Characterization of Heterogeneity Among Individual Neural Precursor Subpopulations in the Embryonic Mouse Ventricular Zone

By Elizabeth K. Stancik

B.S., May 2001, Hope College

A Dissertation Submitted to

The Faculty of the Columbian College of Arts and Sciences of The George Washington University in partial satisfaction of the requirements for the degree of Doctor of Philosophy

January 31, 2011

Dissertation directed by

Tarik F. Haydar Associate Professor of Anatomy and Neurobiology Boston University School of Medicine

and

Anne Chiaramello Associate Professor of Anatomy and Regenerative Biology

The Columbian College of Arts and Sciences of The George Washington University

certifies that Elizabeth K. Stancik has passed the Final Examination for the degree of

Doctor of Philosophy as of August 27, 2010. This is the final and approved form of the

dissertation.

Characterization of Heterogeneity Among Individual Neural Precursor Subpopulations in the Embryonic Mouse Ventricular Zone

Elizabeth K. Stancik

Dissertation Research Committee:

Tarik F. Haydar, Associate Professor of Anatomy and Neurobiology, Boston University School of Medicine; Dissertation Co-Director

Anne Chiaramello, Associate Professor of Anatomy and Regenerative Biology; Dissertation Co-Director

Sally A. Moody, Professor of Anatomy and Regenerative Biology; Committee Member

Joshua Corbin, Associate Professor of Pediatrics, Pharmacology and Physiology; Committee Member

Dedication

To Mom and Dad, Grandma and Grandpa Stancik, Grandma Kreiger and Uncle

Mark, for always believing in me and showing such great interest in my research. To

Dave, for putting up with my whining. And to Fran and Turk, for keeping me (mostly) sane as I struggled to write this.

iii

Acknowledgements

I owe my most sincere thanks to a number of people for their time and support in guiding me throughout my doctoral work. First and foremost, I thank my mentor Tarik

Haydar. It has been an interesting five year journey and knowing that I could count on him to lighten the mood, or be serious, or sit back and just let me fume has made all the difference. I even appreciate his moving away at the very end of my graduate career, which forced me to grow up—though against my will. Thank you also to his past and present lab members: Ivan Navarro, Lina Chakrabarti, Karine Loulier, Yi-Chun Hsieh and Bill Tyler. Their help at the bench and answers to my endless stream of questions provided me with the tools necessary to complete my work, and the kindness and understanding from these post-docs towards a lowly grad student meant more to me than they will know.

I have greatly appreciated the support and suggestions of my dissertation committee over the past few years. Thank you especially to Dr. Chiaramello for guiding me through the administrative bureaucracy; I would be lost without you. Thank you to my readers, Dr. Moody and Dr. Corbin, whose careful reading and thoughtful critiques made this document better than I could have hoped. Finally, I greatly appreciate Dr.

Chiappinelli’s enthusiasm in agreeing to serve on my committee.

Lastly, the lab members of Center V have provided invaluable insights, as well as emotional support, for the past four years, and for that I cannot thank you enough. To my lunch group, Laura, Jean-Marie, Luis, Elim, Ainoha and Nikkie: our daily gatherings and

irreverent topics of conversation were often the highlight of my day. I wish all of you the best of luck as we go our separate ways.

Abstract of Dissertation

Characterization of Heterogeneity Among Individual Neural Precursor Subpopulations in the Embryonic Mouse Ventricular Zone

The degree of cellular heterogeneity in the rodent neocortical ventricular zone

(VZ) has been a source of contention among developmental neurobiologists for decades.

The widely-held hypothesis that this germinal compartment contains a single, multipotent population of neural precursor cells contrasts with what is known about the human and primate VZ, in which multiple neuron- or glia-restricted precursor populations reside side-by-side. Radial glial cells (RGCs), the supposed sole occupants of the rodent VZ, have been shown to generate excitatory neurons, astrocytes and oligodendrocytes using in vivo and in vitro techniques. However, these studies do not show that RGCs alone comprise the rodent VZ. Indeed, morphological and molecular differences have recently been identified between RGCs and another neural precursor population—short neural precursors (SNPs)—within the mouse VZ, though acceptance of precursor diversity in rodents has not been quick to follow. In order to more fully elucidate the nature and degree of heterogeneity among neural precursor cell types within the rodent VZ, experiments were designed to examine the cell cycle kinetics, neuronal output and lineage potentials of RGCs and SNPs.

In utero electroporation (IUE) was used to label VZ precursors based on expression of specific promoters: the glutamate-aspartate transporter (GLAST) and brain lipid binding protein (Blbp) promoters are expressed in RGCs and the tubulin α-1 (T α1) promoter is expressed by SNPs. The Nestin promoter, thought to label all neural stem cells (NSCs), was used as a control. For acute experiments, here defined as ≤ 48 hr from

IUE to sacrifice, green fluorescent protein (GFP) reporter plasmids driven by cell type-

specific promoters (T α1 or GLAST) were used. For long-term fate mapping experiments

(≥ 72 hr from IUE to sacrifice), co-electroporation of a Cre plasmid, driven by a cell

type-specific promoter (T α1, GLAST, Blbp or Nestin), and a flox-stopped GFP reporter

plasmid was used to mediate Cre/loxP recombination in specific precursor populations.

We found that GLAST + RGCs and T α1+ SNPs have significantly different cell

cycle kinetics, with the length of G1-phase and the total cell cycle duration being 4 hr

shorter in RGCs than in SNPs. Based on a study in which cell cycle kinetics and mode of

division of VZ precursor cells were compared, we hypothesized that SNPs would

undergo more neurogenic divisions than RGCs. Indeed, immunostaining revealed that

SNPs labeled on embryonic day 14.5 (E14.5) generate neurons (GFP/TUJ1+) through divisions in the VZ, while RGCs more often produce intermediate progenitor cells

(GFP/Tbr2 +), a transit amplifying population residing in the subventricular zone, which then divide to generate neurons. This extra step in neuron generation for RGCs results in their neuronal progeny arriving in the neocortical wall later than, and thus laminating superficially to, progeny of SNPs labeled at the same time. Similar mechanisms of neuron generation and lamination patterns were observed when precursors were labeled at E12.5. Interestingly, Nestin + NSCs labeled at E12.5 and E14.5 produced neurons whose laminar allocation was distinct from both RGC- and SNP-derived neurons.

Moreover, there were significant differences in neuronal allocation from Blbp-expressing

RGCs (bRGCs) and GLAST-expressing RGCs (gRGCs) labeled at E12.5 and E14.5.

Lineage analyses of precursors labeled at E16.5 uncovered additional differences in these four populations: all four precursor types produced GFAP + and S100 β+ astrocytes and

vii

ependymal cells; SNPs and NSCs also generated DCX + neuroblasts; but only NSCs produced Olig2 + oligodendrocyte precursors. In conclusion, the data presented here indicate that the mouse VZ is a heterogeneous germinal compartment, comprised by at least two distinct precursor populations which differ in their cell cycle kinetics, mechanism of neuron generation and the phenotypes of their neuronal and glial progeny.

viii

Table of Contents

Dedication………………………………………………………………………………...iii

Acknowledgements……………………………………………………………………….iv

Abstract of Dissertation…………………………………………………………………..vi

Table of Contents…………………………………………………………………………ix

List of Figures……………………………………………………………………………..x

List of Tables…………………………………………………………………………….xii

List of Abbreviations……………………………………………………………………xiii

Chapter 1…………………………………………………………………………………..1

Chapter 2…………………………………………………………………………………40

Chapter 3…………………………………………………………………………………81

Chapter 4………………………………………………………………………………..118

References………………………………………………………………………………138

List of Figures

Figure 1. The Brodmann map. 4

Figure 2. Forebrain development in the mouse. 7

Figure 3. Overview of neocorticogenesis. 9

Figure 4. Neocortical lamination. 12

Figure 5. The developing neocortical wall. 17

Figure 6. The reeler cortex. 27

Figure 7. Radial glial cells and short neural precursors. 36

Figure 8. Cells throughout the VZ and SVZ are exposed to plasmid following in utero electroporation. 52

Figure 9. Temporal properties of in utero electroporation. 54

Figure 10. Short neural precursors have distinct cell cycle kinetics. 57

Figure 11. Promoter characteristics do not affect proliferation kinetics results. 60

Figure 12. Laminar allocation differences in VZ-derived cells at the end of neurogenesis. 63

Figure 13. RGCs and SNPs labeled on the same embryonic day generate neuronal progeny specified to different cortical laminae. 66

Figure 14. Morphological analysis of GFP + SNP- and RGC-derived neurons. 68

Figure 15. RGC progeny are amplified by proliferative IPCs while SNPs generate neurons directly from the VZ. 71

Figure 16. Differential neuronal production from VZ precursor subtypes. 77

Figure 17. Localization of GFP + neuronal progeny from VZ precursor populations electroporated throughout neurogenesis. 91

Figure 18. Laminar allocation of neuronal progeny of separate precursor populations. 93

Figure 19. Distribution of individual GFP + neurons at postnatal day 30. 95

Figure 20. Contrasting modes of neuronal production from gRGCs and SNPs at E12.5. 98

Figure 21. Morphological comparison of deep neurons generated from E12.5 and E16.5 precursors. 101

Figure 22. Lineage potentials of VZ precursor populations. 105-108

Figure 23. Proposed model of neocorticogenesis. 124

Figure 24. Overview of Cre/lox recombination. 128

List of Tables

Table 1. Summary of co-labeling of GFP + SNPs and RGCs. 73

Table 2. Antibodies used for lamination and lineage potential experiments. 86

Table 3. Quantitation of GFP + co-labeled cells at P30. 110

xii

List of Abbreviations

ANOVA Analysis of variance

APC adenomatous polyposis coli

ApoER2 apolipoprotein E receptor-2

Ascl1 achaete-scute homolog 1

ASPM mammalian homolog of abnormal spindle

Blbp brain lipid binding protein

BrdU bromodeoxyuridine bRGC Blbp-expressing radial glial cell

BSA bovine serum albumin

CAG chicken beta actin

CBF1 C promoter-binding factor 1

CBFRE C promoter-binding factor responsive element

Cdc42 cell division control protein 42

CP cortical plate

CPN callosal projection neuron

Cux2 cut-like homeobox 2

DCX doublecortin

DNA deoxyribonucleic acid

DS Down syndrome

DT double transgenic

E embryonic day

FITC fluorescein isothiocyanate

xiii

GFAP glial fibrillary acidic protein

GFP green fluorescent protein

GLAST glutamate aspartate transporter

GOF gain-of-function

gRGC GLAST-expressing radial glial cell

[3H] tritiated

Hes hairy and enhancer of split

Hr hour

IACUC institutional animal care and use committee

IP intraperitoneal

IPC intermediate progenitor cell

ISVZ inner subventricular zone

IUE in utero electroporation

IZ intermediate zone

LGE lateral ganglionic eminence

MCPH autosomal recessive primary microcephaly

Min minute mRNA messenger ribonucleic acid

MZ marginal zone n number

NA numerical aperture

NEC neuroepithelial cell

Neo neomycin

xiv

NF1A/B nuclear factor 1A/1B

NG2 nervous system proteoglycan

Ngn neurogenin(s)

NGS normal goat serum

NICD Notch intracellular domain

NSC neural stem cell

Olig2 oligodendrocyte transcription factor 2

OSVZ outer subventricular zone

Otx1 orthodenticle homeobox 1

P postnatal day

Pax6 paired box gene 6

PBS phosphate buffered saline

PBS-T phosphate buffered saline with TritonX-100

PFA paraformaldehyde

PH3 phosphohistone H3

Q proportion of cells leaving the cell cycle

RC2 radial cell protein 2

RFP red fluorescent protein

RGC radial glial cell

RNA ribonucleic acid

S100 β S100 calcium binding protein β

siRNA short interfering ribonucleic acid

SNP short neural precursor

SOX2 (self-determining region Y) box 2

Svet1 subventricular expressed transcript 1

SVZ subventricular zone

TC length of total cell cycle

TG1 length of G1-phase

TS length of S-phase

Tα1 tubulin alpha one

Tbr1 T-box brain1

Tbr2 T-box brain 2

Tlx human homolog of Drosophila tailless

TNR transgenic Notch reporter

TRITC tetramethyl rhodamine isothiocyanate

Vldlr very-low density lipoprotein receptor

VZ ventricular zone

WT wild-type

XFP (any color) fluorescent protein

xvi

Chapter 1

Introduction

1.1 Overview of neocortical structure, function and ontogenesis

1.1.1 Basic structural organization of the neocortex

The neocortex (also known as cerebral cortex, or isocortex) is a thin sheet of neurons, glia and unmyelinated fibers that forms the outermost layer of the cerebral hemispheres in mammals. It is comprised by six horizontal laminae, each of which contains excitatory projection neurons, inhibitory interneurons, and glial cells.

Neocortical excitatory neurons send information to and receive information from other cortical, subcortical and subcerebral structures, depending on their laminar position

(Molyneaux et al., 2007), while interneurons project locally, synapsing on neighboring excitatory cells as well as other interneurons. Glial cells serve a multitude of functions including structural support, connecting neurons to their blood supply, and removing excess ions and neurotransmitters from the extrasynaptic space (Allen and Barres, 2009).

In addition to its horizontal layers, the neocortex is organized into columns of interconnected neurons oriented radially across the six layers which share the same physiologic properties (Mountcastle, 1997). Cortical columns provide efficient means of transferring ascending and descending signals, discussed in more detail below, via intracolumnar connections.

1.1.2 Function of the neocortex

The neocortex is the seat of information processing related to a number of

functions, including recognition and localization of sensory and pain stimuli, planning

and execution of movements, working memory and positional awareness. Moreover, in

humans and primates, there are regions dedicated to such higher functions as speech and

language processing, judgment and cognition. Much of what we know about the

functional organization of the neocortex is the result of, interestingly, a

cytoarchitechtonic map of the human cortex made in the early twentieth century by

German neurologist Korbinian Brodmann (1909) ( Fig. 1 ). Based on variations in laminar size and neuronal density, Brodmann parceled the neocortex into 52 distinct regions, which have subsequently been shown to correlate closely with specific functions. For example, Brodmann area 4, characterized by its very narrow layer IV and prominent layer V, corresponds to the primary motor cortex.

The laminar organization of the neocortex enables efficient transformation of incoming and outgoing information within functional areas. Sensory signals, for instance, travel from the periphery to the brain by way of ascending pathways, which run through the thalamus before terminating on neurons of layer IV in the primary somatosensory cortex (Brodmann areas 3, 1, 2). Layer IV neurons then relay the signals to layers II and III where projection neurons convey the signals to neighboring association cortices, but the signals are also amplified within layer IV via local connections between inhibitory and excitatory neurons (Shipp, 2007). Signals directing voluntary movement, conversely, are transmitted from the primary motor cortex to specific muscle groups via descending pathways (Shipp, 2007). Both ascending and descending pathways decussate at the level of the medulla, which means that each

Figure 1. The Brodmann map. Neurologist Korbinian Brodmann divided the

neocortex into different areas based on patterns of layer thickness and density. A lateral

view of the human brain with various Brodmann areas superimposed on it is shown,

along with coronal sections through two cytoarchitechtonically distinct areas: 1) the

primary motor cortex, Brodmann area 4 (top left), with its nearly indistinguishable layer

IV and prominent layer V; and 2) the primary visual cortex, Brodmann area 17 (top

right), in which layer IV is dramatically thicker due to the huge amount of efferents

directed there. Modified from Kandel, Schwartz and Jessel . 2000. Principles of

Neural Science, 4 th Ed.

hemisphere receives sensory information from, and mediates movement of, the opposite side of the body.

1.1.3 Neocorticogenesis

Neocorticogenesis is the process by which the anterior neural tube, only one cell layer thick upon closure, develops into the extraordinarily complex, multi-layered neocortex ( Fig. 2 ). This process occurs through the precise regulation of proliferation, migration and differentiation of neural precursor cells by multiple intrinsic and extrinsic factors, many of which will be covered in later sections ( Fig 3 ). A brief overview of the cellular events involved in neocorticogenesis follows.

The mammalian telencephalon is derived from the vesicles that form upon closure of the anterior portion of the neural tube, which occurs on embryonic day 8 (E8) in mice.

The lumen of this portion of the tube will ultimately become the lateral ventricles of the cerebral hemispheres, and the stem cells dorsal to the lateral ventricles will generate the excitatory projection neurons and glia found in the mature neocortex ( Fig. 2 ). Initially these stem cells, called neuroepithelial cells (NECs) because of their epithelial origin, proliferate rapidly and symmetrically to grow up the founder cell population (Rakic,

1974, 1988). At the onset of neurogenesis at E11 in mice, most, if not all, NECs transition to radial glial cells (RGCs) and divide asymmetrically to produce self-renewing

RGCs and neuronal daughter cells with each mitosis (Gotz and Huttner, 2005; Noctor et al., 2001; Noctor et al., 2004; Rakic, 1988). These neurons, the future gray matter of the neocortex, migrate radially away from the ventricles to form the cortical plate through a process known as lamination.

Figure 2. Forebrain development in the mouse. Upon closure of the neural tube, at

embryonic day 8 (E8), the neuroepithelium is only one cell layer thick and is comprised

by neuroepithelial cells (NECs) which divide symmetrically to exponentially grow their

population with each division. A) The anterior neural tube (left) and a coronal section from the level of the cut (right) at E9.5. The NECs will produce all of the dorsal and ventral structures (above and below the red line, respectively) of the mature forebrain.

The lumen of the tube will become the ventricular system in the brain. B) By E12.5, neurogenesis has begun in earnest in the ventricular zone. At the level shown, the lumen has become the lateral ventricles ( LV ). The cortex ( CTX ) and hippocampus ( HP ) are

derived from cells which were located above the red line in ( A), and the lateral and

medial ganglionic eminences ( LGE and MGE , respectively) from cells below the red

line. The LGE produces interneurons for the striatum and amygdala ( AMY ) and the

MGE generates both cortical and striatal interneurons. C) At E18.5, with gestation

nearly complete, neurogenesis is finished and gliogenesis has begun. Dorsal and ventral

structures have developed many of their defining characteristics, including the three cell

layers of the HP and the striated appearance of the STR . Images modified from Tarik

Haydar.

Figure 3. Overview of neocorticogenesis. Cartoon depicting the neocortical wall through the various stages of neocorticogenesis. Also shown are approximate developmental timeframes for these stages in mice, monkeys and humans. These ranges are very rough estimates, as there are significant variations in the timing of all of these events among different neocortical areas. *These “endpoints” for programmed cell death are reported peaks of synapse elimination; apoptosis occurs in the neocortex throughout life. Adapted from Pasko Rakic.

The earliest born neurons form a thin, transient layer, called the preplate, which resides just under the pial surface ( Fig. 4A ) (Rickmann and Wolff, 1981). The next wave of

neurons born migrates to the level of the preplate, splitting the preplate population into

two groups: one that settles superficially to form the marginal zone (MZ), and one that

moves deep, forming the subplate ( Fig. 4B ) (Marin-Padilla, 1978). The MZ will ultimately become the cell-poor layer I of the neocortex, as the vast majority of its resident Cajal-Retzius neurons disappear shortly after lamination is complete (Derer and

Derer, 1990; Konig et al., 1977; Rickmann et al., 1977), while the neurons that split the preplate form the deepest layer, layer VI, of the cortical plate (Raedler and Raedler,

1978). Most subplate neurons will die postnatally, although a small population has been identified persisting in the deepest portion of layer VI (Chun and Shatz, 1989). As successive waves of neurons are generated, they migrate past the earlier-born neurons, expanding the developing cortical wall and forming layers II-VI of the mature neocortex

(Angevine and Sidman, 1961; Rakic, 1974). The resulting “inside-out” laminar pattern— with the oldest neurons found deep, in layer VI, and the youngest neurons located superficially, in layer II/III—is a defining characteristic of the mammalian neocortex

(Fig. 4C ).

After migrating neurons reach their final laminar position they begin the process of differentiation. Because migration and lamination occurs over many days, differentiation begins earlier for deep-layer than for upper-layer neurons. Differentiation involves both sending out axons to the cortical, subcortical and subcerebral targets specific to their lamina (Molyneaux et al., 2007) and the formation of synapses on their dendritic shafts and spines. Studies on monkeys show that synaptogenesis follows

Figure 4. Neocortical lamination. Similar to neocorticogenesis, lamination involves a

series of steps which must occur in the proper order. A) The transient preplate is formed by the first wave of post-mitotic neurons to migrate away from the ventricular zone (VZ), shown here in blue. B) The next wave of neurons to leave the VZ splits the preplate, forming the marginal zone and the subplate. Interestingly, most preplate neurons do not survive past the lamination process, and are not found in the mature neocortex (note the lack of blue cells in ( C), left). The dots shown in the preplate ( A) and marginal zone ( B) represent the protein Reelin, which will be discussed in the text (section 1.4.1). C)

Successive waves of neurons leave the VZ and migrate past previously-generated neurons to form the six layers of the mammalian neocortex. On the left is a cartoon model depicting the inside-out nature of the neocortex, with the oldest neurons deep

(yellow, orange and red dots, derived from yellow and red cells in ( B)) and the youngest neurons in the most superficial layers. To the right is an image of the mouse neocortex; biocytin labeling allows visualization of differences in neuronal size and density among laminae. Modified from Gilmore and Herrup , 1997. Curr Biol 7: R231-4.

the same order as lamination, with synapses first observed on marginal zone and subplate

neurons, and then from deep to superficial layers (Bourgeois and Rakic, 1993). The

formation of synapses is slow at first, but picks up speed once lamination is complete,

such that maximum synaptic density (per unit neuropil) is reached in the early postnatal

period (Bourgeois and Rakic, 1993). After this early peak, synaptic density begins

decreasing, slowly through infancy/childhood then more rapidly during adolescence,

before leveling off in adulthood (Bourgeois and Rakic, 1993). Concurrent with synapse

elimination, axon pruning, defined as the elimination of excess axonal projections, and

programmed cell death further shape the neocortical landscape. Indeed, it has been

shown that these processes are necessary for normal neocortical functions, as disruptions

of either the apoptotic pathway or axon pruning can lead to morphological and functional

brain defects, and even death (Johnston, 2004; Kuida et al., 1998; Kuida et al., 1996).

Gliogenesis begins contemporaneously with the attenuation of neurogenesis in the

germinal zone(s) of the lateral ventricles, at approximately E16 in mice. Studies

examining the morphology and immunoreactivity of radial glial cells in the monkey and ferret cortices have found that some radial glia remaining after neurogenesis transform into astrocytes (Schmechel and Rakic, 1979; Voigt, 1989). Genetic fate mapping studies in mice have shown that oligodendrocyte precursors and oligodendrocytes are generated from dorsal progenitor cells beginning around embryonic day 16.5 (E16.5), though only after first being produced ventrally, from progenitors in the medial and lateral ganglionic eminences (Kessaris et al., 2006). The switch from neuronal to glial production involves a number of intrinsic factors, including downregulation of proneural genes, like the basic

helix-loop-helix transcription factors Neurogenins (Ngn-1 and -2), and activation of

proglial genes, such as nuclear factors 1A and 1B (NF1A/B) (Okano and Temple, 2009).

Epigenetic modifications, such as DNA methylation of the astrocyte-specific glial fibrillary acidic protein ( GFAP ) gene promoter (Takizawa et al., 2001), are also necessary for the differentiation of neural stem cells into mature astrocytes and oligodendrocytes.

Cortical interneurons, unlike projection neurons, are generated in the ventricular zones of the ganglionic eminences in the ventral telencephalon from roughly E9.5 to

E15.5 in mice (Fig. 2 ) (Butt et al., 2005; Miyoshi et al., 2007; Xu et al., 2004). Cortical interneurons migrate tangentially from the ganglionic eminences into the developing neocortical wall, after which they migrate radially into all cortical layers (Anderson et al.,

2001; Corbin et al., 2001; Tamamaki et al., 1997; Tanaka et al., 2006). Interestingly, despite their ventral origins, interneurons follow a similar lamination pattern to projection neurons such that early-born interneurons occupy deep cortical layers and later-born interneurons are found superficially (Butt et al., 2005; Miyoshi et al., 2007).

In order to form a fully- and correctly-functioning neocortex, all of the processes described here—stem cell proliferation, generation of projection neurons dorsally and of interneurons ventrally, gliogenesis, lamination, differentiation, and axon and neuronal pruning—must occur at the proper time and in the proper amounts.

