“Every one is aware of the exquisite delicacy, and
acknowledges the admirable construction of the Eye; its
excellence has been at all times proverbial; there are but
few, however, who understand the real grounds upon which
this excellence depends, and an attempt to elucidate these is
perhaps deserving of notice.”
-JOHN WALKER
taken from “The Philosophy of the Eye”
Cambridge press,1837
i
Intrinsic Mechanisms Governing
Retinal Progenitor Cell Biology:
Retinal Homeobox Transcriptional Regulation and the
Function of Forkhead Transcription Factors During
Eye Development
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate School
of The Ohio State University
By
Holly E. Moose, B.A.
Integrated Biomedical Sciences Graduate Program
*****
The Ohio State University
2009
Dissertation Committee:
Professor Heithem El-Hodiri, Advisor
Professor Christine Beattie
Professor Andy Fischer
Professor John Oberdick
i
Copyright by
Holly E. Moose
2009
i
ABSTRACT
Understanding the development and maintenance of Retinal Progenitor Cells
(RPCs) is critical to understanding normal and disease processes within the neural retina.
To understand RPC biology, it is important to understand the transcriptional regulation
of known intrinsic regulators, and continue to identify new RPC expressed genes to
demonstrate their function during eye development. The studies presented in this work
address the transcriptional regulation of the Xenopus laevis Rx gene product, Rx2A, and
function of two newly recognized RPC genes, FoxO3 and FoxM1.
To further the understanding of transcriptional regulation in RPCs, we
characterized the Rx2A promoter in transgenic embryos. Both the distal portion and the
proximal portion of the Rx2A promoter are sufficient for expression of a GFP transgene
in the developing eyes. We identify a highly conserved element in the distal region of the
Rx2A promoter (UCE). Within UCE, an OTX, SOX and POU site act as cis-elements to
coordinately specify proper gene expression in the developing eye. We show that the
activity of the proximal promoter is dependant on a forkhead-binding element (FBE). In
addition, we have shown that the distal region containing the UCE can cooperate with the
FBE to maintain robust Rx expression throughout all stages of eye development. The
work associated with the transcriptional regulation of Rx furthers our understanding of
how this primary retinal transcription factor is regulated. This is applicable to the
ii
understanding of RPC development since Rx is one of the first eye field transcription
factors expressed in the anterior neural plate as RPCs are specified.
The identification of the FBE within the Rx2A promoter led to further
investigation of the involvement of the forkhead family of transcription factors in
vertebrate eye development. We present a discussion of current data regarding the
expression and function of this family of transcription factors during eye development,
and we present the expression pattern of 5 forkhead transcription factors (FoxG1, FoxN2,
FoxN4, FoxM1 and FoxP1) in the maturing X. laevis retina. These forkhead gene
products have not been previously described in the maturing retina of X. laevis.
We chose to pursue studies of FoxO3 and FoxM1 in the developing neural retina
to further the understanding of RPC regulation by transcription factors of the forkhead
family. These two factors were chosen for co-current studies for the following reasons:
(1) neither gene product had been ascribed a role in developing RPCs, (2) their known
functions suggested they act in opposing ways with regards to the cell cycle, and (3) due
to their expression patterns both FoxO3 and FoxM1 serve as candidate factors to regulate
the Rx2A promoter through the FBE.
To define the role of FoxO3 during vertebrate eye development, we
overexpressed FoxO3 RNA in the anterior neural plate of X. leavis embryos. FoxO3
overexpression results in embryos with small eyes. The small eye phenotype is a result
of decreased proliferation, induction of apoptosis, and changes in RPCs gene expression.
The phenotype can be exacerbated by introducing a threonine to alanine mutation at a
conserved PI3K phosphorylation site, which produces a constitutively nuclear form of
FoxO3. The changes in gene expression suggest that FoxO3 can function to delay the
iii
differentiation of RPCs, although they are properly specified. The data supports our
original hypothesis regarding FoxO3 as cell cycle antagonist with the ability to alter the
differentiation capacity of RPCs.
To investigate the function of FoxM1 in developing RPCs, we performed loss-offunction studies using morpholino oligonucleotides (MO). FoxM1 knockdown in the
anterior neural region of X. laevis results in embryos with slightly smaller eyes and clear
defects in retinal lamination. Our data reveals that FoxM1 is not necessary for the
specification of RPCs, and suggests that differentiation into the Muller glia, and rod
photoreceptor lineages is possible with reduced levels of FoxM1.
The studies presented concern the transcriptional regulation of the Rx gene
product, Rx2A, and a role of FoxO3 and FoxM1 during RPC development. Collectively,
they represent an advancement in the knowledge regarding two important intrinsic
mechanisms governing RPC development: transcriptional regulation and transcription
factor function.
iv
ACKNOWLEDGEMENTS
To the members of my graduate studies committee, thank you for your thoughtful insight
and critical analysis regarding my thesis project. Your expertise has improved the
science in my thesis and my thinking as a scientist. I am very grateful.
To Heithem, thank you for teaching me the science as a graduate student, mentoring me
as a future colleague, and providing me with opportunities to think independently. Thank
you for fostering my ability to ask questions, and for sometimes literally challenging me
to do so. Thanks for the countless ways you have furthered my career, most of which I
imagine I won’t completely appreciate until long after I leave the
bench in Wexner 452.
Thanks to the girls of W452, past and present. Lisa, thank you for all your expertise
regarding specific protocols, and knowledge of the ways of the lab. It is impossible to
avoid being influenced by your dedication to basic science. Yi, thanks for your honest
opinions and helpful attitude. Thanks for your willingness to lend a hand, or run a PCR
when my confidence was shaken. Reyna, what a ride it has been! We’ve been to
Providence and Florida and back- all in the name of RPCs. You have had more effect on
my motivation to become a better bench scientist than you probably know. It has been
both truly enjoyable and productive to be your labmate! Thanks for the constant
inspiration, frequent laughter, and occasional commiseration.
To my family, Johnson and Moose, I am so appreciative of your constant support.
Thanks for both the motivation for and the distraction from my studies. Whether it was
laughing about the kids or eating a homemade meal, your thoughtfulness has made quite
an impact on my success. Thanks, Mom and Dad, for your unwavering interest in my
graduate school life. Asking about experiments, calling about presentations or allowing
me stay a bit longer in the lab have all contributed to achieving this goal.
Thank you Dan, for your perpetual support and confidence in me. For allowing me to
reach for lofty goals, and expecting me to achieve them. I am constantly aware of how
lucky I am to have you walking beside me. You expect me to reach high, yet keep me
grounded. Love you lots. And Zachary, for the hugs and kisses as I go out the door, the
smiling faces when I get home, and all the talks we’ve had about how
“Mommy plays with frogs”. I love you
v
DEDICATION
To my father, who taught me the benefits of education and instilled an appreciation of
scientific endeavors.
To my husband, who strengthens my conviction to pursue my goals by relentlessly
believing in my abilities.
To my son, who will someday realize the impact of his youthful candor
on my current successes.
vi
VITA
November 3, 1980..........................................................................................................Born
Rochester, Michigan
2003..........................................................................................B.A. Biology and Chemistry
Capital University
Bexley, Ohio
2003-2009 ...............................................................................Graduate Research Associate
Integrated Biomedical Graduate Sciences Program
The Ohio State University
Columbus, Ohio
RESEARCH PUBLICATIONS
1. Moose HE, Kelly LE, Nekkalapudi S, El-Hodiri HM. “Ocular forkhead transcription
factors: seeing eye to eye.” Int J Dev Biol. 2009;53(1):29-36.
2. Ignatius MS, Moose HE, El-Hodiri HM, Henion PD. “colgate/hdac1 Repression of
foxd3 expression is required to permit mitfa-dependent melanogenesis.” Dev Biol. 2008
Jan 15;313(2):568-83. Epub 2007 Nov 9.
3. Moose HE, and El-Hodiri, HM. “FoxO3 Perturbs Vertebrate Eye Development by
Affecting Differentiation and Proliferation in Retinal Progenitor Cells.” In preparation.
4. Moose HE, and El-Hodiri, HM. “FoxM1 is Necessary for Normal Development and
Proper Lamination of the Neural Retina by Retinal Progenitor Cells.” In preparation.
5. Stanke JJ, Moose HE, El-Hodiri HM, Fischer AJ. “A comparitive study of Pax2
expression in glial cells in the mature retinas of brids and mammals.” In preparation.
FIELD OF STUDY
Integrated Biomedical Science
vii
TABLE OF CONTENTS
Abstract ............................................................................................................................... ii
Acknowledgments................................................................................................................v
Dedication .......................................................................................................................... vi
Vita.................................................................................................................................... vii
List of Tables .................................................................................................................... xii
List of Figures .................................................................................................................. xiii
List of Abbreviations .........................................................................................................xv
Chapter 1: Development and Regulation of Retinal Progenitor Cells by Known
Intrinsic Mechanisms.........................................................................................................1
1.1 Overview........................................................................................................................1
1.2 Eye development in vertebrates and specification of Retinal Progenitor Cells.............2
1.3 From RPC to retinal cell types: Molecular regulation of retinal
cell types by intrinsic factors .........................................................................................5
1.4 RPC decision making: The balance between proliferation and differentiation .............8
1.5 Xenopus Laevis as a model system to study RPC development ..................................11
1.6 The impact of this work relative to the field of retinal development...........................12
Chapter 2: Regulation of Retinal Homeobox Transcription by Cooperative Activity
Among Cis-elements ........................................................................................................19
2.1 Introduction..................................................................................................................19
2.2 Materials and Methods.................................................................................................22
2.3 Results..........................................................................................................................26
2.3.1 Identification and conservation of the Rx2A regulatory sequence...............26
2.3.2
The
Rx2A
regulatory
sequence
directs
gene
expression
in
regions
of
endogenous
Rx
expression...................................................................28
2.3.3
Spatial
and
temporal
specificity
of
Rx
expression
are
determined
by
regions
within
the
Rx2A
regulatory
sequence ............................................29
viii
Analysis
of
5’
end
promoter
deletions ...........................................29
Analysis
of
internal
deletions ......................................................................30
Analysis
of
Hsp
constructs ............................................................................32
2.3.4
The
SOX
and
OTX
sites
within
UCE
act
cooperatively
to
provide
spatial
specificity
to
Rx
expression........................................................................33
2.3.5
Rx2A
promoter
activity
in
the
mature
retina ....................................................34
2.3.6
Cognate
factors
that
bind
the
Rx2A
cis‐elements ............................................35
2.4 Discussion ....................................................................................................................37
2.4.1 Rx genes in X. leavis are similarly regulated................................................37
2.4.2 Expansion of the knowledge of Rx cis-regulatory elements ........................37
2.4.3 The central region of the Rx2A promoter contains a repressor element ......38
2.4.4 The UCE and FBE cooperate to activate Rx2A promoter activity...............39
2.4.5 Putative trans-acting factors of the Rx2A promoter .....................................41
POU factors..........................................................................................41
Otx family member ..............................................................................42
Sox family members ............................................................................45
Forkhead family members ...................................................................47
2.4.6 Coordinated regulatory activity of gene families implicated in Rx2A
promoter activity...........................................................................................47
2.4.7 Future directions ...........................................................................................50
Chapter 3: Seeing Eye to Eye: Forkhead Transcription Factors During Eye
Development .....................................................................................................................68
3.1 Introduction..................................................................................................................68
3.2 Methods........................................................................................................................69
3.3 Results and Discussion ................................................................................................69
3.3.1 Forkhead transcription factors in anterior eye structures.............................69
3.3.2 Forkhead transcription factors in the developing neural retina ...................71
3.3.3 Forkhead transcription factors in differentiated retinal cell types ...............73
3.3.4 Forkhead transcription factors in retinal progenitor cells ............................74
3.3.5 Retinal forkhead function in eye development ............................................77
ix
3.3.6 Retinal forkhead transcription factors hold enormous promise to advance
our knowledge of retinal progenitor cell biology ........................................82
Chapter 4: FoxO3 Perturbs Vertebrate Eye Development by Affecting
Differentiation and Proliferation in Retinal Progenitor Cells .....................................86
4.1 Introduction..................................................................................................................86
4.2 Materials and Methods.................................................................................................88
4.3 Results..........................................................................................................................92
4.3.1
X.
laevis
FoxO3
protein
is
highly
conserved
and
expressed
in
developing
eye
tissue ...........................................................92
4.3.2
Overexpression
of
xlFoxO3
results
in
a
small
eye
phenotype....................95
4.3.3
Differentiated
cell
types
are
produced
in
FoxO3‐injected
embryos........97
4.3.4
Retinal
Progenitor
Cells
are
specified
but
exhibit
altered
differentiation
in
FoxO3
injected
embryos........................................................................................97
4.3.5
FoxO3
overexpression
results
in
altered
cell
cycle
in
RPCs ........................98
4.3.6
FoxO3
overexpression
results
in
increased
apoptosis ..................................99
4.4 Discussion ..................................................................................................................100
4.4.1 A model for FoxO3 function in RPCs ........................................................100
4.4.2 Production
of
differentiated
cell
types
in
the
presence
of
exogenous
levels
of
FoxO3 .............................................................................................................100
4.4.3 Cell Cycle alterations in FoxO3-injected embryos...................................101
4.4.4 FoxO proteins as regulators of diverse progenitor cell populations .........103
4.4.5 Is RPCs regulation a conserved function for FoxO3 in vertebrates?........104
Chapter 5: FoxM1 is Necessary for Normal Development and Proper Lamination of
the Neural Retina by Retinal Progenitor Cells............................................................118
5.1 Introduction................................................................................................................118
5.2 Materials and Methods...............................................................................................123
5.3 Results........................................................................................................................125
5.3.1 Identification of a FoxM1 Isoform in X. laevis...........................................125
x
5.3.2 A xlFoxM1 homologue is expressed in RPCs during eye development ....126
5.3.3
FoxM1
knockdown
in
X.
laevis
results
in
embryos
with
small
eyes......127
5.3.4
xlFoxM1
is
necessary
of
proper
retinal
lamination .....................................129
5.3.5
Differentiated
cell
types
are
associated
with
areas
of
normal
retina
lamination
and
regions
of
aberrant
stratification........................................130
5.4 Discussion and Future Directions ..............................................................................131
Chapter 6: The significance of exploring Rx regulation and Forkhead transcription
factor function in RPCs .................................................................................................146
6.1 Summary of significant findings................................................................................146
6.2. Exploring the relationship between Rx2A, FoxO3, FoxM1 in RPCs.......................147
6.3 Regulation of RPC development by Homeobox and
Forkhead transcription factors .................................................................................148
6.4 Closing remarks .........................................................................................................149
Appendix A......................................................................................................................150
Bibliography ....................................................................................................................154
xi
LIST OF TABLES
Figure
Page Number
2.1 Expression domains of Rx2A promoter deletion construct ...................................... 53
2.2 PCR primers used to isolate the Rx2A regulatory region......................................... 64
2.3 Generation of Rx2A deletion constructs................................................................... 65
2.4 Primers used in the generation of Rx2A deletion constructs.................................... 66
2.5 Primers used to test cognate factor binding to cis-elements in the Rx2A promoter. 67
3.1 Production of anti-sense probes for gene expression analysis in X. laevis............... 69
3.2 Forkhead gene products expressed in eye structures ................................................ 83
3.3 Retinal forkhead expression...................................................................................... 84
4.1 Production of anti-sense probes for gene expression analysis
in FoxO3-injected embryos....................................................................................... 90
5.1 Production of anti-sense probes for gene expression analysis
in FoxM1 MO-injected embryos ............................................................................ 125
xii
LIST OF FIGURES
Figure
Page Number
1.1 Eye development in vertebrates ..................................................................................14
1.2 Retinal lamination and cell types in Xenopus Laevis...................................................15
1.3 Birthdating in the retina and the competence model of retinal development ..............16
1.4 Molecular regulation of RPC development .................................................................17
1.5 The eukaryotic cell cycle and RPCs ............................................................................18
2.1 Conservation and in silico analysis of the Rx2A regulatory region ............................52
2.2 Rx2A2.8 directs gene expression in the eye fields, hypothalamus and pineal gland ....53
2.3 Deletion constructs produced to investigate transcriptional activity
of the Rx2A regulatory region ............................................................................... 54-55
2.4 The Rx2A promoter contains cis-elements required for proper spatial and temporal
gene expression of the GFP transgene .........................................................................57
2.5 Internal deletions of the Rx2A promoter reveal additional regulatory region that
ensure specificity to the developing eyes.....................................................................58
2.6 Cooperative activity of UCE and the proximal FBE directs gene expression 56
throughout developmental stages.................................................................................59
2.7 POU, Sox and Otx sites within UCE cooperatively regulate Rx expression
in the developing eyes of Xenopus leavis embryos ......................................................60
2.8 Rx2A promoter activity in the mature Xenopus laevis retina ......................................61
2.9 Oct-6, FoxO3 and FoxN4 bind the Rx2A promoter cis-elements in vitro...................62
2.9 Model of Rx2A transcriptional regulation derived from deletion analysis .................63
3.1 Expression of retinal forkhead genes in the CMZ of the maturing neural retina in
Xenopus laevis ..............................................................................................................85
4.1 Xenopus laevis FoxO3 protein is highly conserved and is expressed throughout eye
development.....................................................................................................................105
4.2 xlFoxO3 RNA is targeted to the developing eye fields of X. laevis embryos ...........106
xiii
4.3 Overexpression of xlFoxO3 results in small eye phenotype .....................................107
4.4 T30A mutation in xlFoxO3 increased the frequency of eye phenotypes...................108
4.5 T30A mutation in xlFoxO3 increased the severity of the small eye phenotype........109
4.6 Retinal lamination is intact in the presence of exogenous xlFoxO3..........................110
4.7 Normal differentiated cell types are present in xlFoxO3-injected embryos ..............111
4.8 RPCs are specified in the presence of xlFoxO3 overexpression ...............................112
4.9 Differentiation is altered by xlFoxO3 overexpression...............................................113
4.10 Overexpression of xlFoxO3 results in changes in cell cycle gene expression ........114
4.11 Overexpression of xlFoxO3 results in decreased rate of cell cycle in RPCs...........115
4.12 xlFoxO3 overexpression results in increased levels of apoptosis............................116
4.13 Proposed model of FoxO3 function in RPCs...........................................................117
5.1 Xl184i11 encodes a FoxM1 protein...........................................................................136
5.2 xlFoxM1 is homologous to hFoxM1 isoform C ........................................................137
5.3 FoxM1 is expressed in the developing eye fields and in proliferative zones of the
maturing Xenopus laevis retina..................................................................................138
5.4 FoxM1 morpholino is properly targeted to developing eye fields in Xenopus
embryos......................................................................................................................139
5.5 FoxM1 morpholino specifically knocks down FoxM1 target............................ 140-141
5.6 FoxM1 knockdown causes small eye phenotype in X. laevis....................................142
5.7 FoxM1 morphants with eye phenotype......................................................................143
5.8 FoxM1 knockdown results in small eye with lamination defects..............................144
5.9 Differentiated cell types and abnormal retinal patterning are present in FoxM1
morphant retinas.........................................................................................................145
A.1
Cyclin
D1
is
robustly
expressed
in
the
developing
X.
laevis
retina .........................151
A.2
N‐myc
is
dynamically
expressed
in
the
developing
X.
laevis
eyes
and
retina ..............................................................................................................152
A.3
p27
is
dynamically
expressed
in
the
maturing
X.
laevis
retina .................................153
xiv
LIST OF ABBREVIATIONS
RPC
retinal progenitor cells
MO
morpholino oligonucleotide
pg
picogram
mM
micromolar
µg
microgram
RPE
retinal pigmented epithelium
CMZ
ciliary marginal zone
GCL
ganglion cell layer
IPL
inner plexiform layer
INL
inner nuclear layer
OPL
outer plexiform layer
ONL
outer nuclear layer
EFTFs eye field transctipsion factors
bHLH
basic helix loop helix
HD
homeodomain
CDK
cyclin dependant kinase
CDKI
cyclin dependant kinase inhibitors
FKD
forkhead
RGC
retinal ganglion cells
EST
expressed sequence tag
xv
CHAPTER 1 : Development and Regulation of Retinal Progenitor Cells by Known
Intrinsic Mechanisms
1.1 Introduction
Charles Darwin, when writing his book, “The Origin of Species”, included his
description of the eye in the chapter entitled: “Organs of extreme perfection and
complication.” To this day, scientists are continuing to pursue an understanding of this
fascinatingly complex and intricate organ. The study of the eye, and more specifically
retinal development, has been enhanced by the age of molecular biology and advanced by
techniques of genetic manipulation. Our understanding is still remarkably limited,
however, as new molecules involved in eye development are still being uncovered, and
the mechanisms that underlie their function continue to be elucidated.
Progenitor cells are by definition multipotent, and the neurons and glia of the
retina are derived from a cell population termed retinal progenitor cells (RPCs). RPC
development and function is known to be molecularly regulated, and this regulation
occurs on many levels. Initially, expression of RPC specific genes must be regulated so
that they are correctly expressed as cell fate determinants. As well, translational and post-
1
translational mechanisms exist in order to dictate that proper levels of functional protein
exist within developing RPCs. Mis-regulation of gene expression could result in
inappropriate loss or gain of expression of these genes, while altering protein structure or
the post-translational modifications could alter the levels of functional protein, resulting
in aberrant development or disease. It is therefore of high interest to both (1) study the
molecular mechanism of gene transcription of known intrinsic factors and (2) identify
and investigate the function of proteins novel to the development and maintenance of
RPCs.
1.2 Eye development in vertebrates and the specification of Retinal Progenitor Cells
(RPCs)
The presumptive eyes are specified during neurula stages, and are comprised of
tissue that evaginates from the anterior neural tube (Figure 1.1). This tissue, called the
optic vesicle, grows into the surrounding mesoderm towards the surface ectoderm. When
the optic vesicle makes contact with the surface ectoderm, reciprocal inductive signals
between the optic vesicle and the surface ectoderm, as well as the optic vesicle and
mesoderm, occur. As a result, the surface ectoderm thickens and develops into the lens
placode, while the neuroepithelium of the optic vesicle becomes patterned into three
distinct territories: the dorsal region is the presumptive retinal pigmented epithelium
(RPE), the ventral region is the presumptive optic stalk, and the lateral region is
determined as the presumptive neural retina. The optic vesicle and the lens placode
undergo simultaneous invagination and give rise to the optic cup and the lens vesicle,
respectively. Over time, the lens vesicle completely segments from the surface ectoderm,
forming the lens proper. The RPE completely surrounds and becomes abutted to the
2
thickening neural retina. The surface ectoderm adjacent to the lens gives rise to the sclera
laterally and the corneal epithelium medially.
The cells comprising the presumptive neural retina are specified as retinal
progenitor cells (RPCs). This population of cells rapidly proliferates and subsequently
differentiates to form the neural and non-neuronal cell types of the retina. RPCs are
therefore multipotent and constitute a population of cells that can further our
understanding of the nature of multipotency, and the relationship between cell division
and differentiation.
In the mature vertebrate retina, the cell types produced by the RPCs are highly
organized in a laminar tissue (Figure 1.2). The cell bodies of the retinal neurons and glia
are positioned in three distinct nuclear layers: the ganglion cell layer (GCL) adjacent to
the lens, the inner nuclear layer (INL) and the outer nuclear layer (ONL) nearest the
apposing RPE (Figure 1.2, A). The neuronal processes extended by the cells within these
layers make connections within two plexiform layers: the inner plexiform layer (IPL) and
outer plexiform layer (OPL). The inner plexiform layer is positioned between the GCL
and INL, and the outer plexiform layer between the INL and ONL. In addition, a
specialized region of progenitor cells persists in adults of some vertebrate species,
including X. laevis (Figure 1.2, A, brackets). Termed the ciliary marginal zone (CMZ),
the RPCs of this zone continue to give rise to differentiated retinal cells during the life of
the adult animal. The CMZ is located peripherally, adjacent to the lens on both its dorsal
and ventral side.
Six
neuronal
and
1
glial
cell
type
are
observed
in
the
mature
vertebrate
retina
(Figure
1.2,
B).
The
major
phototransducing
cells
of
the
retina,
the
rod
and
3
cone
photoreceptors,
possess
their
cell
bodies
in
the
ONL.
The
specialized
compartments
for
phototransduction,
the
photoreceptor
inner
and
outer
segments,
are
juxtaposed
to
the
overlying
RPE.
The
photoreceptor
cells
of
the
ONL
extend
processes
laterally
into
the
OPL
synapsing
with
the
three
interneuron
cells
types:
the
amacrine,
bipolar
and
horizontal
cells.
Each
of
these
cell
types
possesses
cell
bodies
that
reside
in
the
INL,
and
extend
outgoing
processes
into
the
IPL.
The
IPL
is
the
layer
that
houses
connections
between
the
interneuron
and
the
ganglion
cell
dendrites.
The
long
axons
of
the
ganglion
cells
extend
into
the
optic
nerve
and
eventually
make
connections
in
the
visual
processing
center
of
the
brain,
the
optic
tectum.
The
cell
bodies
of
the
major
glial
cell
type
in
vertebrates,
Muller
glia,
are
positioned
in
the
INL.
They
extend
processes
to
the
inner
segments
of
photoreceptor
cells,
creating
the
outer
limiting
membrane,
and
likewise,
end
feet
of
Muller
glia
create
the
inner
limiting
membrane
just
past
the
ganglion
cell
layer.
Classic birthdating studies have demonstrated that retinohistogensis by RPCs
occurs in a generally conserved order across species, always first with the generation of
ganglion cells, with rod photoreceptors, bipolar and Muller glia the last cell types being
produced (Carter-Dawson and LaVail, 1979; Cepko et al., 1996; Stiemke and Hollyfield,
1995; Wong and Rapaport, 2009; Young, 1985) (Figure 1.3, A). It was initially
speculated that the generation of these cell types in such a stereotypical order was
dictated by environmental cues (Turner and Cepko, 1987) (Holt et al., 1988). However,
the demonstration that stage specific RPCs predictably produce certain cell types
regardless of environmental influences lead to the competence model of retinal
4
development (Figure 1.3, B) (Cepko et al., 1996)
The competence model proposes that after specification, RPCs progress through
distinct stages. In each stage, RPCs are capable of producing specific subsets of retinal
cell types (Cepko et al., 1996; Livesey and Cepko, 2001). This model assumes that (1)
competence states of RPCs are intrinsically defined and (2) the capacity to produce cell
types are intrinsically regulated by the production of competence states. RPCs of a given
competence state are then governed by both intrinsic and extrinsic factors to generate
specific retinal cell types (Cepko et al., 1996).
1.3 From RPCs to retinal cell types: molecular regulation of retinal cell types by
intrinsic factors
But how are the production of these cell types regulated on the molecular level?
What governs the specification, differentiation and eventual functionality of developing
retinal progenitor cells?
According to the competence model, both extrinsic and intrinsic signals govern
the development of retinal progenitor cells into differentiated cell types through certain
competence states. Certainly, extrinsic factors such as extracellular signaling molecules,
mitogens, and neurotransmitters have been shown to be involved in both the specification
and differentiation of the neural retina (reviewed in Martins et al., 2007). However, for
the purpose of introduction for these studies, it is important to emphasize the intrinsic
factors that have been shown to regulate eye development, both in specification of RPCs
and during the differentiation of RPCs into terminally differentiated neurons.
The group of proteins involved in proper specification of RPCs in vertebrates are
termed eye field transcription factors (EFTFs) (Zuber et al., 2003). The EFTF gene
5
products include: retinal homeobox (Rx), paired type homeobox 6 (Pax6), lim-domain
containing homeobox 2 (Lhx2), sine oculis homologs 3 and 6 (Six3/6), orthodenticle
related homeobox 2 (Otx2), the t-box transcription factor, ET, and the orphan nuclear
receptor, (Tlx).The majority of these proteins belong to the homeodomain family of
transcription factors (the exceptions being Tlx and ET). Their importance during eye
development is underscored by the high level of conservation of these factors from flies
to humans (Zuber et al., 2003); in each vertebrate system investigated, these transcription
factors are highly expressed in RPCs during retinal specification and neurogenesis
(Figure 1.4, A). Gene expression and genetic manipulation experiments investigating
these factors have determined that each of the EFTFs is necessary for eye development
(Bailey et al., 2004; Bovolenta et al., 1997; Chow and Lang, 2001; Esteve and Bovolenta,
2006; Nishida et al., 2003; Porter et al., 1997; Tetreault et al., 2009; Viczian et al., 2006;
Wilson and Houart, 2004; Zuber et al., 2003).
Many EFTFs are required, and often capable, for the initiation of gene expression
of other genes within the group. For example, Rx null mice, do not express Pax6,
suggesting that Pax6 is downstream of Rx (Mathers et al., 1997). The overexpression of
Rx in contrast induces Pax6 and Six3, reduces Otx2 (Andreazzoli et al., 1999). As well,
Pax6 overexpression results in ectopic eyes with induction of Rx, Otx2, an Six3 (Chow et
al., 1999). And Six3 itself is also capable of induction of Rx (Loosli et al., 1999)
and Pax6 (Loosli et al., 1999) (Bailey et al., 2004; Bovolenta et al., 1997; Chow and
Lang, 2001; Esteve and Bovolenta, 2006; Nishida et al., 2003; Porter et al., 1997;
Tetreault et al., 2009; Viczian et al., 2006; Wilson and Houart, 2004; Zuber et al., 2003).
Tlx overexpression induces the expression of Pax6, Six3 and Lhx2, and conversely, Pax6
6
and Six3 overexpression upregulates Tlx (Bailey et al., 2004; Bovolenta et al., 1997;
Chow and Lang, 2001; Esteve and Bovolenta, 2006; Nishida et al., 2003; Porter et al.,
1997; Tetreault et al., 2009; Viczian et al., 2006; Wilson and Houart, 2004; Zuber et al.,
2003). It is apparent from these studies that the EFTFs form a self-regulating network of
factors necessary for the specification of the RPCs in the anterior neural plate.
After specification, a combinatorial code of transcription factors dictates the cell
types that differentiate from RPCs of a given competence state (Wang and Harris, 2005;
Harada et al., 2007)). Two major families of transcription factors are involved: basic
helix-loop-helix (bHLH) (Vetter and Brown, 2001), and homeodomain (HD) proteins
(Lupo et al., 2000). In general, at least one bHLH and HD are co-expressed in progenitors
as the produce a certain cell type (Figure 1.4)(Harada et al., 2007). For example, Ath5
and Pax6 specify the production of ganglion cells (Brown et al., 2001; Marquardt et al.,
2001; Wang et al., 2001), or Ath3 and Pax6/Six6 specify the production of horizontal
cells (Marquardt et al., 2001)(Tomita et al., 2000).
It is well established that perturbing
these combinations can alter the distribution of cell types within the retina. Collectively,
the data concerning the coordination of these protein families suggests that homeodomain
factors regulate layer specificity while bHLH proteins dictate cell fate within the
homeodomain factor-specified layers (Harada et al., 2007).
A significant amount of work has been done to understand the molecular
regulation of eye development. The work presented in this dissertation expands the
knowledge of known RPC regulators by (1) furthering the understanding of the molecular
regulation of Rx genes, (2) definitively recognizing a third family of transcription factors,
the forkhead family, as regulators of RPC development and function, and (3)
7
characterizing a role for two forkhead proteins, FoxO3 and FoxM1, neither of which have
been previously attributed a role during vertebrate eye development.
1.4 RPC Decision-making: The balance between proliferation and differentiation
The RPCs of the neuroepithelium in the optic cup undergo several rounds of rapid
proliferation to produce the cellular diversity of the laminated neural retina. It is during
these rounds of cell division that the retinal cell types are specified and differentiate.
Therefore, it is not surprising that the differentiation of RPCs is tightly linked to
proliferation. Several lines of evidence support this notion: (1) loss of function
experiments concerning the EFTFs demonstrate reduced proliferation rates, (2) EFTFs
can act as direct or indirect modulators of cell cycle machinery and (3) aberrations in cell
cycle genes result in changes in cell type specification in the developing eye.
In mice, targeted mutagenesis of Otx1, Otx2, Lhx2, and Tlx results in reduced
proliferation rates in the developing eye region (Martinez-Morales et al., 2001; Miyawaki
et al., 2004; Porter et al., 1997). The use of morpholino knockdown in Xenopus and
zebrafish has demonstrated reduced proliferation in the absence of Lhx2 and Six3, and
Rx, respectively (Ando et al., 2005) (Nelson et al., 2009). These studies clearly
demonstrate that the EFTFs are responsible for the necessary proliferation of RPCs after
specification in the neuroepithelium.
Progenitor cells, including RPCs, make the decision to terminally differentiate or
continue to proliferate during the G1 phase of the cell cycle (Figure 1.5, A). If a neuron
decides to differentiate and enter G0, it is thought to be restricted from re-entering the
cell cycle. An exception is represented by the Muller glial cells of the retina, which
proceed to G0, but retain the capacity for entering the cell cycle. Several groups of
8
molecules are active in progenitor cells during early G0 acting as intrinsic regulators of
differentiation decisions.
First, specific cyclin/cyclin dependent kinase (CDK) enzyme complexes are
active during early G1 to dictate the decision between G0 or continued proliferation:
Cyclin D with CDK4 or CDK6, and Cyclin E with CDK 1 or 2 (Welcker and Clurman,
2005). Secondly, Cyclin-dependent kinase inhibitors exert an additional layer of
regulation; these molecules directly alter the activity of cyclin/CDK enzymes complexes.
There are two major families of cyclin-dependant kinase inhibitors: the INK family (p15,
p16, p18, p19) and the Cip/Kip family (p21, p27, p57). In Xenopus, only Cip/Kip family
member has been identified, which shares sequence and functional characteristics of all
three Cip/Kip proteins (Su et al., 1995). Cyclins, CDKs and CDK inhibitors are expressed
in dynamic tissue-specific manners. The balance of activity among them determines the
phosphorylation state of members of the Rb family, that ultimately dictate an RPCs
choice between cell division and cell cycle exit. The expanding amount of literature
supporting the link between EFTFs and the cell cycle machinery is presented.
Studies of EFTF function have shown changes in cell cycle gene expression as
well as confirming cell cycle genes as direct targets of EFTFs. In Xenopus, Rx controls
cell proliferation by inhibition of the cell cycle inhibitor p27Xic1 (Andreazzoli et al.,
2003). Moreover, embryos injected with Xrx1 mRNA demonstrate significantly
increased levels of Cyclin D1, the major D-type cyclin expressed in RPCs (Casarosa et
al., 2003). Pax6 null mutant mice have decreased expression of a number of cell cycle
inhibitors, p27(kip1), p57(kip2), and p21(cip1) (Duparc et al., 2007). Six6, binds to the
promoter of the p27Kip1 in retinal cells, suggesting that Six6 may be a direct repressor of
9
p27Kip1 transcription in RPCs. p27Kip1 mRNA and protein is also upregulated in the
Six6-null retina (Li et al., 2002). The related EFTF protein, Six3, has been shown to be
upstream of both cyclin D1 and p27 (Gestri et al., 2005). Collectively, these data suggest
that several of the EFTFs have the ability to alter the cell cycle during RPC development.
It has been shown that altering cell cycle components within the retina directly
results in changes in neuronal diversity. Cyclin D1 null mice have hypocellular retinas,
attributed to decreased proliferation (Sicinski et al., 1995), and it has recently been shown
that early born cell types (ganglion and photoreceptor cells) are increased (Das et al.,
2009). Interestingly, replacing cyclin D1 with Cyclin E1 is sufficient to restore normal
retina structure and cell type specification, suggesting that there is some functional
redundancy between active cyclins in the retina (Das et al., 2009). On data regarding
CKIs, overexpression of the cell cycle inhibitor p27(Xic1) results in the increased
numbers of ganglion cells (Ohnuma et al., 1999), while overexpression of p27 is
sufficient to maintain RPCs in the cell cycle and produce later born cell types (Ohnuma et
al., 1999). Additionally, targeted mutagenesis of p57 in mice alters cell fate specification
by increasing the proportion of amacrine cell sub-populations (Dyer and Cepko, 2000;
Dyer and Cepko, 2001). Together, these studies demonstrate the relationship between
proliferation and differentiation in the RPC population.
As neuroepithelial cells of the developing retina proceed through the cell cycle,
the nuclei migrate between the apical and basal surfaces in phase with the cell cycle, a
process termed interkinetic nuclear migration (Figure 1.5, B) (Baye, 2007). M-phase
nuclei are located at the apical surface, adjacent to the RPE. As the cells proceed to G1phase, the nuclei migrate toward the basal surface, and progress into S-phase at their most
10
basal position. During G2, nuclei advance back towards the apical surface to re-enter M
phase. Heterogeneity of interkinetic nuclei migration, i.e. difference in migration distance
from apical surface, results in a difference in time spent in G1 or G2 between two
daughter cells. Importantly, the difference in distance traveled results in exposure to
different localized extrinsic cues that influences the decision to continue proliferation or
become neurogenic. Neurogenic cells undergo asymmetric mitotic cell division at the
apical surface, producing one post-mitotic daughter cell and one cell with continued
proliferative capacity. The post-mitotic cell nuclei progresses to its final position in the
retinal architecture by a process called nuclear translocation, actively remodeling basal
and apical processes until they reach final targets.
It continues to be a major endeavor in the field of retinal development to
understand how cell fate determinants coordinate with the molecular mechanisms driving
cell cycle. It is a major emphasis of this dissertation to further the knowledge regarding
RPC regulators that have the capacity to affect cell cycle machinery. By studying the
mechanism of known cell cycle regulators FoxO3 and FoxM1 in RPC development, this
work strengthens the connection between proliferation and differentiation in the
vertebrate retina.
1.5 Xenopus Laevis as a model system to study RPC development
Xenopus laevis, the South African clawed frog, has been a model system used in
the study of embryology and molecular biology for over 50 years. Pioneered as a model
system for the study of developmental biology by the German biologist, Firschberg, the
model system is now a prominent organism used to answer fundamental questions
regarding development (Gurdon and Hopwood, 2000).
11
The rapid, external embryonic development of X. laevis, and its capacity to
produce large numbers of synchronously developing embryos makes it a suitable system
for developmental biology applications (Gurdon and Hopwood, 2000). In addition, the
stages of development post-fertilization are well documented and easily recognizable
(Nieuwkoop and Faber, 1994).
A key advantage for the use of X. laevis is the ability to manipulate the large
embryos using advanced genetic techniques. RNA overexpression studies were first
described using Xenopus oocytes (Gurdon et al., 1971). With the addition of molecular
cloning techniques, the ability to overexpress RNA of desired genes is commonplace and
easily achieved. As well, the production of transgenic embryos in X. laevis was
described over ten years ago (Kroll and Amaya, 1996). This technique has several
advantages over other systems including the rapid production of first generation
transgenics for analysis. This system is well positioned to study genomic elements in vivo
using transgene reporter constructs specifically in the retina (Hutcheson and Vetter,
2002). Xenopus laevis provides an excellent model system for the studies presented in
this dissertation.
1.6 The impact of this work relative to the field of retinal development
The work presented in this dissertation is focused on three genes expressed in
developing RPCs. Rx, one of the first genes to be expressed in the presumptive eyes, and
two forkhead proteins, FoxO3 and FoxM1, which have not been previously ascribed a
role during vertebrate eye development. The work presented regarding Rx (Chapter 2)
advances the knowledge of the mechanism of Rx transcription, thereby furthering the
understanding of RPC development and maintenance as multipotent cells. As an
12
extension of primary data elucidated in Chapter 2, data regarding the role of forkhead
proteins in eye development is presented in Chapter 3. This includes both a literature
review of forkhead proteins with known function in the eye and the expression pattern of
6 forkhead proteins in Xenopus laevis retinas that have not been characterized for their
role in eye development. Two of these proteins, FoxO3 and FoxM1, both have extensive
bodies of literature to support their involvement in cell cycle control and differentiation.
FoxO3 has significant literature supporting its role in cell cycle arrest and inhibition of
differentiation (reviewed in Ho et al., 2008), while FoxM1 has been demonstrated to
promote cell cycle progression (reviewed in Weirstra and Alves, 2007). The apparent
opposing functions of FoxO3 and FoxM1 with regards to cell cycle progression and
differentiation partially motivated the studies presented in Chapter 4, and Chapter 5
respectively. Collectively, the studies presented in this work furthers the understanding
of the molecular mechanism of gene transcription in the developing eye, provides
evidence from the literature that distinguishes forkhead proteins as RPC regulators,
identifies two forkhead transcription factors as novel regulators of RPC development, and
advances the knowledge of cell cycle regulators during differentiation of RPCs.
13
Figure 1.1 Eye development in Vertbrates
(A) The eye develops as an extension of the anterior neural plate called the optic vesilcle
(OV). The OV evaginates away from the midline through the mesoderm (MS) towards
the surface ectoderm (SE). (B) The OV comes into contact with the SE, forming the lens
placode (LP). The neuroepithelium can be divided into three distinct region: the dorsal
region is the presumptive retinal pigmented epithelium (RPE), the ventral region is the
presumptive optic stalk (OS), and the lateral region is the presumptive neural retina (NR).
(C) Inductive signalling events stimulate both the presumptive NR and the overlying LP
to invaginate, forming the optic cup and lens vesicle (LV). (D) The LV eventually
seperates from the SE. The NR thickens, and the RPE completely surrounds the NR. E.
In the mature eye, sclera (S) and cornea (C) are derived from the SE the dorsal and
ventral region of the OS become the optic nerve (ON), and the RPE completely surrounds
the NR. (Adapted from Adler and Canto-Soler, 2008).
14
Figure 1.2 Retinal Lamination and Cell Types in Xenopus laevis
(A) Mature retina is Xenopus Laevis, hematoxylin and eosin stained. Laminar structure is
composed of three nuclear layers: ganglion cell layer (GCL), inner nuclear layer (INL),
and outer nuclear layer (ONL, arrowhead); and two inner layers: inner plexiform layer
(IPL) and outer plexiform layer (OPL). Apical to the ONL the outer segments (OS) are
positioned, adjacent to the retinal pigmented epithelium (RPE). Brackets indicate ciliary
marginal zone that continues to give rise to retinal cell types throughout the life of the
animal. Functionally, light enters through the lens, traverses all the retinal layers where it
is phototransduced by the photoreceptors. The photoreceptors relay the information via
electrical impulses through the neurons of the internuclear layers, back through the
ganglion cells layers and out through the optic nerve to the brain. on stage 41 paraffinembedded section. (B) The cell types of the nueral retina are depicted in the layers in
which they reside: ganglion cell (GC), amacrine cells (AC), bipolar cels (BP), horizontal
cells (HC), rod and come photoreceptor (RP) and (CP), and muller glial cell (MC).
15
Figure 1.3 Birthdating in the Retina and the Competence model of RPC
development (A) RPCs of vertbrates produce cell types in a specific order in X. laevis.
The timing of the generation of cell types in the retina overlaps, and the order of
production is generally conserved in vertebrates. (B) Progenitors pass through distinct
states (colored areas under curve) over time. See text for details. (modified from Harada
et al., 2007 (A), and Livesay and Cepko, 2001 (B))
16
Figure 1.4 Molecular regulation RPC devlopment (A) The EFTFs specific the eye
fields by a their dynamic expression in the anterior neural plate. A schematic of neurala
stage (stage 15) embryo in X. laevis. (B) Generation of cell type in the retina by
homeobox and basic helix-loop-helix proteins. In general, the production of each cell
type in the vertebrate retina is dicated by the expression of a homeobox protein(s) and a
basic helix-loop-helix (bHLH) protein(s). (adapted from Harada et al., 2007)
17
Figure 1.5 The eukaryotic cell cycle and RPCs (A) RPCs make the decision to
continue to proliferate or terminally differentiate. An exception is the commitment to
glial cell fate, which maintain the ability to proliferate. Molecular regulation of this
decision is achieved by differential expression of cyclin/CDK complexes and CKIs that
regulate the phosphorylation of Rb proteins at the G1/S checkpoint (red x). (B) Nuclei of
RPCs in the developing retina migrate from the basal to apical surface as they progress
through the cell cycle. (adapted from Livesay and Cepko, 2001 and Bale and Link,
2007).
18
CHAPTER 2: Regulation of Retinal Homeobox Transcription by Cooperative
Activity Among Cis-elements*
2.1 INTRODUCTION
The vertebrate eye field transcription factors (EFTFs) Rx1, Pax-6, Six-3, and Lhx2
form a self-regulating feedback network that specifies the vertebrate eye field (Zuber et
al., 2003). In each vertebrate system investigated, these transcription factors remain
expressed in RPCs during retinal neurogenesis. In this chapter, we focus our attention on
the retinal homeobox gene (Rx), a transcription factor that is critical for the proper
development of vertebrate eyes (Andreazzoli et al., 1999; Chuang et al., 1999; Mathers et
al., 1997). First discovered in X. laevis, Rx homologues have been identified in flies, fish,
mouse, chicken and human (Casarosa et al., 1997; Chuang et al., 1999; Deschet et al.,
1999; Eggert et al., 1998; Furukawa et al., 1997; Mathers et al., 1997; Ohuchi et al.,
1999; Strickler et al., 2002).
The number of Rx genes varies by organism. For example, zebrafish have three (Rx1,
Rx2, Rx3), X. laevis have two (Rx1A, Rx2A), while mice and humans (termed RAX in
these organisms) have only one (Casarosa et al., 1997; Chuang et al., 1999; Deschet et
al., 1999; Furukawa et al., 1997; Mathers et al., 1997). Normal expression of the Rx gene
*Equal contributions to the data and writing of this chapter were contributed by
H. Moose and R.I.Martinez-De-Luna.
19
product begins in the anterior neural plate at a time before eyes are morphologically
distinct (Casarosa et al., 1997; Chuang et al., 1999; Furukawa et al., 1997; Mathers et al.,
1997). Rx continues to be expressed in the mature neural retina in a subset of
photoreceptors, although it is generally downregulated in differentiated cells (Mathers et
al., 1997; Perron et al., 1998; Furukawa et al., 1997; Nelson et al., 2009). Notably, the
expression of Rx is maintained in the region of the neural retina that contains RPCs.
A number of studies have demonstrated the importance of Rx in proper eye
development. Complete loss of Rx function results in the lack of eye formation in both
medaka and mice (Loosli et al., 2001; Mathers et al., 1997). Targeted mutagenesis of the
mouse Rax gene results in animals with no visible eye structures, demonstrating that Rx is
important in the initial events in eye specification (Mathers et al., 1997). The
temperature sensitive medaka mutant eyeless is anophthalamic (Winkler et al., 2000), and
the chokh mutant in zebrafish, having a mutation in the zebrafish Rx3 gene, also fails to
develop neural retina (Loosli et al., 2003).
Conversely, overexpression of Rx produces extra retinal tissue (Andreazzoli et al.,
1999; Chuang and Raymond, 2001; Mathers et al., 1997). Ectopic tissue forms with the
same complex laminar structure as normal retinal tissue (Mathers et al., 1997). This
suggests that Rx plays a role in the early decisions of anterior neural tissue to become
retinal tissue, which consists of RPCs competent to produce the neural retina. Zebrafish
Rx genes, Rx1 and Rx2, induce the formation of ectopic retinal tissue at the expense of
forebrain cells, suggesting a role for Rx in the early specification of optic primordia
(Chuang and Raymond, 2001). The overexpression of Rx also induces the upregulation of
other eye field genes, including Pax6 and Six3, while decreasing expression of Otx2
20
(Andreazzoli et al., 1999). This reiterates that Rx acts early in eye specification by
affecting downstream genes in RPCs that have themselves been shown to be necessary
for eye development. Consistent with this data, Rx was shown to be essential for the
formation of RPCs in mice, in contrast to Pax6 (Zhang et al., 2000). Together these
expression studies clearly demonstrate the importance of Rx in early eye specification and
development. Even with these studies demonstrating the critical importance of Rx
function in eye field specification, the molecular and genetic events leading to Rx
transcription initiation, throughout stages of RPC development, have not been clearly
established.
Two Rx genes have been identified in X. laevis (Mathers et al., 1997). The presence
of the two genes, Rx1A and Rx2A, is most likely a result of the partial genome duplication
that occurred within the species (Bailey et al., 2004). The expression patterns of these
two genes are indistinguishable and the coding region sequences exhibit considerable
identity. For the purpose of this document, Rx expression refers to the combined
expression patterns of Rx1A and Rx2A. Previous work from our lab has shown that the
Rx1A regulatory sequence is organized into two distinct regions (Zhang et al., 2003). The
distal portion of a 3.4KB DNA sequence is capable of transcriptional activation in the
anterior neural plate, while the proximal region of the promoter activates transcription in
later stages of eye development (Zhang et al., 2003). The proximal region is not only
necessary for the late activity of the promoter, but is also sufficient (Zhang et al., 2003).
The Rx1A regulatory sequence also exhibits some redundancy in that the distal region is
sufficient for directing transcription in the photoreceptor cells late in development (Zhang
et al., 2003). The findings in this study demonstrate the overall organization of the Rx1A
21
regulatory sequence. It failed to elucidate specific cis-acting regions that are necessary
for Rx expression, therefore still leaving a gap in our knowledge of mechanisms
determining appropriate Rx transcription. Chapter 2 addresses the investigation of
sequence specific Rx cis-acting regions using the Rx2A regulatory sequence.
2.2 MATERIALS AND METHODS
Animals
Female and male X. laevis were purchased from NASCO. Egg laying was
induced by sub-dermal injection of 150cc of human chorionic gonadotropin (HCG)
(Chorulon; Intervet). Embryos were dejellied in 2.25% cysteine pH 7.9. The
developmental stage of embryos was determined according to Nieuwkoop and Faber
(Nieuwkoop and Faber, 1994).
Isolation of the Rx2A regulatory sequence
We isolated a 2.8KB genomic fragment of the Rx2A regulatory sequence by three
rounds of linker-mediated PCR (LM-PCR) according to manufacturer’s protocol
(Universal Genome Walker Kit, Clontech) (Moritz et al., 2002; Siebert et al., 1995).
Initial primers were designed to be specific the Rx2A coding region (AF001049). The
primers used for LM-PCR are listed in Table 1.
Generation of transgenesis constructs
The full length Rx2A-GFP transgene was generated by ligating the Rx2A
regulatory sequence into the BamHI and EcoRV sites of pBS-GFP (Zhang et al., 2003).
A detailed list of the transgenes used in this study is provided in Figure 3. Constructs 1316, 18, 19, 21-25, 27-32, 33, and 34-36 were generated by PCR using Platinum Pfx DNA
22
polymerase (Invitrogen). All primers sets and templates used for PCR for each of these
constructs are detailed in Tables 3 and 4. PCR conditions were as follows: 94°C for 2
min, 94°C for 15 sec, 55°C for 30 sec, 68°C for 1 min or 2 min depending on the length
of the fragment for 20 or 30 cycles. All PCR products were TA cloned into pCR2.1
(Invitrogen) and sequenced to confirm that point mutations were not introduced during
PCR. Subsequently the amplified fragments were subcloned into pBS-GFP or Hsp-GFP
(Zhang et al, 2003). Constructs 20, 27 and 29 were made using restriction enzyme sites
already present in the Rx2A promoter. The restriction enzymes used are listed in Table
3. The corresponding fragments were then subcloned into pBS-GFP or Hsp-GFP (Zhang
et al, 2003).
Quick Change XL Site-Directed Mutagenesis Kit (Stratagene) was used to
generate constructs 2-7. A detailed description of the primers and templates used for
mutagenesis is listed in Table 3 and 4. PCR conditions for mutagenesis were as follows:
95°C for 30 sec, 95°C for 30 sec, 55°C for 1 min, 68°C for 11 min, repeat 18 cycles.
Success of mutagenesis reactions was confirmed by sequencing or restriction enzyme
digestion using a restriction site introduced during PCR.
Transgenesis
Transgenic X. laevis embryos were generated by the intracytosolic sperm
injection (ICSI) method (Sparrow et al., 2000). Sperm nuclei were isolated as previously
described (Kroll and Amaya, 1996) using digitonin (10mg/ml) instead of lysolecithin.
DNA transgenes were released from their vector by restriction enzyme digestion (detailed
in Table 2), and purified from agarose gel using the Gene Clean kit (QBiogene). ICSI
was performed as previously described (Sparrow et al., 2000), using snap frozen sperm
23
nuclei. For the transgenesis reaction 400,000 sperm nuclei were incubated with 250 ng of
transgene DNA and 2 µl of sperm dilution buffer (SDB) for 15 min at room temperature.
The reaction was then diluted in 22.5 µl and 2.5 µl of this mixture was further diluted in
230 µl of SDB for injection. Cysteine-dejellied eggs were injected with 10nl of
transgenesis reaction in 0.4 X MMR (Marc’s Modified Ringer’s) + 6% Ficoll. Properly
dividing embryos were transferred to 0.1 X MMR + 6% Ficoll and changed to 0.1 X
MMR after 24 hrs. Embryos were raised in 0.1 X MMR until the appropriate stage.
Section in situ hybridization
In situ hybridization was performed on 8µm retinal sections as previously
described (Shimamura et al., 1994; Viczian et al., 2003). In situ hybridization analysis
was performed using an antisense riboprobe against GFP to detect expression of the
transgene. The GFP antisense riboprobe was labeled using digoxygenin (Ambion) as
previously described. To make the antisense riboprobe for GFP, pBS-GFP (Zhang et al.,
2003) was linearized with EcoRI and the probe transcribed with T7 RNA polymerase
(Ambion).
Immunohistochemistry
Stage 41 transgenic embryos were fixed in MEMPFA for 1 hour (Sive et al.,
2000), dehydrated in methanol for 1 hour, embedded in paraffin and sectioned at 8 µm as
previously described (Pan et al., 2006). Immunohistochemistry was performed as
described previously (El-Hodiri et al., 1997). Mouse anti-Myc (Sigma) was used as
primary antibody at a 1:1,000 dilution in blocking solution.
Electrophorectic mobility shift assays
24
In vitro translation of FoxO3, FoxN4 and Oct-3 was performed using the TnT
Quick Couple Transcription Translation kits (PROMEGA) according to manufacturer’s
instructions. Oligonucleotides were designed using specific Rx2A promoter sequences
containing the POU site of the FBE element. Complimentary oligonucleotides were
(45ug each) were added to annealing buffer containing 0.1M Tris-HCL, 0.5M NaCl, and
0.05M EDTA, incubated for 10 minutes as 65°C and let to cool to room temperature
slowly. Annealed oligos were diluted 1:10 from the previous reaction and radio-labeled
with [α-32p] dUTP (Amersham) by incubating at for 30 minutes at 37°C in buffer
containing the following: [α-32p] dUTP; 5µM dATP, dTTP and dGTP (Invitrogen); 2.5 U
Klenow (New England BioLabs); 1/20 total volume of React #2 buffer (GIBCO). After
labeling, unincorporated [α-32p] dUTP was removed using Qiaquick Nucleotide Removal
Kit (Qiagen). Binding reactions were carried out by incubating 20ng in vitro-translated
FoxO3, FoxN4 or Oct-6 with the following for 1 hour on ice: 1x104 cpm/µl labeled oligo,
1x EMSA buffer, 1 mg/ml BSA, 1mg dIdC. 5x EMSA buffer contains: 50mM HEPES
(pH 7.9), 375 KCL, 12.5mM MgCl2, 0.5mM EDTA, 5mM DTT, 15 Ficoll. All reactions
were run on a 5% pre-cast TBE gel (100V for 1hour in 0.5x TBE at in 4°C) (Bio-Rad) for
approximately 3.5 hours at 60V at 4°C. Gels were dried at 80°C for 50 minutes, and
imaged with phosphoimager. For competive binding: Specific probes were comprised of
unlabeled Rx2A UCE POU or Rx2A FBE probe. A radio-labeled mutated Rx2A UCE
POU site probe was used to test specificity of OCT-6 binding, and radio-labeled OCTA
probe was in testing FoxO3 or FoxN4 binding the FBE. See table 5 for detailed
sequences of all EMSA oligonucleotides.
25
Microscopy
A fluorescent dissecting microscope (Leica) was used to detect GFP expression in
transgenic embryos at the appropriate stages.
2.3 RESULTS
2.3.1 Identification and Conservation of the Rx2A Regulatory Sequence
Rx genes can be found in zebrafish to human genomes, and the function of their
gene products is essential for the formation of eyes in vertebrates (See Review: Bailey et
al., 2004). Thus, the regulatory mechanisms that govern the transcription of Rx genes are
important for the understanding of eye specification and determination. We have
previously characterized the regulatory sequence of the X. laevis Rx1A gene (Zhang et
al., 2003), one of two Rx genes present in this species. In order to more fully understand
the transcriptional regulation of Rx genes, we isolated a 2.8 KB genomic region upstream
of the Rx2A coding sequence by linker mediated PCR.
To reveal regions of identity and divergence between the 2.8 KB Rx2A region we
used MultiPipMaker and the previously published 3.2 KB regulatory region of Rx1A.
Alignment analysis using MultiPipmaker (http://pipmaker.bx.psu.edu/pipmaker) plots the
percent identity of DNA segments between two or more designated sequences. The
percent identity plot (PIP) between Rx2A and Rx1A reveals stretches of high
conservation separated by regions with more divergent sequence (Figure 1A). A distal
326 bp region (Rx2A: -2360 to -2035) shares 92% identity between the two sequences,
followed more proximally by a shorter 123 bp segment sharing 93% identity (Rx2A: 1897 to -1775) and a 272 bp region sharing 95% identity (Rx2A: -456 to 84) (Figure 1A).
26
The distal conserved region is highly conserved in vertebrata, and is referred to as the
Ultra Conserved Element (UCE) for the remainder of this document. The sequence
corresponding to UCE has recently been published to function as an enhancer of the
human Rx gene product, RAX (Danno et al., 2008). In the Rx2A regulatory sequence, the
UCE and two proximal regions comprise the three longest, most highly conserved
stretches of sequence shared between the two regulatory regions of DNA.
It is also apparent that regions of divergence exist between the two sequences: a
172 bp region (Rx2A: -2559 to 2387) is present in the Rx2A regulatory region that is not
present in the Rx1A sequence. Interestingly, this region is present in a region of X.
tropicalis upstream of the Rx coding region (data not shown). Lastly, PIP analysis
predicts an inversion event in the region corresponding to Rx2A -1252 to -1347
compared to Rx1A. This region also contains a predicted JH12 repetitive DNA element
(Rx2A:-1024 to -1329) (http://www.repeatmasker.org).
We continued our analysis of the Rx2A regulatory region using bioinformatics to
predict cis-regulatory elements within the 2.8Kb Rx2A sequence using the Transcription
Element Search Software (TESS) (http://www.cbil.upenn.edu/cgi-bin/tess/tess). Binding
motifs for several families of transcription factors are present in the 2.8Kb region
upstream of Rx2A including Forkhead motifs (Kaufmann et al., 1995), Otx binding
motifs (Gan et al., 1995), Sox binding motifs (Kamachi et al., 2000), and POU-domain
binding motifs (Phillips and Luisi, 2000). Phylogenetic analysis of these motifs reveals
varying levels of conservation between Rx regulatory sequences of X. laevis or in
combination with higher vertebrates (Figure 1C). A predicted SOX motif, POU-domain
binding motif and an Otx binding motif are clustered together in the distal conserved
27
region, and are highly conserved even with human sequences (Figure 1C, and Danno et
al., 2008). An additional Otx site is predicted at Rx2A: -1703, and a forkhead binding
element (FBE) at Rx2A: -847 is conserved among Rx regulatory regions in X. laevis
(Figure 1C).
2.3.2 The Rx2A regulatory sequence directs gene expression in regions of endogenous
Rx expression
We tested the ability of the 2.8Kb region upstream of Rx2A to direct gene
expression by cloning this fragment upstream of Green Fluorescent Protein
(GFP)(Rx2A2.8/GFP) and analyzing GFP expression in X. laevis transgenic embryos
(Figure 2, A-F). GFP expression was detected in the anterior neural plate of developing
embryos, and later in the eyes, ventral forebrain and pineal gland of tailbud stage
embryos. Rx2A2.8/GFP transgene is active in all the areas in which endogenous X. laevis
Rx is expressed (Zhang et al., 2003, Mathers et al., 1997).
2.3.3 Spatial and temporal specificity of Rx expression are determined by regions
within the Rx2A regulatory sequence
We performed deletion analysis of the Rx2A2.8 promoter to gain insight into the
regulatory mechanisms governing Rx transcription. We generated 5’, 3’ and internal
deletions of the Rx2A regulatory sequence, and fused these promoter fragments upstream
of GFP. We tested the ability of these constructs to drive expression of GFP in
transgenic X. laevis embryos.
Analysis of 5’ end promoter deletions. Expression analysis of 5’ deletions (Figure 3, 817) reveals regions of the Rx2A promoter that provide spatial and temporal specificity to
the Rx2A promoter sequence. The -2359 fragment results in GFP expression
28
indistinguishable from the full-length construct (Figure 4, A-C). Deleting to -2160 bps
results in the loss of GFP expression in the anterior neural plate (ANP) compared to the
full-length construct; GFP expression is observed beginning in the optic cups (stage 24)
(Figure 3, construct 9; Figure 4, D-F). In addition to the loss of expression in the ANP,
ectopic expression is seen in the Rx2A 2160 construct (construct 9 and Figure 4, D-F).
Ectopic expression is evident in the dorsal hindbrain of these embryos beginning at stage
24 and persists throughout all stages examined (until stage 41) (Figure 4, D-F, red
arrowheads). Both the loss of ANP expression and the ectopic expression in the dorsal
hindbrain is evident in constructs containing successive 5’ deletion from -1954 to -944 of
the Rx2A promoter (Figure 3, 9-16). Expression, both ectopic and in regions of
endogenous Rx expression, becomes progressively weaker in these constructs (Figure 3
constructs 10-16, Figure 4, G-J). Collectively, this data suggests that a cis-acting element
that provides spatial and temporal specificity to the Rx2A promoter is located between 2160 and -2359. As well it suggests that additional cis-elements that regulate the strength
of expression are present between -2160 and -944 of the Rx2A promoter.
As previously mentioned, deleting to -944bps results in ectopic hindbrain
expression, very weak forebrain expression and retinal expression from stage 24 onward
(Figure 3, construct 16; Figure 4, L). In contrast, deleting to -818bps does not result in
visible GFP transgene expression. Sequence analysis reveals the presence of 2 putative
FBEs between -818 and -944 (FBE, -847; FBE II, -858).
To test whether either of these sites functionally acts as cis-regulatory elements,
we added the FBEs individually upstream of the -818 fragment of the Rx2A promoter
(Figure 3; constructs 18-19). Only the FBE at -847 (Figure 3; construct 18) was capable
29
of restoring the eye expression in developing embryos (Figure 4, M). We also tested the
sufficiency of the two FBEs to direct gene expression in developing eye by creating GFP
transgenes containing either FBE upstream of the heat shock promoter (hsp) (Figure 3;
constructs 35-36). Neither FBE was sufficient to drive expression of a GFP transgene in
this context (data not shown). We conclude that the FBE at -847 represents a functional
FBE, and confers expression in the eye in cooperation with the -818 minimal fragment of
the Rx2A promoter.
