxyhypusine Dhps neuronal processes by modifying eIF5A synthase (Huang et al. 2007) It is a regulator of mitogen-activated protein kinase, highly expressed in the brain (Martell et al.1995). The D. melanogaster Dual specificity protein Dusp8 ortholog is Mkp3, which negatively phosphatase 8 regulates the activity of the MAP kinase family and is a modulator of differentiation and proliferation (Uniprot annotation) Its expression is increased in brain Segment polarity metastases from lung cancer (Kafka et al. Dvl3 protein dishevelled 2014), it is a downstream target of Wnt homolog DVL-3 signalling (Bayod et al. 2015) Histone-lysine N- Ehmt2 It inhibits self-renewal of glioma stem cells methyltransferase (Tao et al. 2014) Downregulation of this protein delays Elongator complex Elp3 migration and maturation of cortical protein 3 neurons (Creppe et al. 2009) It is a non-motor actin binding protein: it is an early and highly specific marker of Ectoderm-neural neural induction in vertebrates, expressed Enc1 cortex protein 1 during gastrulation and later on throughout the nervous system and in prospective cortical areas (Hernandez et al. 1997) Ephrin type-A receptor Epha4 It belongs to a family of proteins regulating 4 axonal growth and branching (Holmberg et

214

al. 2002)

It is required for maintenance of Fasn Fatty acid synthase proliferation and neurogenesis of adult neural stem cells (Knobloch et al. 2013) It is a E3-ubiquitin ligase regulating F-box/WD repeat- centrosome duplication (Puklowski et al. Fbxw5 containing protein 5 2011) and implicated in a model of Parkinson's disease (Ha et al. 2014) It is a critical member of the anaphase promoting complex, it is also known as Fizzy-related protein Fzr1 Cdh1 and it is required for terminal homolog neuronal differentiation and cell cycle exit (Delgado-Esteban et al. 2014) It has a critical role during brain ARF GTPase- development, knock out mice have Git1 activating protein microcephaly-like small brain (Hong et al. 2015) It is expressed as early as E14.5 – E.16.5 Glul Glutamine synthetase and it is actually a glial marker (Akimoto et al. 1993) Glucocorticoid It is not very studied, cited in the context of Gmeb2 modulatory element- NDMA stimulation, which increase Gmeb2 binding protein 2 expression (Ortuno et al. 2012) Its mutations in human cause encephalopathy with early onset seizures G protein-regulated (Ohba et al. 2015), it interacts with Sprouty Gprin1 inducer of neurite to modulate activation by MAP-kinase outgrowth 1 following growth factor stimulation (Hwangpo et al. 2012). It is upregulated in microglia following G protein-coupled spinal chord injury in a rat model and its Grk6 receptor kinase 6 expression is altered in Parkinson’s disease (Manago et al. 2012) Glutathione S- It has a neuronal protecting role (Lee et al. Gsto1 transferase omega-1 2015) Hairy and enhancer of Hes5 It regulates self-renewal and proliferation of split 5 neural precursors (Kobayashi et al. 2014)

Inhibitor of growth It has a scattered expression in the mouse Ing2 protein 2 brain (Zhao et al. 2015). It may control cell cycle by regulating p21 (Larrieu et al. 2010)

215

Its function is low in Huntington’s disease and its overexpression ameliorates the degenerative phenotype (Ahmed et al. Inositol polyphosphate Ipmk 2015). Mice without this protein die around multikinase embryonic day E9.5 with abnormal folding of the neural tube as well as other developmental defects. It potentiates Notch signalling in a cell autonomous fashion (Kato et al, 2010). In zebrafish, it acts as part of a feedback loop Lfng Lunatic fringe to maintain Notch signalling and prevent excessive differentiation at the expense of the neural progenitors (Nikolaou et al. 2009) It is expressed at low levels in cortical MAP/microtubule precursors but it is upregulated early on Mark4 affinity-regulating during neurogenesis in differentiating kinase 4 neurons (Moroni et al. 2006) It is not very characterized in mouse brain DNA replication Mcm3 but D. melanogaster Mcm3 is involved in licensing factor Mcm3 neurogenesis (FlyBase) Microspherule protein It interacts with FMRP in neurons Mcrs1 1 (Davidovic et al. 2006) Mutations of this gene in humans may DNA mismatch repair Msh2 cause agenesis of corpus callosum (Baas protein Msh2 et al. 2013) In Xenopus it is an important factor to prime the neural precursors for neurogenesis Max-interacting protein (Klisch et al. 2006). D. melanogaster Mnt: Mxi1 1 transcription factor activity, negative regulator of cell growth, neuron projection morphogenesis (FlyBase) It is required for centriole migration to the apical plasma membrane at the onset of Myh10 Myosin heavy chain 10 cilia formation (Hong et al. 2015). Knockout mice for this gene die at E14.5 due to heart and brain defects (Tuzoivic et al. 2013) It is an extracellular matrix glycoprotein and variations or polymorphism in this protein Ncan Neurocan core protein are associated with mental disorders (Raum et al. 2015) Nascient helix loop It is a neuronal marker (Ratie’ et al. 2014), Nhlh1 helix 1 it is downstream of Neurod1 (Kim, 2012) It is a ribosome biogenesis factor and its Nol11 Nucleolar protein 11 absence causes craniofacial defects in Xenopus (Griffin et al. 2015)

216

It may be differentially expressed in people with AD (Chen et al. 2015); Paxip interacts with Pax proteins such as Pax2. It is also involved in DNA repair (Wang et al. 2010), it promotes double stranded break repair Pax-interacting Paxip1 through homologous recombination, protein1 important for maintenance of pluripotency (Kim et al. 2009). Its ablation in mice is lethal at embryonic day 9.5, development is very abnormal, probably due to faulty DNA repair (Cho, 2003) It is required in the mouse cortex to specify upper layer neurons (Inoue et al. 2015). Also involved in the transition from intermediate progenitors to neuronal PR domain zinc finger identity (Inoue et al 2014). It forms a Prdm8 protein 8 repressive complex with Olig-related transcription factor Bhlhb5 to aid in the establishment of appropriate neuronal circuitry and one of the targets that they regulate is Cadherin11 (Ross et al. 2012) Protein kinase C delta Prkcd It protects cortical neurons from apoptosis type (Jung et al. 2005) Pre-mRNA-processing Prpf6 Its mutations in the retina result in retinitis factor 6 pigmentosa (Tanackoivc et al. 2011) Its expression promotes cancer progression (Yang et al. 2015), apparently it interacts with Tctex-1 (Pavlos et al. 2011). Mice knockout exist for all four Ras-related protein Rab3d family members but only if all of them are Rab-3D ablated there is embryonic lethality. Neurons without all Rab have slightly compromised neurotransmitter release (Schluter et al. 2004) Is involved with the social deficits of Williams syndrome (Hoefft et al. 2014). It is one of the factors activated in response to DNA damage and in breast cancer cells its Replication factor C Rfc2 translation is upregulated by subunit 2 overexpression of EIF4G (Badura et al. 2012). D.melanogaster ortholog is rfc4 and without it checkpoint control is faulty (Krause et al. 2001) This gene may be important for Mitochondrial Rho Rhot2 mitochondrial trafficking in neurons GTPase 2 (Fransson et al. 2006)

217

It is downstream of Erk, implicated as an important downstream target of Bm1 in Ribosomal protein S6 glioma (Baxter et al. 2014). D. Rps6ka2 kinase alpha-2 melanogaster ortholog: S6k, it is involved in neuronal development such as axon growth (FlyBase) Sh3 domain-binding protein 1; it is a GTPase activator of Cdc42 and Rac, thus it SH3 domain-binding is important for actin remodelling (Elbediwy Sh3bp1 protein 1 et al. 2012). It is also a downstream effector of Semaphorin mediated growth cone collapse (Tata et al. 2014) It is involved in encephalopathy (Cohen et Mitochondrial Slc25a22 al. 2014) and early-childhood seizures glutamate carrier (Poduri et al. 2013) It is involved in vescicle transport as a dynein adaptor, involved in the retrograde transport of BDNF-TrkB by recruiting SNARE-associated dynein (Zhou et al. 2012). Snapin Snapin protein Snapin deficiency causes abnormalities in the CNS as assayed by mouse models, its deletion results in apoptosis and decreased cell density in the CP and IZ (Zhou et al. 2011) Sterol-O- Soat1 Not too much about it, involved in remote acyltransferase 1 memory in mice (Matynia et al. 2008) It is involved in the modulation of the Reelin pathway and its absence, together with Suppressor of cytokine Socs6 knockout, results in the inversion of Socs7 signaling 7 cortical layers (Lawrenson et al. 2015), development of hydrocephalus (Krebs et al. 2004)

Sphk2 Sphingosine kinase 2 It has a protective role during ischemia (Pfeilschifter et al. 2011) Its lower expression is a poor-prognosis Speckle-type POZ marker in glioma (Ding et al. 2015). In mice Spop protein it is highly expressed in Purkinje cells (Huang et al. 2014) Upon knockdown, interneuron migration in St8sia2 Sialyltransferase the mouse prefrontal cortex is affected (Krocher at al. 2014) It is a layer specific marker in the mammalian brain (Fauser et al. 2013). In Transducin-like Tle4 the chick neuroepithelium it is involved with enhancer protein 4 regional specification during brain development (Agoston and Schulte 2009)

218

It is preferentially expressed by post-mitotic TNF receptor- neurons (Masson et al. 1998), important for Traf4 associated factor 4 myelination and homeostasis of the cerebellum (Blaise et al. 2012) It is part with Trax of an evolutionarily conserved RNA binding complex and deletion of Translin results in loss of Trax Tsn Translin also. This protein seems to be involved in dendritic trafficking and synaptic function (Li et al. 2008) It is part with Trax of an evolutionarily conserved RNA binding complex and Translin-associated deletion of Translin results in loss of Trax Tsnax protein X also. This protein seems to be involved in dendritic trafficking and synaptic function (Li et al. 2008). It is part of the U5 splicing complex and the mutations of this protein are rare, however, Thioredoxin-like Txnl4a they cause Burn-McKeown syndrome and protein 4A present with craniofacial abnormalities (Wieczorek et al. 2014) Uveal autoantigen with It interacts with Rab39b to promote Uaca coiled-coil domains differentiation of Neuro2a cells (Mori et al. and ankyrin repeats 2013) It is a regulator of p53 levels (Du et al. Ubiquitin conjugation 2016) and it is also required for the Ube4b factor E4 B maintenance of the nervous system in adult mice (Kaneko-Oshikawa et al. 2005) It is a regulator of the release and Unc13b Unc-13 homolog B maturation of synaptic vesicles (van de Bospoort et al. 2012)

Xaa-Pro Knockdown of this gene in mice causes Xpnpep1 aminopeptidase 1 severe growth retardation, microcephaly and modest lethality (Yoon et al. 2012)

219