1.2 Cytoarchitecture and resident cell types of the neocortical germinal zones

1.2.1 The ventricular zone

The ventricular zone (VZ) refers to the single layer of stem cells lining the lateral ventricles of the telencephalon ( Fig. 5 ) (Boulder Committee (1970)). The resident stem

cells, commonly called neuroepithelial cells (NECs), are radially oriented within the VZ,

with apical and basal processes terminating at the ventricular and pial surfaces,

respectively (Gotz and Huttner, 2005). NEC nuclei display a characteristic to-and-from

movement in concert with the cell cycle, termed interkinetic nuclear migration, such that

nuclei migrate back and forth between the ventricular and pial surfaces, where they

undergo M- and S-phases, respectively (Fujita, 1962; Sauer, 1935; Sauer and Walker,

1959; Sidman et al., 1959). Interkinetic nuclear migration gives the single cell layer the

appearance of stratification, hence the VZ is often referred to as a pseudostratified

epithelium.

As stated earlier, NECs are the sole VZ population from closure of the neural tube until the onset of neurogenesis, E8-11 in mice. During this time NECs undergo multiple rounds of symmetric division, doubling the number of stem cells with each mitosis. The length of this proliferative period and the number of mitoses differs among species, and is hypothesized to account for the huge expansion in cortical surface area from rodents to primates to humans, referred to as the radial unit hypothesis, described below (Rakic,

1988, 1995). The initiation of neurogenesis begins when the mode of division in the VZ switches from symmetric to asymmetric, resulting in the production of one proliferative and one post-mitotic daughter, as opposed to two proliferative daughter cells.

Concomitant with the switch from symmetric to asymmetric divisions, NECs lose some epithelial properties while beginning to display certain astroglial characteristics; this is thought to signal the transition from multipotential NECs to more fate-restricted radial glial cells (Gotz and Huttner, 2005), which persist in the VZ for the duration of

Figure 5. The developing neocortical wall. Although nuclei can be seen throughout the depth of the ventricular zone (VZ) at any given time, it is only a single cell layer thick. Radially-oriented biopolar cells (shown in green) undergo mitosis at the ventricular surface but their nuclei travel to the basal VZ for S-phase, a process known as interkinetic nuclear migration. Cells in the VZ first divide symmetrically, to grow up their population, then divide asymmetrically to generate one progenitor and one neuron or neuronal precursor with each round of division. The overlying subventricular zone

(SVZ) contains round or oval multipolar cells which do not undergo interkinetic nuclear migration. These cells are neuronal precursors generated from asymmetric divisions of

VZ progenitors. They usually divide symmetrically within the SVZ to produce two neuronal daughters. The intermediate zone (IZ) is comprised by radial fibers of VZ cells and migrating neurons. The marginal zone (MZ) is the most superficial layer of the developing neocortical wall. Transcription factor expression is also used to define these regions, as VZ cells express Pax6, SVZ cells express Tbr2 and migrating neurons express

Tbr1.

neurogenesis.

Radial glial cells (RGCs) were first observed over one hundred years ago, using

Golgi impregnation, in human cortical tissue (Retzius 1893, 1894). Morphologically,

RGCs are similar to their NEC antecedents in that they are bipolar, undergo interkinetic

nuclear migration, and have processes which terminate at the ventricular and pial surfaces

(Gotz and Huttner, 2005). Molecularly, RGCs share expression of the intermediate

filament protein nestin with NECs, but also express a number of proteins not found in

NECs, such as vimentin, the glutamate-aspartate transporter GLAST, brain lipid binding

protein (Blbp) and the calcium binding protein S100 β (Gotz and Huttner, 2005). RGC

basal processes maintain contact with the pial surface as the cortical mantle expands

during development (Rakic, 1971), even during cell division (Miyata et al., 2001; Noctor

et al., 2001). This is essential for proper cortical development because neurons use these

processes as guides while migrating to the cortical plate (Hatten and Mason, 1990; Rakic,

1971, 1972).

Many studies have been devoted to the potentiality of RGCs. Based on the

expression of GFAP in radial cells in fetal monkey tissue and in astrocytes postnatally, it

was believed that RGCs were committed glial precursors (Levitt et al., 1981, 1983). This

hypothesis was supported by studies in rodents demonstrating the transformation of

RGCs into astrocytes both in vitro and in vivo (Culican et al., 1990; Voigt, 1989).

However, retroviral lineage tracing studies in rodents, which presumably labeled RGCs, found labeled clones composed mostly of one type of cells, that is, neurons or astrocytes or oligodendrocytes but not all three (Grove et al., 1993; Luskin et al., 1988; McCarthy et al., 2001). More recently, time-lapse fluorescence imaging and genetic fate mapping

studies in rodents have confirmed the multipotentiality of the RGC population (Anthony et al., 2004; Casper and McCarthy, 2006; Malatesta et al., 2003), but at the same time demonstrated that, during neurogenesis, RGCs predominantly generate neurons (Anthony

et al., 2004; Noctor et al., 2001; Noctor et al., 2004). In humans, immunohistochemical

analyses have shown that three types of RGCs exist in the early fetal VZ: 1) those

expressing only neuronal markers, such as phosphorylated neurofilament protein (SMI-

31), βIII-tubulin or MAP2; 2) those expressing only RGC-specific markers, such as vimentin or phosphorylated vimentin (4A4); 3) and those co-expressing neuronal and

RGC markers (Howard et al., 2006; Zecevic, 2004). These results, similar to Levitt’s findings in monkeys (Levitt et al., 1981, 1983), indicate that committed neuronal and glial precursors co-habit the human VZ. Interestingly, a population of neuron-restricted progenitors which do not express radial glial markers and display limited proliferative capacities has recently been isolated in the human VZ (Mo et al., 2007), demonstrating that multiple neuron-generating populations exist within the VZ of higher primates. This finding suggests that the greater precursor diversity seen in the human VZ compared to the rodent VZ may account for the greater complexity of the human neocortex. However, it also begs the question of whether additional precursor populations may also be found in the rodent VZ by using more specific labeling techniques.

The radial unit hypothesis, mentioned above, postulates that changes in cortical surface area among mammalian species result from variations in the size of the NEC population in the VZ (Rakic, 1988, 1995). Among the tenets of the radial unit hypothesis are these: 1) radial units consist of all the post mitotic neurons arranged radially in columns, termed ontogenetic columns, which migrated in successive waves to the cortical

plate along the same RGC basal process; 2) the number of neurons in an ontogenetic column is determined by the number and duration of mitotic cycles in the VZ; 3) the number of ontogenetic columns is determined by the number of NECs present at the end of the proliferative period prior to neurogenesis; and 4) the number of neurons within a column determines cortical thickness, but the number of columns determines cortical

surface area. In order to test this hypothesis, low doses of ionizing radiation were

administered to embryonic monkeys at different times of development. When embryos

were irradiated prior to the onset of neurogenesis (E40), cortical surface area was

decreased without similar effects on thickness; however, irradiation after E40 produced

the opposite effect, decreasing cortical thickness but not surface area (Rakic, 1995).

Conversely, significant increases in ventricular and cortical surface area were observed in

mice in which β-catenin, a part of the Wnt signaling pathway thought to promote proliferation, was constitutively expressed (Chenn and Walsh, 2002). These normally smooth-brained rodents developed cortical folds similar to those of gyrencephalic mammals. While competing models of cortical expansion have since been proposed and grouped under the “intermediate progenitor hypothesis” as described below, the radial unit hypothesis remains the most accurate both in predicting how various perturbations to cell proliferation and neuron production will affect cortical size, and in explaining how species-specific differences in NEC proliferation regulate surface area expansion.

1.2.2 The subventricular zone

The subventricular zone (SVZ) is classically defined as the space between the VZ and the intermediate zone, where “small and round or oval” cells proliferate without

undergoing interkinetic nuclear migration ( Fig. 5 ) (Boulder Committee (1970)). More recently it has been shown that SVZ cells, called intermediate progenitor cells (IPCs) or basal progenitors, but referred to as IPCs throughout the remainder of this dissertation, are generated from asymmetric divisions of RGCs in the VZ (Haubensak et al., 2004;

Miyata et al., 2004; Noctor et al., 2004). IPCs migrate from the ventricular surface basally to the SVZ, where they retract their apical processes and adopt a unipolar morphology (Miyata et al., 2004). IPCs generally divide symmetrically within the SVZ to produce two post mitotic neurons (Haubensak et al., 2004; Miyata et al., 2004), though they have also been shown to undergo proliferative symmetric divisions, producing two daughter IPCs (Noctor et al., 2004).

Expression of the T-box transcription factor T-brain 2 (Tbr2) is a hallmark of

IPCs (Englund et al., 2005). Interestingly, the developing cortical wall can be divided into 3 basic regions based on expression of the transcription factors Tbr2, Tbr1 and Pax6.

While Pax6 is expressed in RGCs within the VZ, IPCs in the SVZ and sometimes in the basal VZ express Tbr2, and Tbr1 is expressed in post-mitotic neurons in the intermediate zone and cortical plate ( Fig. 5 ) (Englund et al., 2005; Gotz et al., 1998). While their expression does occasionally overlap at the boundaries between these regions, most often their expression is mutually exclusive (Englund et al., 2005). Recently, studies in which the Tbr2 gene was conditionally inactivated in the developing central nervous system to elucidate the function of Tbr2 in IPCs (Arnold et al., 2008; Sessa et al., 2008) have revealed that Tbr2 expression is required for the specification of IPCs from RGCs, and for IPC proliferation and the differentiation of their neuronal progeny.

In gyrencephalic mammals including monkeys, humans and ferrets, the SVZ is greatly expanded compared to lissencephalic mammals, such as mice and rats (Kriegstein et al., 2006). In early development, the germinal zones of all mammals can be similarly divided into a Pax6-expressing VZ and a Tbr2-expressing SVZ. However, by mid- gestation the SVZ of gyrencephalic mammals has grown to such an extent that it is commonly parceled into inner and outer subventricular zones, ISVZ and OSVZ, respectively (Smart et al., 2002). Recent studies in human and ferret neocortex have shown that a vast majority of OSVZ cells express Pax6 but not Tbr2, while ISVZ cells maintain the Tbr2 +/Pax6 - phenotype characteristic of SVZ cells (Fietz et al., 2010).

Moreover, OSVZ cells can be labeled with the RGC markers nestin, GLAST, vimentin,

SOX2 and Blbp (Fietz et al., 2010; Hansen et al., 2010). Morphologically, OSVZ cells

include a radial glia-like population, which maintain pia-contacting basal processes

during division but lack apical processes (Fietz et al., 2010; Hansen et al., 2010). Using

time-lapse imaging it was shown that these radial glia-like cells are capable of self-

renewing divisions and can generate neuronal precursors similar to IPCs (Hansen et al.,

2010). Altogether, these results indicate that the cellular constituency of the mammalian

OSVZ is more closely related to the VZ than to the SVZ, and that the OSVZ may be an

evolutionary adaptation of gyrencephalic mammals, necessary for production of their

larger neuronal population, as compared to lissencephalic mammals.

Much has been made about potential roles of IPCs in mediating the observed

increases in cortical complexity and surface area from rodents to primates and humans

(Kriegstein et al., 2006; Martinez-Cerdeno et al., 2006; Molnar et al., 2006; Striedter and

Charvet, 2009). This hypothesis, called the intermediate progenitor hypothesis

(Kriegstein et al., 2006), is predicated on two other hypotheses: 1) based on the fact that

upper layer markers, like Svet1 and Cux2 , are expressed in the SVZ during development,

IPCs are thought to generate neurons destined predominantly to upper cortical layers

(Tarabykin et al., 2001; Zimmer et al., 2004); and 2) increased SVZ thickness parallels increases in the thickness of upper layers (II-IV) compared to lower layers (V-VI) from rodents to primates, which may be due to a greater number of IPCs, and thus a greater number of IPC-generated neuronal progeny (DeFelipe et al., 2002; Lukaszewicz et al.,

2006; Molnar et al., 2006). However, a number of findings indicate limitations of the intermediate progenitor hypothesis. First, dividing phosphohistone H3 (PH3)/Tbr2 + IPCs are detectable from E10.5 to P0.5 in mice (Kowalczyk et al., 2009), a period which encompasses the entire neurogenetic epoch, and certainly extending beyond the times when upper layer neurons are generated (Angevine and Sidman, 1961). Moreover, the neurogenic fraction of Tbr2 + IPCs was highest from E10.5-E12.5, indicating that IPCs produce deep layer neurons, and likely in high quantities (Kowalczyk et al., 2009).

Second, the observed increases in SVZ thickness have only been hypothesized, but not demonstrated, to correlate with increases in IPC number. Indeed, the human SVZ contains a greater portion of post-mitotic neurons, glial precursors and interneuron progenitors than does the rodent SVZ (Letinic et al., 2002; Zecevic et al., 2005). The presence of the OSVZ in gyrencephalic mammals, with its radial glia-like progenitors

(Fietz et al., 2010; Hansen et al., 2010), also contributes to increased SVZ thickness. In addition, studies characterizing mutant mice have reported alterations in IPC number mirroring changes in cortical thickness, but not in cortical surface area (Pontious et al.,

2008). For example, deletion of the Cdc42 gene greatly increases the number of IPCs

and cortical thickness, but decreases surface area (Cappello et al., 2006; Chen et al.,

2006). Conversely, deletion of the Tlx gene decreases both IPC abundance and upper

layer thickness without any effect on surface area (Land and Monaghan, 2003; Roy et al.,

2004). In conclusion, these collective studies downplay a role for IPC number and SVZ

thickness on evolutionary cortical expansion.

1.4 Determinants of laminar fate

It is well established that the process of cortical lamination is governed largely by

neuronal birth date. Early autoradiographic studies in both rodents and primates have

shown that early-born neurons form the deepest layer (VI) of the neocortex, and that

successively-generated waves of neurons migrate past the first neurons in the cortical

plate, settling between the previously established layers and the pia (Angevine and

Sidman, 1961; Rakic, 1974). In this way, layers II-VI are formed in the characteristic

“inside-out” manner. In the decades since these initial findings, a number of factors have

been identified which also affect laminar specification. Here I will discuss only three,

however: 1) the reelin gene and its protein product Reelin, which together mediate proper

lamination independent of neuronal birth date; 2) β-catenin signaling in the VZ, the

timing and levels of which must be tightly controlled in order for lamination to occur

correctly; and 3) cell cycle parameters, which seem to regulate neuronal birth date and

laminar fate through separate mechanisms.

1.4.1 The reeler mutant and Reelin

The reeler mouse mutant was first described more than half a century ago, identified by its “reeling” gait. Although the ataxic phenotype was due to severe cerebellar hypoplasia, neuroanatomic defects in the neocortex were also prominent (Rice and Curran, 2001). Most obviously, reeler cortical laminae are reversed in order. Early-

born neurons, with morphological and functional characteristics of layer VI neurons, are

found superficially. Late-born neurons, which in wild-type cortex occupy layers II and

III, are instead found deep in the neocortical wall ( Fig. 6B ) (Caviness and Sidman, 1973).

Interestingly, neurons of the reeler cortex are generated from the VZ in the correct order but fail to laminate properly (Caviness and Sidman, 1973), suggesting that the molecular mechanisms regulating neuronal position are separate from those controlling neuronal number and laminar phenotype. The laminar disorganization observed in reeler mice

occurs early in neocorticogenesis, after formation of the preplate. The cohort of neurons

which normally splits the preplate into the marginal zone and subplate fails to do so in the

reeler cortex, creating a dense “superplate” of Cajal-Retzius cells, subplate neurons and

some layer VI neurons, settling in what should be the sparsely-populated layer I (Fig. 6A )

(Caviness, 1982; Derer, 1985). As subsequent waves of neurons migrate out from the

VZ, they are unable to penetrate this mass of cells and laminate top-down in order of

birth date. The fact that afferent and efferent targets are largely preserved in the reeler

mutant (Simmons et al., 1982), infers that neuronal positioning mechanisms are separate

from those regulating axon guidance.

The mutation responsible for the reeler phenotype has been mapped to the reelin

gene, located on the mouse chromosome 5 (Rice and Curran, 2001). The reelin gene

Figure 6. The reeler cortex. In the developing neocortical wall of the reeler mouse

mutant, cells fail to laminate properly. A) The preplate (blue cells) is not split by

subsequent waves of migrating neurons (yellow and red cells), creating a dense mass of

cells through which later-generated neurons cannot pass. B) Thus, the first neurons

generated remain at the top of the cortical plate and late-born neurons are located deep, in

contrast to the normal inside-out order of lamination. This is caused by a mutation to the

reelin gene, resulting in the absence of the Reelin protein product (note the lack of blue

dots compared to Fig. 4 ). Modified from Gilmore and Herrup , 1997. Curr Biol 7:

R231-4.

encodes a large, extracellular glycoprotein, Reelin, which is secreted by Cajal-Retzius

cells in the developing wild-type neocortex, but not in the reeler mutant (D'Arcangelo et al., 1995; D'Arcangelo et al., 1997; Ogawa et al., 1995). Because of its temporal and cell type-specific expression, a number of functions relating to cortical lamination have been proposed for Reelin: 1) to provide positional information to migrating neurons instructing them where to stop; 2) to serve as an attractant molecule, drawing neurons past their predecessors in the cortical plate; 3) or to repel subplate neurons during and after preplate splitting, enabling formation of the cortical plate (Rice and Curran, 2001). In order to determine the mechanism by which Reelin might mediate these actions, the proteins to which it binds were investigated. It was found that Reelin binds lipoprotein receptors, such as the very-low density lipoprotein receptor (Vldlr) and the apolipoprotein E receptor-2 (ApoER2), after which it becomes internalized to phosphorylate Dab1, the cytoplasmic protein product of the disabled-1 gene (Howell et al., 1997; Rice and Curran,

2001). Studies with mice deficient for either Vldlr or ApoER2 receptor demonstrated that binding of Reelin to one or the other receptor yields different neuroanatomic phenotypes, indicating that these receptors have different functions in neuronal migration and positioning (Forster et al., 2010). Specifically, binding of Reelin to Vldlr is responsible for instructing neurons to stop migrating, whereas binding of Reelin to

ApoER2 promotes migration into the preplate (Forster et al., 2010). Thus, Reelin signaling affects laminar fate of migrating neurons through multiple mechanisms.

1.4.2 β-catenin signaling in the VZ

β-catenin is not only a structural component of adherens junctions, which mediate cell-to-cell adhesion in epithelial cells, but also an important signaling molecule of the

Wnt pathway (McCrea and Gu, 2010). Activation of β-catenin is thought to promote cell

proliferation, demonstrated by the decreased cell cycle exit of VZ progenitor cells,

resulting in overproduction of neurons elicited by constitutive activation of β-catenin in

the developing mouse neocortex (Chenn and Walsh, 2002). Examinations of active β- catenin signaling in vivo showed that signaling is strong during early- and mid- neurogenesis, but not late-neurogenesis, and that signaling is largely confined to the VZ, with little activity in the SVZ or intermediate zone (Mutch et al., 2009; Woodhead et al.,

2006). The temporal and spatial patterns of β-catenin signaling suggest that, in addition to proliferation, it may play a role in deep layer neuron specification.

In order to investigate this possibility, dominant-negative and constitutively active

β-catenin constructs were introduced into the developing mouse neocortex at mid- neurogenesis via in utero electroporation, inhibiting and activating the signaling pathway, respectively (Mutch et al., 2009). Analyses performed at E19.5 and P7 have shown that activation of signaling results in more cells located deep in the cortical plate, while inhibition increases the number of cells located superficially in the cortical plate.

Furthermore, BrdU labeling and layer marker analysis have revealed that these superficial neurons express upper-layer markers despite being generated at a time when deep-layer neurons are normally produced (Mutch et al., 2009). Finally, significantly more electroporated (GFP +) cells were found deep in the cortical plate and expressed deep- layer markers following electroporation with the constitutively active β-catenin construct compared to electroporation with a control GFP construct (Mutch et al., 2009).

Altogether, these results indicate that active β-catenin signaling imparts deep-layer

identity to neurons generated from the VZ, and that the temporal gradient of β-catenin

signaling plays a role in determining laminar fate.

1.4.3 Cell cycle parameters

The previous two sections detail the effects of extrinsic factors on neuronal

laminar fate. However, a number of intrinsic factors exert control over laminar

specification, cell cycle parameters principal among them. The length of total cell cycle

(T C) and of individual cell cycle phases (T S, TG1 , etc.), as well as the proportion of

daughter cells exiting the cell cycle in a given division (Q), are some of the cell cycle

properties which affect layer determination. For example, areal differences in laminar

thickness have been shown to correlate with differences in T C and Q in the underlying

VZ (Polleux et al., 1997). However, studies in which specific cell cycle parameters have

been artificially altered in neocortical progenitor cells have provided a better

understanding of the relationships between these factors and laminar fate.

p27 Kip1 is a cyclin-dependent kinase inhibitor that regulates cell cycle progression

at G1-phase (Sherr and Roberts, 1999). As such, it has been suggested to play a role in

neuronal specification. To investigate this hypothesis, Tarui et al. (2005) examined the

cell cycle properties of VZ progenitors and the laminar allocation of their progeny in

doxycycline-dependent p27 Kip1 -overexpressing mutant mice. After administering

doxycycline from E12-E14, they found that cell cycle kinetics and the length of the

neurogenetic period were unchanged, but that Q was significantly increased in transgenic

mice compared to wild type controls (WT) (Tarui et al., 2005). That is to say, while the

length of the cell cycle was the same in DT vs. WT progenitor cells, more cells underwent terminal division at each mitosis in p27 Kip1 -overexpressing mice. BrdU

injection at E14 followed by analysis at P21 revealed more labeled cells in DT cortices

than in WT, consistent with the increase in Q for DT. Moreover, a greater percentage of

labeled neurons were found in the upper layers (II/III, IV) of DT cortices compared to

controls (Tarui et al., 2005). Thus, neurons generated on the same time scale have

different laminar fates when Q is increased via p27 Kip1 overexpression, as compared to

control values. This study demonstrates that the mechanisms regulating Q and laminar

specification work coordinately.

Using forced overexpression of the cell cycle regulators cyclins D and E, Pilaz et

al. (2009) demonstrated the length of G1-phase (T G1 ) also affects neuronal fate. E15.5

VZ precursors, in which cyclin D1 or cyclin E1 was overexpressed via in utero electroporation of gain-of-function (GOF) constructs, display a four hour-long decrease in cell cycle duration compared to controls, with kinetic analysis revealing a similar decreases in T G1 . Examination of the laminar allocation of electroporated cells at P15 uncovered a significant decrease in cortical depth of GFP + GOF neurons compared to

GFP + controls. Specifically, in control cortices, labeled neurons were located primarily in layer IV, though cells were spread throughout the depth of the cortical wall; in GOF cortices, labeled neurons were found almost exclusively in layer II/III and the top half of layer IV (Pilaz et al., 2009). These results indicate that differences in T G1 among VZ precursors result in different laminar fates for their neuronal progeny. Taken together, these studies demonstrate the ability of cell cycle parameters to regulate laminar determination independent of birth date.

1.5 The debate over VZ heterogeneity

1.5.1 Historic perspectives on VZ composition

Although it has been known for over a century that the dorsal telencephalic VZ contains the stem cells which will produce the neurons and glia of the mature neocortex, the exact composition of the VZ has been a source of contention for just as long. In the late nineteenth century both W. His and A. Schaper identified two morphologically distinct cell types in the ventricular zone: rounded cells with mitotic nuclei near or at the ventricular surface, and elongated cells that were not dividing (His 1889; Schaper 1897).

Their interpretations of these similar observations, though, were at odds with one another.

His believed that the two morphologies represented two distinct cell types: the rounded cells, which he termed “germinal cells,” he thought would generate neurons; and the elongated cells, which he named “spongioblasts,” he believed to generate glia (His 1889).

Conversely, Schaper postulated that the two morphologies were representative of different cell cycle phases for the same cell type, which he called “indifferent cells.”

These indifferent cells, Schaper hypothesized, would migrate to the intermediate zone, and there divide to produce both neurons and glia (Schaper 1897). In the interim, S.

Ramon y Cajal had identified modified astroglial cells in the VZ which were present throughout neurogenesis (Ramon y Cajal 1890). Today we know these cells as radial glia, which are multipotent, but at the time Ramon y Cajal interpreted this as a demonstration of coexistence of neuronal and glial precursors in a heterogeneous VZ.

The debate was pushed back towards a homogeneous VZ decades later by studies examining mitoses in the neural tube (Sauer, 1935) and kinetic properties in the germinal

zone of chick embryos (Fujita, 1962; Sidman et al., 1959). Because all cells in the kinetic study were labeled by [ 3H]-thymidine, they were grouped together as a single population, which was postulated to first produce neurons and later glia (Fujita, 1963). Finally, in the late twentieth century, advances in imaging technology and labeling techniques had improved to such a degree that differences in antigen expression could be analyzed on a cell-by-cell basis. Using immuno-electron microscopy analysis of GFAP expression in the monkey VZ, Levitt et al. (1981, 1983) demonstrated the presence of GFAP + cells dividing alongside GFAP - cells throughout cortical development. This indicated that the

primate VZ was, indeed, heterogeneous, containing both dedicated glial precursors

(GFAP + cells) and committed neuronal precursors (GFAP - cells).

1.5.2 Unsolved mysteries of the rodent VZ

With the matter of VZ heterogeneity seemingly settled in primates, the debate turned to other mammals. As described above, the human VZ has been found to contain a mixture of neurogenic, gliogenic and multipotential RGCs, as well as non-RGC committed neuronal precursors (Howard et al., 2006; Mo et al., 2007; Zecevic, 2004).

For rodents, however, the degree of VZ heterogeneity remained unresolved. Many believe that RGCs are the primary, if not the only, cell type in the rodent VZ. This view is based largely on time-lapse imaging studies showing that RGCs can produce neuronal precursors (IPCs) and neurons (Noctor et al., 2001; Noctor et al., 2004), and on genetic fate mapping studies demonstrating neuronal and astro- and oligodendroglial production

(Anthony et al., 2004; Casper and McCarthy, 2006; Malatesta et al., 2003; Pinto et al.,

2008). While these studies demonstrate that RGCs are capable of generating the three

major dorsally-derived cell classes in the mature neocortex, the methods employed—

retroviral labeling with GFP (Noctor et al., 2001; Noctor et al., 2004), and in vivo and in

vitro analyses of cells expressing GFP driven by the GFAP and Blbp promoters (Anthony

et al., 2004; Casper and McCarthy, 2006; Malatesta et al., 2003; Pinto et al., 2008)—are

inherently biased towards labeling RGCs, and may have excluded other progenitor cell

types.

Employing different methods, our lab previously identified two cell types in the

mouse VZ (Gal et al., 2006). Using in utero electroporation (IUE) of membrane-tagged

enhanced green fluorescent protein (eGFP), mitotic cells with ascending processes

(RGCs) were labeled alongside mitotic cells that lacked basal processes ( Fig. 7 ). This

morphological difference was striking, in that it had been established that RGCs maintain

their basal processes during division (Miyata et al., 2001; Noctor et al., 2001; Rakic,

1971, 1972). To confirm that thin and/or convoluted processes were not simply missed in

confocal Z-stacks, three-dimensional reconstructions of ultra-thin serial electron

micrographs were performed, conclusively showing that no basal processes were present

on these “short cells” (Gal et al., 2006). Furthermore, using IUE of reporter plasmids in

which different promoters are used to drive GFP expression, it was found that,

predictably, GLAST and Blbp promoters drive GFP in RGCs, whereas the tubulin α-1

(T α1) promoter drives reporter expression in the other cell type, termed short neural precursors (SNPs; Fig. 7 ) (Gal et al., 2006). When GLAST or Blbp plasmids were co- electroporated with the T α1 plasmid, there was almost no overlap in labeled cells (Gal et al., 2006). This was the first study to introduce tools to separately label neighboring proliferating cells, thereby showing that RGCs are not alone in the developing rodent VZ.

Figure 7. Radial glial cells and short neural precursors. Using in utero electroporation, it was shown that the mouse VZ contains two morphologically and molecularly distinct precursor cell types. A) Radial glial cells (RGCs) maintain their basal process even during division and can be labeled by reporter constructs utilizing

GLAST and Blbp promoters. B) Short neural precursors (SNPs), conversely, retract their basal process during division, although, like RGCs, they do divide at the ventricular surface. SNPs can be labeled by reporter constructs driven by the T α1 promoter.

More recently, it was found that RGCs and SNPs display different levels of Notch signaling activity. Notch receptors and ligands mediate cell-cell signaling and are important determinants of cell fate (Yoon and Gaiano, 2005). GLAST and T α1 reporter

plasmids were co-electroporated with plasmids in which the CBF1-responsive element

(CBFRE), the transcriptional regulator to which cleaved Notch binds in the nucleus,

drives GFP expression (Mizutani et al., 2007). GFP + cells with active Notch signaling pathway were often co-labeled with the GLAST plasmid, but rarely with the T α1

plasmid, suggesting that GLAST + RGCs utilize the Notch signaling pathway to a greater

extent than do T α1+ SNPs (Mizutani et al., 2007). Further support for diversity among

rodent VZ cell types came in the form of a detailed and exhaustive single-cell gene

profiling study (Kawaguchi et al., 2008). After isolating single cells from the E14 mouse

cortical wall, their individual cDNAs were synthesized, amplified and analyzed using

microarrays. Clusters were formed based on shared gene expression and, interestingly,

also corresponded to general cell type classifications: RGCs/VZ cells, IPCs/SVZ cells,

post mitotic neurons. However, within Cluster I (RGCs/VZ cells), there was

considerable variability in expression of cell fate molecules, such as Hes5, Neurogenin2

and AsclI (Kawaguchi et al., 2008). Altogether, these studies indicate that the rodent VZ

is comprised by at least two cell types that differ in their morphology, Notch pathway

activity and gene expression profiles.

Despite this evidence for cellular heterogeneity in the VZ, other studies have

questioned the validity of SNPs as a resident VZ population (Kowalczyk et al., 2009;

Noctor et al., 2008). In an examination of cleavage plane orientation of ventricular

(RGC) and abventricular (IPC) cell divisions, Noctor et al. (2008) found that mitoses at

the ventricular surface had vertical cleavage planes, while abventricular divisions displayed horizontal and oblique cleavage planes. Identification of a Tbr2 + cell dividing

at the ventricular surface with an oblique cleavage plane next to a Tbr2 - cell with a

vertical cleavage plane suggested to them that SNPs were simply a misclassified subset

of IPCs, because all cells dividing vertically displayed a bipolar RGC morphology

(Noctor et al., 2008). Along similar lines of thinking, Kowalczyk et al. (2009) identified

Tbr2 + cells in the VZ with bipolar morphology but no pial-contacting basal processes,

which sometimes divided at the ventricular surface. Kowalczyk et al. (2009) speculated

that these surface-dividing IPCs may represent a transient VZ population, which had

previously been identified as SNPs. These studies demonstrate the need to clearly

identify and classify the SNP population, in order to solve the mystery of rodent VZ

heterogeneity.

To that end, I have developed the following overarching hypothesis: The rodent

neocortical VZ is a heterogeneous germinal compartment composed of multiple

proliferative populations with distinct characteristics. To test this hypothesis, I have

conducted in-depth analyses examining the cell cycle kinetics, mechanisms of neuron

generation, and lineage potentials of VZ precursors, which are presented in the following

chapters:

• In Chapter 2 , new methods for estimating cell cycle kinetics are used on

GLAST-expressing RGCs and T α1-expressing SNPs. This is followed by

analyses of localization of neuronal progeny of these populations, as well

as the general neural stem cell population (Nestin +), in both the embryonic 38

and postnatal cortices. Finally, utilization of IPCs for neuron generation

by the RGC and SNP populations is assessed as a proposed mechanism to

explain differences in neuronal output.

• In Chapter 3 , neuronal specification studies are extended to include

progeny of BLBP-expressing RGCs, and to examine neuronal progeny of

these precursor populations labeled in early, mid and late neurogenesis.

Additionally, the potential of these populations to produce glia is analyzed

through immunohistochemistry.

Chapter 2

Heterogeneity in ventricular zone neural precursors contributes to neuronal fate diversity in the postnatal neocortex

2.1 Introduction

The ventricular zone (VZ) of the dorsal telencephalon contains the progenitor cells that produce all of the various excitatory neurons of the mature neocortex. This process occurs during prenatal mammalian brain development through a precisely regulated series of proliferative and neurogenic divisions. Whether the diversity in neuronal progeny is generated from a similarly heterogeneous pool of precursor cells remains unclear. In fact, numerous studies suggest that the rodent VZ is composed predominantly of a single, multipotent cell type: radial glial cells (RGCs) (Anthony et al.,

2004; Hartfuss et al., 2001; Miyata et al., 2001; Noctor et al., 2001; Noctor et al., 2002;

Pinto et al., 2008). However, this view contrasts with recent studies in which another cell type, termed short neural precursors (SNPs), was identified in the mouse VZ which differs morphologically and molecularly from RGCs (Gal et al., 2006; Mizutani et al.,

2007). Furthermore, primate studies have clearly demonstrated a mix of committed neuronal and glial precursors in the VZ of human and monkey (Howard et al., 2006;

Levitt et al., 1981; Mo et al., 2007; Zecevic, 2004). Delineating the composition of the rodent VZ is therefore important both for understanding how neural stem cells and progenitors properly form the cerebral cortex as well as for elucidating possible mechanisms of species-specific diversity.

In addition to the neural precursors in the VZ, a separate class of neuronal progenitors has been described in the overlying subventricular zone (SVZ). These

intermediate progenitor cells (IPCs) are generated from RGCs (Noctor et al., 2004) and

are now considered a secondary progenitor population within the rodent and primate

neocortex. Although IPCs reside in the SVZ, there is some physical intermixing between

IPCs and RGCs in the basal VZ (Haubensak et al., 2004; Miyata et al., 2004). A small

number of IPCs have also been shown to divide at the ventricular surface, where they

display morphological characteristics similar to SNPs (Kowalczyk et al., 2009; Noctor et

al., 2008). Therefore, it is unclear if SNPs are a distinct population or whether they

simply represent a subset of the IPC population.

To resolve these issues, we developed powerful new methods allowing cell cycle

and lineage analyses of multiple contiguously cycling populations in vivo . Here we show

that the cell cycle kinetics, specifically G1-phase duration, of SNPs and RGCs are

markedly different. Moreover, genetic fate-mapping demonstrates that SNPs and RGCs

give rise to distinct neuronal lineages due to differential reliance on IPC amplification.

These data for the first time uncover the proliferative differences between SNPs and

RGCs and demonstrate how these differences lead to diversity of neuronal daughter cells

during neurogenesis.

2.2 Materials and Methods

2.2.1 In utero electroporation surgery

In utero electroporation (IUE) surgeries were performed on ICR mice bred at

Children’s National Medical Center. Females were checked daily for vaginal plugs; the

day of plug was considered embryonic day (E) 0.5. Surgeries were performed at E13.5 or

E14.5. Electroporation was carried out as previously described (Gal et al., 2006).

Briefly, dams were anesthetized with ketamine/xylazine cocktail and their uterine horns

exposed by midline laparotomy. One microliter of plasmid DNA (3-4 g/ l) mixed with

0.1% fast green dye (Sigma) in phosphate buffer was injected intracerebrally, via pulled

micropipette, through the uterine wall and amniotic sac. The anode of a Tweezertrode

(Genetronics) was placed over the dorsal telencephalon outside the uterine muscle. Four,

40 V pulses (50 ms duration separated by 950 ms) were applied with a BTX ECM830

pulse generator (Genetronics). When the desired number of embryos had been

electroporated, the uterine horns were placed back inside the abdomen, the cavity was

filled with warm physiological saline, and the abdominal muscle and skin incisions were

closed with silk sutures. Dams were placed in clean cages to recover and monitored

closely. These procedures were conformed to United States Department of Agriculture

regulations, and approved by the Children’s National Medical Center Institutional

Animal Care and Use Committee (IACUC).

2.2.2 Plasmid vectors

The CAG-RFP plasmid expresses red fluorescent protein under the control of the

chicken beta actin promoter (gift from J. LoTurco, U. of Connecticut). The T α1-hGFP

plasmid (gift from S. Goldman, U. of Rochester Medical Center) expressed humanized

GFP under the control of the T α1 promoter in short neural precursors (SNPs). The

glutamate-aspartate transporter (GLAST) promoter (gift from D.J. Volsky, Columbia U.

Medical Center) was subcloned into a promoterless, farnesylated eGFP plasmid

(Clontech) to make the GLAST-eGFPf plasmid to label radial glial cells (RGCs). T α1-

Cre plasmid was made by subcloning the T α1 promoter (from pT α1-hGFP) into the

pBS185 Cre vector: the hCMV promoter was removed at the SpeI (5’) and XhoI (3’) restriction sites, and the T α1 promoter was inserted (Life Technologies); pGLAST-Cre was made by subcloning the GLAST promoter and the Cre fragment from pBS185 into the pBluescript II SK (+/-) vector (Stratagene); GLAST promoter was inserted by adding

KpnI (5’) and XhoI (3’) restriction sites to the vector, and the Cre + polyA fragment was inserted via XhoI (5’) and HindIII (3’) sites. pNestin-Cre was a gift from C.Y. Kuan

(Cincinnati Children’s Hospital), and the GFP reporter plasmid (pCALNL-GFP), in which the GFP sequence lies downstream of a floxed Neomycin (Neo) cassette under the control of a CAG driver, was a gift from T. Matsuda (Harvard Medical School).

2.2.3 Determination of plasmid infiltration

MAX Efficiency DH5 α Competent Cells (Invitrogen) were transformed with

CAG-RFP plasmid. The cells were grown in Luria Broth (LB) containing 50 µg/ml bromodeoxyuridine (BrdU; Sigma) for 24 hours, after which the plasmid DNA was concentrated and purified with the EndoFree Maxi Kit (QIAGEN). One µl of the resulting BrdU-laced plasmid was injected intracerebrally and drawn into the cortical wall via IUE. Immediately following surgery the brains were removed, fixed overnight in 4% paraformaldehyde (PFA; Sigma) and cryopreserved in 30% sucrose (Sigma) solution for sectioning. To determine the depth of plasmid infiltration after IUE, sectioned brains were processed for BrdU immunofluorescence (see below).

2.2.4 Cell cycle phase dependence of transfection

IUE was performed with native CAG-RFP plasmid, and single intraperitoneal

BrdU injections (50mg/kg) were administered 2 hr before (-2 hr), immediately after (0hr), or 2, 4, or 6 hr after the surgery. The 0 hr injection labeled cells in S-phase at the time of electroporation, while the -2 hr injection labeled cells in S-through-M-phase at the time of IUE; other time points labeled cells in S-phase at given intervals after IUE. Animals were allowed to survive 14 hrs after IUE, at which time the mother was sacrificed and the fetuses’ heads were removed, fixed overnight in 4% PFA, cryopreserved in 30% sucrose, and sectioned on a cryostat. Sectioned tissue (20 m) was then processed for BrdU

immunofluorescence (see below), using a FITC-conjugated secondary antibody (goat

anti-mouse IgG1 (Southern Biotechnology Associates) diulted 1:200 in 1.5% normal goat

serum (NGS; Sigma) in 1X phosphate buffered saline (PBS)). CAG-RFP +BrdU + (double

positive) cells were counted in a 100 m x 100 m area on confocal Z-stacks and divided

by the number of RFP + cells in the same area to determine the percentage of double

positive cells.

2.2.5 IUE/BrdU method

IUE was performed at E14.5 first using the CAG-RFP plasmid as a control to

compare our experimental values to cumulative BrdU values, and subsequently with

pT α1-hGFP to label SNPs and pGLAST-eGFPf to label RGCs, once the method had been

validated. Single, IP BrdU injections were given at increasing intervals after IUE (one

injection per mouse), and mice were sacrificed 2 hr later. After sacrifice, fetuses’ heads

were removed, fixed, sectioned and processed for BrdU immunofluorescence (see

below). Double positive cells were counted in 100 m x 100 m area on confocal Z-

stacks obtained with a 40x objective. The number of double positive cells was then divided by the number of XFP + cells in the same area to calculate the labeling indices for the respective populations. The time at which the slope of the labeling index increased from zero was established as the length of G1-phase (T G1 ).

2.2.6 IUE/M-M phase method

Tissue from the IUE/BrdU experiments was used for these analyses. The nucleic acid stains SYTO 24 (1:2000; Invitrogen) or propidium iodide (1:1000; Sigma) were used to visualize condensed chromatin on RFP + or GFP + sections, respectively. Mitotic electroporated cells at the ventricular surface were counted and divided by the total number of electroporated cells within 70 m of ventricular surface to calculate the mitotic

+ index. Re-entry of transfected (XFP ) cells into M-phase (i.e. cell cycle duration, T C) was determined to be the time at which the slope of the line increased significantly from zero.

2.2.7 Postnatal GFP + neuron morphological analysis

Morphological analyses were performed on 50 µm vibratome sections from

postnatal day (P) 10 brains that had been co-electroporated at E14.5 with T α1-, GLAST-

or Nestin-Cre plasmid and a floxed-stop GFP plasmid. The morphology of labeled

neurons was assessed as either pyramidal or stellate as compared to Ramon y Cajal’s

drawings of these subtypes. Zeiss LSM Image Browser software was used to make and

rotate 3-D projections of confocal Z-stacks so that entire cell soma and projections could

be visualized. Cells that were not easily identifiable as either pyramidal or stellate neurons were not counted.

2.2.8 Immunohistochemistry

Immunostaining for BrdU was performed as follows: 20 µm sections were rehydrated in PBS containing 0.1% TritonX-100 (PBS-T; Sigma). After 5 min treatment with 0.1% trypsin (Sigma) in 0.1M Tris (Sigma) buffer containing 0.1% calcium chloride

(Sigma), sections were incubated in 2N hydrochloric acid (Fisher Scientific) for 1 hr at room temperature. Sections were then blocked in 10% NGS in 0.1% PBS-T for 30 min and incubated overnight in mouse anti-BrdU antibody (13 l in 1 ml PBS-T; Becton

Dickinson) at 4 oC. Sections were washed 5 times for 5 min in 0.1% PBS-T, incubated at room temperature for 1 hr in goat anti-mouse IgG1 antibody, washed 5 times for 5 min in

0.1% PBS-T and mounted with Vectashield Mounting Medium for Fluorescence (Vector

Laboratories). Sections containing CAG-RFP + cells were processed for BrdU

immunofluorescence using a FITC-conjugated secondary antibody, while GFP + sections

required a TRITC-conjugated secondary antibody (both 1:200; Southern Biotechnology

Associtates).

Immunostaining for mouse anti-TUJ1 (1:500, Covance), rabbit anti-Tbr2 (1:500,

Abcam), rabbit anti-Pax6 (1:1000, Covance) and mouse anti-Ki67 (1:200, BD

Biosciences) was performed on 20 µm coronal sections from embryonic tissue fixed in

4% PFA and cryopreserved in 30% sucrose. Immunostaining for goat anti-Brn1 (1:100,

Santa Cruz) was performed on postnatal (P10) tissue from mice perfused intracardially

with ice-cold PBS followed by 4% PFA. Tissue was post-fixed overnight in 4% PFA and

then sectioned on a vibratome (50 µm, Leica). In all cases, sections were blocked 1 hr at room temperature in 5% NGS and 1% bovine serum albumin (BSA, Sigma) in 0.2%

PBS-T. Primary antibodies were applied overnight at 4 oC. Sections were then washed three times for 5 min in PBS and incubated for 1 hr at room temperature in secondary antibodies (1:200 for all). Secondary antibodies used were: AlexaFluor 546 donkey anti- goat, AlexaFluor 543 goat anti-mouse IgG2a, AlexaFluor 633 goat anti-rabbit (all

Invitrogen), and TRITC-conjugated goat anti-mouse IgG1 (Southern Biotechnology

Associates). Sections were washed three times for 5 min in PBS and counterstained with

ToPro-3 (1:500, Invitrogen), then coverslipped with Vectashield Mounting Medium for

Fluorescence (Vector Labs).

2.2.9 Image acquisition and analysis

Confocal images were acquired with a Zeiss LSM 510 Meta NLO system equipped with an Axiovert 200M microscope. GFP, SYTO 24 and FITC were excited at

488 nm; RFP, propidium iodide, AlexaFluor 543 and 546 and TRITC were excited at 543 nm; ToPro-3 and AlexaFluor 633 were excited at 633 nm. Z-stacks 5-20 m thick, composed of 1024 by 1024 pixel, 0.7- to 5-m-thick optical sections, were collected using 10X, 25X or 40X objectives depending on the experiment.

In order to achieve continuity across experiments, only sections from the same general area of the dorso-lateral neocortex were analyzed. Using anatomical landmarks such as the ganglionic eminences (fetal tissue) and the fimbria and CA1 of the hippocampus (postnatal tissue), sections from as close to the same rostro-caudal level as possible were used throughout.

Image analysis/cell counting was done with the Zeiss LSM Image Browser program. To determine recombined GFP + cell allocation at E17.5, the depth of the cortical wall was measured and divided by five, yielding five bins corresponding to the

VZ, SVZ, IZ and lower and upper CP. GFP + cells in each bin were counted across a 300

µm-wide column in collapsed confocal Z-stacks, then divided by the total number of

GFP + cells in that column to give the percent of total cells in each bin. For all other experiments, cells were counted in 225 µm 2 areas on images taken with the 40X objective. In the Tbr2/TUJ1/Ki67/Pax6 immunostaining images, cells were only counted within 200 µm of the ventricular surface.

2.2.10 Statistical analysis

Student’s t-tests were performed using Microsoft Excel to determine significance between groups in most experiments. However, ANOVA (in SigmaPlot) was used for the IUE/BrdU labeling index, and a nonparametric (Median) regression was performed for the IUE/M-M phase experiment. In all cases, confidence intervals were set at 95% and p-values less than 0.05 were considered significant.

2.3 Results

2.3.1 The mechanics of in utero electroporation enables VZ kinetic studies

We previously showed that the mouse VZ is heterogeneous with respect to its constituent neural precursors (Gal et al., 2006). Using time-lapse multiphoton imaging, it was determined that SNPs retract their basal processes during division at the ventricular surface, while RGCs maintain their basal fibers during mitosis (Miyata et al., 2001;

Noctor et al., 2001; Noctor et al., 2002). SNPs and RGCs were also distinguished based on their abilities to express GFP driven by the tubulin alpha-1 (T α1) or glutamate

aspartate transporter (GLAST) promoters, respectively. It was also recently found that

VZ cells expressing the T α1 and GLAST promoters differentially utilize the Notch

signaling pathway (Mizutani et al., 2007). However, because these were the only known

characteristics distinguishing SNPs from RGCs, we sought to identify further differences

between these precursor populations, starting with their cell cycle kinetics.

The physical intermixing of SNPs and RGCs in the VZ presented challenges for

measuring potential differences in their proliferation in vivo . Cumulative BrdU labeling, a

common technique for estimating cell cycle kinetics in vivo which labels all S-phase cells

indiscriminately (Nowakowski et al., 1989), cannot distinguish between SNPs and RGCs.

Since SNPs and RGCs were previously distinguished by their preferential promoter

expression via IUE, this method seemed ideal to adapt for in vivo proliferation kinetics

studies. However, despite the increasing use of IUE in developmental neuroscience, very

little is known about the mechanics and temporal characteristics of IUE-induced

transfection.

In order to use IUE in quantitative measurements of cell proliferation, both the

location and the proliferative nature of the starting population of cells needed to be

determined. As most IUE is performed with plasmid vectors requiring 10-18 hr delays

before DNA expression is detectable, it has not been possible to estimate the initial

penetration depth of electroporated plasmid since cell division, migration and apoptosis

can all occur within this time frame. Thus, to uncover the location of DNA immediately

after electroporation, we created a plasmid which did not require DNA transcription and

translation to be detectable by physically tagging the plasmid with BrdU ( Fig. 8B).

Immediately following IUE with this BrdU-laced plasmid, we found that all VZ cells within the electroporated area of the neocortical wall were physically exposed to the delivered plasmid up to a depth of 160 µm from the surface of the ventricle ( Fig. 8C).

The next steps were to determine whether all cells physically exposed to the plasmid also express the exogenous DNA and to uncover which cell cycle phases may be necessary for transfection. To do so, we tagged individual precursor groups by phase location using systemic injections of BrdU at various intervals before or after IUE at E14.5 with the ubiquitously expressed CAG-RFP plasmid. For example, injecting BrdU immediately following the IUE procedure labeled VZ cells in S-phase at the moment of electroporation. In this pulse labeling procedure, quantification of the percent of transfected (RFP +) cells which were also BrdU + revealed that 70.87 ± 6.60% were co- labeled ( Fig. 9Ai, D ), strongly suggesting a link between S-phase and plasmid expression. When BrdU was administered 2 hr prior to IUE, which labeled cells from S- phase into M-phase at the moment of electroporation ( Fig. 9Aii ), there was almost complete co-labeling of RFP + cells with BrdU (95.97 ± 4.00%, Fig. 9B, B’, D ). Thus, despite the fact that the entire VZ appears to be exposed to plasmid DNA upon IUE, only the cells arrayed through S- and M-phases at the moment of electroporation eventually express the plasmid. To determine the length of time after IUE that transfection and plasmid expression is still possible within the VZ, we allowed longer gap periods between IUE and subsequent BrdU injection. We found that increasing the time between

IUE and BrdU administration resulted in steadily decreasing proportions of co-labeled cells ( Fig. 9D). Thus, the further away from S-M phases the cells are, the less likely they

Figure 8. Cells throughout the VZ and SVZ are exposed to plasmid following in

utero electroporation. In utero electroporation of a BrdU-labeled plasmid, amplified in bacteria grown in LB broth supplemented with 50 mg/ml BrdU ( B), was performed at

E14.5. ( A) The cartoon depicts the electroporation of plasmid into the mouse cortical

wall via 4 x 40 V pulses of 50 ms each. Immediately thereafter, the fetuses were

sacrificed, their brains removed, fixed and processed for BrdU immunostaining. ( C)

Within the swath of electroporated neocortical wall, the BrdU-doped plasmid was driven

into the tissue to a depth of 160 m. BrdU is shown in red, cell nuclei stained with DAPI

are shown in blue. This experiment indicates that all VZ cells within the electroporated

area are physically exposed to the plasmid. Scale bar: 100 m. ( V) ventricle; ( VZ )

ventricular zone; ( SVZ ) subventricular zone; ( IZ ) intermediate zone; ( CP ) cortical plate.

Figure 9. Temporal properties of in utero electroporation. BrdU was administered immediately following IUE with pCAG-RFP to label cells in S-phase at the moment of

IUE (0 hr; Ai ). Using this paradigm, the majority of cells transfected by IUE (RFP +) were co-labeled with BrdU (70.87 ± 6.60%; D). Administering BrdU 2 hr prior to IUE labeled cells throughout S- to M-phase at the moment of IUE (-2 hr; Aii ). Nearly the entire RFP + population also expressed BrdU using this protocol (95.97 ± 4.00%; B, B’,

D). The arrow in B and B’ denotes an RFP + cell that was not BrdU + (magnified in inset).

These data demonstrate that IUE preferentially transfects cells that are in S- and M- phases at the time of electroporation. ( C, C’ ) Only a small percentage (23.29 ± 4.16%) of cells were co-labeled with RFP and BrdU when BrdU was injected 6 hr after IUE; arrows point to the minority of cells that were co-labeled, shown at higher magnification in the inset. ( D) The percentage of co-labeled BrdU +RFP + cells decreased as the time between IUE and BrdU administration increased, indicating that plasmid viability is limited. Together, these data indicate that the cohort of electroporated cells is temporally limited to those in S-through-M-phase at the moment of IUE, with the leading edge of the transfected population in M-phase, and that the electroporated plasmid is only viable for

6-8 hr after IUE.

are to become transfected. From these experiments, we conclude that IUE preferentially transfects cells which transit through S- and M-phases of the cell cycle within 8 hr of the surgery, and that the vanguard of the transfected population is in M-phase at the moment of electroporation.

Subsequent tracking of this temporally-limited cell cohort allowed measurements of cell cycle phase durations. We developed an IUE/BrdU protocol in which IUE with the CAG-RFP plasmid was followed by BrdU injections at different intervals to estimate the duration of G1-phase (T G1 ) as the time necessary for the electroporated cells to re- enter S-phase ( Fig. 10A). A labeling index was created by quantifying

(BrdU +RFP +)/RFP + cells in a 100 X 100 µm 2 area in confocal images ( Fig. 10B). Our

IUE/M-M phase protocol estimated the total cell cycle duration (T C) as the time required

for the electroporated VZ cell cohort, the leading edge of which is in M-phase at the

moment of electroporation, to re-enter the next mitosis at the surface of the ventricle ( Fig.

10A). A mitotic index was created by counting the number of mitotic (visibly condensed

chromatin) RFP + cells at the ventricular surface, divided by the total number of RFP +

cells ( Fig. 10C). Estimates of T G1 and T C acquired using the IUE/BrdU and IUE/M-M

phase protocols with CAG-RFP (12 hr and 20 hr, respectively) were consistent with

previous estimates of the general VZ population (11.8 hr and 17.5-18.4 hr, respectively)

(Takahashi et al., 1995). Therefore, we conclude that these novel IUE-based methods

represent sensitive and robust alternatives to cumulative BrdU labeling for analyses of

cell cycle kinetics.

Figure 10. Short neural precursors have distinct cell cycle kinetics. (A) Labeling schemes of IUE/BrdU and IUE/M-M phase methods for estimating T G1 and T C, respectively. T G1 represents the time it takes transfected cells (red) to transit through G1- phase and re-enter S-phase, determined by administering BrdU at 2 hr intervals following

IUE until cells become co-labeled with BrdU (yellow). T C represents the cell cycle duration, determined by identifying the time necessary to find mitotic electroporated cells

(green) at the ventricular surface. ( B) For proof of principle, IUE/BrdU method was first performed with CAG-RFP plasmid to label all cells within the VZ; cells re-entered S- phase 12-14 hr after IUE (arrow). ( B’ ) To measure T G1 for individual precursor populations, IUE was performed with pT α1-hGFP or pGLAST-eGFPf; SNPs entered into the next S-phase 4 hr later than RGCs (T G1 of 16 hr vs. 12 hr, marked with arrows; n ≥3 for all time points; ANOVA, * p<0.0001). ( C) To determine M-to-M-phase duration,

RFP + cells with condensed chromatin at the ventricular surface were counted following staining with SYTO 24. Transfected RFP + cells entered into the next M-phase 20-22 hr following IUE (arrow). ( C’ ) M-phase re-entry following IUE for SNPs was delayed by 4 hr compared to RGCs (T C of 24-26 hr vs 20-22 hr, marked with arrows; n ≥3 for all time points; median regression, * p<0.009). In all graphs, error bars represent standard error

(SE).

2.3.2 VZ precursor populations differ in the length of G1-phase

To measure potential differences in cell cycle kinetics between the individual

SNP and RGC VZ cell types, we used the IUE/BrdU and IUE/M-M phase methods with

cell type-specific pT α1- and pGLAST-GFP reporter plasmids. Quantification of

(BrdU +GFP +)/GFP + cells for both the T α1 and GLAST time courses resulted in significantly different labeling curves ( Fig. 10B’ ; ANOVA, p<0.0001). Notably, S-phase re-entry for SNPs was delayed by 4 hrs compared to RGCs (16 hr vs. 12 hr), indicating that T G1 is 33% longer for SNPs. The IUE/M-M phase analysis uncovered a statistically significant 4 hr delay in M-phase re-entry for SNPs compared to RGCs: 24 hr vs. 20 hr

(Fig. 10C’ ; median regression, p<0.009). Together, these data suggest that the increased cell cycle duration in SNPs is due specifically to a lengthened G1-phase. This is the first demonstration of cell cycle differences in the heterogeneous population of mammalian

VZ precursor cells. In addition, the biphasic curves for the IUE/BrdU and IUE/M-M phase studies ( Fig. 10B’, C’ ) demonstrate that some SNPs are retained in the VZ, since they returned to the ventricular surface for at least two successive divisions after beginning to express the exogenous plasmid.

To ensure that potential differences in strength between the GLAST and T α1 promoters were not biasing the results (i.e. driving transcription and translation of GFP more quickly in one cell type than the other), we examined GFP expression in tissue harvested 10 hr post IUE with either pGLAST-eGFPf or pT α1-hGFP ( Fig. 11 ). Both

RGCs (Fig. 11A) and SNPs ( Fig. 11B) displayed strong, widespread GFP expression.

We therefore conclude that there are no temporal differences in promoter expression that

could bias the proliferation results. Together, these kinetic experiments confirm

Figure 11. Promoter characteristics do not affect proliferation kinetics results.

GFP-expressing cells are seen across a wide swath of electroporated tissue 10 hr after

IUE at E14.5 with pGLAST-eGFPf (n = 3) or pT α1-hGFP (n = 5) to label RGCs ( A) or

SNPs ( B; 25X objective). At higher magnification (40X objective; A’-B’ ), it is clear that none of the labeled cells has returned to the ventricular surface to divide, and in fact, most cell soma are in the basal VZ, consistent with our reported T C of 20-22 hr. Scale bars: 50 µm.

that the murine VZ contains multiple types of resident precursors which are distinguished

by significant differences in proliferation dynamics.

2.3.3 Neuronal output differs for SNPs and RGCs

We next asked how these differences in SNP and RGC cell cycle kinetics affect

the growth of the neocortical wall and the allocation of SNP- and RGC-derived cells to

the cortical mantle. Previous studies showed that VZ cells with longer cell cycles tend

towards neurogenic, rather than proliferative, divisions (Calegari et al., 2005). We

therefore hypothesized that SNPs would undergo more direct neurogenic divisions than

RGCs, quickly producing neurons that migrate away from the VZ. To test this, we used

Cre/Lox-based genetic fate mapping to label the progeny of SNPs and RGCs at mid-

neurogenesis and analyzed their distribution across the neocortical wall. Co-

electroporation of promoter-specific Cre plasmids and a floxed stop GFP reporter

plasmid was used at E13.5 to label RGCs (GLAST promoter), SNPs (T α1 promoter), or

all neural stem cells (NSCs, Nestin promoter; Fig. 12A). By E17.5, there were noticeable differences in the allocation of labeled (GFP +) progeny of RGCs and SNPs ( Fig. 12B).

While less than a third of the total cells produced from each precursor cell type remained in the proliferative areas of the cortical wall (VZ and SVZ; Fig 12C), the percent of RGC progeny in the SVZ was nearly twice that of SNP progeny (18.12 ± 2.66% vs. 9.09 ±

3.43%; Fig. 12C). SNP progeny were primarily localized in the lower half of the cortical plate (40.76 ± 7.05% of GFP + cells; Fig. 12C) whereas RGC progeny occupied the upper half (35.23 ± 2.96% of GFP + cells; Fig. 12C). By comparison, Nestin + NSCs generated

Figure 12. Laminar allocation differences in VZ-derived cells at the end of

neurogenesis. (A) Genetic fate mapping experiments were performed by co- electroporation of cell type-specific Cre plasmids together with a floxed stop GFP reporter plasmid. ( B) Embryos electroporated at E13.5 were harvested at E17.5 for

analysis of the distribution of GFP + progeny across the thickness of the neocortical wall.

bins, yielding regions corresponding to the VZ, SVZ, IZ, and lower and upper CP, and

GFP + cells were counted across a 300 µm width in each region. Most GFP + progeny of

SNPs were found in the lower CP (second box from top on all bars) by the end of

neurogenesis, while more RGC progeny were found in the upper CP (top box on all bars).

pNestin + NSCs generated progeny with an intermediate distribution between that of

RGCs and SNPs. Graph shows percent of total GFP + cells (per 300 µm column) in each

region, error bars indicate SE; n = 3 for pNestin and pT α1, n = 4 for pGLAST. Scale bar:

100µm.

cells that were evenly distributed across the depth of the neocortical wall ( Fig. 12B, C ),

as expected for a general precursor population comprised by both SNPs and RGCs.

Given the inside-out laminar specification of the mammalian neocortex (Rakic,

1974), these results suggest either that: 1) SNP-generated neurons are born earlier than

RGC-generated neurons and are thus specified to deeper cortical laminae, or 2) SNP-

generated neurons are born later and have not finished migrating by E17.5. To test these

possibilities, we conducted a longer-term study, performing the IUE-mediated fate

mapping at E14.5 and analyzing the final laminar position of the recombined GFP + neurons on postnatal day 10 (P10; Fig. 13 ). By measuring the depth of GFP + neuronal soma from a reference line drawn at the top of layer II/III (determined by immunohistochemistry for Brn1) we found that neurons generated from SNPs reside significantly deeper than neurons from RGCs ( Fig. 13B-D). The average distance of

GFP + RGC neuronal progeny, which were found in the lower half of layer II/III, was

236.00 ± 8.58 µm ( Fig. 13D), while GFP + SNP progeny resided more than 100 µm deeper, primarily in layer IV (avg. distance 361.83 ± 21.59 µm; Fig. 13D). By comparison, neurons generated from pNestin + NSCs displayed a more diffuse arrangement than the tightly packed bands of RGC or SNP progeny ( Fig. 13A-A” ), with

individual neurons spread throughout the depth of layer II/III (avg. distance 163.60 ±

8.76 µm; Fig. 13D). We also classified all fate-mapped progeny based on morphology to

determine whether individual VZ precursor types generate specific neuron subtypes. We

found that each of the three precursor populations generated both pyramidal and stellate

neurons during this stage of neurogenesis ( Fig. 14 ) and that most of the GFP-labeled pyramidal neurons sent callosum-projecting axons. These data indicate that RGCs and

Figure 13. RGCs and SNPs labeled on the same embryonic day generate neuronal progeny specified to different cortical laminae. Following IUE with Cre and GFP reporter plasmids at E14.5, brains were harvested at P10. Composite, low magnification images of GFP + neuronal progeny and callosal axons are shown for each precursor population in A, B and C. Upon examination at higher magnification ( A’-C” ), variations in cell location became apparent. RGC progeny were grouped tightly in the deep part of layer II/III (labeled with Brn1 staining; B” ). In contrast, SNP progeny formed a thin band deep to layer II/III, in layer IV ( C” ). ( D) The average distance (µm) of all GFP +

neuronal soma from the top of layer II/III was significantly greater for cells generated

from SNPs (pT α1) compared to neurons produced by RGCs (pGLAST) or NSCs

(pNestin). ( E) GFP + pyramidal neurons showed similar distribution patterns compared to

the entire neuronal populations of NSC (pNestin), RGC (pGLAST) and SNP (pT α1)

progeny. Error bars show SE; n=3 for pNestin and pT α1, n=2 for pGLAST. * p<0.01

and ** p<0.001 (Student’s t-test). Scale bars: 900 m A-C; 100 m A’-C” .

Figure 14. Morphological analysis of GFP + SNP- and RGC-derived neurons. Close

examination of GFP + cells from the long-term fate mapping experiments revealed two

distinct morphologies, characteristic of pyramidal ( a) and stellate ( a’ ) neurons. ( b) While

all three precursor types generated neurons with stellate morphology, RGCs and SNPs

produced roughly ten percent more stellate neurons (as a percent of total GFP + neuronal

population) than NSCs. n=3 for pNestin and pT α1, n=2 for pGLAST. Scale bar: 25 m.

SNPs labeled at E14.5 produce a similar variety of neuron subtypes, but that these neurons are surprisingly specified to different cortical laminae with progeny of SNPs

predominantly in layer IV and progeny of RGCs mostly in layer II/III.

2.3.4 Relationship of SNPs and RGCs to intermediate progenitor cells

To uncover the mechanism underlying the different positioning of SNP and RGC

progeny, we expanded our focus to include IPCs, reasoning that the time of specification

for SNP- and RGC-derived neurons may be different if one subgroup is sequestered

within the IPC pool. We performed immunohistochemistry for markers of proliferation

(Ki-67), differentiation (TUJ1) and for the transcription factor T-box brain 2 (Tbr2),

which labels IPCs (Englund et al., 2005), on tissue electroporated at E14.5 with pT α1- or

pGLAST-GFP and sacrificed 24 hr later. We found that 60% more pGLAST + RGCs co-

expressed Tbr2 compared to pT α1+ SNPs ( Fig. 15A, A’, D; Table 1 ). Conversely, two-

thirds more SNPs expressed the neuronal marker TUJ1 compared to RGCs ( Fig. 15B, B’,

D; Table 1 ). Differences in proliferation (GFP +/Ki67 + cells) between RGCs and SNPs

were also apparent, but not significant ( Fig. 15D; Table 1 ). Together with the results

from the fate mapping experiments, these data support the hypothesis that SNPs in the

VZ serve a direct neurogenic role while RGCs primarily generate neurons indirectly, via

IPCs.

Finally, we examined expression of the transcription factor paired box gene 6

(Pax6), which labels VZ precursors (Gotz et al., 1998) and is rarely co-expressed in

Tbr2 + cells (Englund et al., 2005). Previous studies identified Tbr2 + cells dividing at the

ventricular surface which are morphologically similar to SNPs

Figure 15. RGC progeny are amplified by proliferative IPCs while SNPs generate

neurons directly from the VZ. Immunostaining for Tbr2, Ki67, TUJ1 ( βIII-tubulin) and

Pax6 was performed on tissue electroporated at E14.5 with pT α1-hGFP or pGLAST- eGFPf and fixed 24 hr later ( A-C’ ). Arrows in A, B and C point to GFP + RGCs double positive for Tbr2, TUJ1 or Pax6, respectively. Orthogonal views of single Z-sections in

A’-C’ show double labeling of GFP + SNPs (outlined in white boxes). ( D) Nearly twice as many RGCs co-expressed Tbr2 compared to SNPs. Three fold more SNPs co- expressed TUJ1 compared to RGCs. Differences between RGCs and SNPs in coexpression of Ki67 or Pax6 were not statistically significant. Error bars represent SE; n=4 per group. * p<0.01 (Student’s t-test). Scale bar: 20 m.

Table 1. Summary of co-labeling of GFP + SNPs and RGCs with Ki67, Pax6, Tbr2

and TUJ1. Immunostaining was performed on tissue harvested 24 hr after

electroporation at E14.5 with either pT α1-hGFP or pGLAST-eGFPf. Cell counts were

performed on 40x collapsed Z-stacks, across the entire width of the image (225 µm) and

up to 200 µm from the ventricular surface. Values represent the percent of GFP + cells

co-expressing each protein or transcription factor, ± the SE; n=4 per group.

Table 1. Summary of Co-labeling of GFP + SNPs and RGCs

Percent GFP+ cells co-labeled with: Ki67 Pax6 Tbr2 TUJ1 11.73 15.66 7.98 2.96 RGCs ±2.41 ±5.64 ±1.40 ±0.84 7.75 17.60 4.78 9.65 SNPs ±0.83 ±0.69 ±0.63 ±1.39

(Kowalczyk et al., 2009; Noctor et al., 2008), implying that SNPs may be a misclassified

population of IPCs. In tissue electroporated with pT α1- or pGLAST-GFP and stained for

Pax6, we found that both pT α1+ SNPs and pGLAST + RGCs were robustly labeled with

Pax6 ( Fig. 15C-D), supporting our proliferation study that indicates that SNPs are a resident VZ population ( Fig. 10 ). In addition, in our immunostainings for Tbr2, we did not observe even a single pT α1+/Tbr2 + cell dividing at the ventricle; most pT α1+/Tbr2 + cells were in the basal VZ or SVZ (n=1707 total pT α1+ cells, n=75 pT α1/Tbr2 + cells).

Thus, we conclude that SNPs and IPCs are distinct precursor cell types.

2.4 Discussion

In this study we used molecular labeling techniques and novel IUE-based

methods to uncover diversity in the proliferative properties and lineage potentials of

separate VZ precursor populations. We present three main findings that fundamentally

increase our knowledge of the rodent neocortical VZ. First, in contrast to the long-

standing view that all VZ cells exhibit similar proliferation parameters (Cai et al., 1997;

Takahashi et al., 1995), we demonstrate that two populations dividing concurrently at the

ventricular surface have significantly different cell cycle kinetics. Second, we find that

neurons fate-mapped from these two populations migrate to different cortical laminae.

Specifically, SNPs and RGCs fate mapped on E14.5 generate neurons that predominantly

settle in layer IV and layer II/III, respectively. Finally, we show that this differential

laminar allocation is due, at least in part, to an increased reliance of RGCs on

intermediate progenitor cells for neuronal production compared to SNPs. From the

significant and persistent differences in laminar positioning, we conclude that SNPs and

RGCs in the murine VZ represent mutually exclusive populations from E14.5 onwards and generate phenotypically different neuronal progeny via separate mechanisms.

Our data indicate that SNPs labeled at E14.5 transit through one or two cell cycles at the ventricular surface, directly producing TUJ1 + neuron(s) with each division which

migrate away to the cortical plate ( Figs. 10B’, 10C’, 15 ). In comparison to SNPs,

RGCs labeled at E14.5 display greater perdurance in the VZ, undergoing multiple rounds

of asymmetric divisions to generate a proliferative daughter cell and a Tbr2 + IPC with

each mitosis (Noctor et al., 2004). We found that progeny of the labeled RGCs sojourn

in the SVZ as IPCs before migrating to the cortical plate, thus arriving later than—and

laminating superficially to—neuronal daughters of SNPs ( Fig. 16 ). Though our data do indicate that some SNPs utilize IPCs for neuronal amplification, just as some RGCs are capable of generating neurons directly (Miyata et al., 2001; Noctor et al., 2001), the results suggest that direct and indirect neuronal production are the primary mechanisms employed by SNPs and RGCs, respectively. Not surprisingly, these different modes of neuron generation influence the resulting size of the neuronal progeny; there were considerably more GFP + RGC-derived neurons than SNP-derived neurons in the brains

we examined ( Fig. 13B-C” ; avg. number GFP + cells per volume analyzed: pGLAST =

135.25 ± 20.61, pT α1 = 45.5 ± 8.94). Thus, SNPs augment neuronal output from the

RGC population and thereby enhance the overall neuronal production capacity of the VZ.

It is therefore tempting to speculate that the SNP pool may have evolved to boost

neuronal production in a discrete, region-specific manner. We previously found that

SNPs are present in substantial numbers within the murine VZ between E13.5 and E16.5

(Gal et al., 2006), the birth period of neurons destined to the mid- to superficial layers of

Figure 16. Differential neuronal production from VZ precursor subtypes. The

experiments in this study were conducted at mid-neurogenesis (E13.5-14.5), the stage

during which stellate and pyramidal neurons begin to be generated from precursors in the

VZ and SVZ. Our data demonstrate that pGLAST + RGCs and pT α1+ SNPs exhibit

markedly different cell cycle kinetics and that SNPs persist in the VZ for at least two cell

divisions ( Fig. 10 ). The Cre/lox fate-mapping studies show that SNPs and RGCs present

in the VZ at the same time produce different types of neurons, as defined by their laminar

position ( Fig. 13 ). Due to the inside-out nature of cortical layer formation, layer IV

neurons are born before neurons residing in layer II/III. Furthermore, these experiments

demonstrate that RGCs divide in the VZ to directly generate a small number of layer IV

neurons, and that they also indirectly generate layer II/III neurons via IPC divisions in the

SVZ. These IPC divisions take additional time, thus progeny of the RGC antecedants are

spread radially over a large laminar area. In contrast, SNPs within the E14.5 VZ do not

generate appreciable numbers of IPCs but rather generate neuronal progeny directly from

the VZ which are allocated more quickly and specifically to layer IV. Thus, RGCs

contribute more substantially to laminar expansion by producing neurons for multiple

layers over a longer period of time, whereas SNPs generate discrete neuronal populations

over a short time window.

the neocortex. Our current fate mapping results indicate that the SNP pool provides focused and temporally limited neurogenesis during this period, most likely generating multiple and successive cohorts of neurons to each developing cortical layer. Whether

SNPs play a significant role in neurogenesis before this period (E13.5-16.5) is currently unknown, but it is possible that SNP prevalence may differ across developmental ages

and cortical regions and thereby help to finely tune laminar growth across the neocortical

mantle.

The results presented here complement and extend work demonstrating a link

between the cell cycle of VZ precursors and the laminar position of their neuronal

progeny (McConnell and Kaznowski, 1991; Pilaz et al., 2009). McConnell’s seminal

heterochronic transplantation study established that laminar fate is determined prior to the

terminal division of neuronal precursors (McConnell and Kaznowski, 1991). Our fate

mapping data expand upon this important study, showing that the time and location of

terminal division of simultaneously labeled precursors varies among individual precursor

groups. Since the terminal division of a VZ precursor’s lineage can either occur at the

ventricular surface (as in the case of most SNPs) or in the SVZ (as in the case of most

RGC descendant cells), laminar specification is therefore critically dependant on the site

of terminal division on a cell-by-cell basis. Recently, the Dehay lab (Pilaz et al., 2009)

found that forced overexpression of cyclin D1 or E1 in E15 mouse VZ precursors can

shorten T G1 and result in an accumulation of neurons in cortical layer II/III and the upper

half of layer IV at P15. This differed significantly from the neuronal progeny of control

precursors, which primarily resided deeper, in layer IV (Pilaz et al., 2009). Here we

demonstrate that the native embryonic mouse VZ contains a heterogeneous pool of

endogenous neural precursors that differ both in cell cycle kinetics and whether they

undergo secondary proliferation as IPCs. These differences naturally predispose SNPs

and RGCs to settle in different laminae.

While numerous studies have identified variations in morphology and antigen and

gene expression within the RGC population (Hartfuss et al., 2003; Hartfuss et al., 2001;

Malatesta et al., 2003; Malatesta et al., 2000; Pinto et al., 2008), our results add to a

growing body of evidence suggesting that multiple neural precursor cell types contribute

to heterogeneity in the rodent VZ. For example, it has been shown that VZ precursors

differentially utilize the Notch signaling pathway (Kawaguchi et al., 2008). In particular,

Notch pathway activation is present in pGLAST + RGCs but is largely absent in pT α1+

SNPs (Mizutani et al., 2007). As Notch activity is known to promote proliferation and inhibit differentiation in both invertebrate and vertebrate systems (Gaiano and Fishell,

2002), this finding suggested that pGLAST + RGCs may remain proliferative while pT α1+

SNPs differentiate more quickly. Indeed, the authors demonstrated in vitro that pT α1-

EGFP + cells generated fewer and smaller neurospheres than cells with active Notch signaling (Mizutani et al., 2007). Here we present in vivo confirmation of this hypothesis, and identify a functional consequence of these differences in gene expression: namely, SNPs generate neurons on an accelerated time scale but in more limited numbers compared to RGCs.

Diversity of the neural precursor pool is likely a critical component of cortical morphogenesis and function. For example, the mature human brain contains up to 100 billion neurons, roughly one-fifth of which are found in the neocortex (Pakkenberg and

Gundersen, 1997). The majority of these are excitatory projection neurons that are

further grouped into subtypes based on their morphology, location, connectivity and function. Evidence from prior studies strongly indicates that these major aspects of neuronal fate are specified in the germinal zone (Miyashita-Lin et al., 1999; Rakic, 1988;

Sansom et al., 2005); thus, the full extent of neocortical diversity is thought to require a similarly heterogeneous neuronal precursor population. Data supporting this premise are

found in several studies that demonstrate multiple precursor populations co-existing in

the human and non-human primate dorsal telencephalic VZ that differ in morphology,

antigen expression and regional and temporal prevalence over the course of

corticogenesis (Howard et al., 2006; Levitt et al., 1981, 1983; Mo et al., 2007; Zecevic,

2004). Though the number of cortical neurons in rodents is orders of magnitude smaller

than in humans (an estimated 4 million in mice (Roth and Dicke, 2005)), their degree of

phenotypic diversity is comparable. Here we provide direct in vivo kinetic evidence of two different proliferative populations intermingled within the murine VZ, demonstrating a fundamental level of similarity between the rodent and primate embryonic neuronal precursor pools. VZ precursor diversity is therefore likely a common mammalian trait necessary to generate the full complement of mature neuronal phenotypes in the neocortex.

Chapter 3

Fate mapping reveals substantial differences in neuronal and glial progeny of ventricular zone precursor populations

3.1 Introduction

In the previous chapter we showed that short neural precursors (SNPs) comprise a resident VZ population in mice which differs from the radial glial cell (RGC) population in their cell cycle kinetics and neuronal output. This demonstrated a fundamental level of heterogeneity in VZ composition more similar to that found in higher mammals than previously thought for rodents. Even within the RGC population there is considerable morphological and molecular diversity among individual cells, further increasing the level of precursor diversity. For example, RGC antigen expression profiles change throughout cortical development, with multiple antigen-defined subsets present early in neurogenesis (RC2 +, RC2/GLAST +, RC2/Blbp + and RC2/GLAST/Blbp +) and only two major subtypes—GLAST/Blbp + and RC2/GLAST/Blbp +-- present at the onset of gliogenesis (Hartfuss et al., 2001). Morphological subsets of RGCs, distinguished by the length of the characteristic radial (basal) process, are also present over the course of corticogenesis (Hartfuss et al., 2003). It was recently found that RGCs with high levels of GFAP-driven GFP expression are more likely to generate intermediate progenitor cells through asymmetric divisions than RGCs with low GFP expression, which undergo neurogenic asymmetric divisions (Pinto et al., 2008). Furthermore, these subsets of GFP- high and –low RGCs differ in their gene expression profiles (Pinto et al., 2008).

Curiously, a functional consequence of the high level of VZ heterogeneity in rodents remains to be elucidated.

The mammalian neocortex is comprised by six layers of phenotypically distinct neurons; neuronal morphology, connectivity and gene expression patterns differ between, and even within, laminae (Molyneaux et al., 2007). Classic autoradiography studies demonstrated an inverse relationship between laminar position and neuronal birth date, such that the earliest-born neurons reside deep in the cortical wall while subsequently- generated neurons migrate past them and settle superficially (Angevine and Sidman,

1961; Rakic, 1974). However, date of birth is not the sole factor affecting neuronal identity. Expression of the transcription factors Otx1 and Svet1 by neural progenitor cells

correlates with deep- or upper-layer neuronal production, respectively (Frantz et al.,

1994; Tarabykin et al., 2001). High levels of β-catenin signaling in the VZ predicates the

generation of deep-layer neurons, even late in cortical development (Mutch et al., 2009).

In addition to our work, a separate study also evinced a link between the cell cycle

duration of VZ precursors and the laminar fate of their neuronal progeny (Pilaz et al.,

2009). Moreover, we showed that cell cycle duration corresponds to precursor cell type,

with RGCs having shorter cell cycles than SNPs and producing neurons that laminate

superficially to SNP progeny. Thus, it is possible that multiple VZ cell types in rodents

are necessary to produce the incredible number and variety of mature neocortical

neurons.

Numerous in vitro and in vivo studies established the neurogenic potential of

RGCs (Anthony et al., 2004; Casper and McCarthy, 2006; Hartfuss et al., 2001;

Malatesta et al., 2003; Malatesta et al., 2000; Miyata et al., 2001; Noctor et al., 2001;

Noctor et al., 2004; Tamamaki et al., 2001), and fate mapping experiments showed RGCs

to be gliogenic as well, producing both astrocytes (Anthony et al., 2004; Malatesta et al.,

2003) and oligodendrocytes (Anthony et al., 2004; Casper and McCarthy, 2006).

However, even taken all together, this body of work does not demonstrate that all neocortical glia and projection neurons originate from RGCs. Indeed, in the previous chapter we showed that, at mid-neurogenesis, SNPs generate neurons that contribute to layers 3 and 4 in the postnatal neocortex. Retroviral lineage tracing studies suggest that the rodent VZ contains populations restricted to the production of either neurons or glia, along with a small percent of multipotent progenitors (Grove et al., 1993; Luskin et al.,

1988; McCarthy et al., 2001). This presents another tenable explanation for precursor diversity in rodents: SNPs and RGCs may have different lineage potentials, and they work in concert to generate the full complement of mature neurons and glia.

In this study we combine the power of Cre/lox recombination with the temporal specificity of in utero electroporation (IUE) to address the contributions individual precursor populations make to mature neocortical cellular diversity throughout neurogenesis. Using IUE of cell type-specific Cre plasmids together with a floxed-stop

GFP reporter plasmid, we labeled two subsets of RGCs (GLAST + and Blbp +), SNPs

(T α1+) and the general neural stem cell (NSC, Nestin +) population at the beginning

(E12.5), middle (E14.5) and end (E16.5) of neurogenesis, and analyzed their neuronal and glial progeny at postnatal day 30 (P30). The proportions of Blbp- and GLAST- expressing RGCs change during cortical development (Hartfuss et al., 2001), and we speculated that this would impact the size and location of the neuronal progeny of these

RGC sub-populations. We included NSCs in our analysis to better understand the relationship between this supposed blanket population, as all neural stem cells are belived to express Nestin, and more discrete precursor groups. Here we show that laminar

allocation patterns differ most among neuronal progeny of E12.5 VZ precursors, but that average neuronal depths are significantly different among progeny of all four populations labeled at all three embryonic time points. Our fate mapping data reveal that only E16.5 precursors generate GFP + glia, and that, surprisingly, all precursor populations are

multipotent.

3.2 Materials and Methods

3.2.1 In utero electroporation surgery

Surgeries were performed at embryonic days 12.5 (E12.5), 14.5 (E14.5) and 16.5

(E16.5), and were fully described in Chapter 2 , section 2.2.1 .

3.2.2 Plasmid vectors

MAX Efficiency DH5 α Competent Cells (Invitrogen) were transformed with

plasmids, after which the plasmid DNA was concentrated and purified with the QIAGEN

EndoFree Maxi Kit for use in IUE surgeries. The plasmid constructs used in these

experiments are: pT α1-Cre; pGLAST-Cre; pNestin-Cre; pBlbp-Cre; and pCALNL-GFP

(see Chapter 2, 2.2.2 , for sources). Plasmids were mixed in 0.1% fast green dye (Sigma)

in phosphate buffer to a final concentration of 1.5-2 µg/µl (3-4 µg/µl total plasmid DNA).

3.2.3 Immunohistochemistry

All imunohistochemistry for the lineage analyses was performed on

electroporated brain tissue harvested from ICR mice killed at postnatal day 30 (P30) by

intracardial perfusion with ice-cold 1X phosphate buffered saline (PBS; Gibco) followed

by 4% paraformaldehyde (PFA; Sigma). Brains were post-fixed overnight in 4% PFA

and sectioned using a vibratome (50 µm; Leica). For the E12.5 Tbr2/TUJ1 analysis,

embryos were harvested 24 hr post-IUE, heads were removed, fixed overnight in 4%

PFA and cryoprotected in 30% sucrose (Sigma). Heads were then frozen in Tissue Tek

(Fisher Scientific) and sectioned on a cryostat (20 µm; Leica). Free-floating and frozen

sections were blocked 1 hr at room temperature in 5% normal goat serum (NGS; Sigma)

and 1% bovine serum albumin (BSA, Sigma) in PBS + 0.2% Triton X-100 (0.2% PBS-T;

Sigma) and then incubated in primary antibody overnight at 4 oC (Table 2). The following day, sections were washed three times for 5 min in PBS, incubated for 1 hr at room temperature in secondary antibodies ( Table 2), washed three times for 5 min in

PBS, mounted on microscope slides and coverslipped with Vectashield Mounting

Medium for Fluorescence with DAPI (Vector Labs).

NeuroTrace 530/615 (fluorescent Nissl stain; Molecular Probes) was used to visualize neuronal soma for the layer analyses on P30 tissue. Free-floating slices were permeabilized for 10 min in 0.2% PBS-T, washed two times for 5 min in 1X PBS, incubated 20 min at room temperature in NeuroTrace (1:100), washed 10 min in 0.2%

PBS-T followed by two times for 10 min in PBS, and mounted and coverslipped with

Vectashield Mounting Medium for Fluorescence (Vector Labs).

3.2.4 Image acquisition

Images were acquired using a Zeiss LSM 510 confocal microscope. Excitation wavelengths were as follows: DAPI was excited at 405 nm; GFP at 488 nm; AlexaFluor

Table 2. Antibodies used for lamination and lineage potential experiments.

Antibody Company Species Dilution

Primary Antibodies

Tbr2 Abcam (ab23345) rabbit 1:500

TUJ1 ( βIII-tubulin) Covance (MMS-435P) mouse 1:500

GFAP Abcam (ab4674) chicken 1:500

S100 β Abcam (ab4066) mouse 1:200

DCX Abcam (ab18723) rabbit 1:1000

APC (CC1) Calbiochem (OP80) mouse 1:500

Olig2 Abcam (ab42453) rabbit 1:500

NG2 Abcam (ab87471) rabbit 1:500

Secondary Antibodies

Rabbit IgG Molecular Probes (A11010) goat 1:200

Rabbit IgG Molecular Probes (A21070) goat 1:200

Chicken IgG Molecular Probes (A21103) goat 1:200

Mouse IgG1 Molecular Probes (A21123) goat 1:200

Mouse IgG2a Molecular Probes (A21133) goat 1:200

Nissl Stain

NeuroTrace 530/615 Molecular Probes (N21482) -- 1:100

543 and 546 and NeuroTrace 530/615 at 543 nm; and AlexaFluor 633 at 633 nm. Images used in the neuronal layer analyses were collected by sequential excitation through a 10X

objective (numerical aperture (NA) = 0.3); 512 x 512 pixel images composed of 5, 3.0

µm-thick optical sections were captured. High magnification images of dendritic spines

of deep layer neurons were captured with the 63X objective (NA = 1.4), using 0.5 µm

optical sections (1024 x 1024 pixels). Images used in the fate mapping analyses (1024 x

1024 pixels) were collected using 20X (NA = 0.8) and 40X (NA = 1.3) objectives, with

1.0 µm and 0.7µm optical sections, respectively. Post-acquisition processing was

performed using Zeiss LSM Image Browser and Adobe Photoshop 7.0.1.

3.2.5 Neuronal layer analysis

All counting was performed on coronal sections taken from the same rostro-

caudal level, where the radiatum layer of the hippocampus emerges. Collapsed Z-stacks

of the upper and lower layers of the neocortical wall were combined into single images in

Adobe Photoshop. For each image, the position of GFP + neuronal soma was first

measured as the absolute Y pixel position; neurons were marked using the paintbrush tool

to keep track of which had and had not been counted. Neuronal position was then

corrected by subtracting the depth of the pial surface from the top of the image, estimated

as an average of four points across the 900 µm-wide image, from the absolute position.

Next, the corrected depths were normalized by expressing them as percent depth through

the cortical wall by subtracting the Y position of the pial surface from the Y position of

the border of the white matter (average of 4 points; determined by absence of neuronal

soma). Normalization was necessary to account for brain-by-brain differences in cortical thickness.

Laminar thicknesses were estimated by converting the pixel depth of layer boundaries (based on size and density of neuronal soma; average of 4 points across the images) into percent depths, and averaging those for all images.

3.2.6 Statistics

Graph Pad software was used to perform one-way ANOVAs, with Tukey post hoc comparisons, to determine significance for average neuronal depth and percent of population specified to individual layers. Confidence intervals were 95%.

3.3 Results

In order to evaluate the progeny of precursor cells labeled during cortical development, we utilized IUE-mediated Cre/lox recombination, a technique which we previously used with success ( Figs. 12-13, Chapter 2 ). For this method, plasmids in which cell type-specific promoters (Nestin, GLAST, Blbp and T α1) drive Cre expression

are co-electroporated with a reporter plasmid in which a floxed Neo cassette is situated

downstream from the ubiquitously expressed chicken beta actin (CAG) promoter and

upstream of the GFP transgene; thus, in the presence of Cre, the Neo cassette is removed

and GFP expression is driven by the CAG promoter ( Fig. 12A, Chapter 2 ). The

exogenous reporter gene is passed down through successive cell divisions, such that

CAG-driven GFP expression is seen in all progeny of the initially-labeled precursor

cohort, even though the cell type-specific promoters may no longer be active in these

cells. Though transgenes introduced via IUE remain episomal, we did not observe significant dilution of GFP intensity through multiple rounds of division (compare deep and superficial neurons in Fig. 17A).

3.3.1 Differential distribution of neuronal progeny persists throughout neurogenesis

Neuronal progeny of all VZ precursors labeled at E12.5 were found throughout the depth of the neocortex at P30 ( Figs 17A). While the overarching trend for all E12.5 progeny was a decrease in the percent of total GFP + neurons from deep (VI) to superficial

(II/III) layers, progeny of GLAST + RGCs (gRGCs) were distributed evenly across all layers ( Fig. 18A). SNP and gRGC progeny showed the greatest difference in the shapes of their allocations. Nearly 90% of SNP progeny were found in layer VI, with dramatically fewer neurons in each layer moving superficially (~5% in layer II/III; red line, Fig. 18A). Conversely, between 20% and 30% of total gRGC progeny were allocated to each layer, with slightly higher percentages in layers VI and II/III (yellow line, Fig. 18A). Allocation curves for NSC and Blbp + RGC (bRGC) progeny (green and blue lines, respectively, Fig. 18A) were intermediate between those for SNP and gRGC progeny. By organizing the data in box plots ( Fig. 19A) we were better able to visualize differences in neuronal dispersion among the four groups. For example, the shorter box and whiskers for SNP progeny compared to all of the others indicate that, overall, these neurons are more tightly packed, even though the outliers (dots) span the depth of the cortical wall. The more compact distribution of SNP-derived neurons suggests that SNPs do not have as great a proliferative capacity as other precursor types, and likely undergo terminal division sooner.

Figure 17. Localization of GFP + neuronal progeny from VZ precursor populations electroporated throughout neurogenesis. Co-electroporation of cell type-specific Cre plasmids together with a floxed-stop GFP reporter plasmid at E12.5 ( A), E14.5 ( B) and

E16.5 ( C) enabled localization of neuronal progeny of four individual VZ precursor populations at P30. Staining with fluorescent Nissl (NeuroTrace 530/615; red cells on images) was performed to visualize all neuronal soma. Size and density of soma were used to identify laminae. Scale bars: 200 µm.

Figure 18. Laminar allocation of neuronal progeny of separate precursor populations. In order to determine the contributions of individual precursor populations to each neuronal lamina, the number of GFP + neurons per layer was divided by the total number of GFP + neurons throughout the entire depth of the neocortical wall, for each precursor population. A) When precursors are labeled on E12.5, the vast majority of

GFP + neurons are found deep in the cortical wall, mostly in layer VI, at P30. Except for

gRGC progeny, which are evenly distributed among the layers, fewer GFP + neurons were

found in superficial layers of the cortex. B) For precursors labeled at E14.5, however, the

reverse is true: most GFP + neuronal progeny are located superficially, in layers II/III and

IV. C) Precursors labeled at E16.5, the end of neurogenesis, produce neurons which

laminate almost entirely in layer II/III. Error bars represent SE. For E12.5, n = 2 for

NSCs, n=3 for all RGCs and SNPs; for E14.5, n= 2 for NSCs, n=6 for gRGCs, n=3 for

bRGCs and n=5 for SNPs; for E16.5, n = 4 for NSCs, n=2 for all RGCs and n=5 for

SNPs.

Figure 19. Distribution of individual GFP + neurons at postnatal day 30. The absolute depth of GFP + neurons was measured and then normalized to percent depth

based on the thickness of the neocortical wall, as measured from the pial surface to the

corpus callosum, for each sample. The pia represented “0% depth,” while the white

matter was “100% depth.” Similarly, layer thicknesses were estimated by converting the

absolute depths of layer boundaries into percent depths, and averaging these across all

samples. Box plots show the distribution of individual neurons generated from Nestin +

NSCs, GLAST + and Blbp + RGCs and T α1+ SNPs labeled at E12.5 ( A), E14.5 ( B) and

E16.5 ( C). Dots represent outliers (from the top 5 th and 95 th percentiles of all depth values). One-way ANOVAs followed by Tukey post hoc comparisons were performed in

Graph Pad to determine significant differences in average neuronal depths among the populations; confidence intervals 95%. * P-value < 0.01; ** P-value < 0.001; *** P- value < 0.0001.

Based on this observation, we predicted that E12.5 SNPs preferentially generate

neurons directly from the VZ as opposed to utilizing intermediate progenitor cells (IPCs)

for neuronal production, as they do at E14.5. To test this hypothesis, we performed IUE

at E12.5 using pT α1- and pGLAST-GFP reporter plasmids, harvested the pups 24 hours later, and probed the tissue with antibodies against Tbr2 and TUJ1 ( Fig. 20 ). Although there appears to be a trend towards more Tbr2 +/GFP + gRGCs than SNPs, the results were not significantly different (10.6 ±2.3% vs. 13.8 ± 3.4%, respectively; Fig. 20C). Roughly twice as many SNPs expressed Tbr2 compared to the neuronal marker TUJ1 (10.6 ±2.3% vs. 5.9 ±2.0%; Fig. 20C), which is the opposite of the situation at E14.5, when twice as many SNPs are TUJ1 + (Table 1, Ch. 2 ). No TUJ1/GFP double-positive gRGCs were found. Together with our results from the previous chapter, these data show that SNPs may change their mechanism for generating neurons from the beginning (indirect, via

IPCs) to the middle (direct) of neurogenesis. However, there are many more SNPs compared to gRGCs which utilize direct neuronal production at E12.5, similar to what we found at E14.5.

VZ precursors labeled at E14.5 generated neurons that were much more closely grouped than E12.5 progeny ( Figs. 17A,B and 19A,B ). GFP + neurons were located

primarily in layers II/III and IV (Fig. 11B), though individual neurons from each precuror

population could be found throughout the depth of the neocortical wall ( Fig. 19B). The significant difference in the average depth of gRGC and SNP progeny observed at P10

(Fig 13D, Chapter 2 ) persists at P30 ( Fig. 19B), indicating that final neuronal position is determined by P10 in mice, and that the electroporated cohorts analyzed in these studies are consistent in their behavior from one electroporation to the

Figure 20. Contrasting modes of neuron production from gRGCs and SNPs at

E12.5. Immunostaining for the IPC marker Tbr2 and the early neuronal marker TUJ1

(βIII-tubulin) was performed on tissue harvested and fixed 24 hr after electroporation at

E12.5 with pGLAST-eGFPf ( A) or pT α1-hGFP ( B). White arrows point to Tbr2/GFP

double-positive cells; white boxes surround TUJ1/GFP double-positive cells. C) The

number of double-positive cells was quantified on 40X images and are expressed as the

percent of total GFP + cells from those images. Differences in Tbr2 coexpression were

not statistically significant between gRGCs and SNPs. No TUJ1 + gRGCs were found.

Error bars represent SE. * p < 0.02 (Student’s t-test). n = 3 for gRGCs, n = 5 for SNPs.

Scale bar: 20 µm.

next. Interestingly, progeny of E14.5 precursors at P30 seem to split into two groups:

NSC and gRGC progeny show almost identical allocation patterns (green and yellow

lines, Fig. 18B), and patterns for SNP and bRGC progeny are closely matched (red and blue lines, Fig. 18B). The percentage of labeled NSC and gRGC progeny increases from layer IV to layer II/III, while the percentage of SNP (and bRGC) progeny does not.

GFP + neuronal progeny of precursors labeled at E16.5 were specified almost exclusively to layer II/III, where they formed tight bands in the most superficial part of the layer ( Figs. 17C, 19C). gRGC progeny again proved the exception, however, and

were spread out across layer II/III ( Fig. 19C). This resulted in significant differences in the mean depth of gRGC progeny compared to NSC-, bRGC- and SNP-derived neurons

(Fig. 19C). We were surprised to again see a scattering of GFP + neurons deep in the

neocortical wall ( Fig. 19C), as multiple reports have shown that late cortical precursors

are unable to produce mature cells with early phenotypes (Desai and McConnell, 2000;

Frantz and McConnell, 1996; Shen et al., 2006). Morphologically, these seemingly

misspecified neurons were usually pyramidal projection neurons, similar in size and

extent of arborization to deep GFP + neurons generated from E12.5 precursors ( Fig. 21 ).

Under high magnification, dendritic spines appeared slightly smaller and less dense on both apical and basal dendrites of E12.5- compared to E16.5-derived deep neurons ( Fig.

21A’,A”,B’,B” ). Though quantitative measurements were not performed, these observations suggest that the deep progeny of E16.5 precursors may still be undergoing synaptogenesis at P30.

All together, these neuronal allocation data show that, from early- to late- neurogenesis, the four VZ precursor populations examined here generally conform to the

Figure 21. Morphological comparison of deep neurons generated from E12.5 and

E16.5 precursors. A few GFP + neuronal progeny of precursors labeled at E14.5 and

E16.5 were surprisingly found in layer VI, even though precursors at these times largely produce neurons destined for more superficial layers. A morphological comparison of deep neurons generated from E12.5 ( A) and E16.5 ( B) precursors (both T α1+ SNPS) reveals similar sizes and extents of local arborization for these two cells. The apical dendrites of both cells extend, mostly unbranched, into layer IV (out of field of view), where they either terminate or disappear from view. Basally projecting axons (thin basal projections with no spines) are visible for both neurons. Dendritic spines are clearly visible on the basal ( A’, B’ ) and apical ( A”, B” ) dendrites at higher magnification (63X).

NeuroTrace 530/615 shown in red. Scale bars: 50 µm ( A, B ) and 10 µm (A’, A”, B’,

B” ).

100

101

inside-out pattern of lamination ( Figs. 17, 18). However, by comparing the distributions

of neuronal progeny generated from precursors labeled at the same time, we find

significantly different average depths, for all four populations, at the three time points

investigated ( Fig. 19). Similar to our findings in Chapter 2 (Figs. 12, 13 ) these data suggest that cortical lamination is a complex process in which a neuron’s laminar position is determined not only by its birthdate, but also by the type of VZ precursor from which it originates. Moreover, these data indicate that individual VZ precursor

(sub)populations work together to produce the correct number and phenotype of mature neocortical neurons.

3.3.2 Lineage potentials differ among VZ precursor populations

Using tissue from the same brains sectioned for the laminar specification experiments, we next performed immunostaining with multiple, lineage-specific antibodies in order to determine the lineage potentials of individual precursor cell types.

The presence of co-labeling was only detected on tissue that had been electroporated at

E16.5. Additionally, GFP + cells with the characteristic highly-branched morphology of protoplasmic astrocytes were easily discernible in the gray matter of tissue electroporated at E16.5, but were absent from tissue electroporated at E12.5 or E14.5. Gray matter astrocytes were visible in tissue electroporated with each of the four Cre plasmids, indicating that, at E16.5, NSCs, gRGCs, bRGCs and SNPs are all, at the least, bipotent precursor populations.

In our analyses of tissue electroporated at E16.5 we found that all precursor types produced cells which were immuno-positive for glial fibrillary acidic protein (GFAP) and

102

for the calcium binding protein S100 β. GFP +/GFAP + double-labeled cells were found in

and near the white matter tracts ( Fig. 22A,B ), in the SVZ of the lateral ventricles ( Fig.

22C) and in the gray matter as protoplasmic astrocytes ( Fig. 22D). Surprisingly, T α1+

SNPs produced more GFAP + progeny than NSCs and RGCs (30.0% (GFP +GFAP +)/GFP +

cells per 450 µm 2 vs. 11.2% for NSCs, 12.8% for bRGCs and 13.5% for gRGCs; Table

3). Most GFP +/S100 β+ cells were protoplasmic astrocytes ( Fig. 22F-H), though some

were SVZ ependymal cells ( Fig. 22E). Even though S100 β is supposed to mark both

astro- and oligodendroglial progenitors, in double immunostainings with S100 β and

Olig2, an oligodendrocyte lineage marker, we did not see any GFP + cells triple labeled

with these markers.

We next investigated whether SNPs, RGCs and NSCs generate neuroblasts using

an antibody against the microtubule binding protein doublecortin (DCX). DCX

immunostaining was present in the hippocampus and in the SVZ of the lateral ventricles,

and small clusters of cells were seen at a distance from the SVZ, in the presumptive

rostral migratory stream. Only SNPs and NSCs produced GFP/DCX double-positive

progeny ( Fig. 22I,L ), and these were fairly infrequent (7.2% and 2.0%

(GFP +DCX +)/GFP + cells per 450 µm 2 for SNP and NSC progeny, respectively; Table 3 ).

Both gRGCs and bRGCs produced GFP + cells in the same areas as, and even adjacent to,

DCX + neuroblasts ( Fig. 22J,K ), though no co-labeled progeny were found.

Previous fate mapping studies showed that, at P30, oligodendrocyte lineage cells

within the cortex and cortical white matter are derived from both ventral precursor cells,

in the lateral and caudal ganglionic eminences, and from endogenous neocortical VZ

precursors (Gorski et al., 2002; Kessaris et al., 2006). In order to determine whether

103

Figure 22. Lineage potentials of VZ precursor populations. Immunohistochemistry

was performed on fixed, vibratome-sectioned tissue to identify GFP + cells which also expressed the lineage-specific markers GFAP, S100 β, DCX or Olig2. Co-labeling of

GFP + cells with these markers was only seen when electroporations were performed at

E16.5; thus, all images show progeny of E16.5 precursors. All precursor types produced cells which co-labeled with the astrocyte marker GFAP and the glial lineage marker

S100 β ( A-H). Only Nestin + NSCs and T α1+ SNPs generated DCX + neuroblasts ( I, L);

GFP + progeny of GLAST + and Blbp + RGCs were often contiguous to DCX + cells, but

were never DCX + themselves ( J-K). Very few GFP + cells were double-labeled with the

oligodendrocyte lineage marker Olig2, and all of these were progeny of NSCs ( M); N-P

show GFP + progeny of RGCs and SNPs in close proximity to, but not co-labeled with,

Olig2. White boxes highlight co-labeled cells. In all images DAPI is shown in blue.

Scale bar: 20 µm. Single channel micrographs of merged images are shown on pages

immediately following the merged images.

104

105

106

107

108

Table 3. Quantitation of GFP + co-labeled cells at P30. GFP + cells co-labeled with

GFAP, S100 β, DCX or Olig2 were counted on collapsed Z-stacks, in two separate 450 X

450 µm 2 areas, for each slice. Values represent the percentage of double-positive cells

divided by the total number of GFP + cells (GFP +anibody +/GFP + cells), for each precursor

type and antibody, ± the SE. Sections from at least four electroporated brains were

analyzed for each plasmid construct/antibody combination. Number of GFP + cells per combination counted are as follows: Nestin/GFAP n=27; Nestin/S100 β n=19;

Nestin/DCX n=10; Nestin/Olig2 n=5; GLAST/GFAP n=21; GLAST/S100 β n=32;

Blbp/GFAP n=21; Blbp/S100 β n=22; T α1/GFAP n=26; T α1/S100 β n=50; T α1/DCX

n=47.

109

Table 3. Quantitation of GFP + co-labeled cells at P30

Nestin-Cre GLAST-Cre Blbp-Cre Tα1-Cre (NSCs) (RGCs) (RGCs) (SNPs) GFP +/GFAP + 11.2 ± 5.1% 13.5 ± 7.7% 12.8 ± 7.7% 30.3 ± 15.1% GFP +/S100 β+ 11.2 ± 5.6% 11.8 ± 6.5% 22.1 ± 5.2% 6.9 ± 2.9% GFP +/DCX + 2.0 ± 1.0% -- -- 7.2 ± 2.4% GFP +/Olig2 + 5.0 ± 4.5% ------

110

any of the VZ precursor populations examined here serve as oligodendrocyte precursors, we first used an antibody against adenomatous polyposis coli (APC, CC1 clone) which labels the cell bodies of mature oligodendrocytes. No GFP/APC double-positive cells were found, so next we tried markers that label cells earlier in the oligodendrocyte lineage: the proteoglycan NG2, and the transcription factor Olig2. No GFP + cells co- localized with NG2 + cells, and only a few GFP + cells, all of NSC descent, were co- labeled with Olig2 ( Fig. 22M). Thus, our lineage analyses indicate that the four VZ precursor cell types investigated are all multipotent populations which contribute differently to the range of glial cells in the postnatal neocortex.

3.4 Discussion

In the previous chapter it was shown that at mid-neurogenesis the murine VZ contains two separate populations, gRGCs and SNPs, which differ in their cell cycle kinetics, modes of neuronal production and phenotype of neuronal progeny. The fate mapping studies here expand upon those findings, exploring the roles of multiple precursor cell types by labeling them at different points throughout neurogenesis and characterizing their progeny at P30. The results presented in this chapter further our understanding of the lineage properties of resident VZ precursor populations, and raise important questions about how various groups of neural precursors combine to facilitate

VZ output and cortical lamination.

The lineage analyses of the four precursor populations presented a number of surprising results: 1) SNPs generate glia; 2) SNPs do, while RGCs do not, produce DCX + neuroblasts; and 3) no mature oligodendrocytes at P30 were found to be derived from

111

these populations. The T α1 promoter is active in neuronal precursors and early

postmitotic neurons (Gloster et al., 1999), thus we hypothesized that SNPs would be

committed neuronal precursors. With this understood role, it was not entirely

confounding to find that SNPs produce neuroblasts. However, discovering that they also

generate GFAP + and S100 β+ glia was quite astounding. One possible explanation for this unexpected finding is that the T α1 promoter sequence in our plasmids—the 1.1-kb 5’ flanking region from the Tα1 gene (Gloster et al., 1999), a minimal region of the endogenous sequence—may label a broader population of precursors than the endogenous promoter.

On the basis that RGCs have previously been shown to generate SVZ astrocytes, more specifically type B cells (Kriegstein and Alvarez-Buylla, 2009), which are the main cell type responsible for producing type A neuroblasts in rodents (Doetsch et al., 1999), the absence of GFP +/DCX + cells in tissue electroporated with the GLAST- and Blbp-Cre plasmids was puzzling. Interestingly, numerous GFP + RGC progeny of similar size,

morphology and location compared to neuroblasts derived from NSCs and SNPs were

observed. Though we did not retain olfactory bulbs from brains sectioned for these

experiments, it would be instructive to examine bulbs in future experiments to determine

whether progeny from any of these precursor populations would become integrated as

olfactory interneurons.

Finally, based on genetic fate mapping experiments demonstrating the presence of

oligodendrocyte precursors and mature oligodendrocytes derived from Emx1 neocortical

progenitor cells at P9 (Kessaris et al., 2006), we expected similar results at P30.

Moreover, since all of the precursor cell types labeled in our experiments should be Emx1

112

lineage, it was rather perplexing that more GFP + progeny did not co-label with

oligodendrocyte lineage markers. It is possible that our IUE/Cre labeling technique was

too narrow—both temporally and in number of cells electroporated—to label sufficient

precursors in order to visualize a greater number of GFP + oligodendrocyte lineage cells.

Another possibility is that most dorsally-derived oligodendrocytes arise from a

(sub)population of precursors not investigated in this study. However unexpected the

combined results of the lineage analyses may have been initially, they do confirm the

necessity of multiple VZ precursor populations for generating the variety of dorsally-

derived cell types in the mature neocortex.

Based on our previous finding that SNPs and gRGCs labeled at E14.5 generate

neurons largely specified to different laminae, we had postulated that multiple precursor

cell types might be fundamental for producing the huge diversity of neocortical neurons.

The laminar allocation data presented in this study ( Fig. 18) show that, while neurons

generated from the four precursor populations investigated follow the same general

patterns, there may be subtle neuronal specification differences among the populations.

This was confirmed by comparing the entire cohorts of labeled progeny from the E12.5,

E14.5 and E16.5 precursor groups, which revealed striking differences in the dispersions

of these cohorts. The greatest differences were seen among progeny of E12.5 precursors.

The average depth of SNP progeny was significantly greater (that is, deeper) than those

from other VZ cell types. Furthermore, the bulk of SNP progeny were more closely

grouped than other progeny. At E14.5, the average depths of SNP- and bRGC-derived

neurons were significantly greater than those of NSC- and gRGC-derived neurons. Some

of the differences, both between groups and between ages, can be attributed to the

113

method of neuronal production utilized by specific precursor populations, as determined by co-labeling of electroporated cells with TUJ1 (direct production) or Tbr2 (indirect

production) ( Figs. 15, 20 ). The greater utilization of IPCs by SNPs at E12.5 compared to

E14.5 accounts for the wider distribution of the majority of neuronal progeny at E12.5.

Additionally, the lack of direct neuronal production by gRGCs at E12.5 helps explain the

relative paucity of deep layer gRGC progeny compared to that of SNPs. For future

studies, it would be useful to develop pBlbp- and pNestin-GFP plasmids in order to

perform similar Tbr2/TUJ1 analyses on the E12.5 and E14.5 bRGC and NSC

populations, respectively. This would offer a more complete picture of how the mode of

neuron production correlates to the observed neuronal distributions.

One question left open by these experiments is the presence and timing of lineage

restricted vs. multipotent progenitors in the rodent VZ. In vitro studies exploring cell fate

of neural stem cells found that cells isolated between E10 and E13 generated neurons

before glia (astrocytes or oligodendrocytes), but importantly, that both neurons and non-

neurons were produced by individual stem cells (Qian et al., 1998; Qian et al., 2000). We

show here that, when labeled at E16.5, NSCs, RGCs, and SNPs are all multipotent

populations. However, the same populations labeled at E12.5 and E14.5 produce only

neurons. This suggests a number of possibilities: 1) the potentiality of these precursor

populations may change, from neuronal restricted to multipotent, as cortical development

proceeds; 2) cells electroporated at E12.5 and E14.5, but not at E16.5, undergo terminal

division without generating glia; and 3) each precursor population is comprised by both

committed neuronal precursors and multipotent progenitors. Unfortunately, because of

the labeling technique used here, we cannot differentiate between the progeny of

114

individual precursor cells. It would be interesting, though technically difficult, to create a

Cre (or Cre-like) system in retroviruses, so that smaller numbers of precursors could be

labeled and fate-mapped at specific times during cortical development. By adjusting the

virus titer so that only a few VZ cells become infected, the density of labeled progeny

would be much reduced from what is seen using IUE, and identifying lineal relationships

might then be possible.

One of the most fascinating findings of this study was that deep neurons can be

generated from precursors at mid- and late-neurogenesis. At E14.5, all four precursor

cell types produced at least one neuron specified to layer V or VI. In the case of gRGCs

and SNPs there were multiple GFP + progeny allocated to these layers. At E16.5, only bRGCs did not produce any deep layer neurons. The relative consistency with which these neurons were observed suggests that they are not artifacts. Morphologically, these large pyramidal neurons appear to be developing normally. Indeed, even the observation of fewer and slightly smaller dendritic spines on E16.5-derived deep neurons compared to E12.5 derived neurons parallels findings in a recent paper from the Gage lab analyzing synaptogenesis during adult hippocampal neurogenesis (Toni et al., 2007).

Characterizing the functional properties—membrane and spiking properties and efferent targets—of deep cells produced from E12.5, E14.5 and E16.5 precursors would be an informative next step.

Interestingly, heterochronic transplantation studies showed that late- and even mid-stage VZ progenitors are incapable of generating layer VI neurons. Ferret VZ cells, isolated at time points when mid- (Desai and McConnell, 2000) or upper- (Frantz and

McConnell, 1996) cortical layers are normally generated, are unable to respond to

115

environmental cues when transplanted into younger hosts in which layer VI neurons are being produced. Indeed, transplanted upper-layer progenitors were observed to migrate through layer VI to the top of the cortical plate, where they maintained the elongated morphology of migrating neurons, and to keep “migrating” until such time as the host animal generated layer II/III neurons. Only then did the transplated cells adopt the morphology of differentiating neurons (Frantz and McConnell, 1996). How do we reconcile our findings with these previous studies that indicate that late-born neurons are intrinsically fated to upper layers? One major difference between this study and the transplantation experiments is that, with the exception of the IUE procedure, which has

been shown not to affect neuronal differentiation, migration, or functional properties

(Navarro-Quiroga et al., 2007), we are following VZ cells which divide and differentiate

unperturbed in their native environment Thus, the results presented here are more likely

to reflect what happens during normal cortical development, and may elucidate how

intrinsic factors can be overcome in order for signals in the endogenous milieu to

respecify upper layer neurons. This could occur via a molecular mechanism, for example over-expression of early progenitor genes, like Otx1 or Fezf2 (Chen et al., 2005; Frantz et al., 1994), in individual late precursors cells could result in their progeny acquiring the identity of deep layer neurons. Alternatively, respecification may take place through extrinsic means, such as cues along the migratory pathway which signal neurons to stop and differentiate in place, instead of reaching their “normal” upper layer destination.

Regardless of how it occurs, our data demonstrate that late VZ precursors generate deep layer neurons, suggesting a greater level of innate plasticity in the mammalian

116

neocortex—similar to the hippocampus, where newborn neurons are regularly integrated

into an established environment (Altman and Das, 1965)—than previously thought.

117

Chapter 4

Conclusive discussion and clinical perspectives

4.1 Summary of Results

The back-and-forth debate over whether the mammalian neocortical VZ is a homo- or heterogeneous germinal compartment was settled for primates nearly thirty years ago with the discovery of GFAP-positive and –negative cells, corresponding to glial and neuronal precursors, intermingled in the monkey VZ (Levitt et al., 1981, 1983).

The question remained unanswered, though, for rodents. Early retroviral lineage tracing studies suggested that multiple, lineage-restricted precursor populations most likely co- habited the rodent VZ (Grove et al., 1993; Luskin et al., 1988; McCarthy et al., 2001).

However, when time-lapse microscopy of individual radial glial cells showed (Noctor et al., 2001; Noctor et al., 2004), and genetic fate mapping studies further confirmed

(Anthony et al., 2004; Malatesta et al., 2003), that these cells generate neurons in addition to glia (Culican et al., 1990; Voigt, 1989), it became generally accepted that the rodent

VZ contained predominantly a single multipotent progenitor cell type. The work presented herein rejects that supposition and, building on recent studies which uncovered morphological (Gal et al., 2006), molecular (Mizutani et al., 2007) and genetic

(Kawaguchi et al., 2008) diversity among individual VZ cells, demonstrates that the mouse VZ is comprised by multiple precursor populations which differ in their cell cycle kinetics, modes of neuron production, phenotype of neuronal progeny, and lineage potentials.

118

In order to compare the cell cycle kinetics of two contiguously cycling VZ cell populations, we needed to devise new methods for tracking the cell cycle progression of

labeled progenitors in vivo . This was accomplished by first elucidating the mechanics of in utero electroporation (IUE): we found that the majority of cells which become transfected by IUE are in S- or M-phase at the moment of electroporation, but also that cells transiting through S- and M-phases within eight hours of the IUE procedure can also become transfected. We used this new understanding to develop protocols for estimating the length of G1-phase (T G1 ) and the cell cycle duration (T C) of electroporated cells.

+ Here we show that T G1 and T C of T α1 SNPs labeled at E14.5 are significantly different from those of GLAST + RGCs (gRGCs)—both are four hours longer in SNPs. Moreover, the biphasic labeling curves used to generate these data demonstrate that SNPs are not a transient VZ population, as some labeled cells remain proliferative in the VZ for at least two cell cycles after IUE. Taken together, these data definitively show that SNPs are a resident VZ population, along with gRGCs

Subsequent immunostaining and fate mapping experiments showed that E14.5 precursors differ in their neuronal output—both in the manner in which they produce neurons and in the laminar fate of those progeny. Twenty-four hours after IUE at E14.5, there were significantly more TUJ1 + SNPs than gRGCs, and more Tbr2 + gRGCs than

SNPs. This suggests that SNPs generate neurons directly from divisions in the VZ while gRGCs produce intermediate progenitor cells (IPCs), which amplify their neuronal progeny via divisions in the SVZ. This finding was further supported by comparing the laminar allocation of GFP + neuronal progeny of SNPs and gRGCs at P10. SNP progeny were primarily localized in layer IV while the majority of gRGC progeny were found in

119

the lower half of layer II/III. Thus, even though SNPs and gRGCs were labeled on the

same day (E14.5), the additional time required for IPC amplification of gRGC neuronal

progeny resulted in ther laminating superficial to SNP progeny. These data demonstrate

that precursor identity affects the location of terminal division, which, in turn, impacts

laminar position and phenotype of neuronal progeny.

In order to investigate whether these differences in lamination are unique to

progeny of E14.5 precursors, or if they extend to neurons produced over the course of

neurogenesis, we performed IUE at distinct phases of embryonic neurogenesis, E12.5

(early), E14.5 (mid) and E16.5 (late) and assessed the allocation of GFP + neurons

throughout the cortical wall at P30. In addition to SNPs and gRGCs, we examined the

progeny of another subset of the RGC population, Blbp + RGCs (bRGCs), as well as

Nestin + NSCs. The results were at once predictable and confounding. By organizing the

data as the percent of total GFP + neurons per layer for each precursor type at each age of

IUE, we found that all populations generally conform to the inside-out lamination process. That is, for each population, E12.5 precursors generate more deep- than upper- layer neurons, E14.5 precursors generate neurons found mostly in the mid and upper layers, and the vast majority of progeny of E16.5 precursors are located in the most superficial part of layer II/III. However, looking at individual GFP + neurons, the output

of each precursor population spanned the entire neocortical wall for each age of IUE.

These data indicate that E12.5 precursors remain proliferative through multiple rounds of

division, generating neurons from layers VI to II/III; although E14.5 and E16.5

precursors produce upper-layer neurons, they are also competent to generate deep-layer

neurons. Interestingly, the distributions of bRGC progeny were most similar to those of

120

SNP progeny, while the distributions of NSC and gRGC progeny more closely resembled each other. Significant differences in average neuronal depth among precursor types allude to differences in proliferation capacity and method of neuron production (direct vs. indirect), though we did not fully explore these possibilities for each of the four precursor types.

Lineage analyses were performed on tissue electroporated at E16.5 and harvested at P30, because cells characterized as glia based on their morphology and immunohistochemical profiles were only seen at this (start) time point. Each of the four precursor types produced GFP + protoplasmic astrocytes; often the cell bodies were

S100 β+, and occasionally the processes were GFAP +. GFP +/S100 β+ ependymal cells were seen lining the lateral ventricles in tissue electroporated with each of the four cell type-specific Cre plasmids. Moreover, GFP +/GFAP + glia, not just protoplasmic astrocytes, were found in the white matter tracts and/or SVZ of tissue from all precursor types. Thus, NSCs, gRGCs, bRGCs and SNPs are all astroglial precursors at E16.5.

Although no GFP + mature oligodendrocytes (APC +) were detected, a few oligodendrocyte lineage cells (GFP/Olig2 double-positive) were identified, all of which were generated from NSCs. Therefore NSCs are the only oligodendrocyte precursors, at least at E16.5, from the cell types examined in this study. Finally, in order to characterize numerous GFP + cells which appeared migrating away from the SVZ, we tested for DCX expression by immunohistochemistry to investigate whether these cells were neuroblasts—the migrating neuronal precursors which take part in postnatal neurogenesis in the olfactory bulbs. We found that only SNPs and NSCs produced GFP + cells that co- localized with DCX staining; GFP + progeny of gRGCs and bRGCs were similarly

121

localized with chains of DCX + cells without being DCX + themselves. Thus, based on our

lineage analyses, the RGC populations are bipotent neuronal and astroglial precursors;

SNPs are multipotent neuronal, astroglial and SVZ neuroblast precursors; and finally

NSCs are multipotent neuronal, astro- and oligodendroglial and SVZ neuroblast

precursors.

All together, the data presented here establish that the rodent neocortical VZ

contains numerous resident precursor populations and subpopulations, each of which

plays a nuanced role in neocorticogenesis by contributing to specific neuronal and glial

phenotypes at different times during development ( Fig. 23 ). In the next sections I will discuss the historic and evolutionary implications, as well as the clinical relevance, of these findings.

4.2 Contributions to and impact on the field of developmental neurobiology

4.2.1 Technical advances

The experiments performed for my doctoral research have yielded a number of significant technical advances. In seeking to develop methods for tracking the proliferation kinetics of multiple, concomitantly-dividing populations, we uncovered the mechanism of transfection by in vivo electroporation. Transfection, introducing viral or

bacterial DNA that becomes incorporated into the host cell’s DNA, is one of the most

common methods of gene delivery. Initially, transfecting cells by electroporation, in

which an electric field is applied to the cell to permeabilize the membrane (Neumann et

al., 1982), was solely performed on cultured cells. The first published report of “ in vivo ”

electroporation in the mouse actually involved electroporating cultured embryos

122

Figure 23. Proposed model of neocorticogenesis. Based on the data presented in

Chapters 2 and 3, we propose the following model of neocorticogenesis in the mouse.

The embryonic VZ (left) houses at least four subpopulations of neural precursor cells:

RGCs which are labeled using plasmids with GLAST (yellow; gRGCs) or Blbp (blue; bRGCs) promoters; SNPs (red) which are labeled with T α1 promoter-plasmids; and

NSCs (green) which are labeled using plasmids with the Nestin promoter. These

precursor populations divide within the VZ throughout neurogenesis, producing both

intermediate progenitor cells (IPCs; SVZ) and migrating, post-mitotic neurons (IZ).

Immunohistochemical analyses demonstrated that gRGCs generated primarily IPCs

whereas SNPs produced mostly post-mitotic neurons. Thus we show more yellow IPCs

and more red migrating neurons in this schematic. Using Cre/lox fate mapping of

precursors labeled embryonically, we found that SNPs and bRGCs generated more deep-

layer (V and VI) neurons compared to gRGCs and NSCs, but that all four populations

produced upper-layer (VI and II/III) neurons ( Figs. 18, 19 ). Therefore we show more red and blue neurons in deep layers, and roughly even numbers of neurons from all four precursor populations in the upper layers. Moreover, we found that all four populations produced GFAP + and/or S100 β+ protoplasmic astrocytes, SNPs and NSCs generated

DCX + migrating neuroblasts, and only NSCs generated Olig2 + oligodendrocyte precursor

cells. These data demonstrate that multiple VZ precursor populations work in concert to

generate the dorsally-derived cells which comprise the mature neocortex.

123

124

(Akamatsu et al., 1999), but within a few years the technique had progressed to electroporating in utero (Saito and Nakatsuji, 2001; Tabata and Nakajima, 2001). The

use of in utero electroporation (IUE) in developmental neuroscience has grown

exponentially in the decade since the procedure was first described, but not until this

study was the method of transfection truly understood: electroporated cells must pass

through M-phase in order to take up and express the exogenous DNA. Moreover, we

showed that this must happen within 8 hours of the electroporation procedure because of

limited plasmid viability in vivo .

This newfound understanding of the mechanics of IUE enabled us to develop

methods for analyzing cell cycle kinetics of specific populations of dividing cells in vivo ,

using a combination of IUE, BrdU injections, and condensed chromatin staining to

estimate the durations of G1-phase and the entire cell cycle. Earlier methods for cell

cycle analysis involved cumulative labeling of cells with thymidine analogs, such as

tritiated ([ 3H]) thymidine or BrdU, which become incorporated into DNA during S-phase

(Hughes et al., 1958; Miller and Nowakowski, 1988). Though powerful, these methods

do not provide the specificity necessary for analyzing heterogeneous cell populations, as

they label all dividing cells uniformly. Thus, previous estimates of cell cycle kinetics in

the rodent VZ represent averages of multiple populations (Cai et al., 1997; Takahashi et

al., 1995). The application of our IUE/BrdU and IUE/M-M-phase methods allowed us,

for the first time, to detect with confidence and reproducibility differences in the cell

cycle properties of two populations co-habiting the mammalian VZ. These methods will

be invaluable for future analyses of heterogeneous populations, for example in the

developing amygdala, striatum, or cerebellum.

125

Genetic fate mapping is a powerful and popular tool whereby the progeny of cells expressing a specific gene are indelibly marked with a reporter molecule, such as GFP or

β-galactosidase ( Fig. 24 ). Mating transgenic Cre animals, in which a gene of interest drives Cre expression, with reporter animals, in which a STOP sequence is inserted between two loxP sites upstream of a reporter gene, to achieve Cre/lox recombination is currently one of the most widely used means of genetic fate mapping (Sauer, 1998).

Inducible Cre systems, in which tamoxifen-sensitive estrogen receptor ligand binding domains are fused with Cre (i.e. CreER, CreER T, etc.), allow greater temporal control of labeling than in straight Cre systems (Joyner and Zervas, 2006). However, using IUE to induce recombination enables even tighter temporal control, as progeny of roughly an 8- hour cohort of precursors will be labeled via IUE ( Fig. 9D, Chapter 2 ), whereas tamoxifen persists up to 36 hours in vivo (Joyner and Zervas, 2006). Furthermore, IUE- mediated recombination can be targeted to specific regions of the cortex. Thus, comparing output from GLAST-expressing precursors in the subpallial and neocortical germinal zones, for example, becomes a matter simply of switching the orientation of the paddles during electroporation.

126

Figure 24. Overview of Cre/lox recombination. Schematic depicting the genetic

machinery and conditions necessary in order for recombination to occur. When Cre is

present, the LoxP sites recombine, resulting in the excision of the intervening DNA. In

this example, a Stop sequence is removed leading to constitutive expression of the

reporter protein GFP. Modified from Pechisker, A. 2007-8. The Science Creative

Quarterly, Is. 3.

127

128

4.2.2 Ideological advances

Although development of new methods for estimating cell cycle phases is a

considerable technical advance, the outcome of applying these methods—demonstration

of the existence of two separate precursor populations within the mouse VZ— represents

a significant paradigm shift in how we think about the rodent VZ. For nearly a decade,

radial glial cells have been held as the predominant, even the sole, precursor type in the

rodent VZ (Kriegstein et al., 2006; Noctor et al., 2001; Noctor et al., 2002; Pinto and

Gotz, 2007). Any diversity demonstrated among VZ cells was passed off simply as

diversity among radial glia (Hartfuss et al., 2003; Pinto and Gotz, 2007; Pinto et al.,

2008), and attempts to show otherwise were met with skepticism. Here, by uncovering

significantly different cell cycle kinetics among VZ populations which had previously

been shown to differ in promoter expression, Notch signaling activity, and morphology

(Gal et al., 2006; Mizutani et al., 2007), we have effectively nullified the argument that

RGCs alone comprise the rodent VZ. Henceforth, the rodent VZ must be treated as a

heterogeneous germinal compartment, with future analyses taking into consideration the

different proliferative properties and lineage potentials of RGCs, SNPs and NSCs.

Results from the laminar specification analyses underscore the differences in

neuronal output between VZ populations, as significant differences were observed in the

distributions of neuronal progeny of the four populations (NSCs, gRGCs, bRGCs and

SNPs) and three labeling timepoints (E12.5, E14.5 and E16.5) examined. Contributing to

the overall differences in distributions are specific findings which challenge long-

standing doctrines concerning cortical lamination. The inside-out order of lamination,

with first-born neurons residing deep in the cortical plate and later-born neurons

129

migrating past them to settle superficially, has been firmly established in seminal studies

utilizing labeling methods which were state-of-the-art at the times they were performed:

[3H]-thymidine incorporation (Angevine and Sidman, 1961; Rakic, 1974) and a

combination of [ 3H]-thymidine and BrdU labeling (Takahashi et al., 1999). However, as

described earlier, these methods label all dividing cells indiscriminately. By tracking the

neuronal progeny of individual precursor populations via IUE-mediated Cre/lox

recombination, we uncovered previously unknown complexity in the lamination process

in rodents. For example, GLAST + RGCs (gRGCs) labeled at E12.5 and E14.5 allocate a

greater percentage of their progeny to superficial layers (II/III and IV) than do Blbp +

RGCs (bRGCs) or SNPs labeled at the same times. This suggests that mature neuronal

laminar position is a function not only of birthdate, but also of precursor cell type of

origin. Moreover, Tbr2 immunostaining showed that gRGCs rely more heavily on IPCs

for neuronal output than do SNPs, suggesting that the location of terminal division (VZ

or SVZ) is also important for conveying laminar identity. The finding that precursors are

able to generate neurons “out of order,” that is, that precursors labeled late in

neurogenesis (E16.5) produce neurons fated to deep layers, further muddles the inside-

out lamination dogma, and provides an exception to the reported restriction of neuronal

fate over the course of cortical development (Desai and McConnell, 2000; Frantz and

McConnell, 1996; Shen et al., 2006). The results presented here depict a more plastic,

combinatorial lamination process than previously described, and suggest either that

neocorticogenesis in rodents is fundamentally different from that in primates and higher

mammals, or more likely that, because of the labeling techniques used, previous studies

130

have not fully characterized the intricacy of the relationship between VZ cellular output and neuronal specification.

4.3 Making sense of heterogeneity in the rodent VZ

4.3.1 Significance of multiple VZ precursor populations in rodents

In this dissertation we demonstrate that the murine VZ contains multiple precursor cell types and subtypes, similar to the VZ of humans and non-human primates

(Howard et al., 2006; Levitt et al., 1981, 1983; Mo et al., 2007; Zecevic, 2004).

Interestingly, all four precursor populations examined here produced neurons throughout neurogenesis, in addition to generating different classes of glial cells on or after E16.5.

However, only one population, NSCs, produced a glial cell type (oligodendrocyte precursor cells) not produced by another precursor population. What could be the advantage of having multiple populations, which seem to perform nearly redundant roles?

The slight differences in laminar specification at P30 among the four precursor

populations labeled at all three time points considered suggests there may be inherent

differences in the neuronal progeny of these populations. The finding that neuronal

progeny of SNPs and gRGCs labeled at E14.5 largely occupied different laminae at P10

prompted an examination of GFP + neuronal morphology. Both precursor populations

generated both pyramidal and stellate neurons, thus we concluded that these populations

produced a similar variety of neuronal subtypes. However, it is possible that there are

differences among neurons within the subtypes that we did not detect. In an elegant and

comprehensive gene expression analysis, the Macklis lab recently identified sets of genes

expressed by callosal projection neurons (CPN; (Molyneaux et al., 2009)). Some genes

131

were expressed in a laminar-specific manner, while others identified subpopulations of

CPN within the same laminae. Interestingly, a number of genes subparceled the layer

II/III, labeling only neurons in the upper, middle or lower portion of the layer

(Molyneaux et al., 2009). This separation is similar to what we observed for the neuronal

progeny of E16.5 precursors: NSCs, bRGCs and SNPs produced neurons found in the

most superficial part of layer II/III while gRGCs generated neurons specified to the

middle and lower regions. Though morphologically similar, it is likely that these

subpopulations of projection neurons are molecularly and functionally distinct. Thus, the

necessity for multiple VZ precursor populations may be an evolutionary response to the

increased diversity of CPN from lower vertebrates to mammals.

Alternatively, the presumed redundancy of the precursor cell types examined, as

pertains to their neuron generating capabilities, may be just that. It is possible that these

“separate but equal” VZ populations evolved as a failsafe, to ensure that cortical

development could proceed in the face of defect(s) in one population which prevent it

from functioning normally. Examples making this possibility seem more likely are given

in section 4.4 .

4.3.2 Potential mechanism for generating heterogeneity among VZ precursors

Notch proteins, which are expressed in the mammalian VZ, are transmembrane

receptors that enable cell-to-cell signaling and mediate numerous cellular functions. In

mammals there are four Notch receptors (Notch1-4) and five ligands (Delta-like1, 3 and 4

and Jagged1 and 2) (Yoon and Gaiano, 2005). Activation via ligand binding causes

proteolytic cleavage of Notch receptor(s) by γ-secretase/presenilin-1, followed by release

132

of the Notch intracellular domain (NICD). NICD then translocates to the nucleus and complexes with the DNA-binding protein C promoter-binding factor 1 (CBF1) (Selkoe

and Kopan, 2003). The NICD/CBF1 complex activates an array of genes including the

Hes family, notably Hes1 (Iso et al., 2003), which suppress neurogenesis by antagonizing the proneural basic helix-loop-helix transcription factors Mash1 and the Neurogenins, among others (Yoon and Gaiano, 2005). Thus, Notch activation inhibits neurogenesis and maintains a progenitor state during cortical development.

Additional studies show that Notch signaling may also mediate functions in a

CBF1-independent manner. One such study examined progenitor pools, and the progeny they generated, in the lateral ganglionic eminence in various strains of mutant mice (Yun et al., 2002). It was determined that three lineally-related progenitor populations were present: P1 and P2 in the VZ, and P3 in the SVZ. Progression from one subtype to the next (i.e. P1 P2) was inhibited by Notch activation, consistent with its role in maintaining a progenitor state. However, P2 cells expressed Mash1, contrary to the normal repression of proneural transcription factors via Hes activation (Yun et al., 2002).

Thus, in the presence of active Notch signaling, “traditional” outcomes were effected via non-traditional means.

As mentioned above, it was recently found that Notch/CBF1 activity is not uniform across progenitor cells of the neocortical VZ, even though all cells respond to

Notch receptor activation (Mizutani et al., 2007). The lower Notch/CBF1 activity in

SNPs compared to RGCs is caused by a block in signal transduction. Similar to the Yun et al. study (2002), differentiation of both RGCs and SNPs was inhibited by receptor activation, but occurred through different mechanisms (Mizutani et al., 2007). Together,

133

these studies and many others (Basak and Taylor, 2007; Mason et al., 2005; Ohtsuka et al., 2006) demonstrate that the Notch signaling pathway may present a means of generating heterogeneity among progenitor populations.

4.4 Relevance of results to neurodevelopmental disorders

Neurodevelopmental disorders affect millions of children every year and lead to a

wide array of cognitive impairments including deficits in language, learning and memory.

The broad category “neurodevelopmental disorders” covers dysfunctions which result

from factors ranging from genetic mutations to environmental toxins to poor maternal

nutrition, and which manifest over a wide time span, from early in gestation to early

childhood in humans. The findings presented in this dissertation, however, are most

germane to disorders presenting with congenital neocortical malformation(s) resulting

from aberrations and/or deficits in neural precursor proliferation or migration.

Down syndrome (DS), for instance, is a neurodevelopmental disorder caused by

triplication of chromosome 21 (HSA21) and occurs in 1 in every 700-800 live births per

year (Contestabile et al., 2010). DS is the most common genetic cause of mental

retardation, with significant impairments in language, learning and memory (Contestabile

et al., 2010). Reduced volume and neuron number in areas associated with these

functions—the hippocampus and neocortex—have been observed in DS fetuses

(Contestabile et al., 2010). The Ts65Dn mouse model of DS is trisomic for the distal

portion of mouse chromosome 16 (MMU16), the region most syntenous to HSA21

(Gardiner et al., 2003). Postnatal similarities between Ts65Dn mice and DS individuals

include reduced hippocampal and cerebellar volumes, behavioral and cognitive deficits,

134

and decreased dendritic arborization and spine density in hippocampal and neocortical pyramidal neurons (Contestabile et al., 2010). Recent studies from our lab demonstrated

developmental correlates for these postnatal deficits in Ts65Dn mice (Chakrabarti et al.,

2007). In the embryonic neocortex there is reduced thickness of the intermediate zone

and cortical plate from E13.5-E16.5 which disappears by E18.5, at which point cortical

thickness is no different from controls. These findings are paralleled postnatally, with

decreases in cell density in layers VI-IV but not in layer II/III. Increased cell cycle

duration in the VZ at E13.5 diminished by E16.5. Furthermore, an increase in Tbr2 +

IPCs in the SVZ at E15.5 and E16.5, accompanied by a substantial increase in SVZ

mitoses, was observed in the Ts65Dn brains compared to controls (Chakrabarti et al.,

2007). In light of our discovery that SNPs and RGCs together comprise the embryonic

neocortical VZ, it is possible that proliferation defects in either of these populations could

account for hypocellularity of the cortical plate embryonically and the deep layers

postnatally. However, based on our findings that RGCs have a shorter cell cycle than

SNPs and preferentially utilize IPCs for neuronal production, it appears that the observed

inconsistencies in neurogenesis and postnatal laminar density in Ts65Dn may result

primarily from defects in the RGC population. For example, delayed development of the

RGC population, resulting in fewer RGCs early in neurogenesis, would explain the early

decreases in Tbr2 + IPCs and cortical plate thickness, as well as the subsequent increases

in Tbr2 + and mitotic IPCs and concomitant thickening of the cortical plate.

Microcephaly, unlike DS, is not a specific disorder, but rather a clinical

description of head circumference which is significantly less than the average for age and

sex (Mochida, 2009). Though there are multiple causes and wide ranges of severity

135

among microcephaly cases, the prototype syndrome is autosomal recessive primary

microcephaly (MCPH), also known as microcephaly vera and primary microcephaly, a

congenital disorder characterized by 1) head circumference ≥ 4 standard deviations below the mean; 2) mental retardation in the absence of other neurological findings such as seizures or spasticity; and 3) normal height, weight and chromosome analysis (Woods et al., 2005). Despite significantly smaller brain size, brain function and architecture are normal in most MCPH patients (Woods et al., 2005). Seven loci associated with MCPH have been mapped, MCPH1-7, and from these loci 5 genes have been identified

(Mochida, 2009). The most common cause of MCPH is mutations to the ASPM gene

(mammalian homolog of Drosophila abnormal spindle (asp) gene; (Kouprina et al.,

2005)). In mice, Aspm mRNA is expressed embryonically in the neocortical VZ at

E14.5, with expression levels decreasing as neurogenesis proceeds (Kouprina et al.,

2005), and Aspm protein is localized to the spindle poles of mitotic cells and to the

midbody of cells during late cytokinesis (Paramasivam et al., 2007). Similar

localizations have been observed for fly and human mRNA and proteins (Kouprina et al.,

2005), suggesting that ASPM serves an evolutionarily conserved role in mitotic spindle

function in the developing neocortex. Indeed, using siRNA to knockdown Aspm in the

embryonic mouse telencephalon, Fish et al. (2006) found that lack of Aspm did not

prevent VZ cells from dividing, but did increase the percent of asymmetric divisions

(defined as cleavage plane not perpendicular to the ventricular surface) compared to

control cells. Neural precursor cells in which Aspm was knocked down were more likely

to undergo neurogenic than proliferative divisions in the VZ, resulting in an increase in

young neurons in the cortical plate compared to control conditions (Fish et al., 2006).

136

Taking these results in context with our findings, this suggests that, in the absence of

Aspm in the developing VZ, RGCs adopt the proliferative profile of SNPs, generating neurons directly from the VZ. This would likely deplete the VZ population sooner, as well as decrease the total number of neurons generated from VZ precursors due to a lack of IPC amplification, ultimately resulting in the smaller—though architechtonically normal—cerebral cortex characteristic of microcephalic individuals.

These examples typify how studying the processes involved in normal neocorticogenesis can better our understanding of deviations from these processes in mouse models of disease. The undeniable demonstration of rodent VZ heterogeneity will change both how we interpret previous results and the ways in which we move forward investigating neurodevelopmental disorders. Certainly, caution should be exercised in extrapolating mechanisms of cortical malformation in humans based on those in mice.

However, the findings presented in this dissertation underscore basic similarities in cortical development between rodents and humans, namely, that multiple VZ precursor populations are fundamental to normal development and function of the mature neocortex in both species.

137

References

(1970). Embryonic vertebrate central nervous system: revised terminology. The Boulder Committee. Anat Rec 166 , 257-261.

Akamatsu, W., Okano, H.J., Osumi, N., Inoue, T., Nakamura, S., Sakakibara, S., Miura, M., Matsuo, N., Darnell, R.B., 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.

Allen, N.J., and Barres, B.A. (2009). Neuroscience: Glia - more than just brain glue. Nature 457 , 675-677.

Altman, J., and Das, G.D. (1965). Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats. J Comp Neurol 124 , 319-335.

Anderson, S.A., Marin, O., Horn, C., Jennings, K., and Rubenstein, J.L. (2001). Distinct cortical migrations from the medial and lateral ganglionic eminences. Development 128 , 353-363.

Angevine, J.B., Jr., and Sidman, R.L. (1961). Autoradiographic study of cell migration during histogenesis of cerebral cortex in the mouse. Nature 192 , 766-768.

Anthony, T.E., 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.

Arnold, S.J., Huang, G.J., Cheung, A.F., Era, T., Nishikawa, S., Bikoff, E.K., Molnar, Z., Robertson, E.J., and Groszer, M. (2008). The T-box transcription factor Eomes/Tbr2 regulates neurogenesis in the cortical subventricular zone. Genes Dev 22 , 2479-2484.

Basak, O., and Taylor, V. (2007). Identification of self-replicating multipotent progenitors in the embryonic nervous system by high Notch activity and Hes5 expression. Eur J Neurosci 25 , 1006-1022.

Bourgeois, J.P., and Rakic, P. (1993). Changes of synaptic density in the primary visual cortex of the macaque monkey from fetal to adult stage. J Neurosci 13 , 2801-2820.

Brodmann, K. (1909). Vergleichende Lokalisationslehre der Grosshirnrinde in irhen Prinzipien dargestellt auf Grund des Zellenbaues. Leipzig: Berlin.

Butt, S.J., Fuccillo, M., Nery, S., Noctor, S., Kriegstein, A., Corbin, J.G., and Fishell, G. (2005). The temporal and spatial origins of cortical interneurons predict their physiological subtype. Neuron 48 , 591-604.

138

Cai, L., Hayes, N.L., and Nowakowski, R.S. (1997). Local homogeneity of cell cycle length in developing mouse cortex. J Neurosci 17 , 2079-2087.

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

Cappello, S., Attardo, A., Wu, X., Iwasato, T., Itohara, S., Wilsch-Brauninger, M., Eilken, H.M., Rieger, M.A., Schroeder, T.T., Huttner, W.B. , et al. (2006). The Rho- GTPase cdc42 regulates neural progenitor fate at the apical surface. Nat Neurosci 9, 1099-1107.

Casper, K.B., and McCarthy, K.D. (2006). GFAP-positive progenitor cells produce neurons and oligodendrocytes throughout the CNS. Mol Cell Neurosci 31 , 676-684.

Caviness, V.S., Jr. (1982). Neocortical histogenesis in normal and reeler mice: a developmental study based upon [3H]thymidine autoradiography. Brain Res 256 , 293- 302.

Caviness, V.S., Jr., and Sidman, R.L. (1973). Time of origin or corresponding cell classes in the cerebral cortex of normal and reeler mutant mice: an autoradiographic analysis. J Comp Neurol 148 , 141-151.

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

Chen, B., Schaevitz, L.R., and McConnell, S.K. (2005). Fezl regulates the differentiation and axon targeting of layer 5 subcortical projection neurons in cerebral cortex. Proc Natl Acad Sci U S A 102 , 17184-17189.

Chen, L., Liao, G., Yang, L., Campbell, K., Nakafuku, M., Kuan, C.Y., and Zheng, Y. (2006). Cdc42 deficiency causes Sonic hedgehog-independent holoprosencephaly. Proc Natl Acad Sci U S A 103 , 16520-16525.

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

Chun, J.J., and Shatz, C.J. (1989). Interstitial cells of the adult neocortical white matter are the remnant of the early generated subplate neuron population. J Comp Neurol 282 , 555-569.

Contestabile, A., Benfenati, F., and Gasparini, L. (2010). Communication breaks-Down: from neurodevelopment defects to cognitive disabilities in Down syndrome. Prog Neurobiol 91 , 1-22.

139

Corbin, J.G., Nery, S., and Fishell, G. (2001). Telencephalic cells take a tangent: non- radial migration in the mammalian forebrain. Nat Neurosci 4 Suppl , 1177-1182.

Culican, S.M., Baumrind, N.L., Yamamoto, M., and Pearlman, A.L. (1990). Cortical radial glia: identification in tissue culture and evidence for their transformation to astrocytes. J Neurosci 10 , 684-692.

D'Arcangelo, G., Miao, G.G., Chen, S.C., Soares, H.D., Morgan, J.I., and Curran, T. (1995). A protein related to extracellular matrix proteins deleted in the mouse mutant reeler. Nature 374 , 719-723.

D'Arcangelo, G., Nakajima, K., Miyata, T., Ogawa, M., Mikoshiba, K., and Curran, T. (1997). Reelin is a secreted glycoprotein recognized by the CR-50 monoclonal antibody. J Neurosci 17 , 23-31.

DeFelipe, J., Alonso-Nanclares, L., and Arellano, J.I. (2002). Microstructure of the neocortex: comparative aspects. J Neurocytol 31 , 299-316.

Derer, P. (1985). Comparative localization of Cajal-Retzius cells in the neocortex of normal and reeler mutant mice fetuses. Neurosci Lett 54 , 1-6.

Derer, P., and Derer, M. (1990). Cajal-Retzius cell ontogenesis and death in mouse brain visualized with horseradish peroxidase and electron microscopy. Neuroscience 36 , 839- 856.

Desai, A.R., and McConnell, S.K. (2000). Progressive restriction in fate potential by neural progenitors during cerebral cortical development. Development 127 , 2863-2872.

Doetsch, F., Caille, I., Lim, D.A., Garcia-Verdugo, J.M., and Alvarez-Buylla, A. (1999). Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell 97 , 703-716.

Englund, C., Fink, A., Lau, C., Pham, D., Daza, R.A., Bulfone, A., Kowalczyk, T., and Hevner, R.F. (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.

Fietz, S.A., Kelava, I., Vogt, J., Wilsch-Brauninger, M., Stenzel, D., Fish, J.L., Corbeil, D., Riehn, A., Distler, W., Nitsch, R., and Huttner, W.B. (2010). OSVZ progenitors of human and ferret neocortex are epithelial-like and expand by integrin signaling. Nat Neurosci 13 , 690-699.

140

Fish, J.L., Kosodo, Y., Enard, W., Paabo, S., and Huttner, W.B. (2006). Aspm specifically maintains symmetric proliferative divisions of neuroepithelial cells. Proc Natl Acad Sci U S A 103 , 10438-10443.

Forster, E., Bock, H.H., Herz, J., Chai, X., Frotscher, M., and Zhao, S. (2010). Emerging topics in Reelin function. Eur J Neurosci 31 , 1511-1518.

Frantz, G.D., and McConnell, S.K. (1996). Restriction of late cerebral cortical progenitors to an upper-layer fate. Neuron 17 , 55-61.

Frantz, G.D., Weimann, J.M., Levin, M.E., and McConnell, S.K. (1994). Otx1 and Otx2 define layers and regions in developing cerebral cortex and cerebellum. J Neurosci 14 , 5725-5740.

Fujita, S. (1962). Kinetics of cellular proliferation. Exp Cell Res 28 , 52-60.

Fujita, S. (1963). The matrix cell and cytogenesis in the developing central nervous system. J Comp Neurol 120 , 37-42.

Gaiano, N., and Fishell, G. (2002). The role of notch in promoting glial and neural stem cell fates. Annu Rev Neurosci 25 , 471-490.

Gal, J.S., Morozov, Y.M., Ayoub, A.E., Chatterjee, M., Rakic, P., and Haydar, T.F. (2006). Molecular and morphological heterogeneity of neural precursors in the mouse neocortical proliferative zones. J Neurosci 26 , 1045-1056.

Gardiner, K., Fortna, A., Bechtel, L., and Davisson, M.T. (2003). Mouse models of Down syndrome: how useful can they be? Comparison of the gene content of human chromosome 21 with orthologous mouse genomic regions. Gene 318 , 137-147.

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

Gorski, J.A., Talley, T., Qiu, M., Puelles, L., Rubenstein, J.L., and Jones, K.R. (2002). Cortical excitatory neurons and glia, but not GABAergic neurons, are produced in the Emx1-expressing lineage. J Neurosci 22 , 6309-6314.

Gotz, M., and Huttner, W.B. (2005). The cell biology of neurogenesis. Nat Rev Mol Cell Biol 6, 777-788.

Gotz, M., Stoykova, A., and Gruss, P. (1998). Pax6 controls radial glia differentiation in the cerebral cortex. Neuron 21 , 1031-1044.

141

Grove, E.A., Williams, B.P., Li, D.Q., Hajihosseini, M., Friedrich, A., and Price, J. (1993). Multiple restricted lineages in the embryonic rat cerebral cortex. Development 117 , 553-561.

Hansen, D.V., Lui, J.H., Parker, P.R., and Kriegstein, A.R. (2010). Neurogenic radial glia in the outer subventricular zone of human neocortex. Nature 464 , 554-561.

Hartfuss, E., Forster, E., Bock, H.H., Hack, M.A., Leprince, P., Luque, J.M., Herz, J., Frotscher, M., and Gotz, M. (2003). Reelin signaling directly affects radial glia morphology and biochemical maturation. Development 130 , 4597-4609.

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

Hatten, M.E., and Mason, C.A. (1990). Mechanisms of glial-guided neuronal migration in vitro and in vivo. Experientia 46 , 907-916.

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

His, W. (1889). Die Neuroblasten und deren Entstehung im embryonal Marke. Abh Math Phys Cl Kgl Sach Ges Wiss 15 , 313-372.

Howard, B., Chen, Y., and Zecevic, N. (2006). Cortical progenitor cells in the developing human telencephalon. Glia 53 , 57-66.

Howell, B.W., Gertler, F.B., and Cooper, J.A. (1997). Mouse disabled (mDab1): a Src binding protein implicated in neuronal development. EMBO J 16 , 121-132.

Hughes, W.L., Bond, V.P., Brecher, G., Cronkite, E.P., Painter, R.B., Quastler, H., and Sherman, F.G. (1958). Cellular Proliferation in the Mouse as Revealed by Autoradiography with Tritiated Thymidine. Proc Natl Acad Sci U S A 44 , 476-483.

Iso, T., Kedes, L., and Hamamori, Y. (2003). HES and HERP families: multiple effectors of the Notch signaling pathway. J Cell Physiol 194 , 237-255.

Johnston, M.V. (2004). Clinical disorders of brain plasticity. Brain Dev 26 , 73-80.

Joyner, A.L., and Zervas, M. (2006). Genetic inducible fate mapping in mouse: establishing genetic lineages and defining genetic neuroanatomy in the nervous system. Dev Dyn 235 , 2376-2385.

142

Kawaguchi, A., Ikawa, T., Kasukawa, T., Ueda, H.R., Kurimoto, K., Saitou, M., and Matsuzaki, F. (2008). Single-cell gene profiling defines differential progenitor subclasses in mammalian neurogenesis. Development 135 , 3113-3124.

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

Konig, N., Valat, J., Fulcrand, J., and Marty, R. (1977). The time of origin of Cajal- Retzius cells in the rat temporal cortex. An autoradiographic study. Neurosci Lett 4, 21- 26.

Kouprina, N., Pavlicek, A., Collins, N.K., Nakano, M., Noskov, V.N., Ohzeki, J., Mochida, G.H., Risinger, J.I., Goldsmith, P., Gunsior, M. , et al. (2005). The microcephaly ASPM gene is expressed in proliferating tissues and encodes for a mitotic spindle protein. Hum Mol Genet 14 , 2155-2165.

Kowalczyk, T., Pontious, A., Englund, C., Daza, R.A., Bedogni, F., Hodge, R., Attardo, A., Bell, C., Huttner, W.B., and Hevner, R.F. (2009). Intermediate Neuronal Progenitors (Basal Progenitors) Produce Pyramidal-Projection Neurons for All Layers of Cerebral Cortex. Cereb Cortex 19 , 12.

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

Kriegstein, A., Noctor, S., and Martinez-Cerdeno, V. (2006). Patterns of neural stem and progenitor cell division may underlie evolutionary cortical expansion. Nat Rev Neurosci 7, 883-890.

Kuida, K., Haydar, T.F., Kuan, C.Y., Gu, Y., Taya, C., Karasuyama, H., Su, M.S., Rakic, P., and Flavell, R.A. (1998). Reduced apoptosis and cytochrome c-mediated caspase activation in mice lacking caspase 9. Cell 94 , 325-337.

Kuida, K., Zheng, T.S., Na, S., Kuan, C., Yang, D., Karasuyama, H., Rakic, P., and Flavell, R.A. (1996). Decreased apoptosis in the brain and premature lethality in CPP32- deficient mice. Nature 384 , 368-372.

Land, P.W., and Monaghan, A.P. (2003). Expression of the transcription factor, tailless, is required for formation of superficial cortical layers. Cereb Cortex 13 , 921-931.

Letinic, K., Zoncu, R., and Rakic, P. (2002). Origin of GABAergic neurons in the human neocortex. Nature 417 , 645-649.

143

Levitt, P., Cooper, M.L., and Rakic, P. (1981). Coexistence of neuronal and glial precursor cells in the cerebral ventricular zone of the fetal monkey: an ultrastructural immunoperoxidase analysis. J Neurosci 1, 27-39.

Levitt, P., Cooper, M.L., and Rakic, P. (1983). Early divergence and changing proportions of neuronal and glial precursor cells in the primate cerebral ventricular zone. Dev Biol 96 , 472-484.

Lukaszewicz, A., Cortay, V., Giroud, P., Berland, M., Smart, I., Kennedy, H., and Dehay, C. (2006). The concerted modulation of proliferation and migration contributes to the specification of the cytoarchitecture and dimensions of cortical areas. Cereb Cortex 16 Suppl 1 , i26-34.

Luskin, M.B., Pearlman, A.L., and Sanes, J.R. (1988). Cell lineage in the cerebral cortex of the mouse studied in vivo and in vitro with a recombinant retrovirus. Neuron 1, 635- 647.

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

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

Marin-Padilla, M. (1978). Dual origin of the mammalian neocortex and evolution of the cortical plate. Anat Embryol (Berl) 152 , 109-126.

Martinez-Cerdeno, V., Noctor, S.C., and Kriegstein, A.R. (2006). The role of intermediate progenitor cells in the evolutionary expansion of the cerebral cortex. Cereb Cortex 16 Suppl 1 , i152-161.

Mason, H.A., Rakowiecki, S.M., Raftopoulou, M., Nery, S., Huang, Y., Gridley, T., and Fishell, G. (2005). Notch signaling coordinates the patterning of striatal compartments. Development 132 , 4247-4258.

McCarthy, M., Turnbull, D.H., Walsh, C.A., and Fishell, G. (2001). Telencephalic neural progenitors appear to be restricted to regional and glial fates before the onset of neurogenesis. J Neurosci 21 , 6772-6781.

McConnell, S.K., and Kaznowski, C.E. (1991). Cell cycle dependence of laminar determination in developing neocortex. Science 254 , 282-285.

McCrea, P.D., and Gu, D. (2010). The catenin family at a glance. J Cell Sci 123 , 637- 642.

144

Miller, M.W., and Nowakowski, R.S. (1988). Use of bromodeoxyuridine- immunohistochemistry to examine the proliferation, migration and time of origin of cells in the central nervous system. Brain Res 457 , 44-52.

Miyashita-Lin, E.M., Hevner, R., Wassarman, K.M., Martinez, S., and Rubenstein, J.L. (1999). Early neocortical regionalization in the absence of thalamic innervation. Science 285 , 906-909.

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

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.

Miyoshi, G., Butt, S.J., Takebayashi, H., and Fishell, G. (2007). Physiologically distinct temporal cohorts of cortical interneurons arise from telencephalic Olig2-expressing precursors. J Neurosci 27 , 7786-7798.

Mizutani, K., Yoon, K., Dang, L., Tokunaga, A., and Gaiano, N. (2007). Differential Notch signalling distinguishes neural stem cells from intermediate progenitors. Nature 449 , 351-355.

Mo, Z., Moore, A.R., Filipovic, R., Ogawa, Y., Kazuhiro, I., Antic, S.D., and Zecevic, N. (2007). Human cortical neurons originate from radial glia and neuron-restricted progenitors. J Neurosci 27 , 4132-4145.

Mochida, G.H. (2009). Genetics and biology of microcephaly and lissencephaly. Semin Pediatr Neurol 16 , 120-126.

Molnar, Z., Metin, C., Stoykova, A., Tarabykin, V., Price, D.J., Francis, F., Meyer, G., Dehay, C., and Kennedy, H. (2006). Comparative aspects of cerebral cortical development. Eur J Neurosci 23 , 921-934.

Molyneaux, B.J., Arlotta, P., Fame, R.M., MacDonald, J.L., MacQuarrie, K.L., and Macklis, J.D. (2009). Novel subtype-specific genes identify distinct subpopulations of callosal projection neurons. J Neurosci 29 , 12343-12354.

Molyneaux, B.J., Arlotta, P., Menezes, J.R., and Macklis, J.D. (2007). Neuronal subtype specification in the cerebral cortex. Nat Rev Neurosci 8, 427-437.

Mountcastle, V.B. (1997). The columnar organization of the neocortex. Brain 120 ( Pt 4) , 701-722.

145

Mutch, C.A., Funatsu, N., Monuki, E.S., and Chenn, A. (2009). Beta-catenin signaling levels in progenitors influence the laminar cell fates of projection neurons. J Neurosci 29 , 13710-13719.

Navarro-Quiroga, I., Chittajallu, R., Gallo, V., and Haydar, T.F. (2007). Long-term, selective gene expression in developing and adult hippocampal pyramidal neurons using focal in utero electroporation. J Neurosci 27 , 5007-5011.

Neumann, E., Schaefer-Ridder, M., Wang, Y., and Hofschneider, P.H. (1982). Gene transfer into mouse lyoma cells by electroporation in high electric fields. EMBO J 1, 841- 845.

Noctor, S.C., Flint, A.C., Weissman, T.A., Dammerman, R.S., and Kriegstein, A.R. (2001). Neurons derived from radial glial cells establish radial units in neocortex. Nature 409 , 714-720.

Noctor, S.C., Flint, A.C., Weissman, T.A., Wong, W.S., Clinton, B.K., and Kriegstein, A.R. (2002). Dividing precursor cells of the embryonic cortical ventricular zone have morphological and molecular characteristics of radial glia. J Neurosci 22 , 3161-3173.

Noctor, S.C., Martinez-Cerdeno, V., Ivic, L., and Kriegstein, A.R. (2004). Cortical neurons arise in symmetric and asymmetric division zones and migrate through specific phases. Nat Neurosci 7, 136-144.

Noctor, S.C., Martinez-Cerdeno, V., and Kriegstein, A.R. (2008). Distinct behaviors of neural stem and progenitor cells underlie cortical neurogenesis. J Comp Neurol 508 , 28- 44.

Nowakowski, R.S., Lewin, S.B., and Miller, M.W. (1989). Bromodeoxyuridine immunohistochemical determination of the lengths of the cell cycle and the DNA- synthetic phase for an anatomically defined population. J Neurocytol 18 , 311-318.

Ogawa, M., Miyata, T., Nakajima, K., Yagyu, K., Seike, M., Ikenaka, K., Yamamoto, H., and Mikoshiba, K. (1995). The reeler gene-associated antigen on Cajal-Retzius neurons is a crucial molecule for laminar organization of cortical neurons. Neuron 14 , 899-912.

Ohtsuka, T., Imayoshi, I., Shimojo, H., Nishi, E., Kageyama, R., and McConnell, S.K. (2006). Visualization of embryonic neural stem cells using Hes promoters in transgenic mice. Mol Cell Neurosci 31 , 109-122.

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

Pakkenberg, B., and Gundersen, H.J. (1997). Neocortical neuron number in humans: effect of sex and age. J Comp Neurol 384 , 312-320.

146

Paramasivam, M., Chang, Y.J., and LoTurco, J.J. (2007). ASPM and citron kinase co- localize to the midbody ring during cytokinesis. Cell Cycle 6, 1605-1612.

Pilaz, L.J., Patti, D., Marcy, G., Ollier, E., Pfister, S., Douglas, R.J., Betizeau, M., Gautier, E., Cortay, V., Doerflinger, N. , et al. (2009). Forced G1-phase reduction alters mode of division, neuron number, and laminar phenotype in the cerebral cortex. Proc Natl Acad Sci U S A 106 , 21924-21929.

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

Pinto, L., Mader, M.T., Irmler, M., Gentilini, M., Santoni, F., Drechsel, D., Blum, R., Stahl, R., Bulfone, A., Malatesta, P. , et al. (2008). Prospective isolation of functionally distinct radial glial subtypes--lineage and transcriptome analysis. Mol Cell Neurosci 38 , 15-42.

Polleux, F., Dehay, C., Moraillon, B., and Kennedy, H. (1997). Regulation of neuroblast cell-cycle kinetics plays a crucial role in the generation of unique features of neocortical areas. J Neurosci 17 , 7763-7783.

Pontious, A., Kowalczyk, T., Englund, C., and Hevner, R.F. (2008). Role of intermediate progenitor cells in cerebral cortex development. Dev Neurosci 30 , 24-32.

Qian, X., Goderie, S.K., Shen, Q., Stern, J.H., and Temple, S. (1998). Intrinsic programs of patterned cell lineages in isolated vertebrate CNS ventricular zone cells. Development 125 , 3143-3152.

Qian, X., Shen, Q., Goderie, S.K., He, W., Capela, A., Davis, A.A., 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.

Raedler, E., and Raedler, A. (1978). Autoradiographic study of early neurogenesis in rat neocortex. Anat Embryol (Berl) 154 , 267-284.

Rakic, P. (1971). Guidance of neurons migrating to the fetal monkey neocortex. Brain Res 33 , 471-476.

Rakic, P. (1972). Mode of cell migration to the superficial layers of fetal monkey neocortex. J Comp Neurol 145 , 61-83.

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

Rakic, P. (1988). Specification of cerebral cortical areas. Science 241 , 170-176.

147

Rakic, P. (1995). A small step for the cell, a giant leap for mankind: a hypothesis of neocortical expansion during evolution. Trends Neurosci 18 , 383-388.

Ramon y Cajal, S. Histologie du systeme nerveux de l'homme et des vertebres. Vol 2. A. Maloine, Paris. (Reprinted by Consejo Superior de Investigaciones Cientificas, Instituto Ramon y Cajal, 1952).

Retzius, G. (1893). Studien über Ependym und Neuroglia. Biol Untersuch. Stockholm, N.S. 5, 9-26.

Retzius, G. (1894). Die Neuroglia des Gehirns beim Menschen und bei Säugenthiern. Biologische Untersuchungen 6. 1-24.

Rice, D.S., and Curran, T. (2001). Role of the reelin signaling pathway in central nervous system development. Annu Rev Neurosci 24 , 1005-1039.

Rickmann, M., Chronwall, B.M., and Wolff, J.R. (1977). On the development of non- pyramidal neurons and axons outside the cortical plate: the early marginal zone as a pallial anlage. Anat Embryol (Berl) 151 , 285-307.

Rickmann, M., and Wolff, J.R. (1981). Differentiation of 'preplate' neurons in the pallium of the rat. Bibl Anat, 142-146.

Roth, G., and Dicke, U. (2005). Evolution of the brain and intelligence. Trends Cogn Sci 9, 250-257.

Roy, K., Kuznicki, K., Wu, Q., Sun, Z., Bock, D., Schutz, G., Vranich, N., and Monaghan, A.P. (2004). The Tlx gene regulates the timing of neurogenesis in the cortex. J Neurosci 24 , 8333-8345.

Saito, T., and Nakatsuji, N. (2001). Efficient gene transfer into the embryonic mouse brain using in vivo electroporation. Dev Biol 240 , 237-246.

Sansom, S.N., Hebert, J.M., Thammongkol, U., Smith, J., Nisbet, G., Surani, M.A., McConnell, S.K., and Livesey, F.J. (2005). Genomic characterisation of a Fgf-regulated gradient-based neocortical protomap. Development 132 , 3947-3961.

Sauer, B. (1998). Inducible gene targeting in mice using the Cre/lox system. Methods 14 , 381-392.

Sauer, F. (1935). The cellular structure of the neural tube. J Comp Neurol, 13-23.

Sauer, M.E., and Walker, B.E. (1959). Radioautographic study of interkinetic nuclear migration in the neural tube. Proc Soc Exp Biol Med 101 , 557-560.

148

Schaper, A. (1897). The earliest differentiation in the central nervous system of vertebrates. Science 5, 430-431.

Schmechel, D.E., and Rakic, P. (1979). A Golgi study of radial glial cells in developing monkey telencephalon: morphogenesis and transformation into astrocytes. Anat Embryol (Berl) 156 , 115-152.

Selkoe, D., and Kopan, R. (2003). Notch and Presenilin: regulated intramembrane proteolysis links development and degeneration. Annu Rev Neurosci 26 , 565-597.

Sessa, A., Mao, C.A., Hadjantonakis, A.K., Klein, W.H., and Broccoli, V. (2008). Tbr2 directs conversion of radial glia into basal precursors and guides neuronal amplification by indirect neurogenesis in the developing neocortex. Neuron 60 , 56-69.

Shen, Q., Wang, Y., Dimos, J.T., Fasano, C.A., Phoenix, T.N., Lemischka, I.R., Ivanova, N.B., Stifani, S., Morrisey, E.E., and Temple, S. (2006). The timing of cortical neurogenesis is encoded within lineages of individual progenitor cells. Nat Neurosci 9, 743-751.

Sherr, C.J., and Roberts, J.M. (1999). CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 13 , 1501-1512.

Shipp, S. (2007). Structure and function of the cerebral cortex. Curr Biol 17 , R443-449.

Sidman, R.L., Miale, I.L., and Feder, N. (1959). Cell proliferation and migration in the primitive ependymal zone: an autoradiographic study of histogenesis in the nervous system. Exp Neurol 1, 322-333.

Simmons, P.A., Lemmon, V., and Pearlman, A.L. (1982). Afferent and efferent connections of the striate and extrastriate visual cortex of the normal and reeler mouse. J Comp Neurol 211 , 295-308.

Smart, I.H., Dehay, C., Giroud, P., Berland, M., and 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.

Striedter, G.F., and Charvet, C.J. (2009). Telencephalon enlargement by the convergent evolution of expanded subventricular zones. Biol Lett 5, 134-137.

Tabata, H., and Nakajima, K. (2001). Efficient in utero gene transfer system to the developing mouse brain using electroporation: visualization of neuronal migration in the developing cortex. Neuroscience 103 , 865-872.

149

Takahashi, T., Goto, T., Miyama, S., Nowakowski, R.S., and Caviness, V.S., 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., Nowakowski, R.S., and Caviness, V.S., Jr. (1995). The cell cycle of the pseudostratified ventricular epithelium of the embryonic murine cerebral wall. J Neurosci 15 , 6046-6057.

Takizawa, T., Nakashima, K., Namihira, M., Ochiai, W., Uemura, A., Yanagisawa, M., Fujita, N., Nakao, M., and Taga, T. (2001). DNA methylation is a critical cell-intrinsic determinant of astrocyte differentiation in the fetal brain. Dev Cell 1, 749-758.

Tamamaki, N., Fujimori, K.E., and Takauji, R. (1997). Origin and route of tangentially migrating neurons in the developing neocortical intermediate zone. J Neurosci 17 , 8313- 8323.

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.

Tanaka, D.H., Maekawa, K., Yanagawa, Y., Obata, K., and Murakami, F. (2006). Multidirectional and multizonal tangential migration of GABAergic interneurons in the developing cerebral cortex. Development 133 , 2167-2176.

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.

Tarui, T., Takahashi, T., Nowakowski, R.S., Hayes, N.L., Bhide, P.G., and Caviness, V.S. (2005). Overexpression of p27 Kip 1, probability of cell cycle exit, and laminar destination of neocortical neurons. Cereb Cortex 15 , 1343-1355.

Toni, N., Teng, E.M., Bushong, E.A., Aimone, J.B., Zhao, C., Consiglio, A., van Praag, H., Martone, M.E., Ellisman, M.H., and Gage, F.H. (2007). Synapse formation on neurons born in the adult hippocampus. Nat Neurosci 10 , 727-734.

Voigt, T. (1989). Development of glial cells in the cerebral wall of ferrets: direct tracing of their transformation from radial glia into astrocytes. J Comp Neurol 289 , 74-88.

Woodhead, G.J., Mutch, C.A., Olson, E.C., and Chenn, A. (2006). Cell-autonomous beta- catenin signaling regulates cortical precursor proliferation. J Neurosci 26 , 12620-12630.

Woods, C.G., Bond, J., and Enard, W. (2005). Autosomal recessive primary microcephaly (MCPH): a review of clinical, molecular, and evolutionary findings. Am J Hum Genet 76 , 717-728.

150

Xu, Q., Cobos, I., De La Cruz, E., Rubenstein, J.L., and Anderson, S.A. (2004). Origins of cortical interneuron subtypes. J Neurosci 24 , 2612-2622.

Yoon, K., and Gaiano, N. (2005). Notch signaling in the mammalian central nervous system: insights from mouse mutants. Nat Neurosci 8, 709-715.

Yun, K., Fischman, S., Johnson, J., Hrabe de Angelis, M., Weinmaster, G., and Rubenstein, J.L. (2002). Modulation of the notch signaling by Mash1 and Dlx1/2 regulates sequential specification and differentiation of progenitor cell types in the subcortical telencephalon. Development 129 , 5029-5040.

Zecevic, N. (2004). Specific characteristic of radial glia in the human fetal telencephalon. Glia 48 , 27-35.

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

Zimmer, C., Tiveron, M.C., Bodmer, R., and Cremer, H. (2004). Dynamics of Cux2 expression suggests that an early pool of SVZ precursors is fated to become upper cortical layer neurons. Cereb Cortex 14 , 1408-1420.

151