Analysis of internal deletions. We further characterized the Rx2A regulatory region by
generating a series of internal deletions (Figure 3; constructs 20-22, 24, and 25). The
internal deletions include the -818 to 0 bp minimal fragment determined by 5’ deletion
analysis. The internal deletion constructs then have serially deleted portions toward the 5’
end (Figure 2; constructs 20-22, 24, and 25). All internal deletions lack the critical region
between -944 and -818 bp that contains the functional FBE regulatory element (Figure 2;
constructs 20-22, 24 and 25).
GFP expression of -818 to -1208 bp (Figure 3; construct 20), the shortest internal
deletion, is similar to the expression of the full-length construct (Figure 5A and B). It is
expressed in the eye field at the ANP stage and in the presumptive eye at stage 24 (Figure
5A and B). In contrast, we found that internal deletions spanning from -1703 to -2400 bp
(Figure 3; constructs 21, 22, 24 and 25) resulted in ectopic GFP expression in a domain
ventral to the cement gland at the ANP stage (Figure 5C, F and I). Ectopic expression of
GFP ventral to the cement gland was first observed at ANP stages and persisted
throughout development (Figure 5D, E, G). At tadpole stages, the ventral domain of
ectopic GFP expression seemed to be located in the heart (Figure 5G). Immunolabeling
30
with anti-Myc antibody confirmed expression of the transgene in the heart of embryos
transgenic for construct 25 (Figure 5H; red arrowheads). Taken together, these results
suggest that additional elements between -818 and -2400 also impart spatial and temporal
specificity to the Rx2A promoter.
Using the Transcription Element Search Software (TESS) we found two predicted
Otx sites located at -1706 and -2183. In X. laevis, Otx2 is expressed at the anterior neural
plate and its expression domain forms a ring that surrounds Rx expression in the eyefield
(Andreazzoli et al., 1999). This mutually exclusive expression pattern suggests that Otx
and Rx negatively regulate their expression at early neurula stages (Andreazzoli et al,
1999).
Internally deleting up to -1703 eliminates half of the Otx site located at -1706
(Figure 3; construct 22), possibly rendering it non-functional. To test whether
elimination of the -1706 Otx site resulted in ectopic expression of GFP ventral to the
cement gland, we added back the Otx site to construct 22 by specifically adding back the
Otx sequence (construct 21) or a 126bp region that contains the Otx site (construct 23).
We found that addition of the Otx site did not restore expression specificity. 45.8%
(n=24) of embryos transgenic for construct 22 had ectopic GFP expression ventral to the
cement gland at the ANP stage (Figure 5I).
These results suggest that elimination of the Otx site does not lead to ectopic
expression of GFP ventrally and that other regulatory elements present within this region
are responsible to maintain specificity of promoter activity. Since this region functionally
acts as a repressor of ventral expression, we term this region the Central Repressive
Element (CRE).
31
Analysis of Hsp constructs. To ask which portions of the Rx2A regulatory region are
sufficient for GFP expression, we produced a series of constructs in which the Rx2A
minimal fragment (-818 to 0 bp) was replaced with the minimal heat shock protein (hsp)
promoter (Figure 3, constructs 27-30, 32). Adding the 1703, 1823 or 2136 distal
fragments to the hsp minimal promoter results in transgenic embryos that transiently
express GFP in the optic cups (Figure 6A-C). Expression of GFP for these constructs
only occurs from stage 24 to stage 28 (Figure 6A-C). In contrast, Rx2A 2400-hsp
(Figure 3, construct 34), does not result in GFP expression in embryos (data not shown).
These data suggest that the distal portion of the promoter is sufficient to confer transgene
expression in Rx expression domains, but that more proximal regions are necessary to
sustain the expression during later stages of development.
We had determined that the -847 FBE could restore GFP expression during late
stages. For this reason, we tested whether the same FBE element could sustain late
expression of the hsp transgenes that showed transient expression. Adding the FBE to
2136-hsp (Figure 3; construct 31) resulted in sustained expression of GFP from neurula
stages until stage 41 (Figure 6D and E). However, addition of the same FBE to 2400-hsp
(Figure 3; construct 33) did not result in detectable expression of GFP (Data not shown).
This result suggests a cooperative activity of the FBE with regulatory elements present in
the distal 2136 promoter fragment (which contains UCE) to sustain promoter activity
until late developmental stages.
32
2.3.4 The SOX and OTX sites within UCE act cooperatively to provide spatial
specificity to Rx expression
The distal region of the Rx2A promoter contains a highly conserved region
identified by bioinformatic analysis. We termed this highly conserved region UCE and
tested the ability of specific TF sites within this region to regulate the expression of a
GFP transgene. We mutated the SOX site (-2170) and OTX (-2183) site and deleted the
POU site (-2201) within UCE (Figure 3, Constructs 2-4) in the context of the full-length
Rx2A promoter. We also analyzed whether the more proximal OTX site (-1703) could
functionally compensate for the OTX within UCE by mutating OTX (-1703) alone or in
combination with the OTX (-2183).
Deleting the POU site within UCE results in ectopic GFP expression in
developing neural tissue. The ectopic expression appears to be pan-neural in nature as
opposed to the focal expression in the dorsal hindbrain observed in other constructs. The
ectopic expression as a result of POU mutation begins in early tailbud stage embryos and
persists through all stages examined. This data suggests that the POU site contributes
specificity to the Rx2A promoter, and appears to provide regional specificity to Rx
expression within the developing brain.
Mutating both of the OTX sites in the context of the full-length promoter
produces ectopic expression in the forebrain of transgenic embryos (Figure 3, construct 7;
Figure 4, C-E, red arrows). Ectopic forebrain expression is apparent beginning at stage
24, and persists through stage 41. Mutation of either of the OTX sites alone does not
result in aberrant expression in the hindbrain when compared to the full-length construct
(data not shown). This suggests that either OTX is capable of compensating for the
33
absence of the other. Mutating the SOX site alone is sufficient to produce ectopic
expression of the transgene in the hindbrain region, albeit in a small percentage of
embryos (15%, n=12, Figure 5, G-H). Expression in the eyes appears indistinguishable
from the full-length construct. However, mutating both the SOX and OTX sites of UCE
(Construct 5) in the same construct increases the frequency of ectopic expression to that
observed in Rx2A 2160 transgenics (100%, n=12 construct 5 vs. 100%, n=24, construct
9) (Figure 4, D-F). This data suggests that the SOX and OTX sites within UCE are
required for spatial specificity within the forebrain, resulting in eye-specific expression.
The OTX and SOX sites act cooperatively as the construct harboring mutations in both
TF sites results in a higher frequency of embryos with ectopic hindbrain expression than
either single mutation alone. It also suggests that the proximal OTX site can compensate
for the loss of the distal OTX in the presence of an intact SOX site.
2.3.5 Rx2A promoter activity in the mature retina
To investigate whether the full-length Rx2A promoter was capable of directing
expression to endogenous regions of the mature retina, we examined expression of Rx2A
2823 by in situ hybridization on retinal sections. Rx2A 2823 is active in the PRL, INL,
and partially in the CMZ at stage 41 (Figure 8A). The retinal progenitor cells at the
ciliary marginal zone (CMZ) can be subdivided the CMZ into four zones based on the
maturity of the cells (Perron et al, 1998). In this model of CMZ organization, the most
peripheral cells adjacent to the lens are considered retinal stem cells (Zone 1), followed
by proliferating retinoblasts (Zone 2), differentiating retinal cells (Zone 3) and postmitotic retinal cells (Zone 4). Due to the absence of GFP expression in the very distal tip
34
of the CMZ, it appears that the full-length Rx2A promoter construct is expressed in zones
2-4 of the CMZ (Figure 8A, brackets).
To investigate whether cis-acting elements of the Rx2A promoter differentially
regulate gene expression in the INL, PR or CMZ, we examined expression of Rx2A
deleting constructs in the mature retina by in situ hybridization on retinal sections (Figure
8). The expression pattern for Rx2A2160/GFP and Rx2A1764/GFP is very similar to the
expression of the full-length promoter construct (Figure 8B and C). Both transgenes are
expressed in the PRL, INL and zones 2-4 of the CMZ. On the other hand, expression of
the GFP transgene is not observable in the INL or CMZ with Rx2A944 /GFP construct
(Figure 8D).
We also analyzed the expression of transgenes in which the putative Otx and Sox
cis-regulatory elements were mutated in the context of the full-length promoter.
Mutation of the -2183 Otx site contained within the UCE results in an expression pattern
similar to Rx2A-GFP (Figure 8E). Interestingly, mutating both the -2183 and -1703 Otx
sites also results in expression in PRL and CMZ, but not INL expression (Figure 8F).
Doubly mutating the Sox and Otx cis-regulatory elements within the UCE abolished
expression of the transgene in the INL and CMZ (Figure 8G).
Lastly, we determined the expression GFP for the internal deletions Rx2A 8181208 and Rx2A 818-2136. Both Rx2A 818-1208 and Rx2A 818-2136 are expressed in
the CMZ and PRL, but do not seem to have robust expression in the INL (Figure 8H,I).
2.3.6 Cognate factors that bind the Rx2A cis-elements
The UCE contains a functional POU site that acts to maintain expression of Rx
expression to the forebrain. It has recently been shown that the Class III POU domain
35
transcription factor, Oct-3 (Brn-2) binds another homeobox-containing transcription
factor expressed in RPCs, Chx10 (Rowan et al., 2005). This data along with the
overlapping expression pattern between Rx and Oct-6 led to investigate whether Oct-6
acts as a cognate factor the POU site within UCE. By performing electrophoretic
mobility shift assays (EMSAs) using in vitro- translated Oct-6 and radio-labeled POU site
probe, we show that Oct-6 specifically binds the POU site within UCE of the Rx2A
promoter.
We have shown that the -847 FBE functions as a weak activator of Rx
transcription beginning at the optic vesicle stage. We have recently reviewed the
expression and function of the forkhead family of transcription factors during eye
development (Moose et al., 2008). From this analysis, several forkhead proteins provide
substantial candidates as trans-acting factors to bind the FBE, including FoxO3 and
FoxN4. Each of these forkhead gene products are expressed from neural plate to late
tailbud stages in regions overlapping with endogenous Rx expression. We tested the
ability of in vitro- translated FoxO3 and FoxN4 protein to bind the FBE sequence of the
Rx2A promoter using a radio-labeled FBE probe using EMSAs (Figure 2.9). Both
FoxO3 and FoxN4 are able to bind the FBE in this context. The binding of FoxO3 or
FoxN4 to the FBE sequence is specific evidenced by introduction of 10x unlabeled FBE
probe.
36
2.4 DISCUSSION
2.4.1 Rx genes in X. laevis are similarly regulated
The regulatory region of Rx1A had been previously published (Zhang et al.,
2003). Several lines of evidence suggest that Rx1A and Rx2A are similarly regulated:
(1) the regulatory sequence directly upstream of the Rx1A gene is sufficient to direct
expression of a GFP transgene in regions of endogenous Rx expression, (2) 5’ deletion
analysis of the Rx1A promoter sequence indicates that the distal region of the Rx1A
promoter is necessary for Rx expression in the ANP, (3) and a cis-element in the
proximal region works in concert with a minimal promoter to specify expression at late
stages of retinal development. In contrast to the Rx2A promoter, the Rx1A promoter was
not sufficient to drive gene expression in the CMZ in the mature retina (Zhang et al.,
2003). However, endogenous Rx1A and Rx2A are both expressed in this region. This
discrepancy could be due to differences in gene regulatory mechanisms between the two
genes.
2.4.2 Expansion of the knowledge of Rx cis-regulatory elements
The deletion analysis of Rx2A has furthered the knowledge of Rx gene
regulation. We have demonstrated that Rx2A possesses a repressive cis-element in the
central region we termed the CRE. The CRE provides expression specificity to the Rx2A
gene product, ensuring Rx is not expressed ventral to the cement gland. In addition, we
have shown, using Rx2A internal deletions, that a distal fragment (corresponding to the
region necessary for ANP expression of Rx1A) is expressed only transiently during early
neural stages. A newly identified FBE in the proximal region of Rx2A is responsible for
the late expression in the eye from tailbud stage, and our data indicate a synergistic action
37
between the distal UCE and the proximal FBE. Fragments of the Rx2A promoter that
have either the UCE or the FBE are sufficient to direct gene expression from early
neurula onward. However, constructs containing both the UCE and FBE, lead to robust
gene expression in the ANP and onward. Although no data from the studies of the Rx1A
promoter contradict the possibility of this synergism, the authors did not directly test it.
We have expanded the knowledge of Rx regulation by identifying new cis-regulatory
elements and demonstrating cooperativity between these elements.
2.4.3 The central region of the Rx2A promoter possibly contains a repressor element
From our analysis of internal deletions of the Rx2A promoter, we defined the
region from -1553 to -1703 as containing a functional repressive element for tissue
ventral to the cement gland, the CRE. We ruled out that a conserved Otx site was
responsible for the ventral ectopic expression within the CRE. We cannot currently
ascribe the ventral ectopic expression to the loss of a specific cis-regulatory element.
Contrary to the findings of our internal deletion analysis of the Rx2A promoter, we never
observed ectopic expression of GFP ventral to the cement gland with 5’ end deletions that
lack the CRE. If the CRE acts alone, similar ectopic expression of GFP should have been
observed with the 5’ end deletions that lack the central region of the promoter. This
discrepancy is suggestive of a transcriptional “ON” and “OFF” regulatory switch that
regulates Rx transcription ventral to the cement gland: (1) a region of the Rx2A
regulatory sequence turns “ON” Rx expression ventral to the cement gland, (2) while the
CRE acts as the “OFF” switch, thus ensuring eye specific expression.
Based on our observations, it is likely that a cis-regulatory element for an
activator (the “ON” switch) is present at the 5’ end of the promoter distal to the UCE. In
38
agreement with this, internal deletions likely have the activating cis-regulatory element,
but lack the CRE. Consequently, Rx transcription is activated and not turned off
appropriately, thus resulting in ectopic expression of GFP. We can rule out that UCE
acts as the “ON” switch since an internal construct lacking UCE still exhibits ectopic
ventral expression. With this same construct we determined that the “ON” activator is
present in the promoter region distal to UCE (-2400 to -2823) and for this reason we
termed this putative regulatory element the distal activator element (DAE). Moreover, 5’
end deletions that lack the activator, or lack both the activator and CRE, do not exhibit
ventral ectopic expression. Further dissection of the promoter is needed to identify
specific cis-regulatory elements responsible for regulating Rx transcription within the
CRE and DAE.
2.4.4 The UCE and FBE cooperate to activate Rx2A promoter activity
In our analysis of the internal deletion constructs, we found that most of the
central region of the promoter is not required for its activity in the developing eyes. All
of the internal deletion constructs are expressed from ANP stage to late tailbud stage.
Interestingly, all of these internal deletions include the UCE, but lack the FBE, with the
exception of Rx2A 818-2400. This suggests that in the absence of the FBE, a region
containing UCE is sufficient to activate the promoter in the ANP. Currently, we cannot
discount the fact that the DAE is not able to functionally replace UCE in Rx2A 818-2400.
We also tested whether the 5’ fragment of each internal deletion was sufficient for
promoter activity. To do this, we placed the 5’ fragments upstream of the heat shock
protein (hsp) minimal promoter. We found that the 1208-hsp construct recapitulated
expression of the full-length promoter. In sharp contrast to these results, only transient
39
promoter activity was observed using constructs with shorter 5’ fragments. Transient
expression of GFP commenced at stage 24 and was extinguished by stage 28. Lastly, we
found that the shortest hsp construct containing the 2400 distal fragment was not
sufficient to activate promoter activity at any developmental stage. These results suggest
that the UCE is necessary for activation in the ANP, but additional elements are required
to maintain expression.
In the 5’ end deletion analysis, elimination of the -847 FBE resulted in
abolishment of promoter activity. The 1208-hsp promoter construct lacks the FBE but it
is still expressed at all stages analyzed. A possible explanation is that the 1208 fragment
contains an FBE-like element that compensates for the loss of the -847 FBE. This FBElike element within the 1208 fragment could be sufficient to functionally replace the -847
FBE, thus enabling promoter activity both early and late. The transient expression of the
1556, 1703, and 2136-hsp constructs could be due to loss of the FBE-like element that
promotes expression of the 1208-hsp construct. It is likely that in these hsp constructs the
UCE is sufficient to initiate promoter activity at stage 24, but that an FBE is required to
sustain this activity. However, we cannot explain the expression of GFP at all stages
analyzed with the UCE-hsp construct.
We hypothesize based on the internal deletion constructs, that the region distal to
the UCE contains a DAE for ventral expression. All internal constructs were made using
the 818 minimal fragment. However, constructs containing the DAE upstream of the hsp
minimal promoter do not exhibit ventral ectopic expression. It appears that the 818
fragment of the Rx2A promoter and the hsp minimal promoter have different capacities
for transcriptional activation. This notion is further supported when comparing the
40
construct that contains the FBE and the minimal fragment (Rx2A818+FBE) to one that
contains the FBE upstream of the hsp minimal promoter (Rx2A FBE+HSP).
Rx2A818+FBE is active, albeit weakly, while Rx2A FBE-HSP is not. This is also
consistent with findings concerning Rx1A promoter constructs (Zhang et al., 2003).
The majority of the hsp construct data suggests that without additional input, the
presence of a UCE does not result in sustained expression past stage 28. However,
addition of the -847 FBE to the 2136-hsp construct was sufficient to initiate and maintain
robust expression of GFP throughout development. Addition of the same -847 FBE to
2400-hsp, which lacks UCE, did not result in promoter activity at any stage. Collectively,
this is consistent with cooperation of the FBE and UCE to initiate and sustain promoter
activity. Taken together, the results of the 5’ deletion, internal deletion and hsp construct
analysis strongly suggest that the UCE and the FBE are able to independently initiate Rx
transcription, and that the cooperation between UCE and FBE maintains strong
expression of Rx throughout eye development.
2.4.5 Putative trans-acting factors of the Rx2A promoter
We have demonstrated that the POU, Otx and Sox sites within UCE and a
proximal FBE are all functional cis-elements within the Rx2A promoter. Several
members of each of these transcriptions factor families have the potential to regulate the
Rx2A promoter by their known expression in the developing eye.
POU factors
The deletion of the POU site within UCE results in ectopic expression of a GFP
transgene throughout the neural tube from the ANP stages, demonstrating this cisregulatory element acts with repressive function. We have shown that Oct-6 is able to
41
bind the UCE POU site in vitro. We recognize that additional family members may
represent cognate factors to the POU sites. In particular, Oct-1 is a POU domain
transcription factor that is expressed throughout the neural plate where expression of Rx
is initiated. In addition, the expression of Oct-1 at early tailbud stages is strikingly similar
to the ectopic expression of the GFP transgene construct containing a mutated POU site
(Kiyota et al., 2008). Functionally, Oct-1 acts as a repressor until the recruitment of
additional co-factors allow conformational changes in the protein and transcriptional
activation results (Ryan and Rosenfeld, 1997). For these reasons, Oct-1 is a POU domain
transcription that serves as a candidate trans-acting factor for the Rx2A promoter. Future
work will be needed to determine if Oct-1 binds the Rx2A promoter in vitro, and whether
Oct-6 or Oct-1 are located at the Rx2A locus in vivo.
Otx family members
We have determined that the Otx sites within the Rx2A promoter act with
repressive function, since Rx2A 5’ deletion constructs without the Otx site(s) and Otx
mutated constructs exhibit ectopic expression in the dorsal hindbrain. The ectopic
expression is observed in known regions of expression of three members of the Otx
family: Otx1, Otx2, Otx3. Each of these Otx family members have both expression
patterns and functional roles in CNS development that make them potential trans-acting
factors to act on the Rx2A promoter. Also discussed is the potential regulation of Rx
transcription by Otx family members, such as Otx2 and Otx5b, in the mature retina.
Otx1 is first expressed throughout the presumptive forebrain and midbrain
neuroepithelium, and strikingly in precursor cells of sensory organs, including the RPCs
of the optic vesicles and anterior region of the developed mouse eye (Simeone et al.,
42
1993). Otx1 expression is present in the CMZ of X. laevis (HME, unpublished
observations), and this expression in the anterior regions seems to be conserved in
humans, as Otx1 is expressed in the anterior region of the human fetal retina (Larsen et
al., 2009). It is important to note that Otx1 is the only Otx family member not expressed
in differentiated cell types in the retina, and rather only in regions that contain
proliferative multi-potent cells. Since Rx is also expressed in both the ANP and in the
CMZ that contains multi-potent cells in X. laevis, Otx1 is a strong candidate to regulate
the Rx promoter in these regions.
Otx2 is also a strong candidate to regulate Rx transcription via UCE in the ANP.
Otx2 is expressed in the ANP of X. laevis and functional data also suggests an essential
role for Otx2 in eye formation (Andreazzoli et al., 1999, Chuang et al., 1999, Zuber et al.,
2003). In X. laevis, Otx2 is expressed throughout the region of the ANP that gives rise to
the retina (Pannese et al., 1995), slightly earlier than the onset of expression for Otx1.
The domain of Rx expression is within that of Otx2 in the ANP (Zuber et al., 2003;
Andreazzoli et al., 1999). It has been suggested that reciprocal repression between Otx2
and Rx specifies the eye field (Zuber et al., 2003). Indeed, Rx (Mathers et al., 1997) is
involved in repressing Otx2 expression in the anterior neural plate (Andreazzoli et al.,
1999) (Zuber et al., 2003). Otx2 regulation of the Rx2A promoter would provide direct
evidence for the remaining reciprocal repression. Otx2 has been independently shown to
bind the human enhancer of RAX at the same site we identify in these studies (Danno et
al., 2008). Also, Otx2 is expressed and functions to regulate cell fate specification of the
pineal gland, another region of endogenous Rx expression (Danno et al., 2008; Furukawa
43
et al., 1999; Nishida et al., 2003). This collectively demonstrates that Otx2 provides an
excellent candidate to regulate Rx transcription through the Rx2A promoter.
It has been proposed that the expression of Otx2 confers competence to produce
eyes in the region of the ANP where it is expressed. Misexpressing eye field genes to
produce ectopic eyes is possible only in the Otx2-positive region in zebrafish (Chuang et
al., 1999). In addition, in X. laevis, a cocktail of EFTF mRNAs is sufficient to induce
ectopic eyes outside the CNS, but only when Otx2 is included (Zuber et al., 2003). These
studies are concordant with Otx2 being upstream of Rx, since Rx is also necessary for
eye formation (Mathers et al., 1997). One caveat to these studies is that in mice, Otx1
and Otx2 appear to be functionally redundant (Reviewed in Simeone et al., 2001). So, it
cannot be ruled out that Otx1 provides this function during normal development, but that
Otx2 is sufficient to produce the same effect under certain experimental circumstances.
In either case, both Otx1 and Otx2 provide solid candidates to regulate the Rx promoter.
Otx3 appears to also be more caudally expressed throughout the hindbrain than
either Otx1 or Otx2 (Zhang et al., 2002), and Otx3 is also expressed in the developing
optic cups, although it is not clear whether the eye expression is maintained in
differentiated cell types or whether Otx3 is expressed in the anterior region of the eye in
mammals or CMZ of X. laevis. Nonetheless, Otx3 has an expression pattern consistent
with potential to regulate gene expression.
Lastly, Otx5b is expressed in the developing and mature photoreceptor cells of the
neural retina, as well as in the pineal gland (Furukawa et al., 1999; Vignali et al., 2000).
Loss of Otx5b function results in photoreceptor degeneration in mice (Furukawa et al.,
1999), a phenotype that is also observed when Rx translation is knocked down in X.
44
laevis tadpoles (Pan et al; manuscript submitted). In addition, Rx and Otx5b cooperate
to activate the rhodopsin promoter (Pan et al; manuscript submitted). An attractive
hypothesis is that Otx5b positively regulates Rx expression to promote photoreceptor
identity. Therefore, Otx5b represents another Otx family member with the potential to
regulate Rx gene expression, at least in photoreceptor cells.
In summary, since no Otx family member has an expression pattern that
completely recapitulates the expression of Rx, it is likely that the Otx cis-element (s) are
responsive to different Otx family members depending on developmental stage or cell
type. For example Otx1, Otx2, and Otx3 are each robustly expressed in the developing
forebrain during development (Zhang et al., 2002), and are present to act on the Rx2A
promoter in cells of the ANP. In the mature eye, Otx1 appears to be the only Otx gene
product expressed in the proliferative cells of the retinal margin in mice, and may be the
only Otx member with the potential to regulate Rx gene expression in the CMZ. Otx2 and
Otx5b are both expressed in cells of the INL, and in photoreceptors, making them likely
candidates to act alone in or in concert to regulate Rx transcription in these cell types. As
previously mentioned, Otx2 was shown to bind the human RAX enhancer in vivo (Danno
et al., 2008). However, the authors did not explore the ability of additional Otx family
members to bind, nor did they investigate the specificity of binding in different cell types
or developmental stages.
Sox family members
Our site-directed mutations within UCE of the Rx2A promoter suggest an
involvement of Sox family of transcription factors in regulating the Rx2A promoter. Sox
factors from three classes of Sox genes are expressed in early CNS structures (Sasai,
45
2001), and could act as trans-acting factors to the functional Sox site in the Rx2A
promoter: the Sox B1 subclass (Sox 1, 2, 3), SoxB2 subclass (Sox 14 and sox21) and the
group G Sox protein, SoxD (Reviewed in Sasai, 2001). Sox1 is expressed in the ANP of
X. laevis embryos, and becomes restricted to regions of the brain as well as optic vesicle
by tailbud stages (Nitta et al., 2006). Sox3 is robustly expressed throughout the closing
neural tube, but is not maintained in the eye, restricting the possibility for Sox3 to
regulate the Rx2A promoter to early time points (Penzel et al., 1997). Sox 14 and Sox 21
have multiple regions of expression within the developing brain (Cunningham et al.,
2008; Penzel et al., 1997). However, the only region that represents their potential to
regulate Rx transcription is their expression within the mid-hind-brain boundary
(Cunningham et al., 2008). All these Sox family members have the potential to regulate
Rx transcription on the basis of their expression in the ANP.
Despite multiple Sox factors being expressed in regions of Rx expression, Sox2
provides the most convincing candidate for the regulation of the Rx promoter. Sox2 has
been heavily implicated in eye development. Sox2 maintains the pluripotency of neural
progenitor populations, including RPCs (Graham et al., 2003; Van Raay et al., 2005).
Sox2 heterozygous mutations result in 10-20% of all anophthalamic and microphthalmic
cases in humans (Ragge et al., 2005). In animal models, genetic alteration of the Sox2
locus also results in defects in eye formation (Taranova, 2005), and the authors
demonstrate that Sox2 contributes to progenitor cell competence in the neural retina
(Taranova, 2005). Lastly, and most strikingly, Sox2 has been shown to bind upstream of
the RAX promoter in vivo at the ANP stage (Danno et al., 2008). However, these
46
experiments did not rule out the involvement of other Sox proteins at different time
points.
All known Sox transcription factors bind very similar DNA sequences (Kamachi
et al., 1999). Nevertheless, individual Sox proteins have been shown to have distinct
targets (Kamachi et al., 1999; Kamachi et al., 2000). Therefore, it has been proposed that
Sox factors form complexes with other proteins to affect transcription, and that accessory
co-factors provide transcriptional specificity (Donner et al., 2007).
Forkhead transcription factors
The FBE is a functional cis-element in the Rx2A promoter that is capable of
weakly driving expression at optic vesicle stage and onward. We show that FoxO3 and
FoxN4 can both bind the FBE in vitro. Even in light of these results, further
characterization of the sequence defined by the FBE reveals that it is most like a FoxM
family site as compared to binding sites from other forkhead subfamilies (data not
shown). Additionally, we know that the FBE acts as a weak activator for Rx
transcription during stages of rapid proliferation of RPCs. FoxM1 provides an excellent
additional candidate because of its known role in the cell cycle progression and function
as a transcriptional activator (Wierstra and Alves, 2007). Future studies will address
whether these candidates act as the trans-acting factors that activate Rx transcription
through the FBE.
2.4.6 Coordinated regulatory activity of gene families implicated in Rx2A promoter
regulation
It is evident from the deletion analysis that the presence of intact POU, Otx and
Sox sites are essential for both the proper induction and specificity in the ANP region of
47
transgenic embryos. The following section discusses the known cooperativity between
members of these transcription factor families. We also provide a working model to
explain Rx2A transcription regulation by cis-elements based on our deletion analysis.
Firstly, it has been previously shown that Otx2 and Sox2 can physically interact,
and these two proteins are localized at the RAX promoter in vivo (Danno et al., 2008).
Using luciferase constructs containing multimerized Otx and Sox sites based on the RAX
sequence, the authors found that only the introduction of Otx2, not Sox2, is sufficient to
activate a reporter construct in vitro (Danno et al., 2008). Mutating the Sox or Otx sites
abolishes the transactivation seen with Otx2. The authors of this study do not discuss the
presence or importance of the upstream POU site within the RAX regulatory region.
In contrast, our studies suggest additional regulation through the conserved POU
site within UCE. The interaction of POU domain proteins and Sox proteins is well
documented (Reviewed in Dailey 2001). Several POU/Sox complexes have been
identified that function as transcriptional regulatory complexes, including several
containing Sox2. Sox2 is capable of cooperating with Oct-1, Oct-3/4, and Oct-6
(Ambrosetti et al., 1997; Ambrosetti et al., 2000; Nishimoto et al., 1999; Tanaka et al.,
2004). Interestingly, the ability to transactivate gene targets is confined to discrete
partnerships. For example, the Sox2/Oct-3/4 complex can transcriptionally activate the
FGF-4 promoter; transcriptional activation is only possible in the presence of POU/Sox
protein-protein interaction and is DNA binding dependent (Ambrosetti, 1997). In
contrast, a Sox2/Oct-1 complex is not (Ambrosetti, 1997). These data suggest that (1)
complex formation is essential for transactivation and (2) there is functional specificity
between Sox and POU partners. In addition, it has been shown that the activation or
48
continued repression of gene targets by POU/Sox complexes is dependent on the
recruitment of co-factors (reviewed in Dailey et al., 2001).
It is interesting to speculate about whether the POU/Sox site in the Rx2A
promoter coordinate transcription in this manner, and what might provide the additional
co-activator/co-repressor function. One scenario would be the recruitment of Otx family
members to a POU/Sox repressor complex within UCE. This mechanism would allow
specificity to gene expression by the employment of different Otx family members in a
tissue/temporal specific manner. A second intriguing scenario is that the co-activator
recruited by a UCE POU/Sox complex is bound at the functional FBE in the proximal
promoter. A third possibility is that the binding of stage/tissue-specific Otx proteins to
POU/Sox complex at UCE enables an interaction with an activating trans-factor bound at
the FBE.
Either model involving Otx recruitment explains the difference in ectopic
expression that is seen between the site-directed mutation of the POU site (pan-neural
ectopic expression), and the loss of ANP expression using 5’ deletion constructs. In the
case of the site-directed POU mutation, only the repressive function of POU is lost, and
ectopic expression is gained. Intriguingly, the POU domain protein, OCT-1, possesses an
expression pattern strikingly similar to the ectopic expression seen in the site-directed
POU mutation transgenic embryos. In the case of 5’ deletions that lose expression in the
neural plate, the POU site is lost along with the adjacent Sox and Otx sites. We speculate
that the POU/Sox proteins form a transcriptional repressor complex at UCE (as is the
case for many POU/Sox partnerships), until the recruitment of an Otx creates the
potential to transactivate. If this is the case, then the loss of the POU/Sox/Otx sites
49
together would result in a loss of gene expression, as is the case regarding 5’ deletion
constructs in the ANP.
Moreover, this model can account for both the loss of expression in the anterior
neural plate, and the gain of expression in the dorsal midbrain of 5’ deletion constructs.
We consider that different Otx members might regulate the Rx2A promoter in different
cell types, and Otx proteins can act as co-activators or co-repressors. This allows the
possibility to recruit a co-repressor in the dorsal hindbrain (possibly Otx2), and a coactivator in the developing RPCs (possibly Otx1) or photoreceptor cells (possibly Otx5b).
Interestingly, a combinatorial code between POU/Otx family members has recently been
proposed for the production of bipolar cell gene expression (Kim, 2008).
2.4.7 Future directions
Rx2A deletion analysis has lead us to propose the presence of a DAE and CRE
that work together to retain Rx transcription in the developing eyes. We have not,
however, defined specific TF sites that are responsible for their action. In the future,
finer deletion examination of the DAE combined with continued bioinformatic analysis
may reveal specific TF sites.
We have identified four major TF sites that function in the regulation of Rx
transcription. Therefore, the majority of the future work related to these studies should
be focused on defining the trans-acting factors that function as biologically relevant Rx
transcriptional regulators. Continued analysis using EMSAs will be used to confirm
whether candidate trans-acting factors bind the identified cis-regulatory elements in vitro.
To answer whether each putative Rx regulator is pertinent in vivo, we will perform
chromatin immunoprecipitation using antibodies specific to candidate factors and primers
50
specific to the Rx2A regulatory region. Lastly, we will analyze the nature of regulation
by performing gene transcription assays using luciferase constructs containing full-length
or deletion Rx2A fragments in X. laevis embryos. This method will also enable
investigation of the co-regulatory capacity of trans-acting factors.
51
FIGURE 2.1 Conservation and in silico analysis of the Rx2A Regulatory Region
(A) Xenopus Laevis Rx2A genomic sequence isolated by linker mediated PCR aligned
with Rx1A regulatory sequence (Zhang et al., 2004). Segments of DNA with percent
identity above 50% are shown. (B) Schematic Representation of the Rx2A regulatory
region. Scale indicates transcription start to the right. Longest regions of highest identity
revealed by Pip analysis (red boxes) and predicted cis-regulatory elements predicted by
MatInspector are shown. Yellow: POU Domain Motif. Light Green: OTX Motif. Blue:
SOX Motif. Dark Green: Forkhead Motif. Regions of greatest divergence are shown in
teal (see text for details). (C) Alignments showing conservation of predicted TFBS within
the Rx2A regulatory region. ClustalW alignments are shown (MacVector Software).
Completely conserved residues are shown in gray. Colored boxes indicate core binding
motifs transcription factors. Numbers indicate relative position from proximal end of
isolated Rx2A regulatory region.
52
Figure 2.2 Rx2A2.8 Directs Gene Expression in the Eye Fields, Hypothalamus and
Pineal Gland. Fluorescent images taken of Rx2A2.8/GFP transgene. Stages are indicated
in each panel. Views: (A) anterior; (D) dorsal; (B-C, E-F) lateral. Expression of GFP is
first evident in the anterior neural plate (A), and is present in the developing eye during
neurula and tailbud stages (B-F). (D) Dorsal view that demonstrates eye specific
expression. Expression is also evident in other known regions of Rx expression: the
hypothalamus (arrows E, F) and pineal gland (arrowheads, E, F)
53
Figure 2.3 Deletion Constructs Produced to Investigate Transcriptional Activity of
the Rx2A Regulatory Region Deletion name and schematic representation of transgene
constructs are shown: black bar, Rx2A sequence; green bar, GFP; gray bar, HSP
promoter. The entire regulatory sequence isolated is shown in schematic form above and
below the transgene constructs: red, regions of high conservation with Rx1A sequence,
light blue, regions of divergence with Rx1A sequence; blue, predicted inversion event.
Vertical dashed line indicates where distal conserved region aligns in deletion constructs.
54
Figure 2.3
55
56
Figure 2.4 The Rx2A promoter contains cis-elements required for proper spatial
and temporal gene expression of the GFP transgene White arrows, regions of
endogenous expression. Red arrows, ectopic expression. Transgenes represented are
detailed on the right. (A-C) Rx2A 2394, (D-F) Rx2A 2160; (G-H) Rx2A 1823; (I-J)
Rx2A 1208; (K) RX2A 1090; (L) RX2A 944; (M) RX2A 818+FBE.
57
Figure 2.5 Internal deletions of the Rx2A promoter demonstrate additional
regulatory regions that provide spatial specificity Transgenes represented are detailed
on the right. (A-B) Rx2A 818-1208; (C-E) Rx2A 818-1236; (F-H) Rx2A 818-2400; (I)
Rx2A 818-1703+Otx.
58
Figure
2.6
Cooperative activity of UCE and the proximal FBE directs gene
expression throughout developmental stages Transgenes
represented
are
detailed
on
the
right.
(A)
Rx2A
1703‐hsp;
(B)
Rx2A
1823‐hsp;
(C)
Rx2A
2136‐hsp;
(D­E)
Rx2A
2136
+
FBE‐hsp.
59
Figure 2.7 POU, Sox and Otx sites within UCE cooperatively regulate Rx expression
in the developing eyes of Xenopus laevis embryos White arrows, regions of
endogenous expression. Red arrows, ectopic expression. Transgenes represented are
detailed on the right. (A-B) Rx2A UCE delPOU; (C-E) Rx2A mutOtx (-2179/-1703); (FH) Rx2A UCE mutSOX; (I-K) Rx2AUCE mutOtx/mutSox.
60
Figure 2.8 Expression of Rx2A-GFP transgenes in the mature retina. In situ
hybridization on retinal sections of stage 41 transgenic tadpoles showing the expression
of each transgene. (A) Rx2A 2823; (B) Rx2A 2160; (C) Rx2A 1764; (D) Rx2A 944; (E)
Rx2A UCE mutOTX; (F) Rx2A mutOtx (-2179/-1703); (G) Rx2A UCE mutOtx/mutSox;
(H) Rx2A 818-1208; (I) Rx2A 818-2136.
61
62
Figure 2.9 Oct-6, FoxO3 and FoxN4 bind Rx2A promoter cis-elements in vitro
EMSA using Oct-6 in vitro-translated protein and radio-labeled POU site probe (A). S, specific
competitor, unlabeled R2A UCE POU probe. M-mutated Rx2A UCE POU site probe. EMSA using
FoxN4 (B) or FoxO3 (C) in vitro-translated protein. Radio-labeled Rx2A FBE was used as probe. Sspecific competitor, unlabeled Rx2A FBE probe. NS-radio-labeled OCTA probe.
Figure 2.10 Model of Rx2A transcriptional regulation derived from deletion
analysis Schematic representation of the Rx2A locus is shown as a black bar, coding
region indicated by gray bar. Functional sites as defined by deletion analysis.
Candidate factors for each functional site are those discussed in the text. Bold
represents factors with significant data supporting action as a cis-regulatory molecule.
+, suggested cooperative activity between OTX and SOX site due to increased
ectopic expression in construct containing double mutation compared to construct
containing either a OTX or a SOX mutation alone. Abbreviations: ANP, anterior
neural plate; CRE, central repressive element; DAE, distal activating element; DH,
Dorsal hindbrain; PN, pan-neural; UCE, ultra conserved element; VCG, ventral to
cement gland.
63
64
65
66
67
Chapter 3: Seeing Eye to Eye: Forkhead Transcription factors during eye
development
3.1 INTRODUCTION
Recently, accumulating evidence has been demonstrated for a third major family
of transcription factors to be involved in eye development- the forkhead family of
transcription factors. Forkhead transcription factors are an evolutionarily conserved
family of proteins whose functions are diverse in developmental processes. The evergrowing family of forkhead proteins are divided into subclasses A to S according to the
conservation within their signature winged helix DNA binding domain (Kaestner et al.,
2000). They function as transcription factors, modulators of cell cycle machinery
function, cell fate determinants, as well as cell survival factors. Recently it has become
apparent that several subfamilies of forkheads transcription factors are expressed in
developing eye tissue. Here, the accumulating data regarding eye expression of forkhead
gene products to iterate the importance of this family in control of retinal precursors
during development. We highlight several forkhead proteins that regulate the activity of
retinal progenitor cells within the vertebrate eye. We describe the expression patterns of
known retinal forkhead genes across vertebrate species. We also discuss the available
functional data regarding forkhead proteins in the retina as transcription factors and cell
fate determinants. In this review, we will use species-specific conventions for writing
68
forkhead gene and gene product names: the forkhead box abbreviation (fox), followed by
the family designation, followed by the intra-familial individual gene designation with
species-specific capitalization. For example, the forkhead box gene A1 will be
designated as FOXA1 (human), FoxA1 (frog), Foxa1 (mammals), or foxa1 (zebrafish).
3.2 METHODS
In situ hybridization was performed on 8µm retinal sections as previously
described (Pan et al., 2006b; Shimamura et al., 1994; Viczian et al., 2003). Probes were
used at a concentration of 2.5µg/µl diluted in hybridization buffer. The following table
describes the production of antisense probes used in this study.
3.3 RESULTS AND DISCUSSION
3.3.1 Forkhead transcription factors in anterior eye structures
The retina in vertebrates is derived from competent neural ectoderm during development.
A region of the anterior neural plate is specified during neurulation as prospective eye
tissue. Multiple inductive events between this region and the overlying presumptive lens
ectoderm properly determine the eye fields. The differentiation of the neural retinal
precursors in this region occurs in a conserved stereotypical manner to produce proper
69
retinal layering and cell differentiation of the seven neural retinal cell types (6 neuronal
and one glial). Several forkhead transcription factors are expressed in the developing eye
and differentiated retina. Here we discuss the accumulating expression data for retinal
forkhead gene products with a focus on comparison of expression patterns across
vertebrate species.
Forkhead transcription factors are known to play critical roles in the development
of the anterior eye. Although they are not the focus of the current review, they deserve
mention due to the nature of reciprocal interaction of developing neural retina with
developing lens and anterior tissue during eye development in vertebrates. The forkhead
genes thus far that contribute to anterior segment phenotypes belong to the C, E and P
subclasses.
FoxC1 mutations have been linked to Axenfeld-Reiger anomaly (ARA) as wells
as Type I Iridogoniodysgenesis (OMIM:601090). Recently, Tamini et al, demonstrated
that the FoxC1 also directs lens development during zebrafish development (Tamimi et
al., 2006). Mice heterozygous for either Foxc1 or Foxc2 exhibit anterior segment
phenotypes (Smith et al., 2000) and a linkage mapping study showed mutations in
bovine Foxc2 segregated with ocular dysgenesis phenotypes (Abbasi et al., 2006). To
date, however, no human mutation in FoxC2 has been linked to ocular disease to date
despite screening of 32 patients with ARA (Smith et al., 2000).
Allelic variants of human FOXE3 have been linked to congenital eye
malformation syndromes in humans reviewed in (Medina-Martinez et al., 2005; MedinaMartinez and Jamrich, 2007). A homozygous null mutation causes congenital aphakia in
humans (Valleix et al., 2006), and a single nucleotide insertion resulting in a frameshift
70
mutation in the single coding exon of FOXE3 causes anterior segment mesenchymal
dysgenesis (Semina et al., 2001). The mouse dysgenetic lens (Dyl) phenotype was shown
to be caused by two mutations in the DNA binding domain of Foxe3, thus confirming its
role in proper lens development (Blixt et al., 2000; Brownell et al., 2000). The ability of
FoxE family members to contribute to proper lens development is a highly conserved
mechanism. Morpholino knockdown of the Danio rerio foxe3 orthologue results in
morphants with multilayered lens epithelial cells as well as a significant lens fiber cell
dysmorphogenic phenotype (Shi et al., 2006).
To date, no ocular phenotype has been identified in association with mutations in
genes of the FoxP subclass, but a recent report describes the expression of two X. laevis
FoxP orthologues in developing lens tissue. FoxP1 shows lens specific expression in
addition to faint expression in the neural retina (see below) (Pohl et al., 2005).
3.3.2 Forkhead transcription factors in the developing neural retina
Expression of FoxD1 in X. laevis eye fields begins during early tailbud stages and
continues in the temporal region of the retina in subsequent stages (Mariani and Harland,
1998). The foxd1 orthologue in zebrafish exhibits a comparable expression pattern
(Odenthal and Nusslein-Volhard, 1998). In mammals, Foxd1 is also expressed in the
temporal region of the optic cup and retina (Hatini et al., 1994).
Mammalian Foxd1 and Foxg1 are expressed in retinal precursor cells in a distinct,
complementary pattern at the optic vesicle stage of development (Hatini et al., 1994).
Mammalian Foxg1 is expressed in the anterior neural plate and the nasal retina as
development proceeds (Huh et al., 1999). Expression is also evident in the lens and optic
nerve (Pratt et al., 2004). It has been suggested that the reciprocal nature of the Foxd1
71
and Foxg1 expression patterns is related to their function in determining retinal cell fate
(see below). The expression of the FoxG1 homologue in X. laevis is observed in the
telencephalic region of the developing forebrain (Bourguignon et al., 1998). However,
retinal expression of FoxG1 is not apparent in Xenopus embryos (Bourguignon et al.,
1998; HEM and HME, unpublished data).
Three members of the FoxP subclass in Xenopus laevis exhibit eye expression.
FoxP1, FoxP2 and FoxP4 are expressed in X. laevis beginning at mid-gastrula stage.
FoxP2 appears restricted to the dorsal-most cells within the retinal anlage at this stage; it
persists in the eye throughout tailbud stages, and becomes highly specific to the ciliary
marginal zone (CMZ) after differentiation of the neural retina (Schon et al., 2006b).
FoxP4 is expressed in the retinal anlage at neural tube stages and persists in the neural
retina through tailbud and tadpole stages (Schon et al., 2006). Danio rerio foxp2 resides
in the inner plexiform layer after differentiation of neural retinal subtypes (Bonkowsky
and Chien, 2005), demonstrating that FoxP2 expression patterns are not conserved among
vertebrate species. Foxp1 and Foxp2 in mice have been described in developing CNS
structures (Ferland et al., 2003; Lu et al., 2002; Shu et al., 2001; Takahashi et al., 2003;
Tamura et al., 2003), gut (Pohl et al., 2005; Shu et al., 2001), and lung tissues (Lu et al.,
2002; Shu et al., 2001) but neither has been specifically investigated in retinal tissue of
mammals. Noteworthy, however, is the high conservation of the FoxP family members
across species (86% identical between mouse and human orthologues), and their common
expression patterns in other reported tissues such as the cerebellum.
72
A single citation regarding X. laevis FoxK1, describes its expression in the eye
primordia, where its expression is seen in a dorsal to ventral gradient (Pohl and Knochel,
2004). Whether FoxK1 is maintained in the CMZ after differentiation was not reported.
In zebrafish, foxl1 is expressed in the neural retina (Nakada et al., 2006) after
neurulation (33hpf), albeit weakly. Expression patterns of the Foxl1 homologue in
mammals does not reveal any similarities (Fukuda et al., 2003; Kaestner et al., 1997). In
the case of FoxL proteins, there is little conservation beyond the forkhead domain,
requiring more investigation into whether Danio rerio foxl1 and mouse Foxl1 are
functionally equivalent (Nakada et al., 2006). Even so, the knockdown phenotype
(discussed below) reveals a certain involvement for foxl1 in zebrafish eye development.
A recent paper by Pohl et al. describes the initiation of FoxO3 expression in the
Xenopus eye at stage 26, a time in development before retinogenesis occurs (Pohl et al.,
2004). FoxO3 expression is specific to the neural retina in X. laevis, as the lens is devoid
of FoxO3 expression (Pohl et al., 2004). Expression of the mouse homologue of Foxo3
has not been described in retinal tissue. However, four FoxO family members exist in
mammals (Foxo1,3,4 and 6), and Foxo1 is expressed in the photoreceptor layer of P0
mice (Gray et al., 2004). Whether X. laevis FoxO3 and mouse Foxo1 proteins serve
homologous functions in the eye still remains to be investigated.
3.3.3 Forkhead transcription factors in differentiated retinal cell types
Only two forkhead gene products are known to be expressed in the fully mature
neural retina but not during developmental stages. A single citation for Foxs1 describes
its expression in the outer nuclear layer of the adult mammalian retina, as well as in a
subset of ganglion cells at P14 by analysis of a β-galactosidase reporter gene knock-in
73
mouse at the Foxs1 locus (Heglind et al., 2005). However, eye expression has not been
described for either endogenous mouse Foxs1 (Kaestner et al., 1993) or the X. laevis
orthologue, FoxD2 (Pohl and Knochel, 2002).
The Foxf1 β-galactosidase reporter mouse also demonstrates staining in a subset
of the cells within the outer nuclear layer of the retina (Kalinichenko et al., 2003). The X.
laevis homologue of the predominantly mesodermally expressed Foxf1 has not been
detected in the eye, although two reports describe complete expression patterns (Koster et
al., 1999; Tseng et al., 2004). A single EST described as moderately similar to Foxf1
was isolated from a Danio rerio retinal library (Dr.91954), although the expression
pattern has not been reported. Species-specific differences may account for the
discrepancies between these reports.
3.3.4 Forkhead transcription factors in retinal progenitors
Many vertebrate species have a pool of retinal progenitor cells that remain active
as slowly dividing precursor cells for all retinal cell types, even after complete
differentiation of the retinal cell layers. In zebrafish and X. laevis, these cells reside in a
compartment that is termed the ciliary marginal zone (CMZ) (Johns, 1977; Straznicky
and Gaze, 1971). A spatial gradient along the peripheral to central axis of the CMZ
further defines the level of stem cell potential: peripheral cells remain undifferentiated
and are slowly dividing, followed by a zone of proliferating neuroblast cells, a region of
actively differentiating precursors, and, finally, differentiated, post-mitotic neurons. In
addition, cells residing in the inner nuclear layer of fish retina normally contribute to the
rod photoreceptor lineage (Johns, 1982; Julian et al., 1998). In retinal injury models,
these cells have clearly been shown to exhibit multipotency, defining them as a second
74
population of stem cells within the teleost retina (Del Rio-Tsonis and Tsonis, 2003;
Hitchcock and Raymond, 1992; Otteson et al., 2001; Otteson and Hitchcock, 2003). In
rodents, potential retinal stem cells reside within the pigmented ciliary epithelium
(Ahmad et al., 2000; Perron and Harris, 2000; Reh and Fischer, 2001; Tribioli et al.,
2002; Tropepe et al., 2000). The presence of a given factor in the retinal stem cell
population underscores its importance in the developmental processes of cell-cycle
control and differentiation. The gene expression profiles of the subdivisions of the CMZ
in Xenopus correspond to sequential expression of transcription factors during retinoblast
development during embryogenesis (Dorsky et al., 1995; Perron et al., 1998). Thus,
knowledge of the spatial expression patterns in lower vertebrates contributes to the
understanding of embryonic retinoblast differentiation in mammals. For this reason, we
highlight forkhead transcription factors that are expressed in the retinal progenitor
population.
The FoxN subclass demonstrates distinct patterns of eye specificity. Mouse
Foxn2 is first detected at E10.5 in the optic cup (Tribioli et al., 2002). Expression is less
robust in the eye field at 11.5, and is restricted to the nasal ventral region within the eye.
FoxN2 is first detected in the early eye field of X. laevis embryos and persists through
tailbud stages in the differentiating retina (Kelly et al., 2007; Schuff et al., 2006). In the
mature retina, FoxN2 is expressed throughout the CMZ (Figure 1).
FoxN3 is robustly expressed in the prospective eye fields at the time of
neurulation in X. laevis embryos. Expression becomes limited to cells of the inner
nuclear layer of the central retina at stage 38. Later, minimal expression can be seen in a
subset of cells of both the ganglion cell layer and the inner nuclear layer. (Figure 1).
75
Expression is also visible in the lens at stages 38 and 41, appearing in the region of
proliferating lens cells. A mouse ortholog of Foxn3 in mouse has been
identified(Schorpp et al., 1997), although its expression pattern has yet to be published.
Foxn4 expression is predominantly eye-specific in mammals (Gouge et al., 2001).
Homologues in both X. laevis and Danio rerio also exhibit this highly specific pattern of
expression. (Danilova et al., 2004; Kelly et al., 2007; Schuff et al., 2006). In each
species, the expression of FoxN4 begins in the eye field before the time an eye is
morphologically evident. Expression is maintained in the neural retina during
development, but is downregulated as differentiation occurs. In the mouse, this is evident
as Foxn4 is expressed in cells of the ventricular zone, encompassing the entire central
retina at E12.5. By P2, Foxn4 is downregulated in both differentiated ganglion and
photoreceptor cells layers, but maintained in the ventricular zone. The expression of
FoxN4 is similar in zebrafish and X. laevis: high expression in the primordial eye fields
and downregulated in differentiated retina, although maintained in the CMZ of the
respective species (Danilova et al., 2004; Pohl et al., 2005; Schuff et al., 2006). We have
determined that FoxN4 is expressed in all but the most peripheral of the CMZ
compartments (Figure 1).
FoxM1 is expressed in eye fields of X. laevis embryos from neurulation through
tailbud stage, and also evident in CMZ (Pohl et al., 2005). Additionally, FoxM1 is not
expressed in the most peripheral region of the CMZ adjacent to the lens (Figure 1),
suggesting it may not be critical to maintain these cells in a true stem cell state. This is
consistent with reports from mammals where Foxm1 was initially published as a factor
present in all proliferating mammalian cells, although downregulated in terminally
76
differentiated cell (Korver et al., 1997a; Korver et al., 1997b; Yao et al., 1997; Ye et al.,
1997). There has been no report to date of the Foxm1 specifically within the developing
retina within mammals. A zebrafish entry for foxm1-like clone exists (BC054560),
however no expression pattern or functional data has been published.
3.3.5 Retinal forkhead function in eye development
The forkhead family of proteins is defined by the conserved forkhead DNA
binding domain. These factors function to regulate expression of target genes through
their ability to activate or repress transcription in a sequence-specific manner. The
following section summarizes the published data regarding transcriptional regulation by
retinal forkheads. Discussion of targets is limited to those deemed pertinent to eye
development, and does not represent a comprehensive view of known forkhead targets.
FoxD1 was shown to be a transcriptional repressor during neurulation events
(Mariani and Harland, 1998). Fusion proteins comprising FoxD1 and the engrailed
transcriptional repression domain exhibit the same biological activity as the wild type
protein, suggesting that FoxD1 normally functions as a transcriptional repressor. Foxd1
activity is not required for the proper specification of neural retinal precursors, since
Foxd1 deficient mice develop grossly normal eyes (Hatini et al., 1994; Herrara et al.,
2004). However, inactivation of the Foxd1 gene results in an abnormality in optic chiasm
formation, as well as anomalies in kidney, forebrain and adrenal gland development
(Hatini et al., 1994; Herrera et al., 2004). Mouse Foxd1 is expressed in retinal ganglion
cells (RGCs) of the temporal retina, and this expression is required for maintaining the
proper number of RGCs as well as for normal RGC axon projections into the optic tract.
Although this phenotype allows the retinae of Foxd1 deficient embryos to appear grossly
77
normal, the Foxd1 phenotype has a profound impact on the development of binocular
vision by affecting the development of the optic chiasm. In these animals, the Foxg1
expression domain is expanded ventral-temporally. Foxg1 and Foxd1 have been proposed
to regulate downstream targets within the optic cup to determine regional specificity of
axon projections (Yuasa et al., 1996).
FoxG1 had been hypothesized to have both activating and repressive functions
(Ahlgren et al., 2003; Bourguignon et al., 1998; Li et al., 1996; Yao et al., 2001).
Overexpression of fusion constructs of FoxG1 containing a strong activation domain or a
strong repressor domains fail to recapitulate the full phenotype of wild type FoxG1
(Bourguignon et al., 1998). The dual role in transcriptional activity is supported by
sequence data. FoxG1 in a conserved portion of the N-terminus of FoxG1 is highly
similar to the transactivation domain of FoxA2 (Pani et al., 1992) and the C-terminal
region was specifically shown to have repressive function in the chick homologue, qin (Li
et al., 1995). It remains unclear what allows FoxG1 to change between the activating and
repressing function.
Several data suggest retinal expressed targets of FoxG1. FoxG1 induces the
expression of Ephrin A family members (Takahashi et al., 2003). When FoxG1 is
misexpressed in the temporal retina, it represses EphA3, a tyrosine kinase receptor
expressed in the retina, as well as FoxD1. In X. laevis, p27XIC1, the homologue of the
cdk inhibitor p27Kip1, has been shown to be a direct downstream target of FoxG1
(Ahlgren et al., 2003; Hardcastle and Papalopulu, 2000). The activation of Ephrin family
members may indirectly result from regulation of p27XIC1 by FoxG1. The repression of
p27Kip1 removes a growth-inhibitory signal, allowing activation of Ephrin A family
78
members (Ahlgren et al., 2003; Pohl and Knochel, 2005). The repressive functions of
FoxG1 is mediated by interaction with Groucho and Hes transcriptional co-repressors in
the telencephalic progenitors (Marcal et al., 2005; Yao et al., 2001). It will be interesting
to test the possibility that this mechanism is conserved in retinal progenitors with retinal
expressed groucho family members, such as Grg4 or Grg5 (Zhu et al., 2002). The
previous data demonstrate that FoxG1 contributes to the control of neuroectoderm
proliferation in the developing retina.
FoxG1 loss of function studies demonstrate the importance of FoxG1 in proper eye
development. Foxg1 mutant mice never develop an optic stalk, the most ventral of
structures to be derived from the optic vesicle during eye development (Huh et al., 1999;
Xuan et al., 1995). Instead, optic stalk tissue is replaced by neural retina. In these
mutants, Pax6 and Pax2 expression, normally distributed along a dorsal -ventral gradient,
is perturbed, suggesting that Foxg1 acts to control a dorsal-ventral gene expression
program within the neuroepithelium during eye development. A specific loss of sonic
hedgehog (shh) expression in the ventral telencephalic region of neuroepithelium
precedes the phenotypic changes in the eye in the Foxg1-/- mice. This raises the
possibility that Foxg1 acts upstream of shh and that the Foxg1 phenotype is caused by
this local loss of shh signaling (Huh et al., 1999; Xuan et al., 1995). In addition to the
morphological phenotypes seen in Foxg1-/- eyes, retinal ganglion cell (RGC) axon
navigation is perturbed. RGC axons extend along the optic nerve to the ventral surface of
the hypothalamus. Most RGCs subsequently cross the midline at the optic chiasm and
join the contralateral optic tract, while those that do not reside in the ipsilateral optic
tract. In Foxg1-/- eyes, the proportion of RGCs that contribute to the ipsilateral optic
79
tract is significantly increased compared to wild type RGCs (Pratt et al., 2004). The
authors contend that loss of Foxg1 affects the ability of RGCs to respond to attractive or
repulsive cues at the optic chiasm to correctly navigate along optic tracts. These data
suggest a dual role for Foxg1 in eye development: initially, in the control of eye
morphogenesis by control of gene expression in the retinal epithelium and subsequently
in axon guidance of RGCs.
Recently, Xenopus knockdown of FoxN3 has demonstrated its critical role for
proper eye formation (Schuff et al., 2007). Embryos injected with antisense morpholino
oligonucleotides exhibited small eyes with normally laminated retinae. The mechanism
of FoxN3 action during eye development appears to be linked to apoptosis, and not cell
cycle progression, since injected embryos have higher rates of apoptosis while cell
proliferation rates are not affected.
FoxN3 interacts with Sin3 and RPD3, components of the histone deacetylase
complex in Xenopus (Schuff et al., 2007). Sin3 is thought to act as a co-repressor of
histone deacetylases (HDACs) (Laduron et al., 2004). Thus, FoxN3 may act as a
transcriptional repressor by recruiting HDACs to target DNA during eye development.
Inhibition of HDACs has been linked to an increase in apoptosis (Peart et al., 2003;
Sonnemann et al., 2006), revealing why knockdown of FoxN3 increases apoptosis. The
interactions of FoxN3 homologues with components of the HDAC is conserved across
species; it will be interesting to investigate whether the role of FoxN3 in eye development
is also conserved.
A second FoxN family member, Foxn4, is involved in retinal cell fate
specification. Knocking out the Foxn4 gene in mouse results in animals with a reduced
80
number of amacrine and horizontal cells in the differentiated retina (Li et al., 2004). This
data suggest that Foxn4 is necessary to enable RPCs to produce amacrine and horizontal
cells during retinogenesis, but is not critical for the production/specification of progenitor
cells. This is surprising in light of the robust expression of Foxn4 in early retinal
progenitors of all species in which Foxn4 has been identified. In this same report, Li et
al. show that overexpression of Foxn4 results in an abundance of amacrine cells with no
alteration in the horizontal cell subtype. These data indicate that Foxn4 is sufficient for
the commitment to the amacrine cell fate, but not for the production of horizontal cells.
Ocular retardation (or) mutant mice have retinal progenitors that divide at a
slower rate than that of wt mice. In or mice, Foxn4 expression is limited to a few cells of
the central retina at a time point when Foxn4 is normally expressed throughout the
proliferating ventricular zone (Gouge et al., 2001). This is intriguing data that suggests
that Chx10, the gene mutated in or mice, is upstream of FoxN4. Other downstream
targets of Foxn4 include Math3, NeuroD, and Prox1 (Li et al., 2004), although these have
not been demonstrated to be direct targets. Collectively, the data suggest that Foxn4 is
expressed in proliferating progenitor cells, and plays a role in the comittance to amacrine
and horizontal cell fates.
Expression of a zebrafish foxl1 fusion with the engrailed repression domain gives
a similar phenotype to that induced by overexpression of foxl1, suggesting that foxl1 may
normally function as a transcriptional repressor (Nakada et al., 2006). Foxl1 involvement
in eye development is demonstrated by the phenotype of foxl1 morphants which have
small eyes, as well as degenerated brains (Nakada et al., 2006). In addition to aberrations
in eye size, injected eyes do not display proper retinal layer formation. An increase in the
81
number of apoptotic cells was observed in the morpholino-injected embryos, contributing
to the microphthalmic phenotype. The layering defect suggests that foxl1 may play a role
in proper migration of differentiating retinal progenitor cells. Interestingly,
overexpression of the same gene results in a similar, yet more severe, phenotype; very
small or no eyes are observed in injected embryos (Nakada et al., 2006). Additionally
expression of pax6a is absent in tissue where foxl1 is overexpressed, placing foxl1
upstream of the zebrafish pax6 gene. The zebrafish pax6a gene is a downstream effector
of shh signaling. Using both in vivo and in vitro techniques, the authors showed that
foxl1 was able to repress transcription through the shh promoter. This implicates foxl1 as
a negative regulator of the shh pathway in zebrafish.
3.3.6 Retinal forkhead Retinal forkhead transcription factors hold enormous
promise to advance our knowledge of retinal progenitor cell biology
In summary, accumulating data demonstrates that a number of forkhead
transcription factors are present in developing retinal tissue in vertebrates. The expression
pattern data reveal forkhead transcription factors from multiple subfamilies are present in
different subtypes of retinal cells. In addition, the animal model phenotypes exhibit
various roles for forkhead proteins during eye development. They collectively
demonstrate that several forkheads are important determinants of retinal cell fate. The
data and results presented here, certainly only the tip of the forkhead iceberg, underscores
the enormous potential forkhead transcription factors holds for our understanding of the
development and biology of retinal progenitor cells.
82
Table 3.1 Forkhead gene products expressed in eye structures Retinal Forkheads
divided into subcategories on account of expression data: Retinal Progenitor Forkheads
(red box) and Neural Retinal Forkheads (blue box). Alternate names from published
literature are given as well as Accession number for four vertebrate species. NA, Not
applicable (accession not available in GenBank). Underline indicates UniProt accession
numbers.
83
Table 3.2 Retinal Forkhead Expression Retinal Forkheads divided into subcategories
on account of expression data: Retinal Progenitor Forkheads (red box) and Neural Retinal
Forkheads (blue box). Expression during eye development are indicated (+). Expression
is listed for prospective eye tissue (PE), zones of the CMZ, and layers of the
differentiated Neural Retina. Anterior Structure (AS) expression in the lens is also noted.
GCL, ganglion cell layer. IPL, inner plexiform layer. INL, inner nuclear layer. OPL,
outer plexiform layer. ONL, outer nuclear layer. PL, photoreceptor layer.
84
Figure 3.1 Expression of retinal forkhead genes in the ciliary marginal zone of the
maturing neural retina In situ hybridization on sections of paraffin-embedded X. laevis
embryos at stages 38, 41, and 45.
85
Chapter 4: FoxO3 Perturbs Vertebrate Eye Development by Affecting
Differentiation and Proliferation in Retinal Progenitor Cells
4.1 INTRODUCTION
Members of the forkhead family of transcription factors are emerging as potential
regulators of RPC development within the EFTF network due to their presence and
function in vertebrate eyes (reviewed in (Moose et al., 2009). Recent reports have
demonstrated that FoxM1, FoxN2, FoxN3, FoxN4, FoxO3, FoxP1 and FoxP2 are
expressed in RPCs in vertebrates (Pohl et al., 2005; Pohl et al., 2004; Schon et al., 2006a;
Schuff et al., 2006).
FoxO family members have been described in developing retinal precursors and
specific layers of the differentiated retina (Pohl et al., 2004). In Xenopus, FoxO3 is
expressed in the developing eye fields of embryos at stages of both specification and in
the differentiated retina (Pohl and Knochel, 2004). A single study in Drosophila showed
that FoxO overexpression results in the ablation of ommatidia (Junger et al., 2003). To
date, however, there have not been functional studies of FoxO3 in vertebrate retinal
development or RPC maintenance.
In systems other than the retina, FoxO3 functions as a potent inhibitor of the cell
cycle and pro-apoptotic gene product (reviewed in Calnan and Brunet, 2008). FoxO3 has
86
been shown to directly regulate p27Kip1(Chandramohan et al., 2004; Stahl et al., 2002) a
cdk inhibitor that has a well-established role in controlling cell-cycle exit of retinal
progenitors (Cunningham et al., 2002; Dyer and Cepko, 2001; Levine et al., 2000).
Coupled with the expression data, this places FoxO family members as potential
regulators of cell cycle progression in developing retinal precursors.
It is well documented that the PI3K-Akt pathway negatively regulates the
function of FoxO3 (Biggs et al., 1999; Brunet et al., 1999; Kops and Burgering, 1999;
Kops et al., 1999). The overexpression phenotype in Drosophila is exacerbated by a
genetic background with reduced PI3K activity (Junger et al., 2003). This indicates that
the AKT regulation of FoxO3 is an active mechanism controlling FoxO3 function in eye
precursor cells.
As the function of FoxO3 has not been demonstrated in the vertebrate eye, we
sought to investigate the role of FoxO3 during RPC development in Xenopus laevis.
Defining the role that FoxO3 plays during vertebrate retinal progenitor cell development
will further our understanding of basic RPC biology and be applicable to the
understanding of diseased states in the retina. In this report, we establish that FoxO3
functions in the normal differentiation of the RPC population of X. laevis. We show that
overexpression of FoxO3 results in an small eye phenotype caused by altered
differentiation of retinal precursors, decreased proliferation, and increased levels of
apoptosis. This data is the first to define a role for FoxO3 in vertebrate eye development.
87
4.2 MATERIALS AND METHODS
Animals
Male and Female frogs were purchased from NASCO (Fort Atkinson,
Wisconsin). Testes were isolated as previously described (Kroll and Amaya, 1996) and
stored at 16°C degrees in 1xMMR until fertilization.
Plasmids
An expressed sequence tag (EST) encoding FoxO3 was identified using a BLAST
search of Xenopus ESTs (Altschul et al., 1990). We obtained the FoxO3 EST (pCMVSport6/Xl048f24) from The National Institute of Basic Biology, Japan
(http://Xenopus.nibb.ac.jp). pBS/FoxO3 was produced by digesting pCMVSport6/Xl048f24 with XhoI/EcoRI, and inserting into the same sites of pBS (Stratagene).
To produce pCS2/xlFoxO3, the coding region of FoxO3 from pCMV-Sport6/Xl048f24
was digested using NotI, filled using Klenow, digested with XhoI and subcloned into the
pCS2 (REFERENCE) expression vector digested using Stu/XhoI. Mutation of the T30
residue in FoxO3 was performed using the Site-Directed Mutagenesis Kit (Invitrogen)
using the following oligonucleotide primers: forward,
GACCTCGGTCTTGCGCGTGGCCCCTGCAGAGACTAGACTC and
reverse,GAGTCTACTCTCTGCAGGGGCCACGCGCAAGACCGAGGTC using
pCS2/FoxO3 as a template. Nucleotides that result in amino acid change from threonine
to alanine at position 30 in xlFoxO3 are underlined.
RNA preparation
Capped RNAs for microinjection were prepared using mMessage mMachine kits
(Ambion, Austin, TX). RNAs were purified by gel filtration chromatography (RNA Mini
88
Columns; Roche Diagnostics, Indianapolis, IN) to remove unincorporated nucleotides
and cap analogue. FoxO3 and FoxO3T30A RNAs were prepared by linearizing plasmids
pCS2/FoxO3 or pCS2/FoxO3 T30A, with NotI and transcribing with Sp6 enzyme
(Ambion). LacZ RNA was produced by linearizing pING14/beta-galactosidase plasmid
(a gift from J. Yang) (Sharpe and Goldstone, 1997) and transcribed using Sp6 enzyme
(Ambion).
Injection of RNA and Acquisition of Xenopus Laevis embryos
Embryos used for injection experiments were obtained by in vitro fertilization
(Kroll and Amaya, 1996). FoxO3 or FoxO3 T30A was injected at 250 pg, 500 pg or 1 ng
levels. For phenotype analysis and gene expression analysis RNA encoding GFP (50 pg)
as a lineage tracer during early tailbud stages (20-24). Embryos with GFP expression in
one eye field at stage 26 were isolated for further analysis. Embryos were cultured in 0.1x
MBS at 16°C until desired developmental stage (Nieuwkoop and Faber, 1994). Embryos
were fixed in 4% MEMPFA for 1 hour at room temperature, dehydrated in 100% MeOH
for at least one hour, and paraffin embedded. For TUNEL assays, RNA encoding betagalactosidase (50pg) was co-injected as a lineage tracer. Embryos were cultured in
0.1xMBS at 16°C until desired developmental stage (Nieuwkoop and Faber, 1994).
Embryos were fixed in MEPFA containing 4% PFA for one half hour, washed 3 times 5
minutes in staining buffer: 1x PBS containing 0.01% sodium dexycholate, 0.02% NP-40,
and 2mM MgCl2. Embryos were LacZ stained in staining buffer containing 5mM
potassium ferricyanide, 5mM potassium ferrocynide and 1mg/ml X-gal until desired
staining intensity was present. Embryos with LacZ staining covering the developing
eyefield were selected for further analysis.
89
Immunohistochemistry
Consecutive transverse sections were taken of paraffin-embedded embryos at
8µm. The staining procedure for sections has been described previously using R.T.U
Vector staining kit (Vector Labs) (Pan et al., 2006b). Antibody concentrations used in
this study were as follows: mouse anti-rhodposin, 1:50 (RetP1; Biomeda, Foster City,
CA); mouse anti-islet1, 1:50 (39.4D5; Developmental Studies Hybridoma Bank [DSHB],
University of Iowa); rabbit anti-caspase-3, 1:250 (Sigma); rabbit anti-CRALBP (a gift of
Dr. J. Saari, University of Washington, Seattle, Washington), pHH3, 1:200, (Upstate).
In situ hybridization
Whole mount in situ hybridization was performed as follows: In situ hybridization
was performed on 8 µm retinal sections as previously described (Pan et al., 2006b;
Shimamura et al., 1994; Viczian et al., 2003). Probes were used at a concentration of 2.5
µg/µl diluted in hybridization buffer. The following table describes the production of
antisense probes used in this study.
90
Eye size quantification
10 embyros (stage 38) were collected in each of the following experimental
groups: uninjected control, 500 pg FoxO3/50 GFP, 500 pg FoxO3 T30A/50 GFP, 1 ng
FoxO3/50 pg GFP, and 1 ng FoxO3 T30A/GFP. The anterior to posterior length of both
eyes in each embryos were measured in triplicate using dorsal views obtained using Leica
dissecting microscope. Ratios of injected eyes size to uninjected eye size were obtained
by comparing the mean eye length of the injected eye to the mean length of the
uninjected eye for all embryos. The mean ratios of eye size from each group were
compared using Mann-Whitney t-tests in GraphPad Prism (GraphPad Software, Inc).
Quantification of RPC proliferation
Total PHH3+ cells were counted per eye and total retinal area was measured in
both eyes of each embryo. Ratio of the number of PHH3+ cells in the injected eye per
total area to the number PHH3+ cells in the uninjected eye per total area of the same
embryo determined the mitotic index.
TUNEL assays
Embryos with positive lacZ staining in one eyefield were selected and fixed for an
additional one half hour, and washed in the following EtOH washes for 5 minutes each:
100%, 90%, 70%, 50%, 30%. The following washes followed: 100% MethOH (10 min),
1x PBS (2x5 min), 1x PBS containing 0.2% Tween-20 (15 min), 1x PBS containing 0.5%
Tween-20 (30 min), 1x PBS (2x10 min). Embryos were incubated in 0.5x TdT buffer
(Invitrogen) diluted in 1x PBS for 1 hour. Tdt reaction was carried out at room
temperature overnight using 150 U/ml Tdt (Invitrogen), 0.5 µM Dig-UTP (Roche) in 0.5x
Tdt (Invitrogen) diluted in 1xPBS, followed by washes: 1xPBS containing 1mM EDTA
91
at 65° (2x30), at room temperature:1x PBS (4x15 min), 1x PBS containing 0.1% TritonX and 0.2% BSA (15 min). Blocking was performed using 1x PBS containing 0.1%
Triton-X and 0.2% BSA and 20% Lamb Serum (60 min). Addition of anti-DIG antibody
(Roche) at 1:2000 dilution in blocking serum was performed overnight at 4°C. Embryos
were washed in 1x PBS containing 0.1% Triton-X and 0.2% BSA (4x 30 min), followed
by 2x5 min in AP substrate buffer: 1M Tris-Hcl (pH 9.5), 1M MgCl2, 5M NaCl, 10%
Tween-20, 125 mM Levamisole (Vector). The color reaction was performed using 3.5
µl/ml BCIP, and 4.5 µl/ml NBT, stopped with a brief (<5 min) wash in 1x PBS with
100mM EDTA, and embryos were fixed over two night in Bouin’s fixative (70 ml water,
25 ml 37% formaldehyde and 5ml glacial acetic acid). Whole-mount pictures were
obtained using a dissecting microscope (Leica).
4.3 RESULTS
4.3.1 X. laevis FoxO3 protein is highly conserved and expressed in developing eye
tissue
We obtained an EST (XL048f24,GenBank accession:BJ039089), and confirmed
that the sequence aligns with the previously described FoxO3 homologue (Pohl et al.,
2004) using BLAST analysis (Altschul et al., 1990). Sequence analysis also demonstrated
that the XL048f24 EST contained the complete FoxO3 coding region. ClustalW
alignment was performed using MacVector software suite using the following protein
sequences from GenBank: X. laevis FoxO3, X. tropicalis , mouse, rat, and human.
xlFoxO3 and hFoxO3a are conserved at the level of amino acid with 70 % identity.
92
The characteristic functional domains of FoxO family members are present in
xlFoxO3 including the forkhead domain, a nuclear localization sequence and nuclear
export sequence (Figure 4.1, A). These domains are highly conserved between xlFoxO3
and hFoxO3a (Figure 4.1, A). The forkhead domain which comprises a 102 amino acid
(aa:138-240) domain near the N terminus of the protein and is 96% identical to human
sequence. This domain mediates direct DNA binding and is required for the ability of
FoxO gene products to transactivate gene targets (Hannenhalli and Kaestner, 2009). The
xlFoxO3 nuclear localization sequence partially overlaps the forkhead domain in the Nterminal portion of the protein. This nuclear localization sequence, present in all FoxO
family members, consists of three arginine residues and three lysine residues that are
separated by 19 amino acids (reviewed in (Huang and Tindall, 2007). This stretch of
xlFoxO3 sequence (aa: 232-256) is 100% conserved between X. laevis and human
sequences. Lastly, a nuclear export sequence is contained in the C-terminal region of
xlFoxO3 (aa: 351-363); this sequence is slightly less well conserved with human
sequence (77% identity).
In addition, post-translational modification sites that are involved in determining
subcellular localization of FoxO family members are conserved within xlFoxO3.
xlFoxO3 contains three highly conserved AKT recognition sequences (consensus
sequence RXRXXS/T where X is any residue. These sequences are located near the Nterminus, C-terminus and within the forkhead domain (Fig 1A, red bars). Sequence
alignment demonstrates that the N-terminal AKT site is perfectly conserved among
vertebrates (Figure 4.1,B). The phosphorylation state of these sites modulates the
subcellular localization of FoxO proteins, and are responsive to insulin signaling (Calnan
93
and Brunet, 2008). Phosphorylation of the N-terminal and forkhead domain AKT sites
are required for nuclear to cytoplasmic shuttling of FoxO proteins (Brownawell et al.,
2001). Phosphorylation at the threonine residue in the N-terminal PKB site: 1) affects the
NLS located in the forkhead domain when phosphorylated with the forkhead domain
AKT site (Brownawell et al., 2001), and 2) affects the interaction of the FoxO3 with 143-3 proteins that facilitate nuclear export (Brunet et al., 1999). Although mutations at the
other AKT sites affect DNA binding directly, mutations of the N-terminal AKT site has
not been shown to affect DNA binding. Mutation of the serine within the N-terminal
AKT site has been shown to inhibit nuclear export of FoxO3 by disrupting 14-3-3
binding (Brunet et al., 1999). We have mutated the conserved N-terminal PKB site of
xlFoxO3 at the threonine residue (T30) to alanine (Figure 4.1, A). We demonstrate that
xlFoxO3 T30A alters the sub-cellular localization of xlFoxO3, resulting in a more
nuclear localization compared to wt xlFoxO3 (Figure 4.1, C-D).
To examine the expression of xlFoxO3 in developing eye tissue during
development, we performed in situ hybridization on whole-mounted embryos and retinal
sections (Figure 4.1, E-H). xlFoxO3 expression is seen as early as stage 16 in the anterior
neural plate before the eye fields separate completely. Expression is apparent in the
developing eyes during early retinogenesis, and continues to be expressed in the
developing retina during tailbud stages (Figure 4.1, E-G, (Pohl et al., 2004)). In retinal
sections at stage 38, FoxO3 is expressed in the central region of the retina, and is
excluded from the photoreceptor layer and retinal margins (Figure 4.1, H). xlFoxO3 is
(Brunet et al., 1999) expressed in the eyes throughout neural and tailbud stages, making it
94
present in a spatial and temporal pattern to function during retinal progenitor cell
development.
4.3.2 Overexpression of xlFoxO3 results in a small eye phenotype
We utilized a gain-of-function approach to analyze the function of xlFoxO3 in
developing retinal progenitor cells. RNA encoding a wild type xlFoxO3 or xlFoxO3
T30A of was injected in one dorsal blastomere at thee four-cell stage; co-injection of
GFP as a lineage tracer allows sorting of embryos for properly targeted expression of
RNA (Figure 4.2). Overexpression of FoxO3 or FoxO3 T30A in the anterior neural
region of X. laevis embryos results in embryos with a visibly smaller eye, and delayed
retinal pigmentation (Figure 4.3 and 4.6, brackets in B-D). An otherwise normal body
morphology was observed in injected embryos. The small eye phenotype becomes
apparent at the earliest stages of retinal pigmentation (stage 30). Normal retinal
pigmentation occurs in a dorsal to ventral pattern in the eye field. Pigmentation of the
retina in injected embryos appears to be delayed in reaching the ventral side of the eye
field. Even in the cases when pigmentation in the affected eye is present in the correct
pattern, pigmentation is reduced in the control eye of the same embryo. FoxO3
overexpression results in a small eye phenotype, with delayed retinal pigmentation that is
evident at early tailbud stage in Xenopus laevis.
The small eye phenotype was noticeably more severe in embryos injected with
FoxO3 T30A compared to FoxO3 injected embryos (compare Figure 4.3, H and Figure
4.3, F). We quantified the increase in severity of the eye phenotype by analyzing eye size
ratio in embryos injected with FoxO3 or FoxO3 T30A. We compared the ratio of eye size
in 10 FoxO3 T30A injected embryos to 10 FoxO3 injected embryos at both 500pg and
95
1ng levels of RNA (Figure 4.5, A-C). The ratio of eye size ratio is defined as the mean
of three measurements along the rostral to caudal axis of the injected eye over to the same
measurement of the uninjected eye at stage 41 (Figure 4.5, D). We found that the
reduction in eye size in FoxO3 T30A injected embryos compared to the reduction in
FoxO3 injected embryos was statistically significant at both 500pg and 1ng RNA levels
(Figure 4.5, D) (two tailed Mann-Whitney test; 500pg FoxO3 vs. 500pg FoxO3 T30A,
p=0.0004; 1ng FoxO3 vs. 1ng FoxO3 T30A, p=0.0004).
Injection of FoxO3T30A RNA also results in a higher frequency of embryos with
small eyes compared to embryos injected with wtFoxO3. 72% of embryos injected with
250pg wtFoxO3 exhibited a small eye phenotype while 81% of embryos injected with
250pg of FoxO3 (Figure 4.4). Similar increases in frequency were observed at 500pg and
1ng concentrations of injected RNA. There is a significant increase in the proportion of
embryos affected in the presence of FoxO3 T30A compared to FoxO3 (Fisher’s exact
test; 250 pg, p=0.0126; 500 pg, p=0.0004; 1 ng, p=<0.0001.). Embryos injected with
50pg GFP do not exhibit small eye phenotype (n=28) (data not shown). We have shown
that FoxO3 T30A RNA affects both the severity and the frequency of the small eye
phenotype in X. laevis.
We next investigated whether introduction of FoxO3 disturbs the laminar
structure of the retina. Histological analysis of retinae from FoxO3 or FoxO3 T30A
injected embryos during late retinohistogenesis (stage 32), and in the mature eye (stage
38) demonstrated the presence of intact retinal layering, although the overall eye is
smaller (Figure 4.6). At stage 28, the major phenotype evidence is the lack of
pigmentation in the RPE of retinas of injected embryos (Figure 4.6, B and D, brackets).
96
The optic nerves of the injected retinas (both wt and T30A) were frequently displaced to
the ventral side of the retina compared to the optic nerve of the uninjected retina of the
same embryo. Concordant with this, the retinal tissue ventral to the optic nerve was often
reduced or absent in retinae of embryos injected with FoxO3 or FoxO3 T30A. This data
collectively demonstrates that FoxO3 can perturb normal development of the retina in
vertebrates without significantly affecting lamination.
4.3.3 Differentiated cell types are produced in FoxO3 injected embryos
By stage 41, all major retinal cell types, 6 neuronal and 1 glial, are present in the
Xenopus retina. To investigate whether differentiated cell types are present in FoxO3
injected embryos, we performed immunohistochemistry using an antibody to rhodopsin
(RetP1) (Figure 4.7, A-D), a marker of rod photoreceptor cells, and Islet1 (Figure 4.7, EH), a marker of amacrine, ganglion and horizontal cells (Pan et al., 2006b). At stage 38,
photoreceptors, amacrine, ganglion and horizontal cells are present in FoxO3 and FoxO3
T30A injected embryos. RPCs of FoxO3 injected embryos produce differentiated rod
photoreceptor, amacrine, ganglion and horizontal cells.
4.3.4 Retinal Progenitor Cells are specified but exhibit altered differentiation in FoxO3
injected embryos
The retina develops as an extension of the anterior neural tube in vertebrates.
RPCs arise from neural cells that express genes such as Sox2, and Otx2, and N-Myc
(Adler and Canto-Soler, 2007). We analyzed the effects of FoxO3 on neural
specification in the eye fields by performing whole mount in situ hybridization for Sox2,
Otx2 and N-Myc. The expression levels of Sox2 and Otx2 appear similar in embryos
injected with FoxO3 or FoxO3 T30A RNA. However, the regions of expression appear
97
smaller on the injected side of the embryo (Figure 4.8, G-J,brackets). N-Myc expression
levels appear both decreased and in a smaller field (data not shown). These data indicate
neural precursors are specified, albeit in a smaller region, in FoxO3 injected embryos.
To investigate whether there is proper specification of RPCs in FoxO3 injected
retinae, we performed whole mount in situ hybridization using the RPC markers: Pax6
and Rx. There is a decrease in the region of Pax6 (Figure 4.8, A-D) and Rx (Figure 4.8,
E-F) expression in retinae of embryos injected with FoxO3 or FoxO3 T30A. This data
suggests that RPCs are properly specified in the presence of exogenous FoxO3.
Since RPCs were properly specified as indicated by Rx and Pax6 expression, we
asked whether the differentiation profile of RPCs is altered by FoxO3 overexpression.
We performed in situ hybridization on stage 26 embryos using the markers to indicate
various stages of RPC differentiation. Notch was used as a marker of proliferative
retinoblast cells (Figure 4.9, A-D), NeuroD as a marker of post-mitotic differentiating
retinal cells (Figure 4.9, E-H) and Otx5b as an early marker of photoreceptor cells
(Figure 4.9, I-L). In embryos injected with FoxO3 or FoxO3 T30A, there is a marked
decrease in expression of the differentiation markers Notch, NeuroD and Otx5b,
indicating that differentiation of RPCs is aberrant in the presence of FoxO3.
4.3.5 FoxO3 overexpression results in altered cell cycle in RPCs
Since the regions of expression for both neural precursor and retinal progenitor
markers were decreased, we hypothesized that FoxO3 was affecting proliferation in
RPCs. To investigate this possibility, we analyzed the expression of the cell cycle genes,
cyclin D1 and p27, both downstream targets of FoxO3 (Ho et al., 2008; Medema et al.,
2000). In embryos injected with FoxO3 or FoxO3 T30A, cyclin D1 expression is
98
decreased (Figure 4.10, E-F). Fewer percent embryos exhibited a decreased in cyclin D1
expression when injected with FoxOT30A as compared to injection of FoxO3 (FoxO3
T30A 20%, n=10, FoxO3 80%, n=10 vs.). p27 expression was increased in the majority
of embryos injected with FoxO3 (60%, n=6) (Figure 4.10, A-B). In contrast, embryos
injected with FoxO3 T30A resulted in a decrease in p27 expression (89%, n=9) (Figure
4.10, C-D). FoxO3 affects the expression of known cell cycle target genes normally
expressed by developing RPCs.
We next examined if FoxO3 reduces the amount of proliferating cells in the eye
fields by immunostaining embryos injected with FoxO3 or FoxO3 T30A with the mitotic
marker Phosphohistone H3 (PHH3). A decreased number of PHH3+ cells were observed
in the retinae of injected embryos when compared to control retinae (Figure 4.11, A-D).
The proliferation rate of FoxO3 T30A-injected embryos is decreased compared to
proliferation rate in FoxO3-injected embryos (Figure 4.11, E). FoxO3 functions to reduce
the rate of proliferation in RPCs, and control the expression of retinal cell cycle genes
during vertebrate eye development.
4.3.6 FoxO3 overexpression results in increased apoptosis
To test whether exogenous levels of FoxO3 results in changes in cell survival, we
investigated the levels of apoptotic cell death. We analyzed apoptosis in FoxO3 injected
embryos using TUNEL assays and immunohistochemistry using a caspase-3 antibody
(Figure 4.13). TUNEL assays were performed on stage 28 whole mounted embryos
injected with FoxO3 or FoxO3 T30A. FoxO3 or FoxO3 T30A results in an increase in
TUNEL staining on the injected side of the embryo compared to the uninjected side
(Figure 4.12, A-C). Apoptosis is not attributed to injection of RNA as no visible
99
apoptosis is apparent in embryos injected with lineage tracer alone (data not shown). In
addition, the number of caspase-3 positive cells are increased in the presence of FoxO3 or
FoxO3 T30A (data not shown). FoxO3 is sufficient to cause an increase in apoptotic cell
death in developing RPCs in Xenopus laevis.
4.4 DISCUSSION
4.4.1 A Model for FoxO3 function during RPC development
We have established that overexpression of FoxO3 in Xenopus laevis results in a
small eye phenotype that is both more severe and more frequent when a nonphosphorylatable FoxO3 is introduced. The phenotype is a result of aberrant
differentiation of RPCs as evidenced by decreased expression levels of differentiation
markers. The cell cycle is altered in the presence of exogenous levels of FoxO3 or
FoxO3 T30A RNA, including both expression of cell cycle genes and numbers of
proliferating cells. In addition, FoxO3 overexpression caused increased apoptotic cell
death in developing RPCs. Collectively, these alterations during RPC development
results in reduced eye size in the presence of exogenous FoxO3. A model describing the
changes in FoxO3 injected embryos is presented (Figure 4.13).
4.4.2 Production of differentiated cell types in the presence of exogenous levels of
FoxO3
The eventual production of differentiated cell types in the presence of FoxO3 is
remarkable in light of the marked changes in expression of differentiation genes in RPCs.
The loss-of-function phenotypes of Notch and NeuroD result in cell fate changes of
RPCs. Loss of Notch in mice results in a gain of ganglion and cone photoreceptors at the
100
expense of rod photoreceptors and bipolar cells (Ahmad/Dooley 1991, and
Austin/Feldman, 1995). Targeted mutation of NeuroD results in and increase in later
born cell types, Muller glia and bipolar cells (Yan and Wang, 1998). We have not
detected a cell fate change in our model. WOULDN”T IT BE NICE IF WE HAD???
However, our system represents only a partial loss-of-function of Notch, NeuroD; and
this knockdown is transient as it is dependent upon the presence of overexpressed
FoxO3. The partial expression that is maintained in the presence of FoxO3 (Wt or T30A)
must be sufficient to produce differentiated amacrine, horizontal, ganglion and rod
photoreceptors cell types. In addition, gradual RNA degradation of FoxO3 RNA over
time may release the RPCs from inhibition of differentiation, allowing them to eventually
produce the observed differentiated cell types.
A third possibility exists that may explain the surprising presence of regular
lamination and differentiated cell types in FoxO3-injected embryos. We have
demonstrated an increase in the levels of cell death in the presence of FoxO3. Apoptosis
resulting from FoxO3 overexpression reduces the number of retinal cells affected by
FoxO3 injected over time. Therefore it is plausible that these differentiated cells are
derived from wt cells, and contribute to the regular laminar structure seen in FoxO3
injected embryos.
4.4.3 Cell Cycle alterations in FoxO3-injected embryos
FoxO3 has been shown to be upstream of the cell cycle inhibitor, p27, and
demonstrated that FoxO3 directly transactivates the p27 promoter (Chandramohan et al.,
2004; Stahl et al., 2002). We demonstrate that in the presence of exogenous FoxO3
there is a increase in expression of p27 in the eyes of X. laevis; this data is consistent
101
with previous reports that FoxO3 function as a direct activator of p27 transcription (Lees
et al., 2008). Interestingly, FoxO3 T30A results in a decrease in p27 in our system. The
phosphorylation state of specific FoxO3 residues has been shown to affect DNA binding
and resulting transcriptional control (Reviewed in Ho et al., 2008). Although this has not
been shown for AKT phosphorylation status specifically, it is possible that the threonine
to alanine mutation in FoxO3 affects the ability of FoxO3 to transactivate its target gene
p27. Alternatively, the decrease in p27 expression in FoxO3 T30A-injected embryos
could be a secondary effect of other demonstrated gene expression changes such as Notch
or NeuroD. In addition, p27 in X. laevis appears to have roles distinct from the cell cycle
inhibition roles of its mammalian homologue (Dyer, 2001). P27 loss of function in
Xenopus retina results in cell fate changes to earlier born cell types at the expense of late
born cell types such as Muller glia (Dyer et al., 2003). It is possible that the change in
p27 expression in embryos injected with FoxO3 T30A is due to its’ role in differentiation
rather than cell cycle inhibition.
It is interesting that expression levels of the both the cell cycle inhibitor, p27, and
the cell cycle checkpoint protein, cyclin D1 are decreased in our overexpression model.
One possible explanation is that one or both of these genes are direct targets for FoxO3 in
RPCs. We analyzed the expression levels of these genes of RNA by whole mount in situ
hybridization. Although, levels of p27 protein do not fluctuate as cells progress through
mitosis, levels of cyclin D1 protein vary depending on the phase of the cell cycle. Our
current methods of detection have not analyzed p27 or cyclin D1 protein, and therefore
do not account for changes in protein levels. Lastly, the developing eye tissue is a
heterogeneous cell population containing cells at differing stages of both cell cycle and
102
differentiation. This has been confirmed by the differential expression of cell cycle
genes, such as p27 and p57 in the mouse (From Dyer et al., 2001), leaving the possibility
that the observed changes in gene expression are specific to a subset of these progenitors.
It is possible that the cells normally expressing cyclin D1 and p27 are discrete subpopulations and gene expression in each of these sub-populations is altered, resulting in
the changes seen in both cyclin D1 and p27.
4.4.4 FoxO proteins as regulators of diverse progenitor cell populations
FoxO family members have defined roles in other precursor cell populations such
as muscle progenitor cells, endothelial precursors and hematopoietic precursor cells
(Bakker et al., 2004; Kitamura et al., 2007; Mogi et al., 2008; Tothova and Gilliland,
2007). The current study provides the first evidence that FoxO3 is involved in the normal
differentiation of retinal progenitor cells. We find that FoxO3 negatively regulates the
expression of the retinogenesis genes, Notch, NeuroD and Otx5b. In endothelial
precursors, overexpressed Foxo3a restricts differentiation (Mogi et al., 2008) consistent
with our data. Kitamura et al showed that mouse Foxo1 works in the Notch signaling
pathway to regulate myoblast differentiation (Kitamura et al., 2007). The authors also
demonstrated an increase in Notch activation in FoxO3 loss-of-function experiments. It
would be interesting to investigate whether Notch function would be increased in the
RPCs in the presence of FoxO3 knockdown in X. laevis. The present study adds to the
literature that demonstrates that Foxo3 functions in multiple precursor populations to
regulate differentiation capacity.
103
4.4.5 Is regulation of RPCs a conserved function for FoxO3 in vertebrates?
We suggest a model in which FoxO3 functions to inhibit differentiation. FoxO3
overexpression does not completely eliminate the ability of RPCs to differentiate, as
neuronal sub-types are observed in the mature retina. Fully differentiated amacrine,
horizontal, ganglion and photoreceptors cell types are present in retinae of FoxO3injected embryos. However, the differentiation program is altered, evidenced by the
reduction in expression of the RPC differentiation markers, Notch, NeuroD, and Otx5b.
Coupled with the changes in proliferation and increased apoptosis, FoxO3 overexpression
results in retinae with less tissue.
The data presented here represent the first to indicate FoxO3 functions in
vertebrates during normal eye development. Overexpression in the prospective anterior
neural ectoderm in X. laevis results in embryos with small eyes. Previous work has
shown that the FoxO3 orthologue in drosophila functions in the development of the
specialized cells of the compound eye (Junger et al., 2003). They showed that driving
expression in the developing eye resulted in complete ommatidia ablation. These
analogous results in Drosophila suggest that this may be a conserved function for FoxO
family members. It will be interesting to investigate this possibility in other vertebrate
animals.
104
Figure 4.1. Xenopus laevis FoxO3 protein is highly conserved and is expressed
throughout eye development (A) FoxO3 Protein Schematic. Conserved domains are
shown with percent identity between X. laevis and human amino acid sequences:
forkhead domain; NLS, nuclear localization signal, NES, nuclear export signal. AKT
phosphorylation sites are shown as striped bars. (B) Sequence alignment of N-terminal
AKT site among vertebrate species. Red bar indicates AKT site residues. Asterisk
indicates threonine (T30) residue specifically phosphorylated by AKT, and same residue
mutated to alanine in FoxO3 T30A. (C-D) Immunoflourescence showing FoxO3 or
FoxO3 T30A (green) localization in transfected HER10 cells. (E-H) Whole mount in situ
hybridization using antisense xlFoxO3 probe (E-H); Stage indicated in the bottom right
of each panel. E, anterior view, dorsal up. F-G, lateral view. In situ hybridization on
sectioned retinal material , stage 38 (H). Transverse section shown. L-lens, G-ganglion
cell layer, P-photoreceptor layer
105
Figure 4.2 xlFoxO3 is targeted to the developing eye fields of X. laevis embryos
Anterior views of stage 26 embryos injected with 250 pg FoxO3 (A,C,E) or FoxO3 T30A
(B,D,F).
106
Figure 4.3 Overexpression of xlFoxO3 results in small eye phenotype Embryos at
stage 41 injected with 250 pg FoxO3 (A-B, E-F) or FoxO3 T30A (C-D, G-H).
107
Figure 4.4 T30A mutation in FoxO3 increases the frequency eye phenotype
Frequency of small eye phenotype is increased in T30A xlFoxO3-injected embryos.
Proportion affected embryos with small eye phenotype at each concentration are shown
(from 5 experiments). There is a significant increase in the proportion of embryos
affected in the presence of FoxO3 T30A compared to FoxO3 (Fisher’s exact test; 250 pg,
p=0.0126; 500 pg, p=0.0004; 1 ng, p=<0.0001.). Embryos injected with 50pg GFP do not
exhibit small eye phenotype (n=28) (data not shown). 500 pg, p=0.0004, 1ng, p=0.0004.).
108
Figure 4.5 T30A mutation in FoxO3 increases the severity of the small eye
phenotype (A-E) Eye phenotype is more severe in FoxO3 T30A injected embryos. (A)
Eye size is represented as an average of three measurements of the eye along the rostralcaudal axis using dorsal views of stage 38 embryos. (B-D) Representative embryos from
eye size analysis using 500 pg RNA. Uninjected embryos were used as a control group,
and compared to embryos injected with FoxO3 or FoxO3 T30A. Analysis was performed
with 500 pg or 1 ng RNA. There is a significant difference between FoxO3 and FoxO3
T30A xlFoxO3-injected embryos at both 500 pg and 1 ng concentration (Mann-Whitney
test; 500 pg, p=0.0004, 1 ng, p=0.0004.).
109
Figure 4.6 Retinal lamination intact in the presence of xlFoxO3 Transverse retinal
sections of embryos injected with 250 pg FoxO3 (A-B, E-F) or FoxO3 T30A (C-D, G-H)
stained with Hematoxylin and Eosin. Stage 32, A-D; Stage 38, E-H. Abbreviations: L,
lens; G, ganglion cell layer; I, inner nuclear layer; P, photoreceptor layer. Brackets
indicate areas with decreased levels of pigmentation.
110
Figure 4.7 Normal differentiated cell types are present in xlFoxO3 injected embryos
Immunohistochemistry for Islet1 (A-D) or Rhodopsin (E-H) in embryos injected with
250 pg FoxO3 (A-B, E-F) or FoxO3 T30A (C-D, G-H). Transverse sections on stage 38
embryos.
111
Figure 4.8 RPCs are specified in the presence of xlFoxO3 overexpression Whole
mount in situ hybridization for Sox2 (A-B), Otx2 (C-D), N-Myc (E-H), Pax6 (I-L) and
Rx (M-O) in embryos injected with 250pg FoxO3 (A,C,E-F,I-J, and M) or FoxO3
T30A (B,D,G-H,K-L, and O). Stages: Sox2, 24; Otx2, 24; N-Myc, 28; Pax6, 28; Rx, 24.
Brackets indicate eye field. Lateral views of the head are shown.
112
Figure 4.9 Differentiation is altered by xlFoxO3 overexpression Whole mount in situ
hybridization using antisense probes for Notch (A-D), NeuroD (E-H), and Otx5b (I-L)
on embryos injected with 250 pg FoxO3 (A-B, E-F, I-J) or FoxO3 T30A (C-D, G-H, KL). Lateral views of the head of stage 26 are shown.
113
Figure 4.10 Overexpression of xlFoxO3 results in changes in cell cycle gene
expression Whole mount in situ hybridization using antisense probes for p27 (A-D),
Cyclin D1 (E-H), on embryos injected with 250 pg FoxO3 (A-B, E-F) or FoxO3 T30A
(C-D, G-H). Lateral views of the head of stage 28 are shown.
114
Figure 4.11 Overexpression of xlFoxO3 results in decreased rate of cell cycle
Phosphohistone H3 (PHH3) immunostaining (red) on paraffin-embedded retinal sections
from embryos injected with 250 pg FoxO3 or FoxO3 T30A RNA (A-D). Arrowheads,
PHH3+ cells. (E) Quantification of proliferation rates in FoxO3-injected embryos. Total
PHH3+ cells were counted per eye in both eyes of each embryo. Ratio of the number of
PHH3+ cells in the injected eye to the number PHH3+ cells in the uninjected eye of the
same embryo determined the mitotic index (paired t-test; p=.0087).
115
Figure 4.12 xlFoxO3 overexpression results in increased levels of apoptosis Stage 24
embryos stained with LacZ as a lineage tracer (turquoise), and TUNEL to detect
apoptosis (dark purple) (A-B). LacZ alone does not result in TUNEL positive cells (data
not shown).
116
Figure 4.13 Proposed model of FoxO3 function in retinal progenitor cells Exogenous
FoxO3 results in increased amount of FoxO3 available to regulate transcriptional targets
(Panel A), although some FoxO3 protein is exported from the nucleus due to Akt
phosphorylation. In the presence of T30A FoxO3, Akt is unable to phosphorylate FoxO3
T30A, and exogenous protein remains in the nucleus (Panel B). Overexpression of
xlFoxO3 results in a small eye phenotype in X. laevis. FoxO3 affects expression of cell
cycle and differentiation genes, thereby regulating the development of RPCs. The
alterations in gene expression affects the capacity of the RPC population to differentiate
and proliferate, and increases the amount of cell death in RPCs. Combined these effects
reduce the RPCs available to contribute to the neural retina, thus resulting in a reduced
retinal size (Panel C).
117
Chapter 5: FoxM1 is Necessary for Normal Development and Proper Lamination of
the Neural Retina by Retinal Progenitor Cells
5.1 INTRODUCTION
The study of RPC regulation by cell cycle gene products is strengthening the
connection between proliferative signals and normal retinal development. Alterations in
cell cycle components has detrimental effects on several aspect of neural retinal
development, including proliferation, cell fate decisions and proper lamination (reviewed
in Dyer et al., 2004 and Baye et al., 2007). In this chapter, work is presented that
demonstrates a novel role in RPC development for the proliferation-associated forkhead
transcription factor, FoxM1.
FoxM1 was originally isolated in a screen for M-phase phosphorylated proteins
(Westendorf et al., 1994). FoxM1 was subsequently isolated by several groups, each of
which identified it as a forkhead-domain containing transcription factor: HFH-11 (Ye et
al., 1997), MPP2 (Luscher-Firzlaff et al., 1999; Westendorf et al., 1994), Trident (Korver
et al., 1997a; Korver et al., 1997b), and WIN (Yao et al., 1997). FoxM1 proteins have
been described in human, mouse and rat (Korver et al., 1997a; Korver et al., 1997b; Yao
et al., 1997; Ye et al., 1997), and recently in frog (Pohl et al., 2005). In each organism,
FoxM1 is expressed exclusively in proliferating cells.
118
The fact that FoxM1 is ubiquitously expressed in proliferating cells makes it
unique within the forkhead family, since most forkhead genes are tissue-specific or cell
type-specific. FoxM1 levels are down regulated in terminally differentiated or quiescent
cells (Yao et al., 1997). FoxM1 protein levels increase during S-phase and persist until
the end of mitosis (Korver et al., 1998). FoxM1 is a proliferation-associated transcription
factor (Wierstra and Alves, 2007a).
FoxM1-null mice die during embryogenesis, and have cardiac, hepatocyte and
lung phenotypes (Kim et al., 2005; Korver et al., 1998; Krupczak-Hollis et al., 2004).
Embryonic Foxm1-/- hearts, livers, and lungs have significantly reduced numbers of
cells. Normal cell orientation of hepatocytes and cardiomyocytes was also affected in
FoxM1-/- embryos (Korver et al., 1998). Even with the alterations that are seen in the
FoxM1-/- hearts, lungs and livers, these organs initially develop normally suggesting that
FoxM1 is dispensable for early stage initiation and development of these organs (Korver
et al., 1998; Krupczak-Hollis et al., 2004; Laoukili et al., 2007; Ramakrishna et al.,
2007).
FoxM1 loss-of-function results in cellular polyploidy, and failure to progress
through M-phase (Korver et al., 1998). In the hepatocytes and cardiomyoctyes of FoxM1
null mice, enlarged nuclei with up to 50-fold more DNA content are present (Korver et
al., 1998). The polyploid defects are specific to tissues that are prone to polyploidy in
adult mice. It is thought that the majority of tissues do not display a phenotype associated
with loss-of-function (Wonsey and Follettie, 2005), since terminally differentiated, nonproliferating cells express very low levels of FoxM1 (Yao et al., 1997). Liver-specific
knockout (Wang et al., 2002), endothelial-specific knockout and hepatocyte-specific
119
knockout (Krupczak-Hollis et al., 2004) mice also exhibit defects in M-phase progression
(Wang et al., 2002). Collectively, the loss-of-function studies demonstrate that FoxM1
play an essential role in the G2/M transition.
FoxM1 depleted cells display multiple mitotic defects. These defects include
defective spindle fiber checkpoint, defects in mitotic spindle formation, chromosomal
misalignment during metaphase, mis-segregation of chromosomes during mitosis, and
aberrant cytokinesis (Laoukili et al., 2005; Wang et al., 2005; Wierstra and Alves, 2007a)
(Wonsey and Follettie, 2005). As well, shRNA knockdown in a breast cancer cell line
results in mitotic catastrophe (Wonsey and Follettie, 2005). Therefore, in addition to its
role in M-phase stimulation, FoxM1 also plays a critical role in the execution of mitosis.
Overexpression studies have confirmed the role of FoxM1 in cell cycle
progression. FoxM1 overexpression in regenerating hepatocytes results in acceleration of
DNA replication and mitosis (Ye et al., 1999). As well, FoxM1 is one of the most highly
up-regulated genes in human solid tumors (Wierstra and Alves, 2007a). Interestingly,
FoxM1 overexpression in non-regenerating livers was not sufficient to produce
hepatocyte proliferation, due to the retention of the overexpressed FoxM1 in the
cytoplasm (Ye et al., 1999). The authors showed that additional mitogenic signals
induced by hepatectamy are required for the nuclear translocation and subsequent early
progression into S-phase (Ye et al., 1999). Nonetheless, these studies still suggest that in
cells stimulated to proliferate, FoxM1 is essential for the execution of mitosis.
The FoxM1 proteins that have been reported in vertebrates represent three splice
variants of FoxM1 (A-C) (Wierstra and Alves, 2006). FoxM1b and FoxM1c act as
transactivators (Wierstra and Alves, 2007a) and several groups have shown that both
120
FoxM1b and FoxM1c bind to a consensus DNA sequence with the same affinity (Korver
et al., 1997a; Wierstra and Alves, 2006; Ye et al., 1997). In contrast, FoxM1a binds to
DNA, but does not have transactivation activity (Ye et al., 1997). This difference is due
to an inclusion of an additional exon (A2) that eliminates the transcriptional activation
ability (Ye et al., 1997). The studies of transcriptional targets of FoxM1 have come from
studies using FoxM1b and FoxM1c (Wierstra and Alves, 2007a).
Over 50 FoxM1 genes targets have been described, including 22 direct targets
(Wierstra and Alves, 2007a).The majority of these genes, whether direct or indirect
targets, are specifically involved in G1/S transition, S-phase, G2/M transition, and
mitosis. These targets comprise several cyclins (A2, B1, B2, D1, D2, E, and F), cdk1, and
the cell cycle inhibitors, p21 and p27 (Wierstra and Alves, 2007a). FoxM1 expression
down-regulates p21 and p27 via direct activation of the c-myc gene (Wierstra and Alves,
2007b). The downstream targets of FoxM1 clearly illustrate that FoxM1 functions in cell
cycle progression.
Surprisingly, very little is known about the function of FoxM1 in lower
vertebrates. A chicken FoxM1 gene (NP_001012973), and a zebrafish foxM1-like gene
(NP_957391) are annotated, but neither expression nor functional data have been
described. In Xenopus, the expression pattern of FoxM1 has been described (Pohl et al.,
2005), and functional analysis in the early embryo suggests FoxM1 links proliferation
and differentiation in neuronal precursors (Ueno et al., 2008). Expression is consistent
with other homologues of FoxM1; it is present in tissues undergoing active proliferation,
including the developing eye fields (Pohl et al., 2005). Functionally, xlFoxM1 is not
required for the specification of the anterior neural plate, but is essential for the
121
proliferation and subsequent differentiation of neural precursors in that region (Ueno et
al., 2008).
We are just beginning to understand the role of FoxM1 in the developing CNS. In
concert with the data regarding the neural plate precursors in Xenopus, Schuller and
colleagues investigated the function of FoxM1 during the development of cerebellar
granule neuron precursors (CGNPs)(Schuller et al., 2007). By studying loss-of-function
mutations of FoxM1, they were the first to define a CNS precursor population that
expressed FoxM1, the cerebellar anlagen. They also demonstrated that in CGNPs FoxM1
is upstream of G2/M genes, cyclin B1 and Cdc25b, and that FoxM1-deficient CGNPs
show spindle fiber defects and centrosome amplification. Their data established that in
CGNPs of the CNS, FoxM1 functions at the G2/M transition as well as in the execution
of mitosis, consistent with data from other organ systems (Wierstra and Alves, 2007a).
Due to the limited knowledge about FoxM1 function in both lower vertebrates
and in the developing CNS, we initiated studies to understand the effects of FoxM1
knockdown in the developing eye in Xenopus laevis. Based on the current literature, we
hypothesized that FoxM1 loss-of-function would result in reduced eye size and defects in
cell cycle progression in RPCs. We have identified a FoxM1c isoform in Xenopus laevis,
and tested this hypothesis by using translation-blocking morpholino oligonucleotide
(MO) to knockdown the expression of FoxM1 in the anterior neural region of Xenopus
embryos. We have shown that FoxM1 knockdown in the anterior neural region of X.
laevis results in embryos with small eyes that exhibit lamination defects. These studies
represent the first characterization of FoxM1 loss-of-function in lower vertebrates, and
the first indication that FoxM1 function is essential for proper retinal development.
122
5.2 MATERIAL AND METHODS
Animals
Male and female Xenopus laevis frogs were purchased from NASCO (where).
Sub-dermal injections of 150cc human chorionic gonadotropin were used to induce egglaying from female frogs. Testes were isolated as previously described (Kroll and
Amaya, 1996), and stored at 4°C in 1xMBS until use for in vitro fertilization.
Plasmids
An expressed sequence tag (EST) encoding FoxM1 was identified using a
BLAST search of Xenopus ESTs (Altschul et al., 1990). We obtained the FoxM1 EST
(pCMV-Sport6/XL184i11) from The National Institute of Basic Biology, Japan
(http://Xenopus.nibb.ac.jp). pBS/FoxM1 was produced by digesting pCMVSport6/XL184i11 with XhoI/EcoRI, and inserting into the same sites of pBS.
Sequence analysis
All sequence analysis and phylogenic analysis was performed using MacVector
Software (MacVector Software Inc.).
Morpholino design and injection in Xenopus laevis embryos
Translation blocking morpholino oligonucleotides were designed by GeneTools
(www.Gene-Tools.com). The control MO sequence was designed against the same
upstream region of FoxM1, but with 6 mis-matches compared to the target sequence.
Both the control and antisense FoxM1 MOs were lissamine labeled to allow visualization
of MO containing tissues in the living embryo.
Morpholino injection and acquisition of embryos
123
MOs (control or antisense) were injected into one dorsal blastomere of four-cell
stage fertilized X. laevis embryos; 0.1mM or 0.2mM concentrations were used. Embryos
with correctly targeted expression in the anterior neural region and developing eyes were
selected using fluorescent microscopy with a txRed filter (Leica) to detect the lissamine
label. Injected embryos were sorted at early neural stages (20-24), and cultured to stage
41 at 16°C in 0.1x MBS until the appropriate stage (Nieuwkoop and Faber, 1994).
MEMPFA containing 4% PFA was used to fix embryos for 1 hour at room temperature,
dehydrated in 100% MeOH for at least one hour, and paraffin embedded.
Hematoxlin/Eosin staining
H/E was performed on 8µm consecutive transverse sections taken with a Leica RM 2255
microtome as previously described (Pan et al., 2006).
Immunohistochemistry
8µm consecutive transverse sections of paraffin-embedded embryos were
prepared. The staining procedure for sections has been described previously (Pan et al.,
2006b). Antibody concentrations used in this study were as follows: mouse antirhodopsin (RetP1; Biomeda, Foster City, CA), 1:50; mouse anti-islet1, 1:50 (39.4D5;
Developmental Studies Hybridoma Bank [DSHB], University of Iowa), Cralbp, 1:1000 (a
gift from Dr. Jack Saari, University of Washington, Department of Ophthalmology); antiPhosphohistone H3, 1:200 (Upstate).
In situ hybridization
In situ hybridization was performed on whole embryo using anti-sense riboprobes
labeled with digoxigenin as previously described (Pan et al., 2006b). Section in situ
hybridization was performed using 8µm consecutive transverse sections taken with Leica
124
RM 2255 microtome. The protocol has been previously described (Shimamura et al.,
1994; Viczian et al., 2006) (Pan et al., 2006b). Probes were used at 2.5 µg/µl
concentration. The following table describes the production of antisense probes used in
this study.
Statistical analysis
All statistical analysis was performed using GraphPad Prism (GraphPad Software, Inc).
5.3 RESULTS
5.3.1 Identification of a FoxM1 Isoform in X. laevis
We identified an EST clone (Xl184i11) that aligned with mouse FoxM1 using
BLAST analysis (Altschul
et
al.,
1990). The clone contains a full-length X. laevis
FoxM1 homologue (subsequently referred to as xlFoxM1) (data not shown).
Phylogenetic analysis (MacVector Software) revealed that the protein encoded by
xlFoxM1 aligned with FoxM1 homologues from zebrafish, chicken, rat, mouse, and
human (Figure 5.1).
125
At
the
level
of
protein
sequence,
xlFoxM1
exhibits
41%
identity
with
hFoxM1c,
a
rather
low
overall
percentage.
The
presence
of
two
stretches
of
inserted
sequence
(AA
31‐43;
and
AA
66‐91)
in
the
xlFoxM1
protein
contributes
to
the
disparity
in
identity.
Notably,
sequence
comparison
between
xlFoxM1
and
hFoxM1
within
three
major
functional
domains
reveal
much
higher
identity:
the
N‐terminal
repression
domain
(88%),
the
forkhead
domain
(87%),
and
the
c‐terminal
transactivation
domain
(66%)
(Figure
5.3,
A).
The
presence
and
high
conservation
within
these
domains
suggests
they
are
functionally
important
in
the
X.
laevis
homologue.
To determine which of the three FoxM1 isoforms is represented by xlFoxM1, I
performed ClustalW multiple sequence alignment using translated sequence from
xlFoxM1, and protein sequence from the three human FoxM1 isoforms: Isoform A
(NM_021953), Isoform B (NM_202002), and Isoform C (NM_202003)) (Figure 5.2, B).
Class A FoxM1 isoforms are characterized by the presence of an additional exon (A2) at
the C-terminal end of the forkhead domain (Yet et al., 1997); xlFoxM1 does
not
contain
sequence
to
encode
exon
A2.
Additionally,
xlFoxM1
contains
a
partial
exon
A1,
which
is
absent
in
class
B
transcripts
(Ye
et
a.,
1997).
Thus,
xlFoxM1
represents
a
class
C
transcript.
5.3.2 A xlFoxM1 homologue is expressed in RPCs during eye development
Although the published expression pattern suggests that FoxM1 is expressed in
the developing eye (Pohl et al., 2005), the expression pattern in RPCs has not been
rigorously described. We analyzed the expression of FoxM1 in the developing eye fields
and maturing retinas by whole mount in situ hybridization and in situ hybridization on
126
paraffin-embedded sections (Figure
5.3).
FoxM1
is
first
expressed
in
the
eye
fields
and
the
neural
tube
of
early
neural
stage
embryos
(Figure
5.3,
A).
FoxM1
is
strongly
expressed
in
the
developing
eyes
and
neural
tube
during
early
tailbud
stages
(Figure
5.3,
C).
As
development
proceeds,
the
expression
of
FoxM1
begins
to
be
down‐
regulated
in
many
regions
of
the
developing
CNS
(Figure
5.3,
D),
but
remains
expressed
in
the
midbrain,
hindbrain
and
the
developing
ear.
In
situ
hybridization
on
paraffin‐embedded
sections
reveals
a
more
detailed
expression
pattern
of
FoxM1
within
the
developing
retina.
FoxM1
is
originally
expressed
throughout
the
entire
neuroepithelium
(Figure
5.3,
E).
The
expression
of
FoxM1
is
first
down‐regulated
in
the
central
retina,
and
is
progressively
limited
to
the
ciliary
marginal
zone
(CMZ)
of
the
X.
laevis
retina.
At
this
time‐point,
the
CMZ
can
be
divided
into
zones
on
the
basis
of
maturity
of
RPCs.
The
most
stem
cell‐like
RPCs
are
present
in
Zone
1,
which
is
adjacent
to
the
lens.
Zone
1
RPCs
are
a
very
slowly
dividing
population
of
cells.
RPCs
of
Zone
2
and
3
are
actively
proliferating
and
becoming
specified
to
particular
retinal
cell
fates.
In
zone
4,
RPCs
are
exiting
the
cell
cycle
as
they
differentiate
into
specified
cell
types.
In
the
mature
retina,
FoxM1
appears
limited
to
zones
2
and
3
of
the
CMZ
(Figure
5.3,
H‐I).
The
expression
pattern
in
developing
X.
laevis
embryos
and
retina
is
consistent
with
regions
of
known
FoxM1
expression
in
actively
proliferating
cells
(Wierstra and Alves, 2007a).
5.3.3
FoxM1
knockdown
in
X.
laevis
results
in
embryos
with
small
eyes
To
examine
the
functional
requirement
during
retinal
development,
I
performed
loss‐of‐function
studies
of
FoxM1
in
X.
laevis
embryos.
I
acquired
a
127
translation
blocking
morpholino
oligonucleotide
against
FoxM1
(GeneTools).
The
MO
was
designed
against
a
region
directly
upstream
of
the
translation
start
of
FoxM1
(Figure
5.4,
A).
The
control
MO
(COMO)
was
designed
against
the
same
sequence,
but
containing
6
mis‐match
nucleotides.
Injection
of
FoxM1
MO
and
FoxM1
COMO
into
one
dorsal
blastomere
of
fertilized
embryos
at
the
four‐cell
stage.
Injecting
in
this
manner
results
in
morpholino
targeted
to
the
anterior
neural
region
from
which
the
eye
structures
are
derived;
the
FoxM1
COMO
and
MO
were
synthesized
with
a
lissamine
tag
to
facilitate
sorting
of
embryos
with
properly
targeted
expression
(Figure
5.4,
C‐H).
To
test
that
FoxM1
MO
acts
specifically
through
the
FoxM1
target
sequence,
we
performed
the
following
in
vivo
control:
We
cloned
the
FoxM1
5’
UTR
containing
the
FoxM1
MO
target
upstream
of
a
GFP
expression
cassette
(Figure
5.5
A‐B).
We
co‐injected
the
FoxM1
MO
control
construct
DNA
with
FoxM1
MO
or
COMO
(Figure
5.5,
C‐D
and
G‐H).
We
also
co‐injected
FoxM1
MO
or
COMO
with
an
untargeted
DNA
construct
containing
the
GFP
expression
cassette
(Figure
5.5,
A‐B
and
E‐F).
FoxM1
MO
injected
reduced
the
expression
of
GFP
when
injected
with
the
targeted
construct;
GFP
was
visible
in
embryos
injected
with
untargeted
DNA
construct.
In
addition,
GFP
expression
was
visible
in
embryos
co‐injected
with
FoxM1
COMO
and
either
the
targeted
or
untargeted
DNA
construct.
This
demonstrates
that
FoxM1
MO
acts
specifically
through
the
target
sequence,
and
that
FoxM1
COMO
is
does
not
act
to
knockdown
translation
through
this
sequence.
128
Injection
of
0.1mM
FoxM1
MO
results
in
reduced
eye
size
in
X.
laevis
(Figure
5.5,
upper
panel).
The
FoxM1
MO
phenotype
is
highly
specific,
as
control
MO‐
injected
embryos
do
not
exhibit
an
eye
phenotype.
90.4%
(n=186)
of
embryos
injected
with
0.1mM
FoxM1
MO
demonstrated
a
small
eye
phenotype
compared
to
3%
(n=46)
in
FoxM1
COMO‐injected
embryos(Figure
5.6,
A).
The
severity
of
the
phenotype
is
concentration
dependent;
increasing
the
concentration
of
FoxM1
MO
to
0.2mM
results
in
obliteration
of
eye
tissue
by
tailbud
stage
(Figure
5.5,
bottom
panel).
At
0.2mM
concentration,
96.7%
(n=30)
of
embryos
display
a
small
eye
phenotype
compared
to
control
MO‐injected
embryos
(0%,
N=16)
(Figure
5.6,
B).
Embryos
with
residual
eye
tissue
typically
demonstrate
a
colobomatous
phenotype
(Figure
5.5,
bottom
panel
ASMO
injected).
These
data
demonstrate
that
FoxM1
is
necessary
for
normal
eye
development
in
X.
laevis.
5.3.4
xlFoxM1
is
necessary
of
proper
retinal
lamination
I
next
sought
to
determine
if
FoxM1
knockdown
results
in
changes
in
the
laminar
structure
of
the
retina.
Histological
analysis
reveals
that
FoxM1
knockdown
results
in
defects
in
the
normal
retinal
patterning
in
X.
laevis
embryos
(Figure
5.7).
I
examined
H/E
stained
embryos
at
a
stage
28,
a
time‐point
during
retinal
histogenesis,
and
in
mature
retinas
of
stage
41
FoxM1
MO‐injected
embryos.
In
the
neural
epithelium
at
stage
28,
cells
are
typically
oriented
radially
away
from
the
lens.
In
the
FoxM1
MO‐injected
(0.1mM)
embryos,
nuclei
appear
oriented
in
multiple
directions
(Figure
5.7,
D‐
D’).
These
changes
in
orientation
are
present
in
129
the
central
retina,
and
several
distinct
regions
of
laminar
disruption
are
often
discernable.
In
mature
retinas,
FoxM1
MO
results
in
variable
lamination
defects.
The
integrity
of
the
nuclear
layers
is
often
disrupted
by
misplaced
photoreceptors
identifiable
by
the
morphology
of
outer
segments.
These
aberrantly
placed
photoreceptors
are
typically
juxtaposed
to
additional
layers
that
resemble
outer
plexiform
and
inner
nuclear
layers
(Figure
5.7
H‐H’).
These
areas
of
irregular
lamination
are
most
commonly
found
in,
but
not
limited
to,
the
central
retina.
They
are
observed
as
deviations
perpendicularly
spanning
the
normal
layers
or
as
circular
rosettes.
In
addition,
some
retinas
displayed
complete
loss
of
patterning
in
regions
of
the
retina
(Figure
5.7,
J‐J’).
In
embryos
injected
with
0.2mM
FoxM1
MO,
retinas
demonstrate
complete
loss
of
normal
lamination,
although
stratified
cell
types
are
occasionally
apparent
(Figure
5.7,
N‐N’).
It
is
clear
that
loss
of
FoxM1
in
the
RPCs
results
in
abnormal
eye
development,
and
deviations
from
ordinary
retinal
patterning.
5.3.5
Differentiated
cell
types
are
associated
with
areas
of
normal
retina
lamination
and
regions
of
aberrant
stratification
I
ruled
out
that
aberrant
differentiation
was
responsible
for
the
defects
in
retinal
lamination
by
performing
whole
mount
in
situ
hybridization
at
stage
28.
I
tested
whether
there
was
a
change
in
gene
expression
for
the
markers
of
retinal
progenitor
cells,
Rx
and
Pax6,
and
markers
of
RPC
differentiation
by
in
situ
130
hybridization
(Figure
5.7).
FoxM1
knockdown
does
not
result
in
a
change
in
expression
of
these
markers
at
stage
28.
To
confirm
that
differentiated
cells
types
are
generated
in
FoxM1
MO‐
injected
embryos,
and
to
examine
which
cell
types
are
present
in
the
lamination
defects,
I
performed
immunohistochemistry
on
paraffin‐embedded
sections.
Rod
photoreceptors
are
present
in
retinas
of
FoxM1
MO‐injected
embryos,
as
evidenced
by
RetP1
staining.
Rod
photoreceptors
are
also
associated
with
the
areas
of
irregular
lamination
(Figure
5.9,
D‐D’).
In
addition,
the
Muller
glia
cell
marker,
CRALBP,
reveals
that
Muller
glia
are
evident
in
areas
with
normal
retinal
patterning,
and
are
also
present
in
areas
of
irregular
layering.
Muller
glia
endfeet
are
ordinarily
positioned
beyond
the
ganglion
cell
layer
and
at
the
apical
surface
of
the
neural
retina
adjacent
of
the
RPE,
spanning
the
entire
retinal
layers.
It
appears
that
the
Muller
glia
associated
with
regions
of
aberrant
layering
also
span
multiple
layers.
Both
photoreceptors
and
Muller
glia
are
produced
in
FoxM1
MO‐injected
embryos.
In
addition,
both
these
cell
types
are
observed
within
the
areas
of
defective
lamination.
5.4 DISCUSSION AND FUTURE DIRECTIONS
Discussion
I have described the initial characterization of a FoxM1c homologue in X. laevis.
FoxM1 represents as additional factor with the potential to regulate the proliferation of
RPCs during development. I demonstrated that the phenotype associated with FoxM1
knockdown in the anterior neural plate of X. laevis is a slight reduction in eye size with
131
associated lamination defects. We have also shown FoxM1 is not essential for the
specification of RPCs or the generation of differentiated cell types. From the current
studies, I conclude that FoxM1 is necessary for the normal development of the neural
retina, affecting the ability of RPCs to properly laminate.
As mentioned, FoxM1 demonstrates a clear retinal phenotype in X. laevis which
results in embryos with smaller eyes and lamination defects. To date, no retinal
phenotype in mice has been elucidated, although cardiac, hepatic and lung phenotypes
have all been described in FoxM1-/- mice (Korver
et
al.,
1998).
In
light
of
the
current
studies,
it
is
surprising
that
no
overt
eye
phenotype
in
mice
is
noticeable
with
complete
loss
of
function.
It
is
possible
that
the
FoxM1
‐/‐
retinas
have
lamination
defects
that
have
not
been
detected.
It
would
be
interesting
to
directly
ask
this
question
in
FoxM1‐/‐
mice.
Alternatively,
species
differences
in
function
could
result
in
the
increased
severity
of
FoxM1
knockdown
in
X.
laevis
compared
to
FoxM1
loss
of
function
in
mice.
