GENE EXPRESSION PROFILING REVEALS NOVEL ATTRIBUTES OF THE MOUSE DEFINITIVE ENDODERM
by
KRISTEN DAWN MCKNIGHT
B.Sc. (Honors), Simon Fraser University, 2002
THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
in
THE FACULTY OF GRADUATE STUDIES
(Genetics)
THE UNIVERSITY OF BRITISH COLUMBIA
(Vancouver)
December 2008
© Kristen Dawn McKnight, 2008
ABSTRACT
Gastrulation is one of the most critical events of embryogenesis, generating the three primary germ layers (definitive endoderm, mesoderm, and ectoderm) and establishing the embryonic body plan. The definitive endoderm, which generates the lungs, liver, pancreas, and digestive tract, has become a tissue of particular interest in recent years. Understanding definitive endoderm formation and patterning will greatly aid progress in the in vitro differentiation of embryonic stem cells to definitive endoderm for use in treatment of diseases such as diabetes and hepatitis as an alternative for whole organ replacement.
Gene targeting studies have demonstrated a critical role for the Nodal signaling pathway and the forkhead transcription factors Foxh1 and Foxa2 in specification of a group of cells referred to as the anterior primitive streak (APS). However, the transcriptional targets of Foxh1 and/or Foxa2 other than Nodal that regulate specification of this group of cells are currently unknown. Fate mapping and lineage tracing experiments have shown the APS to be the source of the definitive endoderm. However, many questions regarding specification and patterning of the definitive endoderm remain. The study of this tissue has been hampered by the lack of genetic markers specific for the definitive endoderm as many of the current markers, including Cerl,
Foxa2, and Sox17, are also expressed in the visceral endoderm, an extraembryonic tissue.
To further investigate the role of Foxh1 in specification of the anterior primitive streak and to address the lack of genