CHARACTERlZATtON OF THE FUNCTION OF THE OTX-2

GENE IN EARLY MAMMALIAN DEVELOPMENT

Ou Jin

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Graduate Department of Molecular and Medical Genetics University of Toronto

@ Copyright by Ou Jin, 1997

National Library Biblioth&que nationale du Canada

Acquisitions and Acquisitions et Bibliographic Services services bibliographiques

395 Wellington Street 395. nre welrigton OttawaON K 1 A W OttawaON K 1 A W Canada Canada

The author has granted a non- L'auteur a accorde me licence non exclusive licence allowing the exclusive permettant a la National Library of Canada to Bibliotheque nationale du Canada de reproduce, loan, distribute or sell reproduire, preter, distribuer ou copies of this thesis in microform, vendre des copies de cette these sous paper or electronic formats. la forme de microfiche/film, de

reproduction sur papier ou sur format electronique.

The author retains ownership of the L'auteur conserve la propriete du copyright in this thesis. Neither the droit d'auteur qui protege cette these. thesis nor substantial extracts fiom it Ni la these ni des extraits subsbntiels may be printed or otherwise de celle-ci ne doivent Stre imprimes reproduced without the author's ou autrement reproduits sans son pemission. autorisation.

Dedication

to my father

CHARACTERlZATlON OF THE FUNCTION OF OTX-2 GENE IN

EARLY MAMMALIAN DEVELOPMENT

by Ou Jin

A thesis submitted in conformrty with the requirements for the degree of Doctor

of Philosophy in the Graduate Department of Molecular and Medical Genetics at

the University of Toronto, 1997

ABSTRACT

How the anterior region of the vertebrate embryo is specified is still an

unsolved question in early pattern formation. In this thesis, I have undertaken a

study of the roles of a homeobox gene, Otx-2, in the early murine development,

with particular reference to its role in anterior patterning.

Otw-2, one of the two mammalian Otx genes, related to the Drosophila otd

gene, has been cloned and characterized. It encodes a protein containing a 60

amino acid homeodomain motif with only two amino acid differences from the

sequence of Drasophila Otd. 00r-2 is one of the earliest genes showing

restricted anterior expression domains. During early embryogenesis, OW2 is

first expressed in the entire epiblast which gives rise to the embryo proper, then

is gradually restricted to the anterior region of the embryo and finally to the

presumptive forebrain and midbrain. This expression profile suggests that Ok-

2 may have multiple roles in the early development and anterior patterning of

the embryo. These studies provide fundamental background information for

further functional analysis.

iii

Functional analysis of Otx-2 was performed by a gene targeting

approach. A loss-of-f unction mutation of Otx-2 leads to loss of forebrain,

midbrain and anterior hindbrain, suggesting the critical role of Otx.2 in head

formation. In addition, homozygous Ok-2 mutant embryos also show early

gastrulation defects and defects in the prechordal mesoderm and notochord

precursors. These defects are consistent with multiple roles for Otx-2 in

gastrulation and the patterning of the rostra1 brain in mice.

To further dissect the function of 0tx-2 in anterior patterning, a double

heterozygous mutant was made with HNF-313, a member of the winged-helix

transcription factor family. Generation of Otx-2+/-; HNF-3&/- double

heterozygous mutants demonstrates a new lethal phenotype in mice. The

phenotype is characterized by varying degrees of holoprosencephaly, cyclopia

with proboscis-like structures, suggesting the genetic interaction of the two

genes in anterior midline patterning. Co-expression of Otx-2 and HNF-3B in the

anterior midline of the embryo further supported such interactions. Furthermore,

the loss of Shh expression, a secreted protein, in the anterior region of 0k-2+'

;HNF-3B+/- embryos indicates that both Ob(-2 and HNF-3B are involved in

regulating the Shh signalling pathway in the anterior midline. This study also

illustrates the way in which dosage-dependent effects of transcription factors

can reveal novel genetic interactions and help define genetic pathways.

iv

ACKNOWLEDGMENTS

First, I sincerely thank my supewisor, Dr. Janet Rossant, for her

consistent support, encouragement and advice throughout the program. I also

would like to take this opportunity to thank Drs. Martin Breitman, Rod Mclnnes

and Andrew Spence for their suggestions and comments during yearly

romm ittee meetings.

I thank people in our lab, old and new, not only for sharing friendships

but also for sharing their expertise. In particular, I would like to thank Lois

Schwartz for aggregation experiments, Ken Harpal for preparing wax sections

and Celine Champigny for some sequencing.

Five years of a life time is short, but this period was an extremely

important time, not only for my career but also for my life. Three years ago, I

lost my father, the most important person in my Me. It was under my father's

influence that I decided to go to medical school. Later, when I became

interested in medical research, he gave me great encouragement. Five years

ago, it was my father who strengthened my decision to pursue a Ph.D. degree. I

wish he could see the day I receive my degree. I would also like to thank my

mother for all her support, especially since she moved to Toronto. Finally, I

would like to thank my wife, Evie, for her understanding and support.

CONTENTS

ABSTRACT .......................................................................................................... ...... ....... .ii ACKNOWLEDGMENTS .......................... ...- ............... .. ................................ iv UST OF FIGURES ............. .., ..................................................................................... vii ... LIST OF TABLES ....................................................................................................... VIII UST OF ABBREVIATIONS ............................................................................................. ix

CHAPTER 1 : INTRODUCTION ....................................................................................... I

Early mouse development ............................................................................................ 2 Gastrulation ....... .... .............................................................................................. 2 Neurulation .............. ., ........................................................................................ 8

Hox genes and vertebrate A-P patterning ................................................................ 14 Otd and ems in Drosophila anterior head development ........................... 4

........................................................................................................................ otd 20 ................................................................................ ems ............................... ...,.. 21

Mammalian homologues of otd and ems and their expression patterns during mouse embryogenesis .................... .. ............................................................ 21

Mouse Otx- 1 and Otx-2 ................................................................................... -22 Mouse Emx- I and Emx-2 ................................................................................. 23

Outline uf the thesis ....................................................................................................... 26

CHAPTER 2: IDENTIFICATiON AND CHARACTERIZATION OF 07X-2 GENE .. 27

...... ........*........................................................*.........-....*.......-..... Introduction ........ ........ .. 28 Materials and Methods .................................................................................................. 29 Resuks .............................................................................................................................. 31

Isolation of Otx-2 cDNAs ................................................................................... 31 Northern blot analysis and identification of 5'-OW-2 cDNA ........................ 33 Exon-intron structure of Otx-2 gene ................................................................ 34

.................................................. Early expression pattern of the OW2 gene -35 Discussion ...................... ....., ......................................................................................... 35

CHAPTER 3: TARGETlNG OF OX-2 ......................................................................... 55

Introduction ......... ,., .......................................................................................................... 56 Materials and Methods ................................................................................................ 58 Resu ks .............................................................................................................................. 60

A targeted disruption of Otx-2 in ES cells by homologous . . recomb~nat~on .................................................................................................... 60 The Obr-2 mutation leads to embryonic lethality ............... ...... ................ 63 Morphological and histological analysis of the Otx-2 homozygous phenotype ........................................................................................................... 64 Deletion of anterior brain regions rostra1 to hornbornere 3 ..................... 36 Defects in gastrulation in Otw-2 homozygous embryos ............................ 66

vii

LIST OF FIGURES

Chapter 1 Figure 1.1 :

Figure 1 2: Figure 1.3: Figure 1.4: Figure 1.5:

Figure 1.6:

Chapter 2 Figure 2.1 :

Figure 2.2:

Figure 2.3: Figure 2.4:

Figure 2.5: Figure 2.6: Figure 2.7:

Figure 2.8:

Chapter 3 Figure 3.1 : Figure 3.2:

Figure 3.3: Figure 3.4: Figure 3.5:

Figure 3.6:

Schematic representation of mouse development from E6.0 to E8.5 ............................................................................................................. 3 Morphological features of the node ...................................................... 6 Neural induction in Xenopus ............................................................... 1 0 Early mouse brain development ......................................................... 12 Genomic organization and collinear expression patterns

............. of Drosophiia HOM-C genes and mammalian Hox genes 17 Comparison of expression of Otx and Emx genes in the mouse

......................................................................................... embryo at El0 24

The nucleotide sequence of the 1.6kb cDNA clone and the amino ........ ............. acid sequence of the putative Otw-2 gene product .. 39

Comparison of Otx-2 homeodomain with Otx-1 , Xotx.2, SpOtx. Otd homeodomains ............................................................................. -41 Transcription of the O&-2 gene ........................................................ 43 Diagram of the 5' RACE and primer extension experiments

......................................... (A) and sequence of 5' RACE product (B) 45 .............................. Result of primer extension experiment 4

Exon-intron structure of Otx-2 gene ................................................ -49 Expression of Otx-2 from prestreak to early somite stages

....................... revealed by whole mount RNA in situ hybridization 51 ................................ Tissue sections analyzed for Otx-2 expression 53

........ Targeted disruption of the Otx-2 gene ................................ .. 75 (A) Southern bolt analysis of DNA from wild-type and

... targeted ES cell lines. 5-1 2. 5-23 and C- 1 2 .................. ....... -37 (B) PCR genotyping of yolk sac biopsies isolated from E8.5

embryos of intercross between Otx-% mice ........................ 79 Phenotype of Otx-2 homozygous ...................................................... -81 Histological analysis of Otx-2 homozygous embryo ....................... 82 Whole mount RNA in situ hybridization of Krox-20 in E8.5 wild-type or heterozygous embryo and homozygous mutant embryo ..................................................................................................... 84 Whole-mount analysis of mesoderm markers in wild-type or heterozygous embryos and homozygous mutant

............. embryos at E7.25E8.5 .. .................................................... -86

Chapter 4 Figure 4.1 : Phenotype of Otx-2 +/- ; HNF-30 +/- embryos at El 2.5 ................. 1 05 Figure 4.2. Phenotype of 0&-2 +/- ; HNF-3B +/- embryos at E9.5 ................... 1 07 Figure 4.3: Histological analysis of Otx-2 +/- ; HNF-38 +/- embryos at 12.5 .. 1 09

viii

Figure 4.4: Comparison expression of OtK-2 and HNF-3B +/- in wildtype mouse embryos between E7.5 and €9.5 ................ ... ............. 1 1 1

Figure 4.5: Expression of Otw-2, Hoxb- 1, Six-3 and Mox- 1 in Otx-2 +/- ; HNF- .......................................................... ..................... 38 +/- embryos .. 1 13

Figure 4.6: Whole-mount in situ hybridization and cross section analysis of Otx-2 +/- ; H W 3 B +I- mutant with BF1, Krm-20 and Shh ....... 115

.............................................. Figure4.7: Phenotypeof Otx-24-embryos 1 1 7 Figure 4.8: Schematic diagram showing that OtK-2 and HNF-3B are

involved in regulating Shh expression in the anterior region ..... 1 1 9

Chapter 5 Figure 5.1 : Topologiocal equivalence between the Xenopus fate map

and mouse fate map at the early gastrula stage ............................ 1 28

LIST OF TABLES

Chapter 3 Table 3.1:

Table 3.2: Table 3.3:

Chapter 4 Table 4.1:

Table 4.2:

Table 4.3:

Results of screens for homologous recombination into OW2 locus .......... ........,. .................................................................................... 62 Frequency of germ-line transmission of targeted ES cells lines ... 62 Genotype of mice resulting from Otw-2 heterozygous intercrosses.. ...... .... .....*.. ...,,. ..,,.. ..-....-........*.... 64

Frequency of genotypes resulting from Otx-2 +/- and HNF-3B 4- intercrosses at weaning stage .......................................................... 102 Phenotypes resulting from OW+/- and HNF-3B +/- intercrosses at E12.5 ............................... .,., ........................................................... 103 Frequency of genotypes resulting from Obr-2 and HNF-38 intercrosses at E9.5 .......................................................................... 1 04

ABBREVIATtONS

A-P Anterior-Posterior

CNS Central Nervous System

D-V Dorsal -Ventral

E Embryonic Day

ES Embryonic Stem cells

RT-PCR Reverse Transcriptase-Polymerase Chain Reaction

RACE Rapid Amplification of cDNA End

Chapter I

INTRODUCTION

The mouse (Mus musculus) is a widely used system for studying

vertebrate development. Not only is it closer to humans than non-mammalian

species, but it also has many valuable genetic resources available such as a

large collection of mutants for investigators to characterize phenotypic variants

and explore the molecular definition of these mutants. However, other systems

have many advantages for experimental embryology. For example, in Xenopus,

chick and zebrafish, embryos are more plentiful and develop outside the

mother. Thus, they are often easier to manipulate and observe. In recent years,

it has become clear that the mechanisms governing early development are

highly conserved between species. Thus, information from other species is

certainly very useful for understanding mouse development. In this section, I will

briefly outline early mouse development with emphasis on those processes

which are relevant to studies described in this thesis.

Gastrulation

Cleavage, gastrulation and organogenesis are three major stages during

mouse embryogenesis. After the stage of cleavage and the formation of a

blastocyst, the embryo enters one of the most critical stages in its development:

the stage of gastrulation. Before gastrulation starts, the embryonic compartment

of the embryo has two components: the primitive ectoderrn or epiblast and

primitive endoderm (Fig.l.1 A). Gastrulation begins at embryonic day 6.5 (E6.5).

It is a process of profound but well-ordered rearrangements of the cells in the

embryo, resulting in the formation of two new layers, mesoderm and definitive

endoderm. Cells at the junction between the primitive ectoderrn and the

Figure 1 -1. Schematic representation of mouse development from E6.0 to E8.5.

(adapted from the following sources with slight modifications: Hogan et al.,

1986, 1994).

xtracmbryontc

Mesodermal cells ingrcsslng into primltlvc streak

E6.0 E7.0

extraembryonic ectoderm undergo an epithelial to mesenchymal transition and

start to migrate and come to lie between the primitive ectoderm and the

endoderm, resulting in the formation of the primitive streak (Fig. 1.1 8, C, D, E).

As mesoderm migration proceeds, the primitive streak gradually elongates and

reaches the distal tip of the egg cylinder. The lateral migration of the mesoderm

through the primitive streak results in a wing-shaped mesodermal layer that

extends around the embryonic ectoderm (Fig. 1.1 C). A specialized structure,

called the node, appears at mid-late streak stages in the anterior end of the

primitive streak (Fig. 1.1 E and Fig. 1.2). It is a localized two-layered area and

there is no visceral endoderm underlying this area (Fig. 12C) (reviewed in

Sulik et al., 1994). Cells from the node migrate fonnrard in the midline, resulting

in the formation of the head process. The cells of the anterior head process

form the socalled prechordal region or prechordal plate which is much more

readily recognizable in the chick. At late streak stage, head process cells meet

mesoderm wings at the anterior end and complete the mesoderm layer. In the

midline, cells displace the overlying primitive endoderm to form definitive

endoderm or gut endoderm (Fig.1 .l E). However, the origin of gut endoderm is

still not clear. HNF4B homozygous mutants lack node formation, and foregut

morphogenesis is severely affected in those mutants (Ang and Rossant, 1994;

Weinstein et al., 1994), suggesting that the node might be one of the cell

sources involved in gut endoderm formation.

The basic body plan along the anteriorposterior (A-P) axis and the future

dorsal-ventral (D-V) axis are established during gastrulation. In addition, the

formation of three germ layers also provides the appropriate tissue interactions

required for further differentiation during organogenesis. The primitive streak

initiates the A-P axis, representing the future posterior portion of the embryo.

Fig. 1.2. Morphological features of the node. (A) Midsagittal section showing the

position of the node. (B) Cross section through a similar stage embryo

immediately anterior of the node. Note that in the midline, there are only two cell

layers. (C) Ventral view of the surface of the node and notochord in €7.5

embryo. (D) Higher magnification view shows cilia-like projections from these

cells (arrows). Abbreviations: .A: allantois; Am: amnion; E: embryonic ectoderm;

En: endoderm; ExEn: extraembryonic endoderm; ExM: extraembryonic

mesoderm of the yolk sac; FG: foregut diverticulum; H: presumptive heart; M:

mesoderm; N: node; NC: notochord; NF: neural folds; PS: primitive streak.

(adapted from Hogan et al., 1994)

The node, which is equivalent to Hensen's node in the chick (Waddington,

1933), the dorsal blastopore lip in Xenopus and the embryonic shield in the

zebrafish (Ha, 1992). plays a prime role in organizing and patterning the

midline axis of the embryo. The organizer function of the dorsal lip was first

suggested by Spemann and Mangold (1 924), who demonstrated the induction

of a secondary axis by grafting a dorsal blastopore lip to the ventral side of a

host embryo. Spemann (1931) also found that the early dorsal lip could induce

a complete axis including a head, suggesting that the early dorsal lip has a

head organizer activity. However, the later stage dorsal lip could only induce a

trunk-tail structure, suggesting that at later stages, it only has trunk organizer

function. In mouse, Beddington (1 994) provided the first direct evidence for the

organizing function of the node, if it is transplanted heterotopically. When the

mid-gastrulation node was grafted to a posterolateral location in a host embryo

at the same developmental stage, a second neural axis was induced. However,

the most anterior end of the second axis is always at hindbrain level. Therefore,

whether the node can induce the most anterior structures such as forebrain still

remains open.

Neurula tion

When mesoderm extends to the anterior end of the embryo, the overlying

anterior ectoderm forms the neural plate which is first recognizable soon after

€7.5. The neural plate gradually folds up to form a neural tube. At E9.0, the

anterior neuropore is completely closed. The formation of the neural plate and

its transformation to neural tube is one of the most dramatic events following

gastrulation. Induction and axial patterning of the neural plate, which has been

mainly investigated in amphibian and chick embryos, is a complex process

involving in planar signals from the organizer region and vertical signals from

underlying axial mesoderm (reviewed by Slack and Tannahill, 1992; Ruiz i

Altaba, 1994; Kessler and Melton. 1994; Kelly and Melton, 1995). A two-step

model of neural induction has been proposed based on a series of

transplantation and recombination experiments in amphibian embryos using

morphological and molecular markers. It suggests that the ectoderm is first

induced to anterior neural fate by the signals present in the mesoderm, followed

by a transformation to a more posterior neural fate by a graded posterior signal

coming from posterior mesoderm (Fig. 1.3). It also has been found that different

regions of the underlying mesoderm have different inducing potential. In mouse,

similar observations have been achieved by explant-recombination assays

(Ang and Rossant, 1993; Ang et al., 1994). These studies suggest the

importance of tissue interactions in neural induction and regional specification

along the A-P axis of the neural plate and neural tube.

Regionalization of the neural tube becomes morphologically apparent as

a result of changes in the shape of the tube. In the anterior region of the neural

tube (the future brain), there are three primary brain compartments: forebrain,

midbrain and hindbrain (Fig. 1.4). The forebrain further subdivides into

telencephalon and diencephalon. The telencephalon will form the cerebral

hemispheres while the diencephalon will form the thalamus, hypothalamus and

retina, each of which has distinct morphologies and histologies (reviewed by

Puelles and Rubenstein, 1993). The midbrain does not undergo further

transverse subdivision. However, the hindbrain further evolves into a series of

repeated undulations or segments. These segments are so-called

rhombomeres(r), rl-r7, each of which has a unique developmental fate

(Lumsden and Keynes, 1989; Lumsden et al.. 1991).

Fig. 1.3. Neural induction in Xenopus. During gastrulation. dorsal mesoderm

involutes and migrates underneath the ectodermal surface of the gastrula. As

dorsoanterior mesoderm involutes, the adjacent ectoderm is initially induced to

form anterior neural tissues. The mesoderm migrates toward the former animal

pole, and newly contacted ectoderrn is also induced to form anterior neural

tissue. Ectoderm is progressively contacted by more posterior mesoderm

resulting in posterior neural elements (darker shades of purple). As a result of

this process, the anteroposterior pattern of induced neural tissue reflects the

anteroposterior character of underlying dorsal mesoderm. Both vertical and

planar signals emanating from dorsal mesoderm have been implicated in the

transformation of overlying embryonic ectoderm to neural ectoderm. 0,

organizer mesoderm. (Adapted from Kessler and Melton, 1994)

Figure 1.4. Early mouse brain development. The three primary brain vesicles

are subdivided as development continues. (adapted from Gilbert. 1994)

In recent years, a number of candidate molecules implicated in axial

patterning have been identified. Signaling molecules such as chordin (Sasai et

al., 1 994), follistatin (reviewed in Kessler and Melton, 1 994) and noggin (Smith

and Harland, 1992; Smith et al., 1993) are candidates for neural inducers.

Nuclear transcription factors such as Otx-2, gsc, HNF-3R and Lim-7 (reviewed

by De Robertis, 1994; BallyGuif and Boncinelli, 1997) and secreted molecules

such as wnt-I (reviewed in Joyner, 1996) and Shh (reviewed by

Hammerschmidt, 1997) have been implicated in the patterning of the neural

tube in vertebrates. The roles of Drosophila horneotic complex (HOW) , and

their vertebrate homologues, Hox (homeobox) genes, in the axial pattern have

been extensively studied. In the following section, l will briefly discuss the role

of Hox genes in the axial patterning of vertebrate embryo body.

ox aenesmd verteb~te A-P oattemmg

One of the most fascinating findings in recent years in developmental

biology is that the molecular mechanisms of axial patterning have been highly

conserved during evolution. H O W genes (reviewed by Akam, 1989) were

discovered in the Drosophila due to the dramatic phenotypes that result from

mutations in these genes. Pioneering work by Lewis (1 978) demonstrated that

the segments of Drosophila differ from each other because of the action of a

small set of homeotic genes along the A-P axis of the body. For example,

Ultrabithorax (Ubx) mutations result in homeotic transformations of the third

thoracic (T3) segment carrying halteres to a second thoracic (T2) segment with

wings (Lewis, 1978, 1981, 1982). As figure I .5 illustrates, by genetic criteria, the

labial (lab), Deformed (Dfd) and proboscipedia @b) genes are involved in

specify the identity of anterior segments, the Sexcombs reduced (Scr) and

Antennapedia (Antp) genes contribute thoracic identity functions, and the

bithorax complex genes (Ubx, abdominal-A and Abdominal-@ are required for

the specification of segmental identities in the posterior thorax and in all

abdominal segments (reviewed in McGinnis and Krumlauf, 1 992). Loss-of-

function mutations and gain-of-function mutations of HOM-C genes lead to

homeotic transformations of the embryonic and /or the adult Drosophila body

plan, suggesting that these genes are involved in the genetic control of A-P

patterning in Drosophila (reviewed by McGinnis and Krumlauf, 1 992; Lawrence

and Morata. 1994). McGinnis and his colleagues (1 984a). and Scott and

Wiener (1984) first discovered that Drosophila H O W genes contain a

conserved sequence called the Homeobox motif. The homeobox motif is not

only present in flies but also found in all vertebrates, such as Xenopus, chick,

mice and human (McGinnis et al., 1984b; Carrasco et al., 1984, ). The

homeobox encodes a 60 amino acid motif (homeodomain). These

homeodomains have sequence-specific DNA binding activities in vifro and

have subsequently been found in many transcription factors (reviewed by

Levine and Hoey, 1988; Scott et al., 1989; Laughon, 1991; Gehring et al.,

1994). However, how these homeobox gene products achieve their in vivo

specificities in axial patterning is still an important and unresolved problem.

The mammalian Hox genes are aefined an the basis of their homology to

the genes of the Drosophila HOMC. Analysis of mouse and human Hox genes

reveals that there are at least 39 genes which are organized in four clusters,

HoxA. HoxB, HoxC, and HoxD. Based on sequence similarities and relative

genomic arrangements, the individual Hox genes with in different clusters can

be aligned with each other and with genes of the Drosophila HomC cluster

(Fig. 1.5). In mammals, Hox genes are expressed in overlapping domains along

A-P axis of the embryo with spatial colinearity according to the order of genes

along the chromosomes. For example, the more 3' located genes are

expressed in more anterior regions whereas the more 5' located genes are

expressed in more posterior regions (Fig. 1.58).

Genetic approaches provide a critical test for the postulated role of

candidate genes in development (Rossant, 1990; Joyner and Guillemont,

1994). The importance of Hox genes in vertebrate axial patterning has been

further demonstrated by loss-of-function mutations and gain-of-function

mutation studies (reviewed by McGinnis and Krumlauf, 1 992; Krumlauf, 1 994).

In Xenopus, these functional studies were usually performed through either

microinjection of mRNA into early embryos. resulting in ectopic production of the

protein, or m icroinjection of antibody into eggs, resulting in interference with the

normal function of the gene. In mouse, most functional studies used either the

approach of generating transgenic mice or the gene targeting approach to

generate null mutant mice (reviewed in McGinnis and Krumlauf, 1992).

Targeted disruption of members of the Hox gene families have demonstrated a

wide spectrum of phenotypes which include structural defects such as in the

formation of the vertebral column and limbs, and the formation of branchial

archderived tissues. For example, loss-of-function mutations of Hoxa-1 and

Hoxa-3 lead to defects in hindbrain and branchial regions of the mouse, but do

not appear to cause homeotic transformations of the affected regions (Chisaka

and Capecchi, 1991 ; Lufkin et al., 1991 ; Carpenter et al., 1993). However, some

phenotypes can be considered as anterior or posterior homeotic

transformations. For example, loss-of-function mutation of Hoxb-4 causes partial

transformation of the axis to atlas identity by the formation of a ventral arch at

Fig. 1.5. Genomic organization and colinear expression patterns of Drosophila

H O W genes and mammalian Hox genes. (A) Example of evolutionary

conservation of horneotic gene organization and expression in Drosophila

embryo and mouse embryo. (B) Schematic representation of the Drosophila

homeotic complex (HOM-C), the four human Hox complexes and a hypothetical

ancestral homeotic complex are displayed showing their possible phylogenetic

relationships. Each gene is represented by a colored box. For the simplicity, the

partial overlap between HOM gene transcripts in thoracic and abdominal

segments of Drosophila and overlapping expression domains of mammalian

Hox genes along the body axis are not represented; therefore, each color

represents the anteriormost expression domain of a given subfamily. (C)

Illustration of the transcription direction of these homeobox genes and

responses to retinoic acid. All these genes are transcribed in the same

direction. Those that are expressed more anteriorly are expressed earlier and

can be induced by low doses of retinoic acid. HOM gene abbreviations are: lab.

labial; pb, proboscipedia; Md, Deformed; Scr, Sex combs reduced; Antp,

Antennapedia; Ubx, Ultrabithoraw; abd-a, abdominaM; Abd-B, Abdominal-8.

(Adapted from the following sources with slight modifications: Gilbert, 1994;

Mark et al., 1 997).

the level of the second cervical vertebra (Ramirez-Solis et al., 1993).

Overexpression of Hoxa-7, analogous to gain-of-function homeotic mutations in

Drosophila. converts the basioccipital bone into a proatlas structure (Kessel, et

al., 1990). Hox mutations in mice often exhibit variably penetrant phenotypes,

suggesting functional redundancy among Hox genes. Consistent with this

hypothesis. mice with combinations of Hox mutations (i.e. double mutations),

show more severe defects than mice with the single mutations (Davis et al.,

1995; Davis and Capecchi, 1996; Fromental-Ramain et al., 1996; Chen and

Capecchi, 1997), suggesting quantitative genetic interactions among Hox

genes in regional pattern formation. For example, Hoxa-9 and Hoxd-9

compound mutants demonstrate synergistically altered phenotypes consisting

of an increase of penetrance and expressivity of malformations present in the

single mutants. The compound mutants also show novel alterations in the

forelimb stylopod (humerus) and additional vertebral transformations. These

observations suggest that Hoxa-9 and Hoxd-9 genetically interacted in forelimb

and axial skeleton patterning (FromentaCRamain et al., 1 996).

The Hox genes are involved in the genetic control of the identity of

specific regions in the hindbrain and spinal cord but they do not specify the

identity of the rostra1 brain region. Therefore, how the anterior region of the

brain is specified becomes one of the most interesting questions in early pattern

formation. However, until recently little was known about this field either in flies

or in vertebrates. A breakthrough has come with the identification of other

homeobox genes in Drosophila, namely orthodenticle (of@ (Finkelstein et al.,

1990) and empty spiracles (ems) (Dalton et al., 1989). In the following section, I

will discuss the roles of atd and ems in Drosophila anterior head development.

development

otd

otd was originally found in a large screen for loci that affect development

of the larval cuticle (Wieschaus et al., 1984). R encodes a homeodomain-

containing protein (Finkelstein et al., 1990). suggesting that it may function as a

transcriptional factor during development. The expression of ofd was detected

as early as 2.5 hr, the stage of blastoderm, by in situ hybridization to wild-type

embryos (Finkelstein et a!.. 1990). It is expressed in a large circumferential

stripe at the anterior end of the embryo. This expression domain includes cells

which will give rise to many of the future head structures (Jurgens et al.. 1986).

According to the fate map, several head structures derived from the anterior

expression domain of otd are deleted in mutant embryos (Cohen and Jiirgens,

1990; Finkelstein and Perrimon, 1990). The early otd expression pattern and

the head defects in mutant embryos suggest that otd plays an important role in

anterior patterning of the Dt-osophila embryo. At about 5-6 hr, shortly after

gastrulation, a second domain of otd expression was detected in a longitudinal

strip of cells along the ventral midline of the embryos. This expression domain

includes cells which are neuronal, glial and epidermal precursors (Jacobs and

Goodman, 1989a.b). In otd mutant embryos, neural and epidermal defects have

been found in the ventral medial region (Finkelstein et al., 1990; Klambt et al.,

1991; Wieschaus et al., 1992), suggesting that otd is also required for the

specification of ventral midline cells in the CNS and epidermis.

ems

ems was found in a search for zygotic patterning mutations (Jurgens et

al., 1984). ems mutant embryos demonstrate severe patterning defects in the

anterior head structures (Jiirgens et al.. 1 984; Dalton et al.. 1 989). suggesting a

role for ems in the patterning of the developing head in Drosaphila. The ems

gene was isolated later using the even-skipped homeo box as a probe (Dalton

et al., 1989). Like the otd gene, ems also encodes a homeodomaincontaining

protein and thus probably functions as a transcription factor during

development. ems is expressed in a stripe at the anterior of the embryo as early

as the blastoderrn stage (Dalton et al., 1989). These ems expressing cells are

fated to give rise to a variety of head stmctures, such as the antenna! sense

organs, the dorsal arms, and the vertical plates and mandibular segments

(Jiirgens et al., 1986). In ems mutant embryos, all of these structures are

missing or disrupted (Jurgens et al., 1984; Dalton et al., 1989), suggesting that

ems plays a role in the patterning of the anterior head. A second expression

domain of ems appears at the beginning of germ-band extension stage in the

anterior lateral ectoderrn and seems to be required for the morphogenesis of

the posterior tracheal tubes (Dalton et al., 1989).

The cloning of the otd and ems genes provided the first molecular

evidence for homeobox-containing genes being involved in the most anterior

patterning of the embryo. Many labs including our lab started to search for their

vertebrate homologues and to explore the molecular mechanisms underlying

patterning of the most anterior regions of the embryo.

an homologlbes of bid and ems and the r exoresslon ~merns durinq

ouse e m b r y ~ e n e s ~ ~

In recent years, vertebrate otd homologues. O k - 1 and &-2, have been

cloned from Xenopus (Bally-Cuif et al.. 1995; Pannese et al., 1995), zebrafish

(Li et al., 1994; Mercier et al.. 1995). chick (Bally-Cuif et al., 1995), mouse and

human (Simeone et al., 1992a. 1993; Ang et al.. 1994). Vertebrate ems

homologues, Emx-1 and Emx-2, have also been isolated and characterized in

the mouse (Simeone et a!., 1992b). In this section, I will summarize the

expression patterns of mammalian homologues of otd and ems during mouse

em bryogenesis.

Mouse Otx-7 and Otx-2

Otx-2 is one of the earliest genes showing restricted anterior expression

domains. At E5.56.0, Otx-2 is expressed throughout the embryonic ectoderm or

epiblast, which gives rise to the embryo proper, but not in the extraembryonic

tissues (Simeone et al., 1993; Ang et al., 1994; Acampora et al., 1995). At E6.0-

6.5, the expression pattern of 00~-2 in the epiblast appears to be the same.

Between E6.5-7.75, Otx-2 expression becomes progressively restricted to the

anterior half of the embryo and later becomes further restricted to the

anterionost third of the embryo. The expression domain includes the anterior

three germ layers. At €8.0-8.5, Obr-2 is expressed in forebrain, midbrain. and

optic eminence of the central nervous system (CNS). Weaker expression is also

found in the notochord, head mesenchyme and foregut at the same axial levels.

as well as in the ectoderm and endoderm cells of the first branchial arch. At

E9.5, the sharp posterior boundary of OtK-2 expression corresponds precisely to

the border between the midbrain and hindbrain. The domain of Otx-2

expression covers almost the entire forebrain and midbrain except the regions

of the optic chiasrna and optic recess. Expression of Otu-1 starts later than Otx-2

and is first observed at approximately at E8.0. R is expressed in similar regions

with Otu-2 except some presumptive ventral forebrain regions and at E8.5, both

genes have the same posterior boundary of expression between the midbrain

and hindbrain. The two genes are also expressed in the developing eyes, ears

and the epithelia cells in the nasal cavities (Simeone et al., 1993).

Mouse Emx-1 and Emx-2

E m - 2 is first expressed in the anterior dorsal neuroectodermal regions

of the embryo at E8.5 (Simeone et al., 1992b). At E9.5, Emx-7 is expressed in

similar regions but the expression domain is smaller than Emx-2. The posterior

boundary of the Emx-1 expression domain coincides with the region between

presumptive diencephalon and telecephalon. Besides dorsal telencephalon,

Emx-2 is also expressed in restricted regions of the diencephalon: the anterior

dorsal and posterior ventral diencephalon. At E9.75, the expression of Emx-2 is

detected in some mesencephalic regions (Simeone. et al., 1992b) and at €13.5,

both genes are expressed in the developing lens, but Emx-2 is also expressed

in the olfactory epithelia in nasal cavities and posterior hypothalamus (Simeone

et al., 1992b).

In summary, the expression domains of Otx and Emx genes are

contained within each other, in the sequence E m x l < E m x - 0 - 2 Figure

1.6 schematically summarizes the expression patterns of the four genes in the

forebrain and midbrain at E10. These expression studies suggest that OW-I and

Otw-2 may cooperate to determine the territory of forebrain and midbrain while

Emx-1 and Emx-2 are involved in the patterning of dorsal telecephalon. The

Fig. 1.6. Comparison of expression of Otx and Emx genes in the mouse embryo

at E10. (modified from Finkelstein and Boncinelli, 1994). For details see text.

Abbreviations: Di: diencephalon; Mes: mesencephalon; Met: metencephalon;

Tel: telencephalon; My: myencephalon.

early expression pattern of Otx-2 may suggest that Obr-2 is also involved in

specifying the developing brain in early embryogenesis.

ne of the thesis

The major goal of the research described in this thesis is to understand

the role of Otx-2 gene, one of the mouse homologues of Drosophila otd, in

mouse embryogenesis. Chapter 2 describes the identification and

characterization of the mouse Otx-2 gene. Chapter 3 describes the functional

analysis of Otx-2 gene by gene targeting. Chapter 4 describes the further

analysis of the role of Mx-2 in anterior midline patterning by making double

heterozygous mutants with HNF-38 , a member of the winged-helix transcription

factor family. Chapter 5 will discuss the significance of the thesis work, the

future experiments and recent important progress in the field of anterior

patterning.

Chapter 2

Identification and Characterization Of the 0txm2 Gene

A portion of this chapter appeared in the following publication:

Ang, S.-L.. Conlon, R.A., Jin, 0. and Rossant, J. (1 994) Positive and negative

signals from mesoderm regulate the expression of mouse Otx-2 in ectoderm

explants. Development 120, 2979-2989.

I performed all work described here, with exception of isolation of the 1 kb Otx-2

cDNA, sectioned in situ hybridization (Fig. 2.8) and most whole-mount in situ

hybridization experiments (Fig. 2.7A. B, C, 0).

INTRODUCTION

How the anterior region of the vertebrate embryo is specified is still an

unresolved question in early pattern formation. Studies on the Drosophila

homeotic complex ( H O W ) genes and their vertebrate homologues Hox

(homeobox) genes, reveal that molecular mechanisms involved in axial

patterning are remarkably conserved during evolution (reviewed by McGinnis

and Krumlauf, 1992). In particular, specific Hox genes are involved in the

regional specification of the hindbrain and spinal cord. However, until recently,

little was known about the development of the most anterior region of the animal

embryo. A breakthrough has come with the identification of other homeobox

genes in Drosophila, namely orthodenticle (atd) (Finkelstein et al., 1990) and

empty spiracles (ems) (Dalton et al., 1989). These two genes are expressed in

overlapping domains in the anterior pole of the blastoderm stage of embryo.

Absence of either of the two genes leads to loss of specific head structures,

suggesting that both otd and ems are involved in the establishment of different

head structures. Recently, mammalian homologues of otd and ems have been

cloned, namely Otu-I, Otx-2, Emx-1 and Emx-2 (Simeone et al., 1992a, 1992b).

Expression analysis at E9.5 showed that these genes are expressed in nested

A-P domains in developing brain (Simeone et al., 1992a).

We cloned Otw-2 independently and analyzed the early expression

pattern of Otx-2 during embryogenesis. At pre- and early streak stages of

embryogenesis (E6.0-E6.5), Otw-2 is expressed throughout the epiblast which

gives rise to the embryo proper. As development proceeds, the expression of

Otx-2 is gradually restricted to the anterior half of the embryo including the

anterior three germ layers. At the somite stage of embryogenesis, the

expression of the Otx-2 is in the presumptive forebrain and midbrain. The

expression profile of Otr-2 suggests multiple roles of Otx-2 gene during early

em bryogenesis.

Methods and Materials

clones

Using a human otd-related cDNA clone EST01828 (Adam et al., 1992).

an 8.5 day embryonic mouse cDNA library (a gift of Brigid Hogan; Frohman et

al., 1987) was screened under low stringency hybridization conditions (5x

Denhart's, 5xSSC. 0.1 % SDS at 42'C). Three independent phage clones were

isolated and inserts were subcloned into pKS plasmid vector. One clone,

pkSotd9, showed specific expression in the anterior part of the embryo by

whole-mount RNA in situ hybridization and, when sequenced, contained a

partial horneodomain with homology to otd. The insert from this clone was used

to screen a 129Sv genomic library in Dash2 (gift of A. Reaume and R Zirngibl)

using high stringency hybridization conditions (5x Denhart's, 5x SSC,

O.l%SDS at 65'C). Two overlaping genomic clones, spanning about 25kb

were isolated, and a detailed restriction map was constructed. To obtain large

cDNA clones, three cDNA libraries, one 11.5 day (a gift of C.C Hui), and two

12.5 day cDNA libraries(gifts of M. Hanks and J. McGlade) were screened with

the 1 kb cDNA probe or with a 170 bp Acc I-Kpn / fragment under high

stringency hybridization conditions (5x Denhart's, 5x SSC, 0.1 %SDS at 65°C).

In total six cDNA clones were isolated and double strand DNA was prepared for

sequencing. Sequence was carried out using Sequenase according to the

method described by United States Biochemical Corporation (USB) .

In situ hybridizations were carried out on whole-mount and sectioned

material as described (Conlon and Rossant, 1992). Single-stranded RNA

probes labeled with digoxigenin-or 3%-labeled UTP were synthesized from

linearized template DNA as directed by the manufacturer (Boehringer

Mann heim Biochemicals). The OfK-2 cDNA containing plasmid, pOtd9, was

linearized with Xbal and transcribed in vitro using T3 polymerase to obtain an

antisense transcript.

Total RNA was isolated from ES cells according to the procedure

described (Chomczynski and Sacchi, 1987). BRACE was performed following

the instructions provided with the B'RAGE system (BRL). Briefly. total RNA

prepared from ES cells was used as template for reverse transcription primed

by Exten 1 (5'CCGCCITACGCAGTCAATGGGCTG-3'). The first strand cDNA

was treated with RNase H for 10 rnin to degrade the RNA template, and it was

then purified using a Glassmax spin cartridge column (BRL). A poly (C) tail was

added to the 5'end of the cDNA using terminal transferase. Second strand

cDNA synthesis and first round PCR were accomplished by an anchor primer

(BRL:5'-CUACUACUACUAGGCCACGCGTCGACTACGGGGGGGGGG-3')

and a Exten 1 primer. A second round PCR was performed using UAS primer

(BRL:5'-CUACUACUACUAGGCCACCGCGTCGACTAGTAC-) and Exten2

primer (5'-CCCTGACCCllTCCATTTCCAGTCGAATCGAGA-3'). DNA

fragments amplified by PCR were cloned and then sequenced to identify 5' 0tx-

2 cDNA sequence.

Northern and orimer extension a n d v s i ~

Total RNA was prepared from ES cells and embryos as described above.

Poly(A) RNA was purified using an mRNA purification kit (Pharmacia). lOpg

total RNA and 1pg mRNA were denatured and electrophoresed on a 1 %

formaldehyde agarose gel and then transferred to a Genescreen Plus

membrane (DuPont). The 1 kb Eco Rl fragment was used as a probe with the

random priming method. Hybridization was carried overnight at 42'C in 50%

formamide, 5x Denhart's, 5x SSC, O.l%SDS, 100rng/ml sheared salmon sperm

DNA. The filter was washed at room temperature twice and then exposed to

Kodak X A R 5 film for two days. After that, the filter was washed again at 55%

twice, and then reexposed for three days. Then the filter was stripped by boiling

and reprobed with a mouse O-actin probe.

Primer extension was carried out as follows. A single stranded

oligonucleotide (Exten 3: 5'-GATAGCTGGCTATITGGAATTTGAAGG-3').

which is complementary to 00r-2 mRNA beginning 194 upstream of the ATG

start codon, was end-labeled using T4 polynucleotide kinase and # 2 ~ ] - ~ ~ ~ .

The labeled oligonucleotides were then hybridized to 5pg and 10pg of poly (A)

RNA or 25pg total RNA derived from ES cells and were extended with reverse

transcriptase at 42°C for 2 hr. The extension products were visualized by

autoradiography after fractionation on a 6 % sequencing gel.

Results

Using a partial human cDNA with homology to the Drosophila Oid gene

(Adam et al.. 1992), S.L. Ang screened a mouse 8.5 day cDNA library and

isolated a single I kb cDNA clone under low stringency hybridization conditions.

I sequenced this clone and found that it contains only a partial 3' homeodomain

(1 1 amino acids) with 100% identity to that of Otd (Finkelstein and Perrimon,

1990). Later, I isolated a 1.6 kb cDNA from a 12.5 day cDNA library.

Analysis of possible reading frames reveals only one open reading frame

(ORF) of significant length (Fig.2.1). The open reading frame is 866

nucleotides long. It is proceeded by a 5' untranslated region of 228 bp and

followed by a 3' untranslated region of 535 bp. At the dend, no polyadenylation

signal (Proudfoot and Brownlee, 1976) is present. However, there is a 13bp

poly(A) tract at the 3' end of cDNA. Sequence analysis from the corresponding

exon (exon 5) also reveals a 15bp poly(A) tract in the genomic region. So it is

likely that the cDNA is the result of internal priming within the 3WTR of the

mRNA. While we were isolating this cDNA, Boncinelli's group published that

there were two mammalian CWs (Simeone et al., 1992a). Sequence

comparison of the horneodomain showed that the cDNA we isolated

corresponded to the gene they called Otx-2. Fig. 2.2 shows a comparison of

a x - 2 with other Obc related homeodomain sequences in different species. The

horneodomains of the predicted mouse Otx-1 and Oh-2 proteins are extremely

similar to those of the Xenopus, sea urchin and Drosophila Otd proteins. In

addition, they all share the lysine residue near to C-terminal end of the

homeodomain. This residue is also found in the products ofDrosophila bicoid

(bcd) and goosecoid (gsc) genes (Frigerio et al., 1986. Blumberg et ai., 1991) . This particular lysine has been shown to play a key role in in vitro DNA binding

specificity (Hanes and Brent, 1989; Treisman et al., 1989).

orthern blot an on of 5'-0tx - 2 cDNq

Fig. 2.3 illustrates a Northern blot analysis using the 1 kb Otx-2 cDNA as a

probe. The probe detected a 2.5kb transcript in ES cells, 10.5 and 12.5d

embryos. However, another 6.6kb band was detected in 10.5d and 12.5d

embryos but not in ES cells, although the signal is weaker than that of the 2.5kb

band. The longest cDNA I have is 1.6kb in length and 900bp shorter than the

mRNA detected in ES cells and El 0.5/€11.5 embryos by Northern blot analysis.

In order to identify regulatory elements upstream of the Otx-2 gene, it is

necessary to identify the transcription initiation site. Two additional 1 1.5 and

12.5 day cDNA libraries were screened in order to obtain more 5' sequence of

Otx-2 cDNA. However, partial sequence analysis of all clones showed that

none extended beyond the 5' end of the 1.6 kb cDNA. In order to get more 5'

sequence of Otx-2, the 5'-RACE technique was employed with ES cell RNA.

Two RACE products were cloned and sequenced. One of the products

contains 240bp 5'-sequence of the Obr-2 gene (Fig. 2.4). Comparing with the

1.6kb cDNA, the product contains an addiiional 70 nucleotides but is still not full

length. Primer extension was used to identify the transcription initiation site(Fig.

2.5). It demonstrated that there are possibly two or three transcription initiation

start sites which are about 523bp upstream of the ATG start codon. Thus, the

largest cDNA and RACE product we isolated suggested that it may be still

missing about 225 bp of 5'-UTR and 576 bp of 3' WR. 6RACE cloning will be

needed to clone the remaining 5'-sequence.

on-lntran strucme of the Otx - 2 gene

After screening the 129Sv genomic library using the 1 kb EcoRl cDNA

fragment as a probe, two overlapping genomic clones, G3 and G9, spanning

about 25kb were isolated. These two 15kb genomic fragments were then

subcloned into plasrnid vectors (pKS), and characterized by restriction

mapping, subcloning, and partial sequencing. The exon-intron organization of

Otr-2 gene was determined by comparison of the corresponding regions of

cDNA and genornic clones and is illustrated in Fig. 2.6. Comparison of the 3'

end of the Ofw-2 cDNA with Exon 5 reveals no consensus 3' splice site,

indicating that the 1.6 kb cDNA is incomplete at 3' end. An intron is present in

Otx-2 immediately upstream from the homeodomain, as is often the case for

homeobox genes (Boncinelli et al. 1991). An additional intron is present in the

Otr-2 gene within the homeobox. The results from 5' RACE and primer

extension experiments (Fig. 2.5) suggest that about 225 bp of 5' UTR of Otx-2

cDNA is missing. Whether the exon(s) encoding this region are present within

the G3 genomic clone is not clear. However, the missing exons may be quite far

away from the following exons, as is the case of SpOtx(B) in sea urchin (Li et

al., 1997).

Fig. 3.7 illustrates the early expression pattern of the OW-2 gene. At €6.0-

6.5, Otr-2 appears to be already expressed throughout the embryonic

ectoderm, or epiblast which gives rise to the embryo proper, but not in the

extraembryonic tissues. Between E6.5-7.75. Otr-2 expression becomes

progressively restricted to the anterior half of the embryo and later becomes

further restricted to the anteriormost third of the embryo. The expression domain

includes the anterior three germ layers (Fig2.8C). At 8.0-8.5 dpc, &-2 is

expressed in forebrain, midbrain, and optic eminence of the central nervous

system (CNS) (Rgs2.7D; 2.8D). Weaker expression is also found in the

notochord, head mesenchyme and foregut at the same axial levels. as well as

in the ectoderm and endoderm cells of the first branchial arch (Fig.2.80 and

data not shown). At E9.5, the sharp posterior boundary of Otx-2 expression

corresponds precisely to the border between the midbrain and hindbrain (Fig.

2.7E).

Discussion

I cloned OtK-2, a gene related to the Drosophila otd homeobox gene, and

analyzed the expression pattern of Otx-2 during early embryogenesis. In my

Northern blot analysis I detected a 2.5kb transcript in the ES cells, E l 0.5 and

12.5 embryos. However, l also found a 6.6 band from €1 0.5 and El25 of

embryos but not in ES cells (Fig. 2.3). Southern blot analysis with the same

probe yield a single or two bands when genornic DNA was digested with three

different restriction enzymes (data not shown). These bands correspond to the

genomic clone. tt suggests that Ofx-2 is a single copy gene, although more

distantly related genes may exist. The finding of two transcripts raises a

question whether different promoter utilization and alternative splicing of Otw-2

gene might be important in mouse embryogenesis. Recently, differential

promoter utilization and alternative splicing of Otx have been found in sea

urchin (Li et al., 1997). Four different SpOtx mRNAs were found during sea

urchin embryogenesis. All of these mRNA were products of a single SpOtx gene

resulting from differential promoter utilization and alternative splicing. They

code for two different SpOtx proteins that differ only in their N-terminal regions.

To further investigate the origin of the 6.6 band, the following experiments will

be performed. First, mRNA from early to late stages of embryogenesis will be

used for Northern blot analysis to get the whole profile of this band. Since five

exons so far have been identified and the previous probe used for Northern blot

was the fifth exon, all the other four exons, especially the third exon which

contains most of the homeodomain sequence, will be used individually for

further Northern blot analysis to search whether these four exons are

responsible for this 6.6 band. If all or the third exon can hybridize to this 6.6

band, it may suggest that this 6.6 band is alternative form of Otx-2. If not. it

would indicate that the 6.6 band represents part of another gene which has

sequence homology to the fifth exon.

Unlike hindbrain and spinal cord, little was known about the molecular

mechanism involved in patterning the anterior regions (forebrain and midbrain)

of the embryo. However, identification of the homeobox genes otd and ems in

Drosophila (Dalton et al.. 1989; Finkelstein et al.. 1990) and their mammalian

otdrelated genes, Otw-1 and Otx-2, along with the ems-related genes, Emx-1

and Emx-2 (Simeone et al., 1992b) open a useful way to investigate how the

anterior of embryo is specified. The mammalian Obc and Emx genes are

expressed in nested anterior-posterior (A-P) domains in the anterior C N S

(Simeone et al., 1992a). The Otx-2 gene shows the earliest restricted anterior

expression domain. It is already expressed in epiblast as early as E6.0 of the

mouse embryo. Between 6.5 - 7.5 dpc, Otx-2 expression becomes progressively

restricted to the anterior half of the embryo and later becomes further restricted

to the anteriormost third of the embryo. At 8.0 - 8.5 dpc, Otx-2 is expressed in

the developing anterior neural tube with a posterior boundary corresponding to

the border between

expression domain

the presumptive midbrain and hindbrain. By 9.5 dpc, the

spans the developing forebrain and midbrain. We also

detected Otr-2 expression in anterior mesoderm and endoderrn during early

gastrulation. The progressive restriction of Otx-2 expression to the anterior of

the embryo by headfold stage correlates with the anterior migration of

mesendoderm. Early expression of Ofx-2 raises the issue of whether the Otw-2

plays a direct role in specifying anterior structure.

The expression pattern of Otx-2 is largely consistent with the recently

reported results in Xenopus (Pannese et al., 1995; Blitz et al., 1995) and chick

(Bally-Cuif et al., 1995). In Xenopus, XOLx-2 is expressed throughout early

development from unfertilized egg to late blastula with increasing level of

expression. At stage 9.5, expression of XOtw-2 is localized in the dorsal region

of the marginal zone and at stage 10.25 it is in dorsal bottle cells and cells of the

dorsal deep zone fated to give rise to prechordal mesendoderm. At stage 10.5,

the expression of XOtx-2 extends to presumptive anterior neuroectoderm and it

persists in subsequent stages. Similarly, in the chick, expression of c-Otx-2 is

first detected in all epiblast and associated with cells with presumptive anterior

mesendoderm fate and it correlates with mesendoderm migration toward

anteriormost regions of the embryo. Subsequently it extends to anterior

neuroectoderm.

Early expression of OW-2 raises the issue of whether Otx-2 plays a direct

role in specifying anterior structure. The study of experimentally manipulated

Xenopus embryos suggests a role for XOtx-2 in development of anterior

structures (Pannese et al.. 1995; Blitz et al., 1995). Microinjection of XOtx-2

mRNA into I-, 2-, and 4cell stage embryos produces embryos with severely

reduced trunk and tail structures and the appearance of secondary cement

glands. The cement gland is one of the most anterior structures of the

developing Xenopus embryo. The induction of a secondary cement gland by

ectopic expression of XOtx-2 indicates a role of XOtx-2 in specifying developing

anterior head structures.

conservation of the expression pattern of Otd-related gene OW-2 in

mouse, Xenopus and chick suggests that the underlying molecular mechanisms

of head patterning may have been conserved throughout evolution.

Overexpression of XObr-2 in Xenopus already suggests a role for this gene in

specifying anterior head structures. Loss-function-mutation of Otx-2 in mouse

will be a powerful way to address its role in anterior patterning of embryo.

Fig. 2.1. The nucleotide sequence of the 1.6kb cDNA clone and the amino acid

sequence of the putative Otx-2 gene product. The underlined sequence

corresponds to the homeodomain, and the asterisk marks the translational

termination codon.

tcgctagaggagctgagtcgccacctctactttgatagctggctctttggaattttgaag gataatttgattttttttttcttttctaacgtccaatgcggctgtaagttccgtcactcc aaatctacccaccaaggaccctgatcctgtccactccaggcgaatcgagaccgtccggct gggtccccccaatttgggccgactttgcgcctaaaaacaaccttagcatgatgtcttatc

M M S Y L taaagcaaccgccttacgcagtcaatgggctgagtctgaccacttcgggtatggacttgc K Q P P Y A V N G L S L T T S G M D L L

tgcatccctccgtgggctaccc~g~~acc~~~~ggaaacagcgaagggagaggacgactt H P S V G Y P A T P R K O R R E R T T F

ttactagggcacagctcgacgttctggaagctctgtttgccaagacccggtacccagaca T R A O T , D V T , E A L F A K T R Y p D L

tcttcatgagggaagaggtggcactgaaaatcaacttgccagaatccagggtgcaggtat F M R E E V A ~ , K I N I , P E S R V O V X

ggtttaagaatcgaagagctaagtgccgccaacagcagcagcagcagcagaatggaggtc F R N R R A K C R O O O G Q ~ Q N G G Q

agaacaaagtgaggcctgccaagaagaagagctctccagctcgggaagtgagttcagaga N K V R P A K K K S S P A R E V S S E S

gtggaacaagtggccagtt~agtcccccctctagtacctcagtcccaaccattgccagca G T S G Q F S P P S S T S V P T I A S S

gcagtgctccagtgtctatctggagccccagcgtccatctccccactgtctgaccccttgt S A P V S I W S P A S I S P L S G P L S

ccacttcctcctcctgcatgcagaggtcctatcccatgacctatactcaggcttcaggtt T S S S C M Q R S Y P M T Y T Q A S G Y

atagtcaaggctatgctggctcaacttcctactttgggggcatggactgtggatcttatt S Q G Y A G S T S Y F G G M D C G S Y L

tgacccctatgcatcacca~cttcctggaccaggggccacactcagtcccatgggtacca T P M H H Q L P G P G A T L S P M G T N

atgctgttaccagccatctcaatcagtccccagcttctctttccacccagggatatggag A V T S H L N Q S P A S L S T Q G Y G A

cttcaagcttgggttttaactcaaccactgattgcttggattataaggaccaaactgcct S S L G F N S T T D C L D Y K D Q T A S

cttggaagcttaacttcaatgctgact9cttggatcataaagatcagacgtcctcatgga W K L N F N A D C L D Y K D Q T S S W K

aattccaggttttgtgaagacctgtagaagctctttttgtgggtgatttttaaatatgct F Q V L *

cggctgaacattccagttttagccaggcattggttaaaaaagttagatggaacgatgctc tcagactcctgatcaaagttaccgagaggcatagaaggaanaaggaaggggccttagaag ggtccatcaaccagcaacctgaaatggacaaaccaatctacttaagattctgttatagtt ctagatcattggtttcctgatttgcaaatgattgatcaaanatattctagcgacatgcaa ccaaataccactcaaaacaaaaatccagcaaaactgagttgtgagggaagggagggaagg tcatggccttcaaagcagaggtgatccggtgttttagccaatctttggttgaatgttagg aatggacaatgtcccaggctcattcacgtttcatgaccaacaggtagttggcactgaaaa acttttcagggctgtgtggttgtgcgactgattgtcctagatgcagtactttatttaaaa aaaaaaaaa 16 2 9

Fig. 2.2. Comparison of Otx-2 homeodomain with Otx-I , Xotx-2(Xenopus),

SpOtx(sea urchin) and Otd (Drosophila) homeodomains. Red amino acids

indicate differences with Otx-2. The lysine residue at position 9 of the third helix

is marked with pink arrow.

Fig. 2.3. Transcription of the OW-2 gene. Lanes 1 and 2 contain 1 pg poly(A)+

RNA derived from El 0.5 and E12.5 mouse embryos, respectively. Lane 3

contains 10 pg of total RNA derived from ES cells. The probe is 1 kb Eco FU

fragment from pKSotd9. The blot first (A) was washed at room temperature

(details see methods and materials) and then (B) at 55°C. The same Northern

blot that had been stripped and rehybridized with an mB-actin probe((;).

Fig. 2.4. Diagram of the 5' RACE and primer extension experiments (A) and

sequence of 5' RACE product (8). (A) Primer Exten 1 was used to synthesize

first strand cDNA. Anchor Primer was added later for first round PCR (details

see method and materials). Second round PCR was performed using primer

Exten 2b (with Eco RI cloning site) and UAP (Universal Amplification Primer with

Sal I cloning site). Primer Exten 3 was used for primer extension experiments.

(6) The nucleotide sequence of 5' RACE product. The underlined sequence

highlighted in red is primer Exten 2b.

Anchor Primer - UAP -

\\ Otx-2 mRNA - Exten 3 (for primer extension)

240bp from RACE

(3 2 5 bp estimated by Primer Extension)

TGCAAATCTCCCTGAGAGCGGAACCTCCTCAGCTCCAACTAAGCCnCCACKTTACTAAAAAAT AAAAATCGCTAGAGGAGCTCAGTCGCCACCTCTACTlTGATAGCTGGCTAmGGAAmGAAGG ATGGmGAmC'TmCTAACGTCCAATGCGGCTGTAAGTCCGTCACTCCAAATCTACC CACCAAGGT CCCTGACCCTTTCCAICCAGTCGAATCGAGA

Exten 2b

Fig. 2.5. Result of primer extension experiment. Lanes 1 and 2 contain 5 and 10

pg of poly(A) RNA derived from ES cells, respectively. Lane 3 contains 25 pg

total RNA derived from ES cells. Primer Exten 3 was used in this experiment.

Fig. 2.6. Exon-intron structure of Ofx-2 gene. (A) The overlapping two genomic

clones are represented by the thin lines above the gene. (B) Lightly shaded

rectangles represent exons, darkly shaded rectangles denote the

homeodomain, striped ban represent open reading frame (ORF), pKSotd9

represents the cDNA clone containing a 1 kb Otx-2 cDNA, pEXloxOl.1

represents the clone containing a 1.6 kb Otr-2 cDNA. pKS-5-RACE2 obtained

from 5'RACE cloning contains 240bp Otw-2 cDNA which was used to identify the

two additional exons upstream of the ATG-startcodon-exon .

Fig. 2.7 Expression of Otw-2 from prestreak to early somite stages revealed by

whole mount RNA in situ hybridization. (A) Pre- to early streak stage (E6.0-E6.5)

embryos showing widespread expression of Otx-2. (B) Mid- to late streak stage

(E7.0-€7.3). Expression became progressively restricted to the anterior half of

the embryo. (C) Headfold stage (E7.5E7.7). Further restriction of Otx-2

expression to the anterior third of the embryo. (D) At the somite stage (E8.0-

E8.5), Otx-2 expression in the neuroectoderm was found in the forebrain and

midbrain regions. (E) At E9.5 the sharp boundary of Otx-2 expression

corresponds precisely to the boundary between the midbrain and hindbrain

(arrows). Scale bars in D and E represent 100 pm, while the scale bar in A

represents I 00 p n for the other embryos.

Fig. 2.8 Tissue sections analyzed for Otx-2 expression. (A,B) Dark-field and

bright-field views of a sagittal section of an early streak stage embryo analyzed

by radioactive RNA in siiu hybridization showing widespread expression ofOtx-

2 in the ectoderm and delaminating mesoderm in the posterior end. (C)

Parasagittal section of a headfold stage embryo previously stained by whole

mount in situ hybridization showing Otx-2 expression in all three germ layers at

the anterior end.(D) Frontal section of a 10- to 15-somite stage embryo assayed

by whole mount RNA in situ hybridization showing Obr-2 expression in optic

eminence (oe) , diencephalon (di), notochord (No), foregut (fg), ectoderm cells of

the first branchial arch (ba) and endoderm cells surrounding the first branchial

pouch (arrowed). Scale bar represent 20 pm. abbreviations: A: anterior; P:

posterior; ect: ectoderm; mes: mesoderm; en: endoderm; ne: neuroectoderrn;

am: amnion.

Chapter 3

Targeting of the Ofx-2 Gene

This chapter is a modified version of the following publication:

In. 0.. Rhinn. M.. Daigle, N., Stevenson, L. and Rossant, J. (1996). A

targeted mouse Otx-2 mutation leads to severe defects in gastrulation and

formation of axial mesoderm and to deletion of rostra1 brain.Development 1 22:

243-252

The first two authors contributed equally to this work.

I am responsible for the following work: generating two targeting vectors (Fig.

3.1), most of the work in the table 3.1, a portion of the work in the table 3.3, the

work in figures 3.2,3.3,3.4 (with exception of the preparing wax sections) and

fig. 3.5.

INTRODUCTION

The patterning and development of the vertebrate neural tube is a

complex process involving both cell extrinsic and cell intrinsic events. Among

the cell extrinsic events, the mesoderm in the organizer region, namely the

dorsal blastopore lip in amphibians, Hensen's node in birds and the node in the

mouse, is able to induce neural differentiation in the surrounding ectoderm

tissue (Spemann, 1938; Waddington, 1933; Beddington, 1994). Furthermore, in

vim experiments have demonstrated that the prechordal mesoderm and

notochord, descendants of the organizer, can also induce and pattern the

neural tube along the anteroposterior(A-P) axis in Xenopus embryos (reviewed

in Slack and Tannahill, 1992; Doniach, 1993; Ruiz i Altaba, 1994). Support for

a role of the prechordal mesoderm in the induction of forebrain and midbrain

has also come from the analysis of the phenotype of hornozygous Lim-1 mutant

embryos: a defect in prechordal mesoderm cells in these mutant mice is

suggested to be responsible for the subsequent loss of anterior brain tissues

(Shawlot and Behringer, 1995). However, the role of the notochord in A-P (A-P)

patterning of the neural tube is still in question, since embryos homozygous for

a mutation in the gene H W B lack an organized node and notochord but show

relatively normal A-P patterning of the central nervous system (CNS) (Ang and

Rossant, 1994; Weinstein et al., 1 994).

A major contribution to the identification of cell intrinsic molecules

responsible for neural tube regionalization has come from cloning of genes

homologous to homeobox-containing genes within the HOM-C complex in

Drosophila (reviewed in Lawrence and Morata, 1994), namely the Hox genes in

mice. The function of these genes in the regional specification of hindbrain and

vertebrae has been investigated using both loss-of-function and gain-of-

function studies (reviewed by Krumlauf, 1994). Hox genes are not expressed in

the forebrain and midbrain, suggesting that some other classes of horneobox

genes are involved in the development of these rostral regions. Recently, two

new classes of homeobox genes, related to the Drosophila orthoden tide (otd)

(Finkelstein and Perrimon, 1990) and empty spiracles (ems) gene (Dalton et al.,

1989), have been cloned. These genes, Otx-I, Obr-2,, Emx-1 and Emx-2? are

expressed in nested domains in the forebrain and midbrain regions (Simeone

et al., IW2a). Since otd and ems have been shown to participate in a

regulatory network required for head formation in flies, it has been suggested

that the conserved murine genes serve similar roles in the patterning of rostral

brain in mice (Finkelstein and Boncinelli, 1 994).

The mouse owrelated genes. Otx-1 and Otx-2, belong to the bicoid-class

of homeobox genes. The amino acid sequences of the homeodomains of their

protein products differ by two and three amino acids from that of the otd gene

product respectively (Simeone et al., 1993). These genes are expressed in

overlapping domains in the anterior GNS with the domain of Otx-1 expression

contained within the Otu-2 domain. Otx-2 is already expressed by embryonic

day 5.5 (E5.5). while the expression of Otr-I mRNA is not detected until early

E8.0. Otx-2 expression at E5.5 is widespread in the epiblast, which gives rise to

the embryo proper. From the early primitive streak to headfold stages, Otr-2

expression in the ectoderrn becomes restricted to the anterior end of the embryo

(Simeone et al., 1993, Ang et al., 1994). We and others have previously shown

that this Otx-2 expression in the anterior ectoderm depends on interactions with

the underlying mesoderm at the anterior end of the embryo in mice and

Xenopus (Ang et al.. 1994; Pannese et al., 1995; Blitz and Cho. 1 995). Obr-2

expression was also found in anterior endomesoderm tissues, including axial

mesoderm tissues, such as notochord and prechordal mesoderm (Ang et al.,

1994; Simeone et al.. 1995). The earlier expression of 0tx-2, compared toOtx-

1, and its expression in axial mesoderm tissues that possess neural patterning

capabilities suggests that this gene could be involved in the patterning of

anterior neural tissues.

To begin to dissect the roles of OW2 in vivo, I have generated a

homeobox deletion in the gene using homologous recombination in ES cells.

This mutation results in early gastrulation defects. Homozygous Otw-2 mutant

embryos also show defects in the prechordal mesoderm and notochord

precursors by the headfold stage. By E8.25, rostral deletion of the neural tube

to rhombornere 3 was clearly apparent. These defects are consistent with

multiple roles for 0tx-2 in gastrulation and the patterning of rostral brain in mice.

MATERIALS AND METHODS

A l kb mouse 00r-2 partial cDNA probe was used to isolate two

overlapping genomic clones, containing the entire coding region of the Otu-2

gene from a 129SVlJ genomic library. To construct the first targeting vector,

pPNT02. a 1.7 kb Bgl 11-Xba I fragment that maps 5' to the OtK-2 homeobox was

subcloned into the BamH I-Xba I site of pPKT (Tybulewicz et al., 1991). A 4.4 kb

Stu I fragment located 3' to the homeobox was subsequently cloned into the

Xho I site of the above vector. The PGKneo and PGKtk cassettes were in the

opposite transcriptional orientation compared to the endogenous Otx-2 gene

(Fig. 3.1). To construct the second targeting vector, pPNTKSN, a 6.5 kb Sma I

fragment from the 5' region of Utx-2 was first subcloned into the Xho I site of

pPNT. A 2.5 kb Kpn I fragment from the 3' end of the gene was then subcloned

into Kpn I site of the same vector. in pPNTKSN, the PGKneo and PGKtk

cassettes were in the same transcriptional orientation with respect to the OtK-2

fragments as the endogenous Otx-2 gene (Fig. 3.1).

Generation of the mutation

The R1 ES cells (Nagy et al., 1993) were cultured and electroporated

with Nod-linearized pPNTO2 and pPNTKSN as described (Wurst and Joyner,

1 993). Doubly resistant cells were selected in a concentration of 1 50 pg/ml of

active G418 and 2 pM gancyclovir for 10-1 1 days before picking. Colonies were

picked onto gelatinized 96-well plates and grown to near confluency before

splitting into two 96-well plates. The master plate was frozen down, and ES cell

genornic DNA from the other plate was analyzed by Southern blotting for

homologous recombination events.

Genomic DNA from these cell lines was digested with EwR I and Xho I

and probed with a 2.0 kb Hindl 11-Xbal 3' flanking probe and a 1.7 kb 8gl 11-Xba I

5' internal probe. Hybridization was carried out overnight at 42OC in 50%

formamide, 5x Denhart's, 5x SSC, O.l%SDS, 100 pg/ml sheared salmon sperm

DNA.

Chimeras were generated by ES-morula aggregation and blastocyst

injection with targeted ES lines. Chimeric males were bred to CD1 females to

establish F1 heterozygotes. Embryos from intercrosses of F1 heterozygotes

were typed either by Southern analysis or by PCR of yolk sac DNA. To

genotype €7.5 embryos, ectoplacentai cones were isolated and cultured for 1

week in 96 well dishes using Dulbeccots modified Eaglets medium plus 15%

fetal calf serum. Cells were lysed in 35 pi of proteinase K buffer (50 mM KCI,

1 OmM Tris.HCI (pH 8.3). 2.0mM MgC12, 0.1 mg/mL gelatin. 0.45% Nonidet p40,

and 0.45% Tween-20). To detect a 1.3 kb wildtype band, the following primers

were used: sense strand (5'-ATGATGTCTTATCTAAAGCAACCGCCTTACG-3')

and an tisen se strand (5'-TCATTGGGTCATCAGTATAAACCA-3'). The OW-2

mutant allele was detected by amplifying a 650 bp neo fragment using a set of

primers corresponding to the sense strand of the neo gene (5'-

ATCTCCTGTCATCTCACClTGC-3') and antisense PGK poly (A) sequence (5'-

ACCCCACCCCCACCCCCGTAGC-3'). Samples were amplified for 35 cycles

(94°C 40 seconds; 55°C 1 minute; 72°C 1.5 minutes) for the wild-type allele and

for 40 cycles (94°C for 1 minute; 65OC for 1 minute; 72°C for 2 minutes) for the

mutant allele. Amplified bands were visualized by agarose gel electrophoresis

and ethidium bromide staining.

Wholemount in situ hybridization was performed as described

previously (Conlon and Herrmann, 1993). For histology and in sifu

hybridization to sections, embryos were fixed overnight in 4%

paraformaldehyde in PBS. They were then processed and embedded in wax

and sectioned at 5-6pm. Slides were then dewaxed, rehydrated and stained

with hematoxylin and eosin. Probes used for in situ hybridization of sections

were: gsc (Blum et al., 1 W2), Lim-1 (Barnes et al., 1 994), Brachyury (Hermann,

1991). Mox-7 (Candia et al., 1992). HNF-3B (Ang et al., 1993). Krox-20

(Wilkinson et al., 1989b).

RESULTS

tarwted disr~ption of Otx 3 in ES cells bv homolo~pus recambtnat~on . . -

Two positivdnegative targeting vectors were made (Fig.3.1). One

targeting vector (pPNTO2). containing 1.7 kb of 5' and 4.4 kb of 3' genomic

sequence, was designed to delete the homeodomain region of Otx-2, and

replace it with the PGKneo cassette from the pPNT vector (Tybulewicz et al..

1991). Therefore the PGKneo insertion should truncate the Otx-2 coding

sequence immediately after the first exon which encodes 32 amino acids. The

other targeting vector, designated pPNTKSN. consisted of 6.5 kb of 5' and 2.5

kb of 3' genomic sequence and was to designed to delete the exon containing

the predicted initiation codon.

The linearized targeting vectors were electroporated into ES cells (Nagy

et al., 1 993) and clones were selected for resistance to 641 8 and gancyclovir.

A total of 11 94 double resistant ES cell colonies were analyzed by Southern

blot using a 5' internal and a 3' external probe, six cell lines, including 5-23 and

C12, were isolated in which the Ob-2 locus was correctly targeted. Table 3.1

summarizes the screening results. All targeted cell lines are from pPNTO2

vector. Using a genornic probe spanning exon 2, the deletion of the homeobox

region was confirmed in the the genomic DNA of homozygous Ofx-2 mutants by

Southern blot analysis (data not shown).

The 5-23 and C-12 targeted ES cell lines were used to generate

chimeras by ES cell-morula aggregation (Nagy et al., 1993) or by blastocyst

injection. These chimeras transmitted the mutation to their progeny. Mice

generated from both lines showed identical phenotypes. All analyses were

carried out on a mixed CD1/129 background.

Table 3.1. Results of Screens for Homologous Recombination into

Otx-2 Locus

Targeting Vectors G418 and Gancyclovir Recombinants

ES Colonies

DPNTO~ 648 6

fable 3.2. Frequency of Germ-Line Transmission of Targeted E S

Cell tines

Cell

Lines

Total Embryos Number of Number of Germ tine

Transmission

* Cell line 5-23, contributed to germ line later in S.L.Angls lab.

he Obr - 2 m w n leads to e w o n i c lethality

When heterozygous animals were crossed with wild-type CD1 females

and their progeny were genotyped at 3 weeks of age, heterozygous mice were

obtained with at a frequency of 142/311, less than the expected 50% ratio.

These heterozygous mice appeared normal and were fertile. However, when

offspring from heterozygous intercrosses were harvested at E9.5 and E10.5, a

small fraction (81149) of normal size embryos showed an open neural tube

defect at the forebrain and midbrain levels. Caudal to the midbrain region, these

embryos looked identical to wild-type embryos (data not shown). All these

embryos have been genotyped to be heterozygous animals. The Otw-2

mutation thus results in a dominant phenotype that is weakly penetrant on the

CD11129 background, and this dominant phenotype may explain the slightly

lower number of heterozygous animals obtained at 3 weeks of age. This

phenotype will be analyzed in more detail elsewhere.

When the progeny of intercrossed of O W heterozygous animals were

analyzed at birth, no homozygous newborn animals were found, indicating that

Oh-2 is required for embryonic development. To characterize the embryonic

lethality, we analyzed litters from heterozygous intercrosses from E7.25 to

E l 0.5. Homozygous mutant embryos were present between E7.25 and E9.5 at

roughly the expected frequency of 2596, however, at €10.5, the proportion of

mutants obtained declined to 17% and these mutant embryos were either

severely growth retarded or being resorbed. Thus, the Otw-2 mutation leads to

embryonic lethality around E10.5 (Table 3.3). The results from a typical

genotyping analysis of yolk sacs from E8.5 embryos generated from

intercrosses of Obr-2 heterozygotes are illustrated in Fig. 3-28.

s of the Obr - m o 7 y o o u s he no

By E7.25-€7.5 (mid- to late-streak stage), homozygous embryos were

morphologically distinguishable from normal embryos (Fig. 3.38). The

Table 3.3. Genotype of mice resulting from 0-2 heterozygous

*Embryos were either severely growth retarded or being resorbed.

i ntercrosses

abnormal looking embryos were smaller than their littermates, and these

embryos were confirmed to be homozygous mutant by PCR analysis. Sections

through mutant embryos at this stage demonstrated clearly that these embryos

had initiated gastrulation and mesoderm had formed all around the embryo

(Fig.3.4B,C,D). Embryonic ectoderm, mesoderm and endoderm cells are

present in mutant embryos. However, the accumulation of mesoderm cells has

been observed in these homozygous embryos (Fig.3.4B,C9D). In most of the

Stage

7.5

8.5

9.5

10.5

Postpartum

+/+

12

24

16

15

18

OtK-2+/-

27

43

32

23

36

Otx-&/- (%)

1 3(25)

22(25)

1 4(23)

8(17)*

0

mutants, the allantois was poorly developed and formed a round ball of cells, in

contrast to its fairly long and extended appearance in normal embryos at this

stage. It was also often disconnected from the embryonic portion of the

conceptus.

At E7.75 (headfold stage), the headfolds forming at the rostra1 end of

wild-type embryos were not apparent in homozygous mutant embryos. At €8.5

(somite stage), variations in the phenotype of the mutants were found (Fig.

3.3D,E,F). We have divided the mutants at this stage into a less severe class,

which includes embryos enclosed in the yolk sac, and a more severe class in

which the embryos are completely excluded from the yolk sac. The less

severely affected class of mutants exhibited fairly good development of

posterior somites and a rudimentary heart. At the anterior end, the neural tube

looked extremely abnormal with either numerous folds or a single fused

structure instead of open neural folds. Compared to wild-type or heterozygous

embryos, no forebrain or midbrain tissue could be identified suggesting that

deletion of neural tissue had occurred anteriorly. The sornites were not always

normal; in some embryos they were fused at the midline or were irregularly

shaped. The more severely affected homozygous mutant embryos were

smaller than the first class of mutants, wrapped up anteriorly in definitive

endoderm and showed improper segmentation of mesoderm into somites.

Some of the mutants in this class were very thin and spiralshaped and showed

no sign of organogenesis and segmentation of mesoderm into somites

(Fig.3.3F). The two different classes of mutants occurred at approximately the

same frequency.

To determine precisely how much rostra1 brain tissue was deleted in

mutant embryos, several A-P regionspecific neural markers such as Krox-20

and En were used to characterize Otx-2 homozygous embryos (Fig. 3.5 and

data not shown). For these studies, we only used E8.25-8.5 mutant embryos

from the less severe class. Krox-20 was expressed in two bands across the

neural tube (rhombomere 3 and 5) in wild-type or heterozygous embryos.

However, in mutants the first band of expression either completely abolished or

weakly detectable at the anterior most end of the embryos (Fig. 3.5 and data not

show).These results demonstrate that in homozygous mutants. neural tissues

anterior to rhombomere 3 have been deleted.

ous embrvo~

The small size and abnormal morphology of E7.25 embryos

demonstrated that defects had already occurred in homozygous mutants at this

early stage. To determine if defects occurred in the primitive streak, we

analyzed the expression of the genes Goosecoid (gsc) and Brachyury.

Goosecoid is a homeobox gene expressed in the anterior primitive streak

region in mid- to latestreak stage embryos (Fig.3.6A and Blum et al., 1992).

Goosecoidexpressing cells were found ectopically located in the proximal

region of homozygous mutants in one case and were absent in three other

embryos analyzed (Fig. 3.68 and data not shown). Previous studies have

demonstrated that gsc is expressed transiently in the anterior primitive streak in

wild-type embryos (Hum et al., 1992). Since it is difficult to accurately stage the

homozygous mutant embryos due to their abnormal morphology, we cannot

distinguish whether the mutant embryos that fail to express gsc do so because

they are older than the late-streak stage or because Otx-2 is required for the

maintenance of gsc expression at this early stage.

The Brachyurygene is expressed along the entire proximaldistal extent

of the primitive streak at the mid40 late-streak stage (Hermann, 1991). In some

homozygous Otu-2 mutants, Brachyury expression was only found in the

proximal region in a position similar to the gsc-expressing cells in Otu-2

homozygous mutants (data not shown). Other Otx-2 mutant embryos showed

almost complete extension but a thickening of the region of the primitive streak

(Fig.3.6F). Together, these results demonstrate that Otw-2 homozygous mutants

show severe early gastrulation defects, characterized by a lack of proper

primitive streak organization.

oderm and n o w o r d are severelv affected tant

embrvos

To determine if the different populations of embryonic mesoderm tissues

were present in mutants at E8.5, the markers Brachyury, Mox-1 (Candia et al.,

1992) and Lim-1 (Barnes et al., 1994) were used to identify axial, paraxial and

lateral mesoderm cells, respectively. Both Mox-1 and Lim-1 expression could

be detected in mesoderm cells of mutant E8.5 embryos, indicating that the

homozygous mutants contain paraxial and lateral mesoderm cells (Fig.3.6 J, L).

Max-1 expression in the somites in posterior regions was similar in

homozygous mutants and their littermates. Anteriorly however, Mox-1

expression spread across the midline, as expected from the observed fusion of

somites. In eight out of ten embryos analyzed, Brachyury expression in

homozygous mutants was absent in the anterior midline, at the normal position

of the notochord. In one case, Brachyuyexpressing cells were present

anteriorly but appeared to bud off and diverge from the axial notochord (Fig.

3.6N). However, in another case, expression of Brachyury seemed normal in

axial notochord (data not shown). Thus axial mesoderm cells are severely

affected in Otx-2 homozygous embryos, while paraxial and lateral mesoderm

cells do develop more normally. The appearance of fused somites is consistent

with axial mesoderm defects since it has also been seen in other notochordless

mouse mutants and notochordless chick embryos (Dietrich et al., 1993; Ang and

Rossant, 1 994; Teillet and le Douarin, 1983; Rong et al., 1992).

We next tested whether axial mesoderm defects occurred at earlier

stages than those examined above. Lim-1, Brachyury, and HNFGl3 (Ang et at.,

1993; Monaghan et al., 1993; Sasaki and Hogan, 1993) are all expressed in the

node and head process in wild-type embryos at E7.5 (Fig.3.6C.E.G). H W 3 B

and Lim-1 are also expressed in midline cells anterior to the notochord known

as prechordal mesoderm cells (Fig.3.6C.G-arrows). In homozygous Otw-2

mutants, all three genes were expressed in the node and in a few ceils

extending anteriorly at a short distance from the node. This was in sharp

contrast to wild-type embryos in which labeled headprocess cells had migrated

much further anteriorly (Ag.C,E.G). In particular, the anterior-most midline

expression of Lim-1 and HNF3B in the prechordal mesoderm cells was

missing in the mutants (arrows in 40 and ti). These results suggest that the

defects in axial mesoderm observed at E8.5 are due to a failure in the

generation of these cells at earlier stages. Prechordal mesoderm fails to

develop properly and notochord development is incomplete in €7.5 OtK-2

homozygous embryos.

DISCUSSION

Deletion of the homeobox region of the Otx-2 gene produced an

embryonic lethal phenotype in mice. The phenotype was characterized by

severe gastrulation and prechordal mesoderm defects, absence or reduction of

the notochord, and severe anterior truncations. All mutant embryos were

severely growth retarded or resorbed by E10.5, presumably because the

separation of embryonic and extraembryonic regions resulted in defective yolk

sac circulation. Given that Otx-2 is broadly expressed at the pre-streak and

early streak stage embryos, and is later restricted to anterior structures, this

phenotype implicates Otx-2 in several different aspects of early postimplantation

patterning.

Earlv ~ t r u l a t i o n defects in Otx-2 homqy~gpus m u m

In the mid- to latestreak stage mutant embryos, the incomplete

elongation of the primitive streak, the accumulation of mesoderm cells between

embryonic and extraembryonic region of the embryo and the ectopic location of

gsc-expressing cells in the proximal region, indicate that OtK-2 is required in

some manner for the normal organization of the streak.

The smaller size of the embryonic portion of mutant embryos at the mid-

streak stage suggests that proliferation of the epiblast tissue could also be

affected or delayed in the absence of Obr-2. To examine whether cell

proliferation in the epiblast is affected at this early stage, BrdU incorporation

experiments will be performed. In addition, the severe constriction observed

between embryonic and extraembryonic regions in Otw-2 mutants suggest that

other processes besides proliferation are also affected in the mutants. A very

similar extraembryonic-embryonic constriction was also observed in HNF-38

mutants (Ang and Rossant, 1994; Weinstein et al.. 1994). to a lesser extent in

Lim-1 mutants (Shawlot and Behringer, 1995), and in nodaldeficient embryos

(Varlet, et al., 1997). All of these three genes are expressed in the visceral

endoderm and recent evidence from mosaic analysis of nodal suggests the

involvement of the endoderrn in the constriction (Varlet et al., 1 997).

regy~red for Drooer orechordal mesoderm and notochord develooment

We have demonstrated that the prechordal mesoderm is severely

affected in Otx-2 homozygous mutant embryos at the headfold stage using Lim-

7 and HNF-3B genes as markers for this tissue. These results demonstrate that

Otx-2 is an essential regulator of prechordal mesoderm development. Recent

phenotypic studies on Lim-7 hornozygous mutants have also demonstrated an

essential role for the Lim-1 gene in prechordal mesoderm development. Since

Oh-2 and Lim-1 are both expressed in prechordal mesoderm in headfold stage

embryos and are required for its development, it will be interesting to study

whether these two genes might function in the same genetic pathway.

Defects in notochord development were also observed in Otx-2

homozygous embryos. In the late streak and headfold stage embryos, there

was limited midline extension of the head process. In later embryos, the

notochord formed only at the most posterior end in some cases, while in other

cases, notochord cells were present but misplaced lateral to the midline.

Anterior notochord normally expresses Otx-2 at the midbrain level (Ang et al.,

1 9941, perhaps explaining the anterior notochord defects observed in mutant

embryos. However abnormalities in the development of notochord cells, which

normally do not express Otx-2 at the hindbrain and trunk levels, cannot be

readily explained. It seems likely that the failure of later notochord development

reflects earlier defects in headprocess development. Interestingly, studies of

Otx-2 in Xenopus and chick have demonstrated Otx-2 expression in the

organizer tissue of these species, namely dorsal blastopore lip and Hensen's

node respectively (Pannese et al., 1995; Blitz and Cho, 1995; Bailly-Cuif et al..

1995). Thus, by analogy to the situation seen in these species, Otr-2 might be

expressed in the presumptive node at the anterior end of the primitive streak in

early streak stage mouse embryos (Lawson et al., 1 991 ), this expression being

obscured by the simultaneous widespread expression of Otr-2 in the epiblast at

this stage. Loss of Otx-2 in this structure may be responsible for the notochord

defects at later stages.

brain develq~rnent during mouse s m a m

Otx-2 homozygous mutants that failed to become enclosed by visceral

yolk sac by E8.5 because of the severe constriction between embryonic and

extraembryonic regions, were extremely abnormal and in some cases seemed

to lack any axial organization. However, specific anterior defects could be

observed in the less severely affected embryos that were still enclosed in the

yolk sac.

By E8.25, deletions in the anterior neural tube rostra1 to rhombomere 3

were clearly apparent in these hornozygous mutants. Loss of forebrain,

midbrain and anterior hindbrain was demonstrated using early molecular

markers for these tissues such as BF1, Emx-2 and En (data not shown). This

phenotype could have been predicted. on the basis of the anterior expression of

Otx-2 and its relationship to the otd gene in Drosophila. Loss of function

mutations in otd in flies lead to deletion of anterior head structures (Finkelstein

et al.. 1990). These results support evolutionary conservation of the function of

these genes in head development in flies and mice.

This interpretation of the phenotype assumes that 00r-2 acts

autonomously in the anterior CNS and is required for the specification of these

regions. However, this hypothesis cannot readily explain why structures

posterior to the expression domain of Otw-2 are also deleted. The caudal

boundary of Otx-2 expression marks the mid-hindbrain boundary. However,

deletions of the hindbrain region, containing the Errexpression domain of the

metencephalon and close to the anterior border of rhomobomere 3, occur in

mutant embryos. This could either be due to early expression of OW-2 in cells

fated to become hindbrain at the late-streak stage or to a dependence of

anterior hindbrain development on more rostra1 neural tube. Comparison of the

fate-map studies of late streak stage embryos with the domain of Otr-2

expression at this stage suggest that this domain does not include hindbrain

territories (Tam. 1989). However a more extensive study is necessary to

exclude this possibility.