It
is
interesting
to
note
that
one
of
the
main
characteristics
of
FoxM1‐/‐
loss
of
function
in
cardiac
and
hepatic
tissue
is
disrupted
cell
orientation
(Korver
et
al.,
1998).
Perhaps
the
disrupted
lamination
in
FoxM1
morphant
embryos
is
analogous
to
the
disrupted
orientation
of
cardiomyocytes
and
hepatocytes
in
mice.
In
retinas
of
FoxM1
knockdown
embryos,
differentiated
cell
type
are
present
in
regions
of
normal
and
abnormal
lamination.
Since
differentiated
cell
types
are
present
in
areas
of
disorganized
cells,
this
suggests
that
the
lamination
defects
are
not
a
result
of
loss
of
retinal
cell
identity.
This
notion
is
corroborated
by
evidence
132
that
retinal
stem
cell
and
early
retinal
neural
differentiation
markers
are
not
affected
in
embryos
injected
with
FoxM1
MO.
Consistent
with
this,
Ueno
and
colleagues
recently
showed
that
FoxM1
is
not
necessary
for
the
specification
of
neural
precursors
(Ueno
et
al.,
2008).
Future Directions
Although I have demonstrated a role for FoxM1 in eye development, the current
studies warrant further investigation into the mechanism of FoxM1 action in developing
RPCs. Future experiments should be focused on detailing the changes in cell cycle and
understanding the lamination defects.
An attractive hypothesis regarding the lamination defects observed is that they are
a result of inappropriate interkinetic nuclear migration and/or inappropriate nuclear
translocation. The nuclei of RPCs migrate from the basal surface near the lens tissue to
the apical surface nearest the RPE in phase with the cell cycle. S-phase occurs at the
basal surface and M-phase at the apical surface (see Figure 1.5). Additionally,
neurogenic cells leave the cell cycle and their nuclei migrate basally to take their position
in the laminar architecture of the retina (nuclear translocation). These post-mitotic cells
actively modify their apical and basal processes as they reach their targets and commence
synaptogenesis (Baye and Link, 2007). If RPCs do not properly progress through Mphase, their nuclei may not migrate away from the apical surface, causing defects in
lamination. Indeed, several studies demonstrate that alterations in cell cycle genes in
neuroepithelia result in slowed interkinetic nuclear migration, and lamination defects in
the eye and brain (reviewed in Baye and Link, 2007). This represents a possible scenario
of FoxM1 morphant RPCs.
133
Furthermore, it has been shown that FoxM1 is directly upstream of at least 6
genes involved in cell adhesion properties of the extracellular matrix (ECM) (Wierstra
and Alves, 2007). These include pro-collagen, E-cadherin, and laminins 2 and 4
(reviewed in Wierstra and Alves, 2007). It is also known that affecting the function of
ECM genes can negatively affect lamination in the retina. For example, disruption of
laminin 1 in the basement membranes of retina cells results in disorganized ganglion cell
layering (Halfter et al. 2001; Semina et al., 2006). Aberrations in the ECM have been
shown to effect the normal migratory properties of RPCs, and FoxM1 downstream targets
include several ECM proteins. Therefore, it is possible that the lamination defects in
FoxM1 knockdown embryos might be a result of deregulation of FoxM1 ECM target
genes. In the future, it will be interesting to investigate whether the ECM in the retina is
altered by FoxM1 knockdown.
Recently, it has been shown that BMP signaling induces the expression of FoxM1
in neural precursors, as well as FoxM1 targets, cdc25b and cyclin B3 (Ueno et al., 2008).
The authors defined that FoxM1 induction is necessary for proliferation and
differentiation, but not specification of neural precursors in early embryos (Ueno et al.,
2008), and that MO knockdown of cdc25b had strikingly similar effects. The authors
went on to show that cdc25b RNA was sufficient to rescue the FoxM1 phenotypes during
early neurogenesis. It would be valuable to test whether cdc25b RNA is sufficient to
rescue the FoxM1 knockdown phenotype in RPCs.
Lastly, FoxM1 gain-of-function studies in X. laevis should be pursued to further
understand the role of FoxM1 in RPCs. By injecting RNA
encoding
a
FoxM1
in
one
dorsal
blastomere
at
the
four‐cell
stage,
targeted
overexpression
could
further
134
investigate
the
role
of
FoxM1
in
X.
laevis.
A
repertoire
of
histological
and
immunochemical
techniques
similar
to
those
to
investigate
the
result
of
FoxM1
knockdown
would
be
necessary.
Based
on
the
literature,
we
hypothesize
that
FoxM1
gain‐of‐function
would
result
in
the
a
mis‐regulation
of
cell
cycle
in
RPCs,
possibly
leading
to
a
tumorogenic
phenotype
(Wang
et
al.,
2008;
Gemenetzidis
et
al.,
2009;
Brezillon
et
al.,
2007)
Together
the
delineated
and
proposed
studies
would
demonstrate
FoxM1
as
an
additional
forkhead
transcription
factor
that
has
an
essential
role
in
normal
eye
development.
Importantly,
FoxM1
represents
a
factor
that
may
play
a
critical
role
in
the
proliferation
of
RPCs.
Understanding
the
factors
that
govern
proliferation
in
RPCs
is
essential
for
understanding
the
link
between
proliferation
and
cell
fate
decisions
in
the
retina.
135
Figure 5.1 Xl184i11 encodes a FoxM1 protein. Multiple sequence alignment of
FoxM1 proteins from various species Mouse Foxi2 was used as an out-group.
Accession numbers are as follows: Human FoxM1, NM_202002; Mouse FoxM1,
AAH06788; Rat FoxM1, NP_113821; zebrafish FoxM1, NP_957391; Chicken FoxM1,
NP_001012973; mouse Foxi2, AAY79175.
136
Figure 5.2 xlFoxM1 (Xl184i11) is homologous to hFoxM1 isoform C (A) Shematif
representation of FoxM1 protein with characteristic conserved domains. NRD, Nterminal repression domain; TAD, transactivation domain. (B) Multiple sequence
alignment of xlFoxM1 (Xl184i11) with three human splice variants of FoxM1: Isoform A
(NM_021953), Isoform B (NM_202002), and Isoform C (NM_202003). Domains and
alternative exons are shaded as indicated. Since xlFoxM1 does not contain Exon A2 (a
transcript-specific), and possesses a partial Exon A1 (excluded in class B transcripts), it is
homologous to the hFoxM1 c isoform.
137
Figure 5.3 FoxM1 is expressed in the developing eye fields and in proliferative
zones of the maturing Xenopus laevis retina (A-D) Whole mount in situ
hybridization using an antisense probe against xlFoxM1. (A-B) anterior view. (C-D)
lateral view of head. (E-I) In situ hybridization on paraffin-embedded, transversesectioned material using antisense probe against xlFoxM1. Embryonic stages are
indicated in bottom left of the panels. Brackets in A, eye fields; brackets in G-I,
CMZ. L, lens; NR, neural retina; G, ganglion cell layer; I, inner nuclear layer; P,
photoreceptor layer.
138
Figure 5.4 FoxM1 morpholino can be properly targeted to anterior neural region of
Xenopus embryos (A) Design and sequence of FoxM1 morpholino oligonucleotides.
(B-I) FoxM1 COMO and MO are properly targeted to the anterior neural region
containing the developing eye. Anterior view of stage 26 embryos are shown.
139
Figure 5.5 FoxM1 morpholino specifically knocks down FoxM1 target. (A-B)
Untargeted construct (A) and construct containing FoxM1 MO target upstream of GFP
CDS used to test FoxM1 MO and FoxM1 COMO specificity. Red thick line, FoxM1 MO
target. Thin black line, FoxM1 5’UTR. (C-J) FoxM1 MO specifically knocks down
expression of a GFP construct containing the MO target, but FoxM1 COMO does not.
(C-F) Lissamine fluorescence of FoxM1 COMO injected embryos (C-D) or FoxM1 MO
injected embryos (E-F). (G-J) GFP fluorescence of FoxM1 COMO (G, I) or FoxM1 MO
injected embryos (H-J) in the presence of untargeted (G-H) or targeted FoxM1 MO
control (I-J) construct.
140
Figure 5.5
141
Figure 5.6 FoxM1 knockdown causes small eye phenotype in X. laevis Embryos
were injected with 0.1mM control or antisense morpholino (top panel) or 0.2mM
control or antisense morpholino (bottom panel) into one dorsal blastomere of
fertilized embryos. Injected embryos were sorted using fluorescence microscopy at
early tailbud stages for embryos with correctly targeted morpholino. Lateral views of
uninjected or injected side of the embryo are shown, as well as a dorsal view of the
same embryo.
142
143
Figure 5.6 FoxM1 morphants with eye phenotype (A) 0.1mM FoxM1 MO. (B) 0.2mM
FoxM1 MO. P-values from Fisher’s Exact test (0.1mM; <0.0001; 0.2mM,<0.0001).
Figure 5.8 FoxM1 knockdown results in small eyes with lamination defects
Transverse retinal sections of embryos injected with 0.1mM FoxM1 (A-J) or 0.2mM
FoxM1 (K-N) stained with Hematoxylin and Eosin. Stage 32, A-D; Stage 41, E-N.
144
Figure 5.9 Differentiated cell types and abnormal retinal patterning are present in FoxM1
morphants Transverse retinal sections of embryos injected with 0.1mM FoxM1 COMO or MO.
Immunohistochemistry for Islet1 (A-D) or Rhodopsin (E-H). Stage 38 retinas are shown.
-
145
Chapter 6: The significance of exploring Rx regulation and Forkhead transcription
factors function in RPCs
6.1 Summary of significant findings
Presented in this dissertation work (Chapter 2) is the characterization of
transcriptional regulation of the prominent transcription factor, Rx. We identify several
functional regions of the Rx promoter (UCE, FBE, CRE and DAE). Our data indicate that
these regions functionally interact to maintain proper Rx expression in the developing
and mature eye. Furthermore, we have identified specific TF sites that function in this
process: OTX, SOX and POU sites within UCE, and the FBE in the proximal promoter.
Since Rx is one of the first EFTFs to be expressed in the anterior neural plate as RPCs are
specified, this work gives insight into the understanding of the EFTF regulatory network,
and to the understanding of eye field specification.
Due to the elucidation of the FBE within the Rx2A promoter, the remainder of the
work is focused on the function of forkhead transcription factors during eye development.
In chapter 3, a discussion of the present literature regarding the expression and function
of forkhead proteins in the developing eye is presented. Chapter 4 and 5 include studies
identifying novel roles for FoxO3 and FoxM1 vertebrate eye development.
146
FoxO3 and FoxM1 have opposing roles on the cell cycle; FoxO3 antagonizes cell
cycle progression while FoxM1 promotes it. Therefore we have taken opposite
approaches to studying their function in RPCs. Overexpression of FoxO3 results in
embryos with small eyes resultant from decreased proliferation, induction of apoptosis,
and RPC gene expression changes. The changes in gene expression suggest that FoxO3
does not inhibit the specification of RPCs, but functions to delay the differentiation of
RPCs. The data supports our original hypothesis regarding FoxO3 as cell cycle
antagonist, with the ability to alter the differentiation capacity of RPCs. In contrast,
knockdown of a known cell cycle progression gene, FoxM1, also results in decreased eye
size in X. laevis embryos. However, the small eyes associated with FoxM1 loss-offunction have lamination defects. Our data suggest that FoxM1 is not necessary for the
specification of RPCs, but required for their ability to produce normal retinal architecture.
6.1 Exploring the relationship between Rx2A, FoxO3, FoxM1 in RPCs
In each chapter, I have discussed future directions regarding the individual
projects. Here, I would like to discuss a few experiments that would address the
relationship between Rx gene products, and the forkhead transcription factors, FoxO3
and FoxM1 in RPCs.
There is a growing body of literature that suggests a cooperative activity between
FoxO3 and FoxM1 in various tissues (Delpuech et al., 2007; Madureira et al., 2006;
McGovern et al., 2009; Zou et al., 2008). First, FoxO3 may act directly upstream of
FoxM1 to modulate proliferation rates, and FoxO3 induced apoptosis can be recovered in
the presence of exogenous FoxM1 (McGovern et al., 2009). These studies suggest a
antagonistic regulation of proliferation and apoptosis by FoxO3 and FoxM1. It would be
147
interesting to see if this mechanism is relevant to RPC biology. This could be tested in
the models of FoxO3 overexpression presented in this work. To investigate whether
FoxO3 acts upstream of FoxM1 in the regulation of RPC proliferation, we could analyze
gene expression of FoxM1 in FoxO3-injected retinas. In addition, we could test whether
addition of FoxM1 RNA is sufficient to rescue the apoptosis observed in FoxO3-injected
embryos.
Secondly, it has been shown that FoxO3 and FoxM1 cooperate to regulate the
estrogen receptor alpha (ERalpha) promoter (Madureira et al., 2006). This cooperative
activity occurs through two closely juxtaposed FBE elements within the proximal
ERalpha promoter. Since both FoxO3 and FoxM1 represent cognate factors to regulate
Rx transcription through the FBE, they could act cooperatively to regulate Rx
transcription. Due to their known roles in the cell cycle, FoxO3 and FoxM1 may act in
opposition, repressing and activating Rx transcription, respectively. If this activity is
confirmed, this mechanism of transcriptional regulation would directly link two cell cycle
genes to the RPC gene, Rx.
6.3 Regulation of RPC development by Homeobox and Forkhead transcription factors
The regulatory network responsible for specification of RPC in vertebrates is
composed of several members of the homeobox gene family (Zuber et al., 2003). These
genes coordinate cell fate in the retina in cooperation with transcription factors of the
bHLH family (Vetter et al., 2001; Lupo et al., 2000). The data described in this work
implicate the forkhead family of transcription factors as a major additional family that
functions during RPC development. Specifically, FoxO3 and FoxM1 are two cell cycle
regulatory genes whose normal function must be maintained to ensure proper RPC
148
development. In addition, we have demonstrated the importance of a forkhead binding
element in the regulation of Rx transcription. Admittedly, it remains to be tested if Rx
regulation by forkhead transcription factors is biologically relevant. However, it is
interesting to speculate whether the role of forkhead transcription factors could be to act
as cell cycle agonist/antagonists upstream of homeobox and bHLH factors. This model
would provide a method for integrating proliferation cues and cell fate cues in
developing RPCs.
6.4 Closing remarks
In this dissertation, we present data to expand the knowledge of transcription
regulation in the RPCs, recognize the forkhead family of transcription factors as RPC
regulators and demonstrate novel roles for two cell cycle regulators in the development of
the neural retina. Thus, this work advances the knowledge of the intrinsic mechanisms
governing RPC development in vertebrates.
As we expand the body of knowledge regarding RPC regulation, it quickly
becomes apparent that the mechanisms governing this process are each integral to the
normal development of the neural retina. Alterations of these mechanisms are
detrimental to the structure and function of the retina, and this notion perpetually
motivates the pursuit of a complete understanding of RPC development. As well, we can
appreciate the complexities of eye development, which has led to the continued
fascination with this organ over centuries.
149
Appendix A
150
Figure
A.1.
Cyclin
D1
is
robustly
expressed
in
the
developing
X.
laevis
retina
(A-H) whole mount in situ hybridization using an antisense probe against cyclin D1. (AB) and (E), anterior view. (C), lateral view. (F-G), lateral view of head. (H-J) in situ
hybridization on paraffin-embedded, transverse-sectioned material using antisense probe
against xlFoxM1. Embryonic stages are indicated in bottom right of the panels. L, lens.
151
Figure
A.2.
N­myc
is
dynamically
expressed
in
the
developing
X.
laevis
eyes
and
retina
(A-G), whole mount in situ hybridization using an antisense probe against nmyc. A, anterior view. B-C, lateral view. D-G, lateral view of head. (H-I) In situ
hybridization on paraffin-embedded, transverse-sectioned material using antisense probe
against n-myc. Embryonic stages are indicated in bottom right of the panels. L, lens.
152
Figure
A.3.
p27
is
dynamically
expressed
in
the
maturing
X.
laevis
retina
(A-C)
In situ hybridization on paraffin-embedded, transverse-sectioned material using antisense
probe against p27. Embryonic stages are indicated in bottom right of the panels. L, lens.
153
Bibliography
Abbasi, A.R., Geriletoya, Ihara, N., Khalaj, M., Sugimoto, Y. and Kunieda, T. (2006) An
integrated radiation hybrid map of bovine chromosome 18 that refines a critical
region associated with multiple ocular defects in cattle. Anim Genet 37, 58-61.
Adler, R. and Canto-Soler, M.V. (2007) Molecular mechanisms of optic vesicle
development: complexities, ambiguities and controversies. Dev Biol 305, 1-13.
Ahlgren, S., Vogt, P. and Bronner-Fraser, M. (2003) Excess FoxG1 causes overgrowth of
the neural tube. J Neurobiol 57, 337-49.
Ahmad, I., Tang, L. and Pham, H. (2000) Identification of neural progenitors in the adult
mammalian eye. Biochem Biophys Res Commun 270, 517-21.
Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Basic local
alignment search tool. J Mol Biol 215, 403-10.
Ambrosetti, D.C., Basilico, C. and Dailey, L. (1997) Synergistic activation of the
fibroblast growth factor 4 enhancer by Sox2 and Oct-3 depends on protein-protein
interactions facilitated by a specific spatial arrangement of factor binding sites.
Mol Cell Biol 17, 6321-9.
Ambrosetti, D.C., Scholer, H.R., Dailey, L. and Basilico, C. (2000) Modulation of the
activity of multiple transcriptional activation domains by the DNA binding
domains mediates the synergistic action of Sox2 and Oct-3 on the fibroblast
growth factor-4 enhancer. J Biol Chem 275, 23387-97.
Ando, H., Kobayashi, M., Tsubokawa, T., Uyemura, K., Furuta, T. and Okamoto, H.
(2005) Lhx2 mediates the activity of Six3 in zebrafish forebrain growth. Dev Biol
287, 456-68.
Andreazzoli, M., Gestri, G., Angeloni, D., Menna, E. and Barsacchi, G. (1999) Role of
Xrx1 in Xenopus eye and anterior brain development. Development 126, 2451-60.
Andreazzoli, M., Gestri, G., Cremisi, F., Casarosa, S., Dawid, I.B. and Barsacchi, G.
(2003) Xrx1 controls proliferation and neurogenesis in Xenopus anterior neural
plate. Development 130, 5143-54.
Bailey, T.J., El-Hodiri, H., Zhang, L., Shah, R., Mathers, P.H. and Jamrich, M. (2004)
Regulation of vertebrate eye development by Rx genes. Int J Dev Biol 48, 761-70.
Bakker, W.J., Blazquez-Domingo, M., Kolbus, A., Besooyen, J., Steinlein, P., Beug, H.,
Coffer, P.J., Lowenberg, B., von Lindern, M. and van Dijk, T.B. (2004) FoxO3a
regulates erythroid differentiation and induces BTG1, an activator of protein
arginine methyl transferase 1. J Cell Biol 164, 175-84.
154
Biggs, W.H., 3rd, Meisenhelder, J., Hunter, T., Cavenee, W.K. and Arden, K.C. (1999)
Protein kinase B/Akt-mediated phosphorylation promotes nuclear exclusion of the
winged helix transcription factor FKHR1. Proc Natl Acad Sci U S A 96, 7421-6.
Blixt, A., Mahlapuu, M., Aitola, M., Pelto-Huikko, M., Enerback, S. and Carlsson, P.
(2000) A forkhead gene, FoxE3, is essential for lens epithelial proliferation and
closure of the lens vesicle. Genes Dev 14, 245-54.
Bonkowsky, J.L. and Chien, C.B. (2005) Molecular cloning and developmental
expression of foxP2 in zebrafish. Dev Dyn 234, 740-6.
Bourguignon, C., Li, J. and Papalopulu, N. (1998) XBF-1, a winged helix transcription
factor with dual activity, has a role in positioning neurogenesis in Xenopus
competent ectoderm. Development 125, 4889-900.
Bovolenta, P., Mallamaci, A., Briata, P., Corte, G. and Boncinelli, E. (1997) Implication
of OTX2 in pigment epithelium determination and neural retina differentiation. J
Neurosci 17, 4243-52.
Brown, N.L., Patel, S., Brzezinski, J. and Glaser, T. (2001) Math5 is required for retinal
ganglion cell and optic nerve formation. Development 128, 2497-508.
Brownawell, A.M., Kops, G.J., Macara, I.G. and Burgering, B.M. (2001) Inhibition of
nuclear import by protein kinase B (Akt) regulates the subcellular distribution and
activity of the forkhead transcription factor AFX. Mol Cell Biol 21, 3534-46.
Brownell, I., Dirksen, M. and Jamrich, M. (2000) Forkhead Foxe3 maps to the dysgenetic
lens locus and is critical in lens development and differentiation. Genesis 27, 8193.
Brunet, A., Bonni, A., Zigmond, M.J., Lin, M.Z., Juo, P., Hu, L.S., Anderson, M.J.,
Arden, K.C., Blenis, J. and Greenberg, M.E. (1999) Akt promotes cell survival by
phosphorylating and inhibiting a Forkhead transcription factor. Cell 96, 857-68.
Calnan, D.R. and Brunet, A. (2008) The FoxO code. Oncogene 27, 2276-88.
Carter-Dawson, L.D. and LaVail, M.M. (1979) Rods and cones in the mouse retina. II.
Autoradiographic analysis of cell generation using tritiated thymidine. J Comp
Neurol 188, 263-72.
Casarosa, S., Amato, M.A., Andreazzoli, M., Gestri, G., Barsacchi, G. and Cremisi, F.
(2003) Xrx1 controls proliferation and multipotency of retinal progenitors. Mol
Cell Neurosci 22, 25-36.
Casarosa, S., Andreazzoli, M., Simeone, A. and Barsacchi, G. (1997) Xrx1, a novel
Xenopus homeobox gene expressed during eye and pineal gland development.
Mech Dev 61, 187-98.
Cepko, C.L., Austin, C.P., Yang, X., Alexiades, M. and Ezzeddine, D. (1996) Cell fate
determination in the vertebrate retina. Proc Natl Acad Sci U S A 93, 589-95.
Chandramohan, V., Jeay, S., Pianetti, S. and Sonenshein, G.E. (2004) Reciprocal control
of Forkhead box O 3a and c-Myc via the phosphatidylinositol 3-kinase pathway
coordinately regulates p27Kip1 levels. J Immunol 172, 5522-7.
Chow, R.L., Altmann, C.R., Lang, R.A. and Hemmati-Brivanlou, A. (1999) Pax6 induces
ectopic eyes in a vertebrate. Development 126, 4213-22.
Chow, R.L. and Lang, R.A. (2001) Early eye development in vertebrates. Annu Rev Cell
Dev Biol 17, 255-96.
155
Chuang, J.C., Mathers, P.H. and Raymond, P.A. (1999) Expression of three Rx
homeobox genes in embryonic and adult zebrafish. Mech Dev 84, 195-8.
Chuang, J.C. and Raymond, P.A. (2001) Zebrafish genes rx1 and rx2 help define the
region of forebrain that gives rise to retina. Dev Biol 231, 13-30.
Cunningham, J.J., Levine, E.M., Zindy, F., Goloubeva, O., Roussel, M.F. and Smeyne,
R.J. (2002) The cyclin-dependent kinase inhibitors p19(Ink4d) and p27(Kip1) are
coexpressed in select retinal cells and act cooperatively to control cell cycle exit.
Mol Cell Neurosci 19, 359-74.
Danilova, N., Visel, A., Willett, C.E. and Steiner, L.A. (2004) Expression of the winged
helix/forkhead gene, foxn4, during zebrafish development. Brain Res Dev Brain
Res 153, 115-9.
Danno, H., Michiue, T., Hitachi, K., Yukita, A., Ishiura, S. and Asashima, M. (2008)
Molecular links among the causative genes for ocular malformation: Otx2 and
Sox2 coregulate Rax expression. Proc Natl Acad Sci U S A 105, 5408-13.
Das, G., Choi, Y., Sicinski, P. and Levine, E.M. (2009) Cyclin D1 fine-tunes the
neurogenic output of embryonic retinal progenitor cells. Neural Dev 4, 15.
Del Rio-Tsonis, K. and Tsonis, P.A. (2003) Eye regeneration at the molecular age. Dev
Dyn 226, 211-24.
Deschet, K., Bourrat, F., Ristoratore, F., Chourrout, D. and Joly, J.S. (1999) Expression
of the medaka (Oryzias latipes) Ol-Rx3 paired-like gene in two diencephalic
derivatives, the eye and the hypothalamus. Mech Dev 83, 179-82.
Donner, A.L., Episkopou, V. and Maas, R.L. (2007) Sox2 and Pou2f1 interact to control
lens and olfactory placode development. Dev Biol 303, 784-99.
Dorsky, R.I., Rapaport, D.H. and Harris, W.A. (1995) Xotch inhibits cell differentiation
in the Xenopus retina. Neuron 14, 487-96.
Duparc, R.H., Abdouh, M., David, J., Lepine, M., Tetreault, N. and Bernier, G. (2007)
Pax6 controls the proliferation rate of neuroepithelial progenitors from the mouse
optic vesicle. Dev Biol 301, 374-87.
Dyer, M.A. and Cepko, C.L. (2000) p57(Kip2) regulates progenitor cell proliferation and
amacrine interneuron development in the mouse retina. Development 127, 3593605.
Dyer, M.A. and Cepko, C.L. (2001) p27Kip1 and p57Kip2 regulate proliferation in
distinct retinal progenitor cell populations. J Neurosci 21, 4259-71.
Eggert, T., Hauck, B., Hildebrandt, N., Gehring, W.J. and Walldorf, U. (1998) Isolation
of a Drosophila homolog of the vertebrate homeobox gene Rx and its possible
role in brain and eye development. Proc Natl Acad Sci U S A 95, 2343-8.
El-Hodiri, H.M., Shou, W. and Etkin, L.D. (1997) xnf7 functions in dorsal-ventral
patterning of the Xenopus embryo. Dev Biol 190, 1-17.
Esteve, P. and Bovolenta, P. (2006) Secreted inducers in vertebrate eye development:
more functions for old morphogens. Curr Opin Neurobiol 16, 13-9.
Ferland, R.J., Cherry, T.J., Preware, P.O., Morrisey, E.E. and Walsh, C.A. (2003)
Characterization of Foxp2 and Foxp1 mRNA and protein in the developing and
mature brain. J Comp Neurol 460, 266-79.
Fukuda, K., Yoshida, H., Sato, T., Furumoto, T.A., Mizutani-Koseki, Y., Suzuki, Y.,
Saito, Y., Takemori, T., Kimura, M., Sato, H., Taniguchi, M., Nishikawa, S.,
156
Nakayama, T. and Koseki, H. (2003) Mesenchymal expression of Foxl1, a winged
helix transcriptional factor, regulates generation and maintenance of gutassociated lymphoid organs. Dev Biol 255, 278-89.
Furukawa, T., Kozak, C.A. and Cepko, C.L. (1997) rax, a novel paired-type homeobox
gene, shows expression in the anterior neural fold and developing retina. Proc
Natl Acad Sci U S A 94, 3088-93.
Furukawa, T., Morrow, E.M., Li, T., Davis, F.C. and Cepko, C.L. (1999) Retinopathy
and attenuated circadian entrainment in Crx-deficient mice. Nat Genet 23, 466-70.
Gestri, G., Carl, M., Appolloni, I., Wilson, S.W., Barsacchi, G. and Andreazzoli, M.
(2005) Six3 functions in anterior neural plate specification by promoting cell
proliferation and inhibiting Bmp4 expression. Development 132, 2401-13.
Gouge, A., Holt, J., Hardy, A.P., Sowden, J.C. and Smith, H.K. (2001) Foxn4--a new
member of the forkhead gene family is expressed in the retina. Mech Dev 107,
203-6.
Gray, P.A., Fu, H., Luo, P., Zhao, Q., Yu, J., Ferrari, A., Tenzen, T., Yuk, D.I., Tsung,
E.F., Cai, Z., Alberta, J.A., Cheng, L.P., Liu, Y., Stenman, J.M., Valerius, M.T.,
Billings, N., Kim, H.A., Greenberg, M.E., McMahon, A.P., Rowitch, D.H., Stiles,
C.D. and Ma, Q. (2004) Mouse brain organization revealed through direct
genome-scale TF expression analysis. Science 306, 2255-7.
Gurdon, J.B. and Hopwood, N. (2000) The introduction of Xenopus laevis into
developmental biology: of empire, pregnancy testing and ribosomal genes. Int J
Dev Biol 44, 43-50.
Gurdon, J.B., Lane, C.D., Woodland, H.R. and Marbaix, G. (1971) Use of frog eggs and
oocytes for the study of messenger RNA and its translation in living cells. Nature
233, 177-82.
Hannenhalli, S. and Kaestner, K.H. (2009) The evolution of Fox genes and their role in
development and disease. Nat Rev Genet 10, 233-40.
Harada, T., Harada, C. and Parada, L.F. (2007) Molecular regulation of visual system
development: more than meets the eye. Genes Dev 21, 367-78.
Hardcastle, Z. and Papalopulu, N. (2000) Distinct effects of XBF-1 in regulating the cell
cycle inhibitor p27(XIC1) and imparting a neural fate. Development 127, 130314.
Hatini, V., Tao, W. and Lai, E. (1994) Expression of winged helix genes, BF-1 and BF-2,
define adjacent domains within the developing forebrain and retina. J Neurobiol
25, 1293-309.
Heglind, M., Cederberg, A., Aquino, J., Lucas, G., Ernfors, P. and Enerback, S. (2005)
Lack of the central nervous system- and neural crest-expressed forkhead gene
Foxs1 affects motor function and body weight. Mol Cell Biol 25, 5616-25.
Herrera, E., Marcus, R., Li, S., Williams, S.E., Erskine, L., Lai, E. and Mason, C. (2004)
Foxd1 is required for proper formation of the optic chiasm. Development 131,
5727-39.
Hitchcock, P.F. and Raymond, P.A. (1992) Retinal regeneration. Trends Neurosci 15,
103-8.
Ho, K.K., Myatt, S.S. and Lam, E.W. (2008) Many forks in the path: cycling with FoxO.
Oncogene 27, 2300-11.
157
Holt, C.E., Bertsch, T.W., Ellis, H.M. and Harris, W.A. (1988) Cellular determination in
the Xenopus retina is independent of lineage and birth date. Neuron 1, 15-26.
Huang, H. and Tindall, D.J. (2007) Dynamic FoxO transcription factors. J Cell Sci 120,
2479-87.
Huh, S., Hatini, V., Marcus, R.C., Li, S.C. and Lai, E. (1999) Dorsal-ventral patterning
defects in the eye of BF-1-deficient mice associated with a restricted loss of shh
expression. Dev Biol 211, 53-63.
Hutcheson, D.A. and Vetter, M.L. (2002) Transgenic approaches to retinal development
and function in Xenopus laevis. Methods 28, 402-10.
Johns, P.R. (1977) Growth of the adult goldfish eye. III. Source of the new retinal cells. J
Comp Neurol 176, 343-57.
Johns, P.R. (1982) Formation of photoreceptors in larval and adult goldfish. J Neurosci 2,
178-98.
Julian, D., Ennis, K. and Korenbrot, J.I. (1998) Birth and fate of proliferative cells in the
inner nuclear layer of the mature fish retina. J Comp Neurol 394, 271-82.
Junger, M.A., Rintelen, F., Stocker, H., Wasserman, J.D., Vegh, M., Radimerski, T.,
Greenberg, M.E. and Hafen, E. (2003) The Drosophila forkhead transcription
factor FOXO mediates the reduction in cell number associated with reduced
insulin signaling. J Biol 2, 20.
Kaestner, K.H., Knochel, W. and Martinez, D.E. (2000) Unified nomenclature for the
winged helix/forkhead transcription factors. Genes Dev 14, 142-6.
Kaestner, K.H., Lee, K.H., Schlondorff, J., Hiemisch, H., Monaghan, A.P. and Schutz, G.
(1993) Six members of the mouse forkhead gene family are developmentally
regulated. Proc Natl Acad Sci U S A 90, 7628-31.
Kaestner, K.H., Silberg, D.G., Traber, P.G. and Schutz, G. (1997) The mesenchymal
winged helix transcription factor Fkh6 is required for the control of
gastrointestinal proliferation and differentiation. Genes Dev 11, 1583-95.
Kalinichenko, V.V., Gusarova, G.A., Shin, B. and Costa, R.H. (2003) The forkhead box
F1 transcription factor is expressed in brain and head mesenchyme during mouse
embryonic development. Gene Expr Patterns 3, 153-8.
Kamachi, Y., Cheah, K.S. and Kondoh, H. (1999) Mechanism of regulatory target
selection by the SOX high-mobility-group domain proteins as revealed by
comparison of SOX1/2/3 and SOX9. Mol Cell Biol 19, 107-20.
Kamachi, Y., Uchikawa, M. and Kondoh, H. (2000) Pairing SOX off: with partners in the
regulation of embryonic development. Trends Genet 16, 182-7.
Kelly, L.E., Nekkalapudi, S. and El-Hodiri, H.M. (2007) Expression of the forkhead
transcription factor FoxN4 in progenitor cells in the developing Xenopus laevis
retina and brain. Gene Expr Patterns 7, 233-8.
Kim, I.M., Zhou, Y., Ramakrishna, S., Hughes, D.E., Solway, J., Costa, R.H. and
Kalinichenko, V.V. (2005) Functional characterization of evolutionarily
conserved DNA regions in forkhead box f1 gene locus. J Biol Chem 280, 3790816.
Kitamura, T., Kitamura, Y.I., Funahashi, Y., Shawber, C.J., Castrillon, D.H., Kollipara,
R., DePinho, R.A., Kitajewski, J. and Accili, D. (2007) A Foxo/Notch pathway
158
controls myogenic differentiation and fiber type specification. J Clin Invest 117,
2477-85.
Kiyota, T., Kato, A., Altmann, C.R. and Kato, Y. (2008) The POU homeobox protein
Oct-1 regulates radial glia formation downstream of Notch signaling. Dev Biol
315, 579-92.
Kops, G.J. and Burgering, B.M. (1999) Forkhead transcription factors: new insights into
protein kinase B (c-akt) signaling. J Mol Med 77, 656-65.
Kops, G.J., de Ruiter, N.D., De Vries-Smits, A.M., Powell, D.R., Bos, J.L. and
Burgering, B.M. (1999) Direct control of the Forkhead transcription factor AFX
by protein kinase B. Nature 398, 630-4.
Korver, W., Roose, J. and Clevers, H. (1997a) The winged-helix transcription factor
Trident is expressed in cycling cells. Nucleic Acids Res 25, 1715-9.
Korver, W., Roose, J., Heinen, K., Weghuis, D.O., de Bruijn, D., van Kessel, A.G. and
Clevers, H. (1997b) The human TRIDENT/HFH-11/FKHL16 gene: structure,
localization, and promoter characterization. Genomics 46, 435-42.
Korver, W., Schilham, M.W., Moerer, P., van den Hoff, M.J., Dam, K., Lamers, W.H.,
Medema, R.H. and Clevers, H. (1998) Uncoupling of S phase and mitosis in
cardiomyocytes and hepatocytes lacking the winged-helix transcription factor
Trident. Curr Biol 8, 1327-30.
Koster, M., Dillinger, K. and Knochel, W. (1999) Genomic structure and embryonic
expression of the Xenopus winged helix factors XFD-13/13'. Mech Dev 88, 8993.
Kroll, K.L. and Amaya, E. (1996) Transgenic Xenopus embryos from sperm nuclear
transplantations reveal FGF signaling requirements during gastrulation.
Development 122, 3173-83.
Krupczak-Hollis, K., Wang, X., Kalinichenko, V.V., Gusarova, G.A., Wang, I.C.,
Dennewitz, M.B., Yoder, H.M., Kiyokawa, H., Kaestner, K.H. and Costa, R.H.
(2004) The mouse Forkhead Box m1 transcription factor is essential for
hepatoblast mitosis and development of intrahepatic bile ducts and vessels during
liver morphogenesis. Dev Biol 276, 74-88.
Laduron, S., Deplus, R., Zhou, S., Kholmanskikh, O., Godelaine, D., De Smet, C.,
Hayward, S.D., Fuks, F., Boon, T. and De Plaen, E. (2004) MAGE-A1 interacts
with adaptor SKIP and the deacetylase HDAC1 to repress transcription. Nucleic
Acids Res 32, 4340-50.
Laoukili, J., Kooistra, M.R., Bras, A., Kauw, J., Kerkhoven, R.M., Morrison, A., Clevers,
H. and Medema, R.H. (2005) FoxM1 is required for execution of the mitotic
programme and chromosome stability. Nat Cell Biol 7, 126-36.
Laoukili, J., Stahl, M. and Medema, R.H. (2007) FoxM1: at the crossroads of ageing and
cancer. Biochim Biophys Acta 1775, 92-102.
Larsen, K.B., Lutterodt, M., Rath, M.F. and Moller, M. (2009) Expression of the
homeobox genes PAX6, OTX2, and OTX1 in the early human fetal retina. Int J
Dev Neurosci 27, 485-92.
Levine, E.M., Close, J., Fero, M., Ostrovsky, A. and Reh, T.A. (2000) p27(Kip1)
regulates cell cycle withdrawal of late multipotent progenitor cells in the
mammalian retina. Dev Biol 219, 299-314.
159
Li, H., Tao, W. and Lai, E. (1996) Characterization of the structure and function of the
gene for transcription factor BF-1, an essential regulator of forebrain
development. Brain Res Mol Brain Res 37, 96-104.
Li, J., Chang, H.W., Lai, E., Parker, E.J. and Vogt, P.K. (1995) The oncogene qin codes
for a transcriptional repressor. Cancer Res 55, 5540-4.
Li, S., Mo, Z., Yang, X., Price, S.M., Shen, M.M. and Xiang, M. (2004) Foxn4 controls
the genesis of amacrine and horizontal cells by retinal progenitors. Neuron 43,
795-807.
Li, X., Perissi, V., Liu, F., Rose, D.W. and Rosenfeld, M.G. (2002) Tissue-specific
regulation of retinal and pituitary precursor cell proliferation. Science 297, 11803.
Livesey, F.J. and Cepko, C.L. (2001) Vertebrate neural cell-fate determination: lessons
from the retina. Nat Rev Neurosci 2, 109-18.
Loosli, F., Staub, W., Finger-Baier, K., Ober, E., Verkade, H., Wittbrodt, J. and Baier, H.
(2003) Loss of eyes in zebrafish caused by mutation of chokh/rx3. EMBO
Reports 4, 894-899.
Loosli, F., Winkler, S., Burgtorf, C., Wurmbach, E., Ansorge, W., Henrich, T., Grabher,
C., Arendt, D., Carl, M., Krone, A., Grzebisz, E. and Wittbrodt, J. (2001) Medaka
eyeless is the key factor linking retinal determination and eye growth.
Development 128, 4035-44.
Loosli, F., Winkler, S. and Wittbrodt, J. (1999) Six3 overexpression initiates the
formation of ectopic retina. Genes Dev 13, 649-54.
Lu, M.M., Li, S., Yang, H. and Morrisey, E.E. (2002) Foxp4: a novel member of the
Foxp subfamily of winged-helix genes co-expressed with Foxp1 and Foxp2 in
pulmonary and gut tissues. Gene Expr Patterns 2, 223-8.
Lupo, G., Andreazzoli, M., Gestri, G., Liu, Y., He, R.Q. and Barsacchi, G. (2000)
Homeobox genes in the genetic control of eye development. Int J Dev Biol 44,
627-36.
Luscher-Firzlaff, J.M., Westendorf, J.M., Zwicker, J., Burkhardt, H., Henriksson, M.,
Muller, R., Pirollet, F. and Luscher, B. (1999) Interaction of the fork head domain
transcription factor MPP2 with the human papilloma virus 16 E7 protein:
enhancement of transformation and transactivation. Oncogene 18, 5620-30.
Marcal, N., Patel, H., Dong, Z., Belanger-Jasmin, S., Hoffman, B., Helgason, C.D., Dang,
J. and Stifani, S. (2005) Antagonistic effects of Grg6 and Groucho/TLE on the
transcription repression activity of brain factor 1/FoxG1 and cortical neuron
differentiation. Mol Cell Biol 25, 10916-29.
Mariani, F.V. and Harland, R.M. (1998) XBF-2 is a transcriptional repressor that
converts ectoderm into neural tissue. Development 125, 5019-31.
Marquardt, T., Ashery-Padan, R., Andrejewski, N., Scardigli, R., Guillemot, F. and
Gruss, P. (2001) Pax6 is required for the multipotent state of retinal progenitor
cells. Cell 105, 43-55.
Martinez-Morales, J.R., Signore, M., Acampora, D., Simeone, A. and Bovolenta, P.
(2001) Otx genes are required for tissue specification in the developing eye.
Development 128, 2019-30.
160
Mathers, P.H., Grinberg, A., Mahon, K.A. and Jamrich, M. (1997) The Rx homeobox
gene is essential for vertebrate eye development. Nature 387, 603-7.
Medema, R.H., Kops, G.J., Bos, J.L. and Burgering, B.M. (2000) AFX-like Forkhead
transcription factors mediate cell-cycle regulation by Ras and PKB through
p27kip1. Nature 404, 782-7.
Medina-Martinez, O., Brownell, I., Amaya-Manzanares, F., Hu, Q., Behringer, R.R. and
Jamrich, M. (2005) Severe defects in proliferation and differentiation of lens cells
in Foxe3 null mice. Mol Cell Biol 25, 8854-63.
Medina-Martinez, O. and Jamrich, M. (2007) Foxe view of lens development and
disease. Development 134, 1455-63.
Miyawaki, T., Uemura, A., Dezawa, M., Yu, R.T., Ide, C., Nishikawa, S., Honda, Y.,
Tanabe, Y. and Tanabe, T. (2004) Tlx, an orphan nuclear receptor, regulates cell
numbers and astrocyte development in the developing retina. J Neurosci 24, 812434.
Mogi, M., Walsh, K., Iwai, M. and Horiuchi, M. (2008) Akt-FOXO3a signaling affects
human endothelial progenitor cell differentiation. Hypertens Res 31, 153-9.
Moose, H.E., Kelly, L.E., Nekkalapudi, S. and El-Hodiri, H.M. (2009) Ocular forkhead
transcription factors: seeing eye to eye. Int J Dev Biol 53, 29-36.
Moritz, O.L., Peck, A. and Tam, B.M. (2002) Xenopus laevis red cone opsin and Prph2
promoters allow transgene expression in amphibian cones, or both rods and cones.
Gene 298, 173-82.
Nakada, C., Satoh, S., Tabata, Y., Arai, K. and Watanabe, S. (2006) Transcriptional
repressor foxl1 regulates central nervous system development by suppressing shh
expression in zebra fish. Mol Cell Biol 26, 7246-57.
Nelson, S.M., Park, L. and Stenkamp, D.L. (2009) Retinal homeobox 1 is required for
retinal neurogenesis and photoreceptor differentiation in embryonic zebrafish.
Dev Biol 328, 24-39.
Nieuwkoop, P.D. and Faber, J. (1994) Normal Table of Xenopus laevis (Daudin). Garland
Publishing, Inc., New York.
Nishida, A., Furukawa, A., Koike, C., Tano, Y., Aizawa, S., Matsuo, I. and Furukawa, T.
(2003) Otx2 homeobox gene controls retinal photoreceptor cell fate and pineal
gland development. Nat Neurosci 6, 1255-63.
Nishimoto, M., Fukushima, A., Okuda, A. and Muramatsu, M. (1999) The gene for the
embryonic stem cell coactivator UTF1 carries a regulatory element which
selectively interacts with a complex composed of Oct-3/4 and Sox-2. Mol Cell
Biol 19, 5453-65.
Nitta, K.R., Takahashi, S., Haramoto, Y., Fukuda, M., Onuma, Y. and Asashima, M.
(2006) Expression of Sox1 during Xenopus early embryogenesis. Biochem
Biophys Res Commun 351, 287-93.
Odenthal, J. and Nusslein-Volhard, C. (1998) fork head domain genes in zebrafish. Dev
Genes Evol 208, 245-58.
Ohnuma, S., Philpott, A., Wang, K., Holt, C.E. and Harris, W.A. (1999) p27Xic1, a Cdk
inhibitor, promotes the determination of glial cells in Xenopus retina. Cell 99,
499-510.
161
Ohuchi, H., Tomonari, S., Itoh, H., Mikawa, T. and Noji, S. (1999) Identification of chick
rax/rx genes with overlapping patterns of expression during early eye and brain
development. Mech Dev 85, 193-5.
Otteson, D.C., D'Costa, A.R. and Hitchcock, P.F. (2001) Putative stem cells and the
lineage of rod photoreceptors in the mature retina of the goldfish. Dev Biol 232,
62-76.
Otteson, D.C. and Hitchcock, P.F. (2003) Stem cells in the teleost retina: persistent
neurogenesis and injury-induced regeneration. Vision Res 43, 927-36.
Pan, Y., Nekkalapudi, S., Kelly, L.E. and El-Hodiri, H.M. (2006a) The Rx-like
Homeobox Gene (Rx-L) Is Necessary for Normal Photoreceptor Development.
Invest Ophthalmol Vis Sci 47, 4245-4253.
Pan, Y., Nekkalapudi, S., Kelly, L.E. and El-Hodiri, H.M. (2006b) The Rx-like
homeobox gene (Rx-L) is necessary for normal photoreceptor development.
Invest Ophthalmol Vis Sci 47, 4245-53.
Pani, L., Overdier, D.G., Porcella, A., Qian, X., Lai, E. and Costa, R.H. (1992)
Hepatocyte nuclear factor 3 beta contains two transcriptional activation domains,
one of which is novel and conserved with the Drosophila fork head protein. Mol
Cell Biol 12, 3723-32.
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, 70720.
Peart, M.J., Tainton, K.M., Ruefli, A.A., Dear, A.E., Sedelies, K.A., O'Reilly, L.A.,
Waterhouse, N.J., Trapani, J.A. and Johnstone, R.W. (2003) Novel mechanisms
of apoptosis induced by histone deacetylase inhibitors. Cancer Res 63, 4460-71.
Perron, M. and Harris, W.A. (2000) Retinal stem cells in vertebrates. Bioessays 22, 6858.
Perron, M., Kanekar, S., Vetter, M.L. and Harris, W.A. (1998) The genetic sequence of
retinal development in the ciliary margin of the Xenopus eye. Dev Biol 199, 185200.
Pohl, B.S. and Knochel, W. (2002) Temporal and spatial expression patterns of FoxD2
during the early development of Xenopus laevis. Mech Dev 111, 181-4.
Pohl, B.S. and Knochel, W. (2004) Isolation and developmental expression of Xenopus
FoxJ1 and FoxK1. Dev Genes Evol 214, 200-5.
Pohl, B.S. and Knochel, W. (2005) Of Fox and Frogs: Fox (fork head/winged helix)
transcription factors in Xenopus development. Gene 344, 21-32.
Pohl, B.S., Rossner, A. and Knochel, W. (2005) The Fox gene family in Xenopus
laevis:FoxI2, FoxM1 and FoxP1 in early development. Int J Dev Biol 49, 53-8.
Pohl, B.S., Schon, C., Rossner, A. and Knochel, W. (2004) The FoxO-subclass in
Xenopus laevis development. Gene Expr Patterns 5, 187-92.
Porter, F.D., Drago, J., Xu, Y., Cheema, S.S., Wassif, C., Huang, S.P., Lee, E., Grinberg,
A., Massalas, J.S., Bodine, D., Alt, F. and Westphal, H. (1997) Lhx2, a LIM
homeobox gene, is required for eye, forebrain, and definitive erythrocyte
development. Development 124, 2935-44.
162
Pratt, T., Tian, N.M., Simpson, T.I., Mason, J.O. and Price, D.J. (2004) The winged helix
transcription factor Foxg1 facilitates retinal ganglion cell axon crossing of the
ventral midline in the mouse. Development 131, 3773-84.
Ragge, N.K., Lorenz, B., Schneider, A., Bushby, K., de Sanctis, L., de Sanctis, U., Salt,
A., Collin, J.R., Vivian, A.J., Free, S.L., Thompson, P., Williamson, K.A.,
Sisodiya, S.M., van Heyningen, V. and Fitzpatrick, D.R. (2005) SOX2
anophthalmia syndrome. Am J Med Genet A 135, 1-7; discussion 8.
Ramakrishna, S., Kim, I.M., Petrovic, V., Malin, D., Wang, I.C., Kalin, T.V., Meliton, L.,
Zhao, Y.Y., Ackerson, T., Qin, Y., Malik, A.B., Costa, R.H. and Kalinichenko,
V.V. (2007) Myocardium defects and ventricular hypoplasia in mice homozygous
null for the Forkhead Box M1 transcription factor. Dev Dyn 236, 1000-13.
Reh, T.A. and Fischer, A.J. (2001) Stem cells in the vertebrate retina. Brain Behav Evol
58, 296-305.
Ryan, A.K. and Rosenfeld, M.G. (1997) POU domain family values: flexibility,
partnerships, and developmental codes. Genes Dev 11, 1207-25.
Sasai, Y. (2001) Roles of Sox factors in neural determination: conserved signaling in
evolution? Int J Dev Biol 45, 321-6.
Schon, C., Wochnik, A., Robetaner, A., Donow, C. and Knochel, W. (2006a) The FoxP
subclass in Xenopus laevis development. Dev Genes Evol 216, 641-646.
Schon, C., Wochnik, A., Rossner, A., Donow, C. and Knochel, W. (2006b) The FoxP
subclass in Xenopus laevis development. Dev Genes Evol 216, 641-6.
Schorpp, M., Hofmann, M., Dear, T.N. and Boehm, T. (1997) Characterization of mouse
and human nude genes. Immunogenetics 46, 509-15.
Schuff, M., Rossner, A., Donow, C. and Knochel, W. (2006) Temporal and spatial
expression patterns of FoxN genes in Xenopus laevis embryos. Int J Dev Biol 50,
429-34.
Schuff, M., Rossner, A., Wacker, S.A., Donow, C., Gessert, S. and Knochel, W. (2007)
FoxN3 is required for craniofacial and eye development of Xenopus laevis. Dev
Dyn 236, 226-39.
Semina, E.V., Brownell, I., Mintz-Hittner, H.A., Murray, J.C. and Jamrich, M. (2001)
Mutations in the human forkhead transcription factor FOXE3 associated with
anterior segment ocular dysgenesis and cataracts. Hum Mol Genet 10, 231-6.
Sharpe, C.R. and Goldstone, K. (1997) Retinoid receptors promote primary neurogenesis
in Xenopus. Development 124, 515-23.
Shi, X., Luo, Y., Howley, S., Dzialo, A., Foley, S., Hyde, D.R. and Vihtelic, T.S. (2006)
Zebrafish foxe3: roles in ocular lens morphogenesis through interaction with
pitx3. Mech Dev 123, 761-82.
Shimamura, K., Hirano, S., McMahon, A.P. and Takeichi, M. (1994) Wnt-1-dependent
regulation of local E-cadherin and alpha N-catenin expression in the embryonic
mouse brain. Development 120, 2225-34.
Shu, W., Yang, H., Zhang, L., Lu, M.M. and Morrisey, E.E. (2001) Characterization of a
new subfamily of winged-helix/forkhead (Fox) genes that are expressed in the
lung and act as transcriptional repressors. J Biol Chem 276, 27488-97.
163
Sicinski, P., Donaher, J.L., Parker, S.B., Li, T., Fazeli, A., Gardner, H., Haslam, S.Z.,
Bronson, R.T., Elledge, S.J. and Weinberg, R.A. (1995) Cyclin D1 provides a link
between development and oncogenesis in the retina and breast. Cell 82, 621-30.
Siebert, P.D., Chenchik, A., Kellogg, D.E., Lukyanov, K.A. and Lukyanov, S.A. (1995)
An improved PCR method for walking in uncloned genomic DNA. Nucleic Acids
Res 23, 1087-8.
Simeone, A., Acampora, D., Mallamaci, A., Stornaiuolo, A., D'Apice, M.R., Nigro, V.
and Boncinelli, E. (1993) A vertebrate gene related to orthodenticle contains a
homeodomain of the bicoid class and demarcates anterior neuroectoderm in the
gastrulating mouse embryo. EMBO J 12, 2735-47.
Sive, H.L., Grainger, R.M. and Harland, R.M. (2000) Early Development of Xenopus
laevis: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring
Harbor, NY.
Smith, R.S., Zabaleta, A., Kume, T., Savinova, O.V., Kidson, S.H., Martin, J.E.,
Nishimura, D.Y., Alward, W.L., Hogan, B.L. and John, S.W. (2000)
Haploinsufficiency of the transcription factors FOXC1 and FOXC2 results in
aberrant ocular development. Hum Mol Genet 9, 1021-32.
Sonnemann, J., Kumar, K.S., Heesch, S., Muller, C., Hartwig, C., Maass, M., Bader, P.
and Beck, J.F. (2006) Histone deacetylase inhibitors induce cell death and
enhance the susceptibility to ionizing radiation, etoposide, and TRAIL in
medulloblastoma cells. Int J Oncol 28, 755-66.
Sparrow, D.B., Latinkic, B. and Mohun, T.J. (2000) A simplified method of generating
transgenic Xenopus. Nucleic Acids Res 28, E12.
Stahl, M., Dijkers, P.F., Kops, G.J., Lens, S.M., Coffer, P.J., Burgering, B.M. and
Medema, R.H. (2002) The forkhead transcription factor FoxO regulates
transcription of p27Kip1 and Bim in response to IL-2. J Immunol 168, 5024-31.
Stiemke, M.M. and Hollyfield, J.G. (1995) Cell birthdays in Xenopus laevis retina.
Differentiation 58, 189-93.
Straznicky, K. and Gaze, R.M. (1971) The growth of the retina in Xenopus laevis: an
autoradiographic study. J Embryol Exp Morphol 26, 67-79.
Strickler, A.G., Famuditimi, K. and Jeffery, W.R. (2002) Retinal homeobox genes and
the role of cell proliferation in cavefish eye degeneration. Int J Dev Biol 46, 28594.
Su, J.Y., Rempel, R.E., Erikson, E. and Maller, J.L. (1995) Cloning and characterization
of the Xenopus cyclin-dependent kinase inhibitor p27XIC1. Proc Natl Acad Sci U
S A 92, 10187-91.
Takahashi, K., Liu, F.C., Hirokawa, K. and Takahashi, H. (2003) Expression of Foxp2, a
gene involved in speech and language, in the developing and adult striatum. J
Neurosci Res 73, 61-72.
Tamimi, Y., Skarie, J.M., Footz, T., Berry, F.B., Link, B.A. and Walter, M.A. (2006)
FGF19 is a target for FOXC1 regulation in ciliary body-derived cells. Hum Mol
Genet 15, 3229-40.
Tamura, S., Morikawa, Y., Iwanishi, H., Hisaoka, T. and Senba, E. (2003) Expression
pattern of the winged-helix/forkhead transcription factor Foxp1 in the developing
central nervous system. Gene Expr Patterns 3, 193-7.
164
Tetreault, N., Champagne, M.P. and Bernier, G. (2009) The LIM homeobox transcription
factor Lhx2 is required to specify the retina field and synergistically cooperates
with Pax6 for Six6 trans-activation. Dev Biol 327, 541-50.
Tomita, K., Moriyoshi, K., Nakanishi, S., Guillemot, F. and Kageyama, R. (2000)
Mammalian achaete-scute and atonal homologs regulate neuronal versus glial fate
determination in the central nervous system. EMBO J 19, 5460-72.
Tothova, Z. and Gilliland, D.G. (2007) FoxO transcription factors and stem cell
homeostasis: insights from the hematopoietic system. Cell Stem Cell 1, 140-52.
Tribioli, C., Robledo, R.F. and Lufkin, T. (2002) The murine fork head gene Foxn2 is
expressed in craniofacial, limb, CNS and somitic tissues during embryogenesis.
Mech Dev 118, 161-3.
Tropepe, V., Coles, B.L., Chiasson, B.J., Horsford, D.J., Elia, A.J., McInnes, R.R. and
van der Kooy, D. (2000) Retinal stem cells in the adult mammalian eye. Science
287, 2032-6.
Tseng, H.T., Shah, R. and Jamrich, M. (2004) Function and regulation of FoxF1 during
Xenopus gut development. Development 131, 3637-47.
Turner, D.L. and Cepko, C.L. (1987) A common progenitor for neurons and glia persists
in rat retina late in development. Nature 328, 131-6.
Ueno, H., Nakajo, N., Watanabe, M., Isoda, M. and Sagata, N. (2008) FoxM1-driven cell
division is required for neuronal differentiation in early Xenopus embryos.
Development 135, 2023-30.
Valleix, S., Niel, F., Nedelec, B., Algros, M.P., Schwartz, C., Delbosc, B., Delpech, M.
and Kantelip, B. (2006) Homozygous nonsense mutation in the FOXE3 gene as a
cause of congenital primary aphakia in humans. Am J Hum Genet 79, 358-64.
Vetter, M.L. and Brown, N.L. (2001) The role of basic helix-loop-helix genes in
vertebrate retinogenesis. Semin Cell Dev Biol 12, 491-8.
Viczian, A.S., Bang, A.G., Harris, W.A. and Zuber, M.E. (2006) Expression of Xenopus
laevis Lhx2 during eye development and evidence for divergent expression
among vertebrates. Dev Dyn 235, 1133-41.
Viczian, A.S., Vignali, R., Zuber, M.E., Barsacchi, G. and Harris, W.A. (2003) XOtx5b
and XOtx2 regulate photoreceptor and bipolar fates in the Xenopus retina.
Development 130, 1281-94.
Vignali, R., Colombetti, S., Lupo, G., Zhang, W., Stachel, S., Harland, R.M. and
Barsacchi, G. (2000) Xotx5b, a new member of the Otx gene family, may be
involved in anterior and eye development in Xenopus laevis. Mech Dev 96, 3-13.
Wang, I.C., Chen, Y.J., Hughes, D., Petrovic, V., Major, M.L., Park, H.J., Tan, Y.,
Ackerson, T. and Costa, R.H. (2005) Forkhead box M1 regulates the
transcriptional network of genes essential for mitotic progression and genes
encoding the SCF (Skp2-Cks1) ubiquitin ligase. Mol Cell Biol 25, 10875-94.
Wang, S.W., Kim, B.S., Ding, K., Wang, H., Sun, D., Johnson, R.L., Klein, W.H. and
Gan, L. (2001) Requirement for math5 in the development of retinal ganglion
cells. Genes Dev 15, 24-9.
Wang, X., Kiyokawa, H., Dennewitz, M.B. and Costa, R.H. (2002) The Forkhead Box
m1b transcription factor is essential for hepatocyte DNA replication and mitosis
during mouse liver regeneration. Proc Natl Acad Sci U S A 99, 16881-6.
165
Welcker, M. and Clurman, B. (2005) Cell cycle: how cyclin E got its groove back. Curr
Biol 15, R810-2.
Westendorf, J.M., Rao, P.N. and Gerace, L. (1994) Cloning of cDNAs for M-phase
phosphoproteins recognized by the MPM2 monoclonal antibody and
determination of the phosphorylated epitope. Proc Natl Acad Sci U S A 91, 714-8.
Wierstra, I. and Alves, J. (2006) Despite its strong transactivation domain, transcription
factor FOXM1c is kept almost inactive by two different inhibitory domains. Biol
Chem 387, 963-76.
Wierstra, I. and Alves, J. (2007a) FOXM1, a typical proliferation-associated transcription
factor. Biol Chem 388, 1257-74.
Wierstra, I. and Alves, J. (2007b) FOXM1c and Sp1 transactivate the P1 and P2
promoters of human c-myc synergistically. Biochem Biophys Res Commun 352,
61-8.
Wilson, S.W. and Houart, C. (2004) Early steps in the development of the forebrain. Dev
Cell 6, 167-81.
Winkler, S., Loosli, F., Henrich, T., Wakamatsu, Y. and Wittbrodt, J. (2000) The
conditional medaka mutation eyeless uncouples patterning and morphogenesis of
the eye. Development 127, 1911-9.
Wong, L.L. and Rapaport, D.H. (2009) Defining retinal progenitor cell competence in
Xenopus laevis by clonal analysis. Development 136, 1707-15.
Wonsey, D.R. and Follettie, M.T. (2005) Loss of the forkhead transcription factor FoxM1
causes centrosome amplification and mitotic catastrophe. Cancer Res 65, 5181-9.
Xuan, S., Baptista, C.A., Balas, G., Tao, W., Soares, V.C. and Lai, E. (1995) Winged
helix transcription factor BF-1 is essential for the development of the cerebral
hemispheres. Neuron 14, 1141-52.
Yao, J., Lai, E. and Stifani, S. (2001) The winged-helix protein brain factor 1 interacts
with groucho and hes proteins to repress transcription. Mol Cell Biol 21, 1962-72.
Yao, K.M., Sha, M., Lu, Z. and Wong, G.G. (1997) Molecular analysis of a novel winged
helix protein, WIN. Expression pattern, DNA binding property, and alternative
splicing within the DNA binding domain. J Biol Chem 272, 19827-36.
Ye, H., Holterman, A.X., Yoo, K.W., Franks, R.R. and Costa, R.H. (1999) Premature
expression of the winged helix transcription factor HFH-11B in regenerating
mouse liver accelerates hepatocyte entry into S phase. Mol Cell Biol 19, 8570-80.
Ye, H., Kelly, T.F., Samadani, U., Lim, L., Rubio, S., Overdier, D.G., Roebuck, K.A. and
Costa, R.H. (1997) Hepatocyte nuclear factor 3/fork head homolog 11 is
expressed in proliferating epithelial and mesenchymal cells of embryonic and
adult tissues. Mol Cell Biol 17, 1626-41.
Young, R.W. (1985) Cell differentiation in the retina of the mouse. Anat Rec 212, 199205.
Yuasa, J., Hirano, S., Yamagata, M. and Noda, M. (1996) Visual projection map
specified by topographic expression of transcription factors in the retina. Nature
382, 632-5.
Zhang, L., El-Hodiri, H.M., Ma, H.F., Zhang, X., Servetnick, M., Wensel, T.G. and
Jamrich, M. (2003) Targeted expression of the dominant-negative FGFR4a in the
166
eye using Xrx1A regulatory sequences interferes with normal retinal
development. Development 130, 4177-86.
Zhang, L., Mathers, P.H. and Jamrich, M. (2000) Function of Rx, but not Pax6, is
essential for the formation of retinal progenitor cells in mice. Genesis 28, 135-42.
Zhang, Y., Miki, T., Iwanaga, T., Koseki, Y., Okuno, M., Sunaga, Y., Ozaki, N., Yano,
H., Koseki, H. and Seino, S. (2002) Identification, tissue expression, and
functional characterization of Otx3, a novel member of the Otx family. J Biol
Chem 277, 28065-9.
Zhu, C.C., Dyer, M.A., Uchikawa, M., Kondoh, H., Lagutin, O.V. and Oliver, G. (2002)
Six3-mediated auto repression and eye development requires its interaction with
members of the Groucho-related family of co-repressors. Development 129,
2835-49.
Zuber, M.E., Gestri, G., Viczian, A.S., Barsacchi, G. and Harris, W.A. (2003)
Specification of the vertebrate eye by a network of eye field transcription factors.
Development 130, 5155-67.
167