Alternatively, loss of anterior neural tissue could be an indirect

consequence of the loss of notochord and prechordal mesoderm in the Otx-2

homozygous mutants. Classical embryological studies have demonstrated a

role for both these tissues in the induction and patterning of the neural tube

(reviewed by Slack and Tannahill, 1992; Doniach, 1993; Ruiz i Altaba, 1994). A

role for the prechordal mesoderm was supported by the phenotype cf Lim-1 null

homozygous mutants (Shawlot and Behringer, 1995). In contrast, a role for

notochord tissue in A-P neural tube patterning has not been confirmed by

mutant studies in mice. Mouse embryos lacking the HNF38 gene do not

develop a notochord, yet they showed expression of rostral brain markers (Ang

and Rossant, 1 994; Weinstein et al., 1 994). Together, these results suggest an

essential role for the prechordal mesoderm, but not the notochord in patterning

of the anterior neural tube in mice. Furthermore, the Lim-1 mutant phenotype is

remarkably similar to that of the Otw-2 homozygous mutants, in that deletions of

the anterior neural tube occur at about the same anteroposterior level (Shawlot

and Behringer, 1995). Thus, loss of rostral brain tissues observed in Otx-2

homozygous mutants could be a consequence of the lack of prechordal

mesoderm development.

To distinguish between these hypotheses, I am planning to recombine

mesoderm tissue from homozygous mutant Otx-2 embryos with ectoderm tissue

from wild-type embryos to determine if the mutant mesoderm tissue can induce

expression of anterior neural markers such as En genes. The reverse

experiment of recombining ectoderm tissue from OW2 homozygous embryos

and mesoderm from wild-type embryos will be performed to determine if the

OW-2 negative ectoderm tissue can respond to inducing signals from the

mesoderm. The use of the in vitro tissue recombination assay (Ang and

Rossant, 1993, 1994) to analyze Otx-2 mutants should allow us to dissect the

roles of 00r-2 in the ectoderm and mesoderm tissues at the latestreak stage.

In conclusion, Otx-2 homozygous mutants show complex defects in

gastrulation, axial mesoderm and rostral brain development that implicate Otx-2

in the pathways of primitive streak organization, axial mesoderm development

as well as anterior head development. Further experiments will be required to

dissect out these different roles.

Figure 3.1. Targeted disruption of the OW2 gene. The first targeting vector,

pPNTO2, contains 7.1 kb of the Otx-2 genomic locus while the second one,

pPNTKSN, contains 10 kb. The open boxes represent the coding region and

the solid boxes indicate the homeodomain. The 5' and 3' probes used for

Southern blot analysis are indicated. The sizes of the expected restriction

fragments from the wild-type and mutated Otu-2 alleles with specific probes are

indicated in the following table. The sequences amplified by PCR to identify the

wild-type and Otx-2-1- alleles are indicated as red boxes.

Expected Restriction

Fragrnents(kb)

5'- Xho I Eco RI

Expected Restriction

Fragments(kb)

5'- Xho I Em RI

Abbreviations: 6: Bglll; E: Eco RI; H: Hind Ill; K: Kpn I; S: Stu I; Sm: Sma I; Xb:

Xba I; Xh: Xho I. m: mutant; wt: wild-type.

Figure 3.2. (A)Southern blot analysis of DNA from wild-type and targeted ES

cell lines, 5-12, 5-23 and C-12. The sizes of the DNA bands are indicated in

kilobases (kb). Both 5' and 3' probes detected predicted restriction fragments for

the wild-type (wt) and mutated (m) allele. (B) PCR gemtyping of yolk sac

biopsies isolated from E8.5 embryos of intercross between Otx-2+/- mice.

Embryos were scored phenotypically as either normal (N) or mutant (M).

3' probe Eco RI

5' probe Xhol

' 8.0 kb (m)

2.6 kb (wt)

I ) C 8 . O kb (m)

5' probe Xhol

genow Pe +I- +/- +I+ +I+ 4 4- +I- phenotype N N N N M M N

Figure 3.3. Phenotype of Otx-2 homozygotes. (A,B) Lateral views of E7.5

embryos. In mutant embryos (B), a constriction is seen between embryonic

regions and the extraembryonic portion, and the embryonic portion is smaller

than that of wild-type or heterozygous embryos (A). (C,D,E,F) Views of E8.5

embryos. (C) A phenotypically wildtype embryo (either +/+ or +/-). (D,E) are

examples of less severely affected homozygous mutants and (F) is an example

of more severely affected ones. (G,H) are examples of some homozygous

mutants which are either completely outside of the yolk sac (G) or partially

outside of the yolk sac (H). Scale bar, 100 pm.

Figure 3.4. Histological analysis of OtK-2 homozygous embryo. Saggittal

sections of E7.5 wild-type embryo (A) and homozygous mutant embryo (B,C,D

serial sections). In the mutant embryo, embryonic ectoderm, mesoderm and

endoderm cells can be seen. Note the accumulation of mesoderm cells. Scale

bar, 100 urn. ect: ectoderm; me: mesoderm; en: endoderm.

Figure 3.5. Whole mount RNA in situ hybridization of Krox-20 in €8.5 wild-type

or heterozygous embryo and homozygous mutant embryo. Krox-20 is normally

expressed in both rhombomeres 3 (r3) and 5 (r5) in this stage. However, in the

homozygous Otr-2 mutant, the expression of Krox-20 in rhombomere 3 is not

detected.

Figure 3.6. Whole-mount analysis of mesoderm markers in wild-type or heterozygous embryos (A,C,E,G,I,K,M) and homozygous mutant embryos (B,D,F,H J,L,N.O) at E7.25E8.5 (anterior is to the left in A-H, and MO or to the top in I-L). (A,@ gsc expression in E7.25 embryos. (A) gsc expression in anterior primitive streak cells in normal embryos. (B) gsc-expressing cells were present but only in the proximal region of the mutant embryo. (C,D) Lim-7 expression in €7.25 embryos. (C) In normal embryos, Lim-7 is expressed in the mesodermal wings, node, head process and prechordal mesoderm cells (arrow in C). In mutants, Lim-1 is expressed in the same areas except there was no expression in the prechordal mesoderm region (arrow in D). (E, F) Brachyury expression in E7.75 embryos at headfold stage. (E) Brachyury is expressed in the primitive streak, node and head process in normal embryos. (F) In mutant embryos, Brachyury expression is expanded in the primitive streak (open arrowhead), in the node (arrow) and in a few headprocess cells extending anteriorly from the node (arrowhead). (G,H) HNF-3B expression in E7.75 embryos. (G) In €7.75 wild-type embryos, HNFGB is expressed in the node, head process and prechordal mesoderm (arrow). (H). In mutant embryos, HNF- 38 is expressed in the node, in a shortened headprocess but not in the prechordal mesoderm area (arrow). (1,J)Mox-1 expression in €8.5 embryos. (1)Mox-1 is expressed in the somitic mesoderm of wild-type embryos. (J) In mutants,Mox-1 expression anteriorly is spread across the midline (arrow). (K,L) Lim-1 expression in E8.5 embryos. Lim-1 is expressed in the lateral mesoderm of normal embryos (K) and mutant embryos (L). (M.N.0) Brachyuryexpression in E8.5 embryos. In normal embryos, Brachyury is expressed in the notochord and in the posterior primitive streak region. In mutant embryos, Brachyury expressing cells are displaced from the midline anteriorly (arrowheads in N), or missing from the anterior and trunk regions (0). However, Brachyury was expressed in the primitive streak of mutants (N,O). Scale bar, 100pm.

Chapter 4

0 t . 2 and HNF-38 Genetically Interact in Anterior Midline Patterning

This chapter will be submitted as the following manuscript:

Jin, O., Harpal, K., Ang, S-L. and Rossant, J. (1997) Otw-2 and HNF-3B

Genetically lnteract in Anterior Midline Patterning. (in preparation)

I performed all work described here, with the exception of preparing the wax

sections.

INTRODUCTION

Genetic and experimental evidence points to the importance of the

midline axial mesoderm and the ventral midline of the neural tube as critical

sources of signals far dorsal-ventral (D-V) patterning of the neural tube and

head structures. Sonic hedgehog (Shh), a vertebrate homologue of the

Drosophila hedgehog (hh) gene, is an important molecule involved in D-V

patterning. Shh encodes a secreted protein. In the mouse embryo, it is

expressed in the node, notochord, floor plate, ventral forebrain and midbrain

and other patterning centres such as the posterior margin of the limb bud

(Riddle et al., 1993; Echelard et al., 1993; Chang et al., 1994; Marti et al., 1995).

Ectopic expression of Shh in the central neural system (CNS) leads to a

ventralization of large regions in the midbrain and hindbrain of the mouse and

zebrafish (Echelard et al., 1993; Krauss et al., 1993). In early neural plate or

intermediate neuroectoderm explants, Shh protein also induces ventral cell

fates in a dose-dependent fashion (reviewed by Placzek, 1995; Tanabe and

Jessell, 1996).

The phenotype that results from lossof-function mutation of Shh in mice

further demonstrates its critical role in D-V patterning (Chiang et al., 1996).

Homozygous embryos show a single fused optic vesicle (cyclopia), proboscis

and other severe P V patterning defects in the neural tube. The phenotype of

cyclopia was also observed when midline signaling was antagonized by

overexpression of protein kinase A (PKA), a negative regulator of Shh

(Hammerschmidt et aI., 1996). In zebrafish, embryos homozygous for the

cyclops mutation also exhibit cyclopia or partial cyclopia, and abnormal

development of forebrain (Hatta et al., 1991, 1994; Macdonald et al., 1995). In

these mutant embryos, no Shh expression is detected in the midline structures

of forebrain and midbrain (Krauss et al., 1993; Barth and Wilson, 1995).

consistently supporting the role of Shh in D-V patterning of head structures.

HNF-30, a transcription factor of the winged-helix family, has been

proposed as a candidate upstream regulator of Shh. It has been suggested that

HNF3B may activate Shh expression in the midline and that they

subsequently regulate each other's expression by a positive feedback

mechanism (Echelard et a(., 1 993; Sasaki and Hogan, 1 994; Hynes et al., 1 995;

Filosa et al., 1997). HNF3B is expressed in the visceral endoderm, node,

notochord, floor plate, and gut in the mouse embryos (Sasaki and Hogan, 1993;

Monaghan et al., 1993; Ang et al., 1993; Weinstein et al., 1994). Ectopic

expression of HNF-30 in mice leads to induction of ventral structures in the

dorsal region of the brain (Sasaki and Hogan, 1994). Similarly, misexpression

of another member of the winged-helix family, pintallavis, in Xenopus embryos

also induces expression of markers of ventral structures in the dorsal hindbrain

region (Ruiz i Altaba and Jessell, 1992). These studies suggest that HNF3a

has roles in D-V patterning.

The phenotypic consequences of loss-of-function mutation of HNF3B in

mice further demonstrate its key role in D-V patterning (Ang and Rossant, 1994;

Weinstein et al., 1994). Homozygous embryos lack node and notochord

structures and show severe abnormalities in D-V patterning of the neural tube.

In addition, homozygous embryos also show some anterior deletions, although

anterior-posterior patterning (A-P) is relatively normal. Mice heterozygous for

this mutation are viable but some show an incompletely penetrant phenotype

affecting development of the lower jaw. This heterozygous phenotype suggests

that H N W is present in limiting amounts. In such a situation, exacerbation of

the HNF-3B heterozygous phenotype might be expected in mutant mice doubly

heterozygous for the HNF-3B mutation and for mutations in genetically

interacting loci. Otx-2, anoMrelated homeobox gene, is normally expressed in

epiblast of pre-and early streak. Its expression is then gradually restricted to the

anterior of embryo and, at somite stages, to the presumptive forebrain and

midbrain (Simeone et al, 1992a. 1993; Ang et al, 1994). Loss-of-function

mutation of Otx-2 leads to deletion of anterior CNS structures up to rhombomere

3 (Acampora et al., 1995; Matsuo et al.. 1 995; Ang et al., 1 W6), suggesting that

Otw-2 is in involved in head organizer function in patterning of the anterior

embryo, but no specific role of Otx-2 in D V patterning of the CNS has been

proposed.

Comparison of the expression patterns of Obr-2 and HNF30 between

E7.5 and E9.5 revealed that they are co-expressed in the anterior midline of

embryos as early as E7.5. To investigate the potential genetic interactions

between the 00-2 and HNF3B genes, we have generated double

heterozygous mutant Otr-2 and HNF-30 embryos. We found that at €12.5,

double heterozygous embryos show varying degrees of holoprosencephaly,

and cyclopia with proboscis-like structures, suggesting genetic interaction of the

two genes in the anterior midline. Further analysis of the double heterozygous

mutant phenotype demonstrates that Shh expression is severely affected in the

anterior of the embryos, suggesting that both 0tx-2 and H H are involved in

regulating anterior midline signaling during embryogenesis.

MATERIALS AND METHODS

Otx-2 heterozygous mice of 129ISvxCDI background (Ang et al., 1 996)

were crossed with HNF-3B heterozygous mice of 129ISVxCD1 background

(Ang and Rossant. 1994) to generate double heterozygous mice. Genotyping of

newborn mice and embryos was performed by Southern blot analysis with

genomic DNA prepared from biopsies of tails and yolk sacs. Each DNA sample

is divided into two parts. One aliquot was analyzed with an Ok-2 probe and the

other with an HNF* probe in two separate blots. Probes used for Southern

blot analysis were described previously (Ang and Rossant, 1994; Ang et al.,

1996). Hybridizations were camed out at 42'C overnight in 50% formamide,

5xDenhart9s, 5xSSC, 1 %SDS. 100pg/ml sheared salmon sperm DNA. The

filters were finally washed in 0.2xSSC at 63'C. The filters were then exposed to

phosphor imager screens overnight.

. . . . . stoloav. wholemount RNA In sriu hvhr~dmt~on and

Mid-day of the day of the vaginal plug was considered as E0.5 in the

timing of embryo collection. Embryos were dissected and staged according to

morphological criteria (Kaufman, 1992). Embryos were photographed on a

Leitz Wild MI0 microscope. For histological analysis, embryos were fixed

overnight in 4% paraformaldehyde at 4'C, processed, embedded in wax and

sectioned. 5-6 pm sections were dewaxed in xylene, rehydrated through an

ethanol series into PBS and stained with hernatoxylin and eosin.

Whole-mount RNA in situ hybridization was performed as described

previously (Conlon and Hermann, 1 993). Single-strand RNA probes were

labeled with digoxigenin as directed by the manufacturer (Boehringer

Mannheim Biochemicals). The probes used for the whole-mount in situ

hybridization studies were as follows: Shh (Echelard et al., 1993); Of%-2 (Ang et

al.. 1994); BF-l (Tao and Lai. 1992); Hoxb-7 (Wilkinson et al.. 1989a); Six-3

(Oliver et al., 1 995); Mox-7 (Candia et al.. 1 992). After RNA in situ hybridization,

embryos were pastfiixed in 4% paraformaldehyde at 4OC overnight and then

followed by whole-mount antibody staining. Whole-mount antibody staining was

performed as described (Davis et al., 1991) using an anti-HNF3R antibody at a

dilution of 1 :I000 (Sasaki et al., 1993). For sectioning of whole-mount stained

specimens, embryos were postfixed in 4% paraformaldehyde at 4OC overnight.

Sections were cut at 5-6pm and some sections were counterstained lightly with

eosin, and photographed using a Leitz Orthoplan compound microscope and

Nomarski optics.

RESULTS

s of Otx - 3 and HNF-3B double heterozyp~us mutant

To investigate the possible interaction between 00~-2 and HNF-313, Otw-2

heterozygous mice of 129ISvxCDl background were crossed with HNF-30

heterozygous mice of 129/SvxCD1 background and offspring were genotyped 3

weeks after birth. Southern blot analysis revealed that the number of double

heterozygous weanlings was significantly reduced below the expected n urn ber

(Table 4.1). suggesting embryonic or post-natal lethality of double heterozygous

mutants. I observed that several double heterozygous pups died the day of

birth and showed mandible defects. Among the survivors, two died after eight

months and two died after eleven months. All these had incisor overgrowth and

jaw defects. The remaining four appeared normal and healthy, although one of

them had overgrowth of the incisors.

To characterize the embryonic lethality, I dissected litters at €18.5 and

found that some embryos had already begun to be resorbed (data not shown).

Taken together, my observations suggests that the doubly heterozygous

condition causes a variable penetrant lethal phenotype. These double

heterozygous mutants may be lost at late gestation or after birth.

I then dissected and genotyped litters at E12.5 and €9.5. Genotyping

results from E12.5 and €9.5 yolk sacs produced genotypes at roughly the

expected Mendelian frequencies (Table 4.2. Table 4.3). At El 2.5, about 56% of

double heterozygous mutants displayed an obvious phenotype which is

characterized by cyclopic or partial cyclopic eye, reduced distance between the

eyes, and proboscis and other defects (Table 4.2, Fig. 4.1 and data not shown).

At €9.5, this phenotype was already apparent and varied in its severity (Fig.

4.2). Some double heterozygotes showed a clear phenotype, demonstrating a

dramatic reduction in the size of the forebrain. In the most extreme case, the

forebrain was lost (Fig. 4.28). However, in most of these mutants, the forebrain

is always present but shows a reduction in size (Fig. 4.4C, D). In those embryos

with an obviously reduced forebrain, the floor and roof of the neural tube were

almost in contact with each other at the diencephalic and mesencephalic

junction (Fig. 4.2C). Although the overall size of most Otw-Z+/-;HNF33R+/- mutant

embryos was smaller than wild-type or heterozygous embryos, the posterior

part of each embryo appeared normal.

vsis of the Otx2+/:HNF3R+/ - - - ohenohrne

To characterize theOtx-Z+';HNF3B+/- phenotype in more detail, El 2.5

and E9.5 mutant embryos were sectioned for histological analysis (Fig. 4.3, 4.6

and data not shown). In wild-type or singly heterozygous embryos there are two

telencephalic vesicles (future lateral ventn'cles Fig. 4.3A.B). However, in Otx-

24-;HNF30+/- mutants these two vesicles were fused into a single

telencephalic vesicle (Fig. 4.3D, E). The size of the telencephalon and

diencephalon was reduced in Ofx-Z+/-;HNF3B+/- mutants (Fig. 4.3D,E). The

more striking phenotype was a fused single eye (cyclopia) and a proboscis-like

structure (Fig. 4.3F,H). In vertebrate, optic vesicles are formed from evagination

of the lateral walls of forebrain (Carlson, 1996). The optic vesicle invaginates

and forms a double-layered optic cup. The outer layer gives rise to the pigment

layer of the retina and the inner layer gives rise to the neural layer of the retina

which is much thickened than the outer layer. The two cups are connected to

the diencephalon by the optic stalks. However, in those Otx-2+/-;HNF-3B+/-

mutants which showed the cyclopia phenotype, the optic vesicles were fused at

midline and the lateral optic stalks were absent (data not shown). Although

some of Otx-2+/-;HNF38+/- mutants had smaller eyes (data not shown), most

of them seem morphologically normal. The lens, neural layer of the retina and

corneal ectoderrn were apparently normal (Fig. 4.3F. H). In addition, the

mandible of Otx-Z+/-;HNF-3B+/- embryos was smaller than wild-type or

heterozygotes (Fig. 4.3H). However, the morphology of the floor plate,

notochord and the other body structures were normal (Fig. 4.6M.N). These

results suggest that defects in Ob-Z+/-;HNF-38+/-= mutants were confined to

anterior structures.

on of T)fx 2 and H N F 3 B exDrssron In W.5-E9.5 mouse embrvos . . -

To investigate further how Otx-2 and HNF-3B could be interacting in the

anterior regions of the embryo, I compared the expression of Obi-2 RNA and

HNF3O protein in E7.5-€9.5 mouse embryos by double-labelling to examine

regions of overlap. In these studies, embryos were first stained for Otx-2 RNA

by whole-mount in situ hybridization and then for HNF38 protein by whole-

mount antibody staining. Further analysis was performed on sections. At E7.5,

I found that Otx-2 and HNF3B coexpressing cells overlapped in the foregut

pocket and anterior midline (Fig. 4.40). Sections through embryos double-

labelled for Otx-2 and HNF3B revealed coexpression of these two genes in

prechordal mesoderm (Fig. 4.4C), anterior ventral neural fold and ventral

endomesoderm (Fig. 4.4E).

At €8.5, I found that sections through forebrain, midbrain and hindbrain

regions of embryos showed Otx-2 and HNF-3B co-expressing cells in the

ventral forebrain and midbrain region (Fig. 4.4G,H). However, by €9.5, the

expression of these two genes only overlapped in the ventral midbrain regions

(Fig. 4-44. My double-labelling studies clearly suggests that Ofx-2 and HNF-

are co-expressed in the anterior midline of mouse embryo.

* . . ~ t ~ e s rn Ofx2r-/:HNFI3R+/ - - - - mutant emhrvo~

To characterize the anterior abnormality of 0&-2+/-;HNF3B+/- mutant

embryos further, the expression of several genes normally transcribed along the

A-P axis of the embryos was examined at €9.5 (Fig. 4.5). Otr-2, Hoxb- I and

Krm-20 were all expressed at the correct level of A-P axis, suggesting that A-P

patterning in the neural tube of OW-2+/-;HNF-33+/- mutants was not affected.

I also analyzed BF-I and S i x 3 expressions in Otx-2+/-;HNF3B+/-

embryos. BF-I, a winged-helix transcription factor, is normally expressed in the

telencephalon of brain. In homozygous mutants of BF-I, the dorsal

telencephalon is reduced in size while the ventral telencephalon is almost

completely absent (Xuan et al.. 1994), suggesting a role for BF-7 in the

development of telencephalon, especially ventral telencephalon. The

expression domain of BF-1 was missing in Otx-2+/-;HNFF-30+/- mutant embryos

(Fig. 4.58 and data not shown). I then checked the expression of Six-3 which is

normally expressed in the ventral forebrain and optic vesicles at this stage. It

has been demonstrated that Six3 can induce ectopic lens in fish embryos

(reviewed in Oliver and Gruss, 1997). suggesting that it plays a key role inlens

induction. In OtK-2+/-;HNF38+/- embryos, the expression domain of Six-3 in

ventral forebrain was missing but expression was still detectable in the reduced

single optical vesicle (Fig.4.5E.F). Those observations suggest that the ventral

forebrain is severely affected.

h h expression in anterior of Otx - 2+/ - : H N F + a n t - em bwos is severelv

ilmax!

Loss of ventral forebrain expression of Six4 and expression of BF-1,

fused telencephalon, cyclopia and mandible defects in Otx-2+/-;HNF-3&/-

mutants suggest that the chief defect is loss of ventral midline signaling in these

mutants, reminiscent of the phenotype of Shh homozygous mutants in the

anterior region (Chiang et al., 1996).Therefore, I analyzed Otx-24-;HNF-3B+/-

mutant embryos at E9.5 for the expression of Shh. At E9.5, Shh is normally

expressed in the ventral forebrain, midbrain, notochord, floor plate and gut. I

found expression of Shh was almost completely lost at the anterior end (Fig. 4.6

and data not shown). Sections through these mutants showed that the

expression of Shh was lost in ventral forebrain and floor plate, although the

floor plate was present morphologically (Fig.. 4.5). The loss of Shh expression

in the floor plate could be explained by lower level of Shh expression in the

underlying notochord. Very weak expression of Shh could still be detected at

ventral mid brain level (Fig. 4.5E, arrows) and anterior notochord (Fig.4.5M.

black arrowhead). The expression of Shh in the rest of the notochord, foregut

and hindgut is still detectable although the level of the expression is reduced,

perhaps due to reduced dosage of HNF-3B (Fig.4.5 M, N and data not shown).

DISCUSSION

Otw-2+/-; HNF-30+/- doubly heterozygous mutant mice exhibited a

variably penetrant lethal phenotype. The major phenotype is characterized by

varying degrees of holoprosencephaly, cyclopia and proboscis with normal

posterior structure. Shh expression in the anterior of the embryo was severely

affected. This new phenotype was observed in &-2+l-;HNF-3B+/- embryos but

not in Otr-2+/- or HNF-3B+/- single mutant embryos.

In previous single-mutant studies, haploinsufficiency of H N F a or Otx-2

has been demonstrated in HNF3B +/- or Otw-2 +/- animals (Ang and Rossant,

1994; Weinstein, et al. 1994; Acampora et al., 1995; Matsuo et al., 1995; Ang et

al., 1996). About 20% of HNF4B heterozygous adults exhibit malocclusion of

jaws and overgrowth of the incisors. Some of them died later although special

care such as providing soft food and cutting the overgrowth of incisors

periodically can significantly reduce mortality of these mice. 5% of Otx-2+/-

heterozygotes exhibit open neural tube defects at €9.5 (Fig. 4.7A) and

subsequently at €12.5 (Fig. 4-78). showing exposure of neural tissue with

characteristic exencep haly . This phenotype contributed to about 5% loss of

heterozygotes. Variation in the threshold levels of factors or modifiers which

may act synergistically with either HNF3B or Otx-2 may contribute to the

variable penetrance of these haploinsufficiencies.

In the double heterozygotes, the phenotype discovered may be

considered as a regionspecific exacerbation of the HNF-3B mutant phenotype.

H N F a is essential for the formation of midline structures, such as notochord

and floor plate. Shh is a prime signalling molecule in the midline for 0-V

patterning and in HNF-3B homozygous mutants, Shh expression is absent in

the midline of the HNF4B homozygotes. However, HNF3B homozygous

embryos died much earlier than Shh homozygous mutants. The early lethality

probably results from circulation defects, since HNF-36 has a role in endoderm

development. HNF-38 homozygous mutants demonstrate severe D-V patterning

defects along the entire body axis including the head. Recent study of

overexpression of HNF-38 in the entire midline exhibited anterior defects with

normal posterior (Dufort and Rossant, unpublished data). It suggests that the

anterior region is more sensitive for the dosage of HN-8 than that of the

posterior. The phenotype observed in HNF3B heterozygous mice is probably

due to mild D-V defects in mandible development. When HNF3B heterozygous

mice are crossed with Otx-2 heterozygous mice, double heterozygotes show

enhanced anterior midline patterning defects.

It is as yet unclear whether OtK-2 and HNF-3B directly interact to regulate

midline signalling. However, they do overlap in expression in the anterior

midline, suggesting a possibility of directly interacting. At E7.5, Otx-2 and HNF-

38 are coexpressed in the cells of the foregut pocket and anterior midline but

not in the posterior of the embryo. At €8.5, the two genes continue to be co-

expressed in the cells of ventral forebrain and midbrain. At E9.5, the expression

domains of the two genes no longer overlap in the ventral diencephalon but still

overlap in ventral midbrain. These studies support the potential interactions

between the two genes in these tissues.

The phenotype of Otx-2 and HNF38 double heterozygotes again

supports the critical role for Shh signalling in the forebrain where patterning is

the most complex and the exact source of signals is not well understood. The

hypothesis that ventral forebrain serves as an important source of signals for D-

V patterning in the anterior GNS is suggested by the phenotypes of zebrafish

embryos homozygous for the cyclops mutation (Hatta et al., 1991, 1994) and

mouse embryos homozygous for the Shh lossof-function mutation (Chiang et

al., 1996).

The molecular identity of cyclops is unknown. However, mutant embryos

show the fusion of two lateral eyes into a single median vsntral cyclopic one,

reduction in size of the diencephalon and loss of the floor plate in posterior

CNS. Cell transplantation experiments suggest that the cyclops gene product is

involved in regulating a signalling pathway in the forebrain midline cells (Hatta

et al., 1994). Consistent with this, there is no Shh expression in the ventral

forebrain or the posterior of ventral midline (Krauss et al., 1993; Barth and

Wilson, 1995).

In mouse, the most striking phenotype of Shh homozygous mutant

embryo is a single fused telencephalic vesicle with a single fused optic vesicle

and proboscis-like structure (Chiang et al., 1996). The phenotype demonstrated

from Otx-Z+/-;HNF4&/- mutants is not only reminiscent of Shh homozygous

mutant mouse embryos but also reminiscent of a variable spectrum in facial

findings in patients with Shh mutations (Erich et al., 1996). The most severe

group of these patients exhibits cyclopia, synophthalmia, proboscis and

microcephaly with normal development of the rest of the body, suggesting the

role of Shh in the anterior midline patterning in humans. In Otx-2+/-;HNF-33a+/-

mutants, the expression of Shh is lost or almost lost in ventral forebrain and

midbrain, suggesting that both Ofx-2 and HNF-3B are involved in regulating this

pathway in the anterior midline (Fig. 4.8).

These studies have revealed an expected role for Otx-2 in regulating

anterior D-V patterning, which is independent of its role in A-P patterining.

Since Otx-2 and HNF-3R are coexpressed in the Shh expressing tissues, the

ventral forebrain in early embryogenesis, it is possible that the two genes

interact directly by binding to the Shh promoter. Recently. three enhancer

elements have been identified in the regulatory region of Shh by transgenic

approaches (Epstein and Joyner, 1997). One enhancer element drives the

expression of a reporter gene in the caudal diencephalon and ventral midbrain.

The other two enhancer elements drive the expression of a reporter gene in the

floor plate from the mid-hindbrain boundary to a posterior level at the hindlim bs.

Weak expression of the reporter gene was also detected in the notochord.

Further analysis demonstrated that several HNF3 binding sites and a

consensus binding site for a bicoid type homeodomain protein have been found

in these enhancer elements. Altering or deleting these sites resulted in either

restriction or the abolishment of reporter gene activity (Epstein and Joyner,

1997). Their studies suggest that HNF3B family members may be involved in

regulating Shh expression in cooperation with other regulatory proteins in a

regionalized fashion. Whether Otx-2 binds to the ventral forebrain enhancer

remains to be determined. The genetic interaction we report here is also

consistent with other possible mechanisms. For example, it is possible that two

different pathways controlled by Otx-2 and HNF-3B may be separately involved

in regulating Shh signalling pathway.

In conclusion, the phenotype of Otx-2+/-;HNF3B+/- mutants

demonstrates genetic interaction of Otx-2 and HNF-38 in the anterior midline of

the embryo. Coexpression of Obr-2 and HNF4B in anterior midline of the

embryo supports a relatively direct interaction of the two genes in the

development of the anterior midline structures in the embryo. The loss of Sh h

expression in anterior of Otx-2+/-;HNF3B+/- embryos suggests that both Otw-2

and HNF-38 are involved in regulating the anterior Shh signalling pathway.

Fig.4.1 Phenotype of Obr-Z+/-;HNF38+/- embryos at €1 2.5. (A) Lateral view of

normal El 2.5 embryo. (B, C, D) lateral views of Otx-2+/-;HNFF-3B+/- mutants,

demonstrating closely spaced or fused eyes with proboscis.

Fig. 4.2 Phenotype of Otx-24-;HNF-38+/- embryos at E9.5. Lateral views of

wildtype embryo (A) and Otx-24-;HNFdB+/- embryos (6, C. D). (8) Otx-2+/-

;HNF-3hk embryo shows forebrain deletion. (C, D) Otw-2+/-;HNF38+/-

embryos display forebrain defects. It is noted that the floor and roof of the neural

tube of the mutants are in contact each other at the caudal diencephalic and

mesencephalic junction (arrowhead). Abbreviations: fb: forebrain; te:

telencephalon; di: diencephalon; mb: midbrain; hb: hindbrain; ov: otic vesicle.

Fig. 4.3 Histological analysis of Otx-2+/-;HNFaB+/- embryos at Ef2.5. Cross-

sections of wild-type (A, B, C) and Ofx-2+/-;HNF38+/- mutants (D, E, F). The

wild-type embryo has two telencephalic vesicles (Mure lateral ventricles) (A, B)

while in double mutant these two lateral ventricles are fused to form a single

ventricle(D, E). (C, F) show cross-sections at level of eyes. The O~X-~+~;HNF-

3&/- mutant has a single eye (cyclopia) (F). Note the two ears are close to the

single eye in the double mutant (hypotelorism). (G, H) Middle sagittal sections

of embryos. (G) wildtype (H) double heterozygotes. Note a single eye in anterior

midline and the smaller mandible in the double mutant (H). Abbreviations: TeV:

telenceph alic vesicle; cyc: cyclopia; 3rdV: the third vertical; 4thV: the fourth

ventricle; md: mandible.

Fig. 4.4. Comparison of Otx-2 and HNF38 in wildtype mouse embryos between

E7.5 and E9.5. (A,D) Anteroventral views of E7.5 embryo, stained for Otx-2 RNA

(purple)(A) and then stained forHNF3B protein (brown) (D). (D) Co-expressing

OtK-2 and HNF4B cells were found in the foregut pocket (fg) and anterior

midline (ml) (B,C,E) are cross sections through different levels of the anterior

portion of an embryo: top of neural fold tissue (B), under neural fold (C) and

foregut pocket region (E), demonstrating the presence of coexpressing cells

Otw-2 and HNF3B in prechordal mesoderm (pm) (C, arrowed), anterior

endomesoderm (em) (E, white arrows), and anterior ventral neural fold (vn) (E,

black arrow). (F.1) Lateral views of E8.5 and €9.5 embryos. stained for the

expression of Otw-2 and HNF-38. (G,H) Cross sections show co-expressing

cells of Otr-2 (purple) and HNF-3B (brown) in the ventral midbrain (vrnb) and

diencephalon (vdi) at E8.5. (J) Cross section shows co-expressing cells of Otx-

2(purple) and HNF-3B(brown) in ventral midbrain (vmb) at E9.5. (K) Cross

section shows expression of Obr-2 in the ventral diencephalon (vdi) and HNF-

38 in the floor plate (white arrow) at E9.5. At this stage, the two genes are no

longer coexpressed in the ventral diencenphalon. Abbreviations: em:

endomesoderm; fg: foregut pocket; fi: floor plate; ml: midline; op: optical vesicle;

pm: prechordal mesoderm; vdi: ventral diencephalon; vmb: ventral midbrain; vn:

ventral neural fold.

Fig. 4.5 Expression of Obr-2, Hox& 1, Six-3 and Mox-7 in Otx-2+/-;HNFF3Bi/-

embryos. (A) Lateral view of wild-type embryo, showing Otw-2 expression is

forebrain and midbrain. Hoxb-l expression in rhomomere 4 (arrowed) and

posterior spinal cord. (B, C) Otu-2+/-;HNF-33Bt/- embryos show reduced Otw-2

expression domain. (D) Lateral view of wild-type embryo, showing Six-3

expression in the optic vesicle (white arrow) and ventral forebrain (black arrow),

and Mox-I expression in second, third, and fourth branchial arches

(arrowheads), and somites. (E, F) Otx-2+/-;HNF-3+/- embryos, showing loss of

Six-3 expression domain in ventral forebrain but resistent weak expression in

the reduced single optic vesicle(arrows) .

Fig. 4.6. Whole-mount in situ hybridization and cross section analysis of Otx-

2+/-;HNF3R+/- mutant with BFI, Krox-20 and Shh. (A, D) lateral view of E9.5

wildtype (A) OtK-2+/-;HNF-3&/- (D) embryos, respectively, stained for BF-7,

Krox-20 and Shh expression. O ~ X - ~ + ~ ; H N F ~ $ + / - embryo demonstrates lack of

BF1 expression in telencephalon. There is much reduced expression of Shh in

the posterior region of the embryo, and expression is absent from the anterior

ventral midline. (B, C) Cross sections of wildtype embryo show Shh expression

in the ventral neural tube, notochord and foregut. Note Krox-20 expression in

the rhombomere 5. (E, F, G) cross sections of 01%-24-;HNF-33B+/- embryo,

showing loss of Shh expression in anterior ventral neural tube. Note very weak

expression of Shh in ventral midbrain (E, white arrows), anterior notochord (G,

black arrowhead) and expression in the tip of Rathkes's pouch (G, white arrow).

Krox-20 is still normally expressed in the rhombomere 5. (H, I, J, K) cross

sections of wildtype embryo through foregut and hindgut level show Shh

expression in the floor plate, notochord, foregut and hindgut. (L, M, N) cross

sections of Otx-2+/-;HNF-30+/- embryo show reduced Sh h expression in the

notochord, foregut and hindgut. However, Shh expression in the floor plate

seems lost but the morphology of the floor plate appears normal. Abbreviations:

fg: foregut; fl: floor plate; hg: hindgut; no: notochord; ot: otic vesicle; r5:

rhombomere 5; vdi: ventral diencep halon; wt: wildtype; 0o:Hh: Otx-2+/-;HNF-

3R+/-.

Fig.4.7 Phenotype of O H +/- embryos. (A) lateral view of E9.5 embryo.

showing open anterior neural tube. (8) Lateral view of €12.5 embryo, showing

characteristic exencephaly phenotype.

Fig. 4.8. Schematic diagram showing that both Otw-2 and HNF-30 are involved

in regulating Shh expression in the anterior region. For details, see text.

Chapter 5

DISCUSSION

The primary aim of the research described in this thesis was to

investigate the molecular mechanisms involved in patterning the anterior region

of the mammalian embryo. Otx-2, a gene related to the Drosophila otd

homeobox gene, has been cloned (Chapter 2). Obr-2 is expressed first in the

epi blast, then gradually restricted to the anterior region of the embryo, incl udi ng

all three anterior germ layers, and later to the presumptive forebrain and

midbrain. The phenotypic analysis of OW-2 homozygous mutants demonstrated

a critical role for Otx-2 in head formation (Chapter 3). At late-streak and

headfold stages, &-2 is expressed in the anterior midline of the embryo. Its

expression overlaps with that of HNF-38 , another organizer gene involved

mainly in D-V patterning of the embryo. The phenotype of Otx-2 and HNF-3B

double heterozygous mutants reveals a role for 01%-2 in midline patterning in

conjunction with HNF-38 (chapter 4). My studies here thus revealed some of

the multiple roles of the Otx-2 gene during mouse embryogenesis. In the

following sections, I will discuss how my findings on Otx-2 function relate to the

current progress in understanding anterior patterning and suggest future

experiments which may further our understanding of the molecular basis and

cellular interactions for the roles of OW2 in different lineage development.

. . Ivat~on and pattern na of the rostra1 b r m

The loss-of-function mutation of Otx-2 leads to early embryonic lethality

and total loss of the rostral brain. These early defects hinder the analysis of the

role of &-2 in the regionalitation and patterning of the rostral brain. Recently,

Otx-1 (Acampora et al., 1 %6), Emx-7 and Emx-2 have been knocked-out

(Pellegrini et al., 1996; Yoshida et al., 1997). About 30% of OW-1 homozygotes

died in the first postnatal month and all homozygotes exhibited epileptic

behavior with the characteristics of focal and generalized seizures. Further

analysis revealed multiple subtle abnormalities affecting the dorsal

telencephalic cortex, mesencenphalon and cerebellum (Acampora et al., 1996).

For Emx-1 hornozygous mice, half of them died in the neonatal stage. The Emx-

I mutation caused no apparent defects at the embryonic stage but most of the

mutant embryos displayed subtle defects as adults. These defects are restricted

to some regions of dorsal forebrain (Yoshida et al., 1997). Emx-2 homozygous

mice died within a few hours after birth, presumably due to the absence of the

kidneys and other parts of the urogenital system (Pellegrini et al., 1996; Yoshida

et al., 1997). Emx-2 is expressed in the primordia of the urogenital system

during embryogenesis (Simeone et al., 1992b). The mutant phenotype and

expression pattern of Emx-2 strongly suggest that Emx-2 is required for the

development of urogenital system. Analysis of Emx-2 homozygous brains

revealed that the dorsal structures of cerebral hemisphere are affected,

especially the medial limbic cortex and hippocampal region. Dentate gyrus is

always absent (Pellegrini et al.. 1 996; Yoshida et al., 1997). The dentate gyrus

is involved in the formation of hippocampus and is a part of the limbic system.

Further analysis of the role of Emx-2 in this system might be achieved by

rescuing development of the urogenital system. Since the two Otx and two Emx

genes are expressed in nested A-P domains in the sequence Emx-IcEmx-

2=0tK-1<0tw-2, one way to analyze how these genes cooperate in the

regionalization and patterning of the rostra1 brain is to generate double

heterozygous mutants.

In fact, the analysis of phenotype in the Otx-I+/-;0tx-2+/- (Yoshida et al.,

1997) or Otx-I-l-;0&-24- (Pellegrini et al.. 1996) mutants have revealed that the

Otx genes play a role in the regionalization and patterning of the rostral brain.

These mutants exhibit variable defects in the forebrain and midbrain, which are

more severe in the dorsal parts than ventral parts of the brain. These mutants

lack mesencephalon, pretectal area, dorsal thymus and exhibit a large

reduction of Ammon's horn. The metencephalon is expanded and the isthrnic-

like structure is rostrally shifted in the presumptive caudal diencephalon. It has

been suggested that the isthmus at the mesencephalic-metencephalic (mes-

met) boundary might have organizer function since it can induce the

surrounding tissues to acquire a mes-met fate (Marin and Puelles, et al., 1994;

Joyner, 1996). The shift of isthmic-like structure rostrally in these compound

mutants suggests that the Obr gene dosage is also required for establishing the

correct position of the isthmic organizer.

Recent evidence from Drosophila suggests that different threshold levels

of Otd protein are required for the formation of specific subdomains of the head

(Royet and Finkelstein, 1995). Similarly in mammals, appropriate threshold

levels of Otx proteins might also be required for the regionalization and

patterning of the rostral brain. These studies reveal that appropriate threshold

levels of Otx proteins are required in the regionalization and patterning of the

rostral brain. It would be of interest to search for potential interactions of Otu-2

with the Emx genes in regional specification. Besides Otx-I, Ok-2, Emx-7 and

Emx-2, a number of genes are now known to be expressed in the different

regions of the rostral brain, such as BF-1, the Wnt family of genes, the parfamily

of genes and the Dlxgenes (Bulfone et al., 1993, 1995; Puelles and Rubenstein

1 993; Figdor and Stein 1993; Rubenstein et al 1994). Since 00r-2 is expressed

in almost the entire forebrain and midbrain, it would be of interest to investigate

the interactions of Ofx-2 with these genes in regionalization of the brain.

s of the late role of 0&? - brain bv condfi~onal gene tarae . .

ting

Since Otx-2% embryos lack the entire rostral brain and die at an early

stage of embryogenesis, one way to further address the role of Ofx-2 in the

patterning of the different regions of the brain is to take a tissue-specific gene

targeting or conditional targeted rnutagenesis approach. This approach may

allow to specifically address the later role of &-2 in the regionalization and

patterning of the rostral brain. The &-2 locus will be targeted such that the

genomic region containing the ATG start codon and most of the homeodomain

region (exon3 and 4) will be flanked by loxP sites. Another line will be

generated that expresses Cre recombinase is under the control of brain (tissue)

specific promoter, such as the nestin promoter (Lendahl et al., 1990;

Zimmerman et al., 1994). When these two lines of mice are crossed, animals

that lack 0tx-2 function specifically in the brain tissue will be generated. This

Cre-loxP system has been successfully used recently in the conditional

targeting of the DNA polymerase 8 gene in T cells (Gu et al., 1994) and the Apc

gene in colorectal epithelium (Shibata et al., 1997). This system may allow us

to perform a fine, detailed analysis of Obr-2 function in the regionalization and

patterning of the forebrain and midbrain.

Analvsis of the role of Otr-2 in the head omanizer formation

As I discussed in Chapter 1, Spemann and Mangold (1924) first

demonstrated the induction of a secondary axis including head by grafting a

dorsal blastopore lip to the ventral side of a host amphibian embryo. Spemann

(1 931 ) also found that the early dorsal lip could induce a complete axis

including a head but at a later stage, the dorsal lip could only induce trunk and

tail structures, suggesting that the early dorsal lip has a head and trunk

organizer activity but as development proceeds, it only has trunk organizer

function. In mouse, Beddington (1994) has provided direct evidence for the

organizing function of the node by performing heterotropic transplantation

experiments. When the midgastrulation of node was grafted to a posterolateral

location in a host embryo at the same developmental stage, a second neural

axis was induced. However, the second axis always lacked forebrain and

midbrain, suggesting that the mouse node has trunk organizer activity but lacks

head organizer function.

Recent evidence suggests the important role of visceral endoderm in the

most anterior patterning. Cerberus, a putative secreted protein, was isolated by

differential screening in Xenopus (Bouwmeester et al., 1996). It is expressed in

the anterior endoderm. Ectopic heads could be induced when cerberus mRNA

was injected into Xenopus embryos. The inductive interactions between the

visceral endoderm and underlying ectoderm have been further demonstrated

by physically removing the anterior endoderm at earlier stages (Thomas and

Beddington, 1 996). A homeobox gene, HesxVRpx (Hermesz et al., 1 996), has

been used as a marker in this study. Hesxl is first expressed in a small domain

of anterior endoderrn at the start of gastrulation but about 24 hours later, it

begins to be expressed in the underlying ectoderm. The ectoderm expression of

Hesxl is almost lost by physical removal of endoderm cells which express

Hesxl at earlier stages, suggesting the ectodermal expression of Hesxl is

mainly dependent on signals from the endoderm.

Furthermore, evidence from chimeric analysis of nodal also supports this

hypothesis (Varlet et al., 1997). Nodal is a member of the TGFB family of

secreted protein and is required for gastrulation (Zhou et al.. 1993; Conlon et

al., 1 994). Nodal is expressed throughout the embryonic ectoderm before streak

formation and then localized within the most proximal embryonic ectoderrn.

However, nodal is also transiently expressed in the visceral endoderm at the

onset of gastrulation. To distinguish nodal signaling in the primitive ectoderm

and endoderm of the embryo. chimeric embryos in which the visceral endoderrn

is exclusively composed of nodakieficient cells have been generated by

injecting wild-type ES cells into nodakdeficient blastocysts. These chimeric

embryos demonstrate the lack of the most anterior structures, suggesting nodal

signaling in the endoderm is required for anterior patterning during mouse

gastrulation (Varlet et al., 1 997).

Taken together, a model for the presence of two organizers in the

Xenopus and mouse embryos has been proposed recently (Fig.5.2)

(Bouwmeester and Leyns, 1997). In Xenopus, the head and trunk organizers

are overlapping or directly adjacent at early stages. However, during

gastrulation, they become physically separated. In the mouse, the topography is

different. The anterior visceral endoderm , which is topological equivalent to the

anterior-dorsal endoderm in the Xenopus, is not adjacent to the node (Fig.5.1).

The spatial separation of the node and anterior visceral endoderm might

explain the absence of head structures in secondary axes induced by

heterotopical transplantation of mouse node.

Fig. 5.1. Topological equivalence between the Xenopus fate map and mouse

fate map at the early gastrula stage. The large arrows project the extremities of

the embryonic axis, A-P and D V , from the Xenopus onto the mouse fate map.

The smaller arrows indicate the movements of the mesoderm through the

blastopore lip (Xenopus) and the primitive streak (mouse). The different colors

represent the prospective embryonic regions as indicated. (Adapted from

Bouwrneester and Leyns, 1997)

Otx-2 is expressed in :he visceral endoderm as early as E5.5, raising the

issue of whether OW2 is required in the visceral endoderm. To address the role

of Otx-2 in endoderrn development, it is of interest to generate chimeric embryos

which have wild-type endoderm and homozygous Otx-2 embryonic ectoderm. If

ES cells are aggregated with eight cell stage embryos or injected into blastocyst

stage embryos, they can contribute to any tissue in the embryo. However, ES

cells contribute poorly to the extraem bryonic lineages including primitive

endoden (Beddington and Robertson, 1989). In contrast, tetraploid embryonic

cells contribute mainly to the extraembryonic tissues (Tarkowski, 1 977). In this

analysis, tetraploid wild-type embryos will be aggregated with O M - ES cell

lines to generate embryos in which the embryonic compartment is derived from

Otx-24- ES cells and the extraembryonic lineages including primitive endoderm

are derived from wildtype tissues. The isolation of Otx-2-1- ES cell lines would

be achieved by high G418 selection (Mortensen et al., 1992), and the method

for tetraploid aggregation was as previously described (Nagy et al., 1993). If the

Ok-2 homozygous phenotype is rescued or partially rescued, it would strongly

support the role of Otx-2 in the endoderm development. The reverse tetraploid

aggregation experiment (wild-type ES cells c-> Otx-24- tetraploid embryos)

would further address whether OW-2 in the endoderm is required for the anterior

patterning. In this case, we would expect that the chimeric embryos may

reproduce the Otw-2 homozygous head phenotype.

s of the role of Otx-2 n the mesoderm and ectoderm

As previously demonstrated (Chapter 2), the progressive restriction of

Otx-2 expression correlates with the anterior migration of mesoderm in the

embryo, suggesting that interactions with mesoderm might be involved in

setting up the anterior domain of Obc-2 expression in the overlying ectoderrn.

Our previous study demonstrated that a positive signal from anterior

mesendoderm is required to stabilize expression of Otx-2 in ectoderm explants

and a negative signal from the later-forming posterior mesendoderm represses

Otx-2 expression (Ang et al., I 994).

Since the anterior mesendoderm cells are involved in the induction of the

rostra1 brain (Ang et al., 1994) and Otx-2 is also expressed in the anterior

mesendoderrn, diploid chimera analysis (generation of mosaic embryos which

contain contributions from OW-2-1- ES cells and wild-type cells) will provide a

powerful tool for in vivo analysis of the role of OW2 in the anterior

mesendoderrn. For chimera analysis, it is important to have an independent cell

lineage marker to follow the distribution of mutant or wild-type cells in the

mosaic embryos. The mouse line carrying ROSA26 transgene, which

constitutively expresses lacZ in all tissues (Friedrich and Soriano, 1991), will be

crossed with Otx-2+/- mice on a l29Sv background. The Oh-2-/-;ROSA26 ES

cells will be isolated directly from cultured blastocysts which come from the

interbreeding of the offspring of ROSA26 and OW-2+/- mouse lines. Potential

chimeras will be recovered at E7.5E9.5 and processed for 8-galactosidase

staining to analysis the distribution of mutant ES cells and the phenotypic

consequences of Otr-2 mosaicism in the embryo. For example, the consistent

exclusion of Otx-2 mutant cells from anterior mesoderm will suggest that Otx-2

plays a cell autonomous role in this tissue. On the other hand, if chimeras with a

high percentage of Oa-U- cells in the anterior mesoderm reproduce the Otr-2

homozygous phenotype, whereas chimeras with a high percentage of wild-type

cells partially rescue this phenotype, it will suggest that Otx-2 plays a role in the

anterior mesoderm.

Alternatively, using the in v h tissue recombination assay (Ang and

Rossant, 1993, Ang et al., 1994) to analyze Otx-2 mutants may also allow us to

dissect the roles of O W in the ectoderm and mesoderm tissues at the late-

streak stage. In such an assay, mesoderm tissue from homozygous mutant Ob-

2 embryo will be recombined with ectoderm tissue from wild-type embryos to

determine if the mutant mesoderm tissue can induce expression of anterior

neural markers such as En genes. The reverse experiment of recombining

ectoderm tissue from Otx-2 homozygous embryos and mesoderm from wild-type

embryos will be performed to determine if the Oh-2 negative ectoderm tissue

can respond to inducing signals from the mesoderm.

Cloning the _aenesg,plated bv Otx-2fHNF-3B

As discussed in Chapter 4, Ofx-2 and HNF-30 synergistically interact in

anterior midline patterning. RNA in situ hybridization experiments suggest that

Shh, a secreted molecule, and Six3, a nuclear transcriptional factor, may be

regulated by Otx-2/HNF3B. It is of interest to delineate the regulatory elements

which may interact with Otx-2 and HNF3B in these genes. To better understand

the molecular basis of anterior midline patterning, it would be also very useful to

identify other potential target genes which are involved in anterior midline

patterning. One way to clone the downstream genes regulated by Otw-2WNF3R

is to combine the method for establishing single cell cDNA libraries (Brady and

Iscove, 1993) with suppression subtractive hybridization (Diatchenko et al.,

1996). The method of generating cDNA libraries from a small number of cells

overcomes the tissue limitation of the embryo and the suppression subtractive

hybridization enriches for rare sequences greater than 1 000-fold and

simdtaneously prevents non-target (not differentially expressed) DNA

amplification in a model system (Diatchenko et al., 1996). Therefore, it would

be reasonable to combine these two methods to clone genes regulated by Otw-

2WNF4B. In this experiment, the tester cDNA will be prepared from the anterior

midline region, where both genes are coexpressed, in double heterozygous

embryos at E7.5-7.7 and the driver cDNA will be prepared from the same region

of wild-type embryos. If the combined method is successful, it can be applied to

clone other genes involved in anterior patterning using anterior tissues

(endoderm or mesendoderm) as testers and posterior tissues as drivers. These

studies will contribute our insight to better understanding the genetic control of

anterior patterning .

REFERENCES

Acampora, D., Avantaggiato, V., Tuorto, F. and Simeone, A. (1997). Genetic control of brain morphogenesis through Ok gene dosage requirement. Development 124, 36393650

Acampora, Dm, Mazan, S., Avantaggiato V., Barone P., Tuorto F., Lallemand Y., Brulet P., Simeone A. (1996). Epilepsy and brain abnormalities in mice lacking the 00rl gene. Nat Genet, 14(2), 21 8-222.

Acampora, D., Mazan, S., Lallemand, Y., Avantaggiato, V., Maury., Simeone, A. and Brulet, P. (1995). Forebrain and midbrain regions are deleted in OtxZ-/- mutants due to a defective anterior neuroectoderm specification during gastrulation. Development 121, 3279-3290.

Adam, M. D., Dubnick, M., Kerlavage, A. R., Moreno, R., Kelly, J. M., Utterback, T. R., Nagle, J. W., Fields, J. W. and Venter, J. C. (1992). Sequence identification of 2,375 human brain genes. Nature 355. 632-634.

Akam, M. (1987). The molecular basis for metameric pattern in the Drosophila embryo. Development 101, 1-22.

Akam, M. (1 989). Hox and HOM: homologous gene clusters in insects and vertebrates. Cell 57, 347-349.

Ang, S.-L. and Rossant, J. (1993). Anterior mesendoden induces mouse Engrailed genes in explant cultures. Development 1 18. 139-1 49.

Ang, S.-L. and Rossant, J. (1 994). HNF-3B is essential for node and notochord formation in mouse development. Cell 78. 561 -574.

Ang, S.-L., Wierda, A., Wong, D., Stevens, K. Am, Cascio, S., Rossant, J. and Zaret, K. S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF-3alfarkhead proteins. Development 1 19, 1301 -1 31 5.

Ang, S.-L:, Conlon, R. A*, Jin, 0. and Rossant, J. (1994). Positive and negative s~gnals from mesoderm regulate the expression of mouse Otx2 in ectoderm explants. Development 120, 979-989.

Ang, Sm-L., Jin, 0.. Rhinn, M., Gaigfe, N., Stevenson, L. and Rossant, J. (1996). A targeted mouse h - 2 mutation leads to severe defects in gastrulation and formation of axial mesoderm and to deletion of rostra1 brain. Development, 1 22, 243-252.

Bally-Cuif L. and Boncinelli E. (1 997). Transcription factors and head formation in vertebrates. Bioessays, 1037, 127-1 35.

Bally-Cuif, L., Gulisano, M., Broccoli, V. and BoncineBi, E., (1995). c- O W is expressed in two different phases of gastrulation and is sensitive to retinoic acid treatment in chick embryo. Mech. Dev. 49,4463.

Barnes, J. D., Crosby, J. L., Jones, C. M., Wright, C. V. E. and Hogan, B. L. M. (1994). Embryonic expression of Lim-1, the mouse homolog of Xenopus XLim-I, suggests a role in lateral mesoderm differnetiation and neurogenesis. Dev. Biol. 168-1 78.

Barth, K. A. and Wilson, S. W. (1995). Expression of zebrafish nk2.2 is influenced by sonic hedgehoghertebrate hedgehog-1 and demarcates a zone of neuronal differentiation in the embryonic forebrain. Development 121, 1755- 1 768.

Beddington, B. S. P. (1994). Induction of a second neural axis by the mouse node. Development 120, 61 3-620.

Beddington, B. S. P. and Robertson, E. J. (1989). An assessment of the developmental potential of embryonic stem cells in the midgestation mouse embryo. Development 105, 733-737.

Blitz, I. C. and Cho K. W. Y. (1995). Anterior neurectoderm is progressively induced during gastrulation: the role of the Xenopus horneobox gene orthodenticle. Development 121, 993-1004.

Blum, M., Gaunt, S. J., Cho, K. W. Y., Steinbeisser, H., Blurnberg, B., Bittner, D. and De Robertis, E. M. (1992). Gastrulation in the mouse: the role of the homeobox gene goosecoid. Cell 69, 1097-1 106.

Blumberg, B., Wright, C. V., DeRobertis, E. M. and Cho, K. W. Y. (1 991 ). Organizer-specific homeoboxes in Xenopus laevis embryos. Science 253, 194-1 96.

Boncinelli, E., Simeone, A., Acampora, D. and Mavilio, F. (1991). Hox gene activation by retinoic acid. Trends Genet 7, 329-334.

Bouwmeester, T. and Leyns L. (1997). Vertebrate head induction by anterior primitive endoderm. Bio 1268, 855-863.

Bouwmeester, T., Kim, S.-h., Sasai, Y., Lu, B. and De Robertis, E. M. (1996). Cerberus is a headinducing secreted factor expressed in the anterior endoderm of Spemann's organizer. Nature 382, 595-601.

Brady, G. and lscove, N. (1 993). Construction of cDNA libraries from single cells. Methods in enzymology 225, 61 1-61 5.

Carlson, B. M. (1996). Pattents Foundations of Embryology. 6th ed. McGraw- Hill. Inc. pp488.

Candia, A. F., Hu, J., Crosby, J., Lalley, P. Am, Noden, Dm, Nadeau, J. H. and Wright, C. V. E. (1992). Mox-1 and Mox-2 define a novel homeobox gene subfamily and are differentially expressed during early mesodermal patterning in mouse embryos. Development 1 16.1 123-1 136.

Carpenter, E. M., Goddard, J. M., Chisada, O., Manley, N. R. and Capecchi, M. R. (1993). Loss of HoxAl(Hoxl.6) function results in the reorganization of the murine hindbrain. Development 1 1 8. 1 063-1 675.

Carraso, A. E., McGinnis, W., Gehring, W. J. and De Robertis, E. M. (1984). Cloning of an X-laevis gene expressed during early embryogenesis coding for a peptide region homologous to Drosophila homeotic genes. Cell 37, 409-41 4.

Chang, D.E., Lopez, A., von Kessler, D. P., Chiang, C., Simandle, 6. K., Zhao, R., Seldon, M. F:, Fallon, J. F. and Beachy, P. A. (1994). Products, genetic linkeage, and 11mb patterning activity of a mouse hedgehog gene. Development 120: 3339-3353.

Chen, F. and M. R. Capecchi (1997). Targeted mutations in hoxa-9 and hoxb-9 reveal synergistic interactions. Dev Biol 181 , 186-1 96.

Chiang, C., Litingtung, Y., Lee, K. Em, Corden, J. L., Westphal, H. and Beach, P. A. (1996). Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature 383, 407-41 3

Chisaka, 0. and Capecchi, M. R. (1991). Regionally restricted developmental defects resulting from targeted disruption of the mouse homeobox gene hox1.5. Nature 350,473-479.

Chomczynski, P. and Sacchi, N. (1987). Single-step method of RNA isolation by acid g uanidinium thiocyanate-phenolchloroforrn extraction. Anal. Biochem. 162, 156-1 59.

Conlon, F. L., Lyons K.M., Takaesu, N., Barth K.S., Kispert A., Herrmann B., and Robertson, Em (1 994). A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Development 120, 191 9-1 928.

Conlon, R and Rossant, J. (1992). Exogenous retinoic acid rapidly induces anterior ectopic expression of murine Hox-2 gene in viva Development 116, 357-358.

Conlon, R. and Hermann, B. G. (1 993). Detection of messenger RNA by in situ hybridization to postimplantation embryo whole mounts. In Methods in Enzymology, Guides to Techniques in Mouse Development Vol. 225. (eds. P. M. Wasssarman and M. L. DePamphiiis) pp. 373-383. New York, USA: Academic press.

Dalton, D., Chadwick, R. and McGinnis, W. (1989). Expression and embryonic functions of empty spiracles: a Drosophila homeobox gene with two patterning function on the anteriorposterior axis of the embryo. Genes Dev. 3. 1 940-1 956.

Davis, A. and Capecchi M. (1996). A mutational analysis of the 5' HoxD genes: dissection of genetic interactions during limb development in the mouse. Dev Suppl 1 22, 1 175-1 1 85.

Davis, Am, Witte, Dm P., Heieh-Li, H. M., Potter, S. S. and Capecchi, M. (1995). Absence of radius and ulna in mice lacking hoxa-11 and hoxd-1 I . Nature 375, 791 -795.

Davis, C. A*, Holmyard, 0. P., Millen, K. J. and Joyner, A. L. (1991). Examining pattern formation in mouse, chicken and frog embryos with an En- specific antiserum. Development I 11, 287-298.

Dessler, D. S. and Melton, D. A. (1 994). Vertebrate embryonic induction: mesodermal and neural patterning. Science 266, 596-604.

De Robertis, E. M. (1994). Dismantling the organizer. Nature 374, 407408.

Diatchenko, L., Lau, Y-F. C., Campbell, A. P. et al., (1996). Suppression subtractive hybridization: A method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc. Natl. Acad. Sci. USA 93, 6025-6030.

Diettich, S., Schubert, F. R. and Gruss, P. (1993). Altered Pax gene expression in murine notochord mutants: the notochord is required to initiate and maintain ventral identity in the somite. Mech. Dev. 44, 189-207.

Don iac h, T. (1 993). Planar and vertical induction of anteroposterior pattern during the development of the amphibian central nervous system. J. Neurobiol. 24, 1256-1275.

Echelard, Y., Epstein, D. J., St-Jauques, B., Shen L., Mohler, J., McMahon, J. A. and McMahon, A. P. (1993). Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 75: 141 7-1 430.

Epstein, 0. J. and Joyner, A. (1997). Regionalization of Sonic Hedgehog expression along the anterior/posterior axis of the mouse CNS. The 62th symposium on pattern formation. Cold Spring Harbor, New York, pp60.

Erich, R., Belloni, E., Gaudenz, K., Jay, P., Berta, P., Scherer, S. W., Tsui, L C and Muenke. M. (1996). Mutations in the human Sonic Hedgehog gene cause holoprosencephaly. Nature Genetics 14, 357-360.

Fietz, M. J., Jacinto, Am, Taylor, A. M., Alexandre, C. and lngham, P.W. (1995). Secretion of the amino-terminal fragment of the Hedgehog protein is necessary for hedgehog signalling in Drosophila. Cur. Biol. 5, 643-651.

Figdor, M. and Stein, C. (1 993). Nature 363, 630-634.

Filosa, S., Rivera-Perez, J. A., Gomez, A. P., Gansmuller, A., Sasaki, H., Behringer, R. R. and Ang, L. gousecoid and HNF-3B genetically interact to regulate neural tube pattern during mouse embryogenesis. Development 124, 2843-2854.

Finkelstein, R. and Perrimon, N. (1990). The orthodeniicle gene is regulated by bicoid and torso and specifies Drosophila head development. Nature 346, 485488.

Finkelstein, R., Smouse, D., Theresa, M. C., Spradling, A. C. and Perrimon, N. (1990). The orthodenticle gene encodes a novel homeo domain protein involved in the development of the Drosophila nervous system and ocellar visual structures. Genes Dev. 4, 151 6-1 527.

Finkelstein, R. and Boncinelli, E. (1994). From fly to mammalian forebrain: the story of otd and Otx. Trends Neurosci. 9,310-315.

Friedrich, 0. and Soriao, P. (1 991). Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice. Genes Dev. 5, 151 3-1 523.

Frigerio, G., Burri, M., Bopp, Dm, Baurngartner, S. and Noll, M. (1986). Structure of the segmentation gene paired and the Drosophila PRD gene set as part of a gene network. Cell 47,735-746

Frohman, M. A., Boyle, M. and Martin, G. (1990). Isolation of the mouse HOX 2.9 gene: analysis of embryonic expression suggests that positional information along the anterior-posterior axis is specified by mesoderm. Development 1 10, 589-607.

Fromental-Ramain C., Warot, X., Lakkaraju, S., Favier, B., Haack, H., Birling, C., Dierich, A., Dolle, P. and Chambon, P. (1996). Specific and redundant functions of the paralogous Hoxa-9 and Hoxd-9 genes in forelimb and axial skeleton patterning. Development 122, 461 -472.

Gehring, W. J. et al. (1 994). Homeodomain-DNA recognition. Cell 78, 21 1 - 223.

Gilbert, S. F. (1994). Developmental Biology, 4th Ed.

Gu, H., Marth, J. Dm, Orban, P. C., Mossmann, H. and Rajewsky, K. (1 994). Deletion of a DNA polymerase A gene segment in T cells using cell type-specific gene targeting. Science 265, 103-1 06.

Guillemot, F. and Joyner, A. L. (1993). Dynamic expression of the murine achaete-scute homolog (MASH-1) in the developing nervous system. Mech. Dev. 42, 171-1 85.

Hamrnerschmidt, M., Bitgood, M. J. and McMahon, A. P. (1996). Protein kinase A is common negative regulator of Hedgehog signaling in the vertebrate embryo.

Hammerschmidt, M., Brook, A. and McMahon, A. P. (1997) The world according to hedgehog. Trends. Genet. 13

Hanes, S. D. and Brent, R. (1989). DNA specificity of the bicoid activator protein is determined by homeodomain recognition helix residue 9. Cell 57, 1 275-1 283.

Hatta, K., Kimmel, C. B., Ho, R. K. and Walker, C. (1991). The cyclops mutation blocks specification of the floorplate of the zebrafish central nervous system. Nature 350,334341.

Hatta, K., Piischel, A. W. and Kimmel, C. B. (1994). Midline signalling in the primordium of the zebrafish anterior central nervous system. Proc. Natl. Acad. Sci. USA 91,2061-2065.

Hermann, 6. 0. (1991). Expression pattern of the Brachyury gene in whole- mount mutant embryos. Development 1 1 3, 91 3-91 7.

Hermesz, E., Mackem, S. and Mahon K. A. (1996). Rpx: a novel anterior- restricted homeobox gene progressively activated in the prechordal plate, anterior neural plate and Rathke's pouch of the mouse embryo. Development 122, 41-52.

Heynes, M., J. A. Porter, C. Chiang, D. Chang, M. Tassier-Levigne, P. A. Beachy, and A. Rosenthal. (1995). Induction of midbrain dopaminergic neurons by Sonic Hedgehog. Neuron 15, 35-44.

Hiroyuki, S., Toyama, K., Shioya, H., Ito, M., et al. (1997). Rapid colorectal adenoma formation initiated by conditional targeting of the Apc gene. Science 278, 1 20-1 23.

Ho, R. (1 992). Axis formation in the embryo of the zebrafish. Semin. Dev. Biol. 3, 53-64.

Hogan, B., Costantini F. and Lacy, E. (1986). Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbor: Cold Spring Harbor Laboratory Press.

Hogan, B., Beddington, R., Costantini F. and Lacy, E. (1994). Manipulating the Mouse Embryo: A Laboratory Manual Cold Spring Harbor: Cold Spring Harbor Laboratory Press.

Hynes, M., Poulsen, K., Tessier-Lavigne, Rosenthal, A. (1 995). Control of neuronal diversity by the floor plate : Contact-Mediated induction of midbrain dopaminergic neurons. Cell 80, 951 01.

Joyner, A. L. (1996). Engrailed, Wnt and Pax genes regulate midbrain- hindbrain development. Trends Genet 12, 15-20.

Joyner, A. and Guillemot, F. (1994). Gene targeting and development of the nervous system. Cum Opin Neurobiol4, 3742.

Jurgens, G., Wieschaus, Em, Nussiein-Volhard, C. and Kluding, H. (1 984). Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. 11. Zygotic loci on the third chromosome. Roux's Arch. Dev. Biol. 1 93, 283-295.

Jurgens, G., Lehmann, R., Schardin, M. and Nusslein-Volhard, C . (1 986). Segmental organization of the head in the embryo of Drosophila melanogaster. Wilhelm Roux's Arch. Dev. Biol. 1 95, 359377.

Kaufman, M. H. (1992). The Atlas of Mouse Development. Academic Press.

Kelly 0. G. and Melton D. A. (1995). Induction and patterning of the vertebrate nervous system. TIG 11.273-278.

Kessei, M., Balling, R. and Gruss, P. (1990). Variations of cervical vertebrae after expression of a Hoxl .I transgene in mice. Cell 61. 301-308.

Kessler, Dm S. and Melton, D. A. (1994). Vertebrate embryonic induction: mesodermal and neural patterning. Science 266, 596-603.

Klambt, C., Jacobs, J. R. and Goodman, C. (1991). Cell 64, 801-815.

Krauss, S., J.-P. Concordet, and P. W. Ingham (1993). A functionally conserved homolog of the Drosophila segment polarity gene h h is expressed in tissues with polyrizing activity in zebrafish embryos. Cell 75, 1431 -1 4444.

Krumlauf, R. (1 994). Hox genes in vertebrate development. Cell 78, 191 -201.

Laughon, A. (1 991). DNA binding specificity of homeodomains. Biochemistry 30, 1 1357-1 1367.

Lawrence, P. A. and Morata, G. (1994). Horneobox genes: their function in Drosophila segmentation and pattern formation. Cell 78, 181-1 89.

Lawson, K. A., Meneses, J. J., and Pedersen, R. A. (1991). Clonal analysis of epiblast fate during germ layer formation in the mouse embryo. Development 1 13, 891 -91 1.

Lendahl, U.! Zimmerman, L. B. and McKay, R. D. 0. (1990). CNS stem cells expression a new class of intermediate filament protein. Cell 60, 585-595.

Levine, M., and Hoey, T. (1988). Homeobox proteins as sequence specific transcription factors. Cell 55, 537-540.

Lewis, E. 6. (1 978). A gene complex controlling segmentation in Drosophila. Nature 276, 565-570.

Lewis, E. 6. (1981). Developmental genetics of the bithorax complex in Drosophila. In Developmental Biology Clising Purified Genes, Vol 23, D. Brown and F. Fox, eds. (New York : Academic Press), 184208.

Lewis, Em 8. (1982). Control of body segment differentiation in Drosophila by the bithorax gene complex. Prog. Clin. Biol. Res. 85, 269-288.

Li, X., Chuang, C-K., Mao, C-Am, Angerer, L. M. and Klein, W. H. (1 997). Two otx proteins generated from multiple transcripts of a single gene in Strongylocen trotus purpuratus. Dev. Biol. 1 87, 253-266.

Li Y., Allende ML, Finkelstein R. and Weinberg ES (1994). Expression of two zebrafish orthodenticle-related genes in the embryonic brain. Mech Dev 43, 229-244.

Lin, T.-P., Labosky, P. A., Grabel, L. B., Kozak, C. Am, Pitman, J. L., Kleemen, J. and MacLeod, (1994). The pem homeobox gene is X-linked and exclusively expressed in extraembryonic tissues during murine development. Dev. Biol. 166, 170-1 79.

Lufkin, T., Dierich, A., Lemew, M., Mark, M. and Chambon, P. (1991). Disruption of the Hoxl.6 homeobox gene results in defects in a region corresponding to its rostra1 domain of expression. Cell 66, 11 05-1 11 9.

Lumsder, A. and keynes, R. (1989). Segmental patterns of neuronal development in the chick hindbrain. Nature 337, 424-428.

Lumsder, A., Sprawson, N. and Graham, A. (1991). Segmental origin and migration of neural crest cells in the hindbrain region of the chick embryo. Development 113

Macdonald, R., Barth, K. A., Xu, Q., Holder, N., Mikkola, I. and Wilson, S. W. (1995). Midline signalling is required for Pax gene regulation and patterning of the eyes. Development 121,3267-3278.

Marin, F. and Puelles, L. (1994). Patterning of the embryonic avian midbrain after experimantal inversions: a polarizing activity from the isthmus. Dev. Biol. 163, 19-37.

Mark, M., Rijci, F. M. and Chambon, D.(1997). Homeobox genes in embryogenesis and pathogenesis. Pediatr Res 42, 421 429.

Marti, E., D. A. Bumcrot, R. Takada, and A. P. McMahon (1995). Requirement of 19K form of Sonic Hedgehog for induction of distinct ventral cell types in CNS explants. Nature 375,322-325.

Matsuo, I., Kuratani, S., Kimura, C., Takeda, N. and Aizawa, S. (1995). Mouse Ot%2 functions in the formation and patterning of rostra1 head. Genes Dev. 9,264G2658.

McGinnis, W. and Krumlauf, R. (1992). Homeobox genes and axial patterning. Cell 68, 283-302.

McGinnis, W., Levine, M., Hafen, E., Kuroiwa, A., and Gehring, W. J. (1984a). A conserved DNA sequence in horneotic genes of the Drosophila Antennapedia and bithoraw complexes. Nature 308, 428-433.

McGinnis, W., Hart, C. P., Gehring, W. and Ruddle, F. H. (1984b). Molecular cloning and chromosome mapping of a mouse DNA sequence homologous to homeotic genes of Drosophila. Cell 38. 657-680.

McMahon, A. P. and Bradly, A. (1990). The Wnt-1 (int-I) protodncogene is required for development of a large region of the mouse brain. Cell 62, 1073- 1085.

Mercier P., Simeone A., Cotelli F. and Boncinelli E. (1995). Expression pattern of two Otr genes suggests a role in specifying anterior body structures in zebrafish. Int J Dev Biol39(4), 559473.

Michio, Yoshida, Yoko Suda, lsao Matsuo, Norimasa Miyamoto, Naoki takeda, Shigeru Kuratani and Shinichi Aizawa (1 997). Emxl and Emx2 functions in development of dorasal telencephalon. Development 124, 101-111.

Monaghan, A. P., Kaestner, K. H., Grau, E. and Schutz, G. (1993). Postimplantation expression patterns indicate a role for the mouse forkheadlHNF-38 , and genes in determination of the definitive endoderm, chordamesoderm and neuroectoderrn. Development 1 19, 567-578.

Mortensen, R. M., Conner, D. Am, Chao, S., Geisterfer-Lowrance, A. 1. and Seidman, J. 0. (1992). production of homozygous mutant ES cells with a single targeting construct. Mol. Cell. biol. 12, 239102395.

Nagy, Am, Rossant, J., Nagy, R., Abramow-Newerly, W. and Roeder, J. C. (1993). Viable cell culturederived mice from early passage embryonic stem cells. Proc. Natl, Acad. Sci. USA 90. 8424-8428.

Oliver, G. and Gruss, P. (1997). Current views on eye development. Trends Neurosci 20, 41 5-421.

Oliver, G., Maihos, A., Wehr, R., Copeland, N. G., Jenkins, N. A. and Gruss, Po (1995). S1x3, a murine homologue of the sine oculis gene, demarcates the most anterior border of the developing neural plate and is expressed during eye development. Development 1 21, 4045-4055.

Pannese, M. Polo, C., Andreazzoli, M., Vignali, R., Kablar, B., Barsacchi, G. and Boncinelli, E. (1995). the Xenopus homologue of Otx2 is a maternal homeobox gene that demarcates and specifies anterior body regions. Development 121, 707-720.

Puelles, L., and Rubenstein, L. R. (1993). Trends in Neurosci 11, 472- 478.

Pellegrini, M., Mansouri, A., Simeone, Am, Boncinelli, Em and Gruss P. (1 996). Dentate gyrus formation requires Emx2. Development 122, 3893- 3898.

Placzek, M. (1995). The role of the notochord and floor plate in inductive interactions. Cum. Opin. Genet. Dev. 5, 499-506.

Proudfoot, N. J. and Brownlee, G. G. (1976). 3' Nontoding region sequences in eukaryotic mRNA. Nature 263, 21 1-214.

Puelles, L. and Rubinstein, J. L. R. (1993). Expression patterns of homeobox and other putative regulatory genes in the embryonic mouse forebrain suggest a neuromeric organization. Trends Neurosci. 16. 472-479

Ramirez-Solis, Zheng, H., Whiting, J., Krurnlauf, R. and Bradley, A. (1 993). Hoxb-4(Hox2.6) mutant mice show homeotic transformation of a cervical vertebra and defects in the closure of the sternal rudiments. Cell 73, 279-294.

Riddle, R.D., Johnson, R. L., Laufer, E. and Tabin C. (1993). Sonic Hedgehog Mediates the Polarizing Activity of the ZPA. Cell 75, 1401 -1 41 6.

Roessler, E., Belloni, E., Gaudenz, K., Jay, P., Berta, P., Scherer, S. W., Tsui, L-P. and Muenke, M. (1996). Mutations in the human Sonic Hedgehog gene cause holoprosencephaly. Nature Genet 14,357-360.

Rong, P.M., Teillet, MrA., Ziller, C. and Le Douarin, N. M. (1992). The neural tubdnotochord complex is necessary for vertebral but not limb and body wall striated muscle differentiation. Development 1 1 5, 657-672.

Rossant J. (1 990). Manipulating the mouse genome: implications for neurobiology. Neuron 2, 323-334.

Royet, J. and Finkelstein R. (1995). Pattern formation in Drosophila head development: the role of the orthodentide homeobox gene. Development 121, 3561 -3572.

Rubenstein, J. L. R., Martinez, S., Shimamura, K.! and Puelles, L. (1 994). The embryonic vertebrate forebrain: the prosomenc model. Science 266:578-780.

Ruiz i Altaba, A. (1 994). Pattern formation in the vertebrate neural plate. TINS 17, 233-243.

Ruiz i Altaba, A. and Jessell, T. M. (1992). Pintallavis. a gene expressed in the organizer and midline cells of frog embryos: involvement in the development of the neural axis. Development 1 1 6. 81-93.

Sasai, Y., Lu, B., Steinbeisser, H., Geissert, D., Gont, L. K. and D e Robertis, E. M. (1994). Xenopus chordin: a novel dorsalizing factor activated by organizer-specific homeobox gene. Cell 79, 774790.

Sasaki, H. and Hogan, B. I. M. (1993). Differential expression of multiple fork head related genes during gastrulation and axial pattern in the mouse embryo. Development 11 8, 47-59.

Sasaki, H. and Hogan, B. L. M. (1994). HNF-3B as a regulator of floor plate development. Cell 76, 103-1 1 5.

Scott, M. P., Tamkun, J. W. and Hartzell, G. (1989). The structure and function of the horneodomain. Biochim. Biophys. Acta 989. 25-48.

Scott, M. P. and Weiner, A. J. (1984). Structural relationships among genes that control development: sequence homology between the Antennapedia, Ultrabithorax and fushi tarazu loci of Drosophila. Proc. Natl. Acad. Sci. USA 81,411541 19.

Serafini, T.! Kennedy, T., Galko, N., Mirzayan, C., Jessell, T. and Tessier-Lawgne, M. (1994). The netrins define a family of axon outgrowth- promoting proteins homologous to C. elegans UNC-6. Cell 78, 409-424.

Shawlot, W. and Behringer, R. M. (1995). Requirement for Urn1 in head- organizer function. Nature 374. 425-430.

Shih, J. and Fraser, S. Em (1996). Characterixing the zebrafish organizer: microsurgical nalysis at the early shield stage. Development 122, 131 3-1 322.

Shibata, H., Toyama, K., Shioya, H., Ito, M., et al. (1997). Rapid colorectal adenoma formation initiated by conditional targeting of the Apc gene. Science 278, 120-1 23.

Simeone, A., Acampora, D., Gulisano, M., Stornaiuolo, A. and Boncinelli, E. (1992a). Nested expression domains of four horneobox genes in developing brain. Nature 358, 687-690.

Simeone, A., Gulisano, M., Acampora, D., Stornaiuolo, Am, Rambaldi, M. and Boncinelli, E. (1 992b). Two vertebrate homeobox genes related to the Drosophila empty spiracles gene are expressed in the embryonic cerebral cortex. EMBO J. 1 1,2541 -2550.

Simeone, A., Acampora Dm, Mallamaci, A., Stornaiuolo, A., DrApice, M. R., Nigro, V. and Boncinelli, E. (1 993). A vertebrate gene related to orfhodenticle contains a homodomain of the bicoid class and demarcates anterior neuroectoderm in the gastrulating mouse embryo. EMBO J. 12, 2735- 2747.

Simeone, Am, Avantaggiato, V., Moroni, M. C., Mavillo, F., Arm, C., Cotelli, R., Nigro, V. and Acampora, Dm (1995). Retinoic acid induces stage-specific antero-posterior transformation of rostra! central nervous system. Mech. Dev. 51,83-98.

Slack, J. and Tannahill, D. (1 992). Mechanism of anteroposterior axis specification in vertebrates. Lessons from amphibians. Development 1 1 4, 285- 302.

Smith, W. C. and Harland, R. M. (1992). Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos. Cell 70, 829-840.

Smith, W. C., Knecht, A. K., Wu, M. and Harland, R. M. (1993). Secreted noggin protein mimics the Spemann organizer in dorsalizing Xenopus mesoderm. Nature 361, 547-549.

Spemann, H. (1 931 ). Embryonic Development and Induction. Yale University Press.

Spemann, H., Mangold, H., (1 924). Uber induction von Embryonanlagen durch Implantation Artfremder Organis Atoren. Roux's Arch Dev Biol 100, 599- 638.

Storey, K. G., Crossley, J. M., de Robertis, E. M., Norris, W. E. and Stem, C. D. (1 992). Neural induction and regionalisation in the chick embryo. Development 1 14. 729-741.

Sulik, K., Dehart, 0. B., lnagaki, T., Vrablic, J. L., Gesteland, K. and Schoenwold, G. C. (1994). Morphogenesis of the murine node and notochordal plate. Dev. Dynamics 201, 260-278.

Tam, P. P. 1. (1989). Regionalitation of the mouse embryonic ectoderm: allocation of prospective ectodermal tissues during gastrulation. Development 107, 55-67.

Tanabe, Y. and Jessell, T. M. (1996). Diversity and pattern in the developing spinal cord. Science 274, 1 1 15-1 123.

Tao, W. and Lai, E. (1992). Telencephalon restricted expression of BF-I, a new member of the HNFaNforkhead gene family, in the developing rat brain. Neuron 8, 957-966.

Tarkowski, A. K. (1 977). Development of cytochalasin 8-induced tetraploid and diploid/tetraploid mosaic mouse embryos. J. Embryol. Exp. Morph. 41, 47- 64.

Teillet, M.-A. and Le Douarin, N. M. (1983). Consequences of neural tube and notochord excision on the development of the peripheral nervous system in the chick embryo. Dev. Bid. 192-21 1.

Thomas, K. R. and Capecchi, M. R. (1 990). Targeted disruption of the murine Wint-1 proto-oncogene resulting in servere abnormalities in midbrain and cerebellum development. Nature 346, 847-850.

Thomas, P. and Beddington, R. S. P. (1996). Anterior primitive endoderm may be reponsible for patterning the anterior naural plate in the mouse embryo. Current Bid. 6, 1 487-1 496.

Treisman, J., Gonczy, P., Vashishtha, M., Harris, E. and Desplan. (1 989). A single amino acid determines the binding specificity of homeodomain proteins. Cell 59, 553-562.

Tybulewicz, V. L. J., Crawford, C. Em, Jackson, P. KO, Bronson, R. To, and Mulligan, R. C. (1991). Neonatal lethality and lymphopenia in mice with a homozygous disruption of the c-abl proto-oncogene. Cell 65, 1153-1 163.

Varlet, I.! Collignon, J. and Robertson, E. J. (1 997). nodal expression in the prim~twe endoderm is required for specification of the anterior axis during mouse gastrulation. Development 1 24, 1 033-1 044.

Waddington, C. H. (1933). Induction by the primitive streak and its derivatives in the chick. J. Exp. Biol. 1 0,3846-

Webb, G. C., Thomas, P. Q., Ford, J. H. and Rathjen P. D. (1993). Hesxl a homeobox gene expressed by murine embryonic stem cells, maps to mouse chromosome 14, bands A3-B. Genomics 18,464-466.

Weinstein, D. C., Ruiz i Altaba, A., Chen, W. S., Hoodless, P., Prezioso, V. R., Jessell, T. M., Darnell, J. Em, Jr. (1 994). The winged- helix transcription factor HNF-38 is required for notochord development in the mouse embryo. Cell 78, 575-588.

Wieschaus, E., Nusslein-Volhard, C. and Jurgens, C. (1 984). Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. I II. Zygotic loci on the X chromosome. Wilhelm Roux's Arch. Dev. Biol. 193, 296- 307.

Wieschaus, E., Perrimon, N. and Finkelstein, R. (1 992). Development 115, 801-81 1.

Wilkinson, D. G., Bhatt, S., Cook, M., Boncinelli, E. and Krumlauf, R. (1 989a). Segemental expression of Hox-2 Homeobox genes in the developing mouse hindbrain. Nature 341, 405-409.

Wilkinson, Dm G., Bhatt, S., Chavrier, P., Bravo, R. and Charnay, P. (1 989b). Segmental -specific expression of a zinc-finger gene in the developing nervous system of the mouse. Nature 337,461464.

Wurst, W. and Joyner, A. L. (1993). Production of targeted embryonic stem cell lines. In Gene Targeting: A Practical Approach, (ed. A L Joyner), pp. 31 - 62. Oxford: Oxford University Press.

Xuan, S., Baptists, C. Am, Balas, G., Tao, W., Soares, V. C. and Lai, E. (1995). Winged helix transcription factor BF-I is essential for the development of the cerebral hemispheres. Neuron 1 4, 1 1 41 -1 1 52.

Yamaguchi, T. P., Dumont, D. J., Conlon, R. A., Breitman, M. L. and Rossant, J. (1993). flk-1, an flhelated receptor kinase is an early marker far endothelial cell precursors. Development 1 18, 489-498.

Yoshida, M., Suda, Y., Matsuo, I., Miyamoto, N., Takeda, N., Kuratani, S. and Aizawa, S. (1997). Emxl and Emx2 functions in development of dorsal telencephalon. Development 1 19, 1 01-1 1 1.

Zhou, X., Sasaki, H., Lowe, L., Hogan, 6. L. M. and Kuehn, M. R. (1993). Nodal is a novel TGF-&like gene expressed in the mouse node during gastrulation. Nature 361, 543-547.

Zimmerman, L., Lendahl, U., Cunningham, M., McKay, R., Parr, B. et al. (1994). Independent regulatory elements in the nestin gene direct transgene expression to neural stem cells or muscle precursors. Neuron 12, 1 1 - 24.

l MAGE EVALUATION TEST TARGET (QA-3)

APPLIED - 4 IMAGE. lnc = 1653 East Main Street - -. , Rochester. NY 14609 USA -- -- - - Phone: 716M82-0300 -- -- - - Fax: 71 6/200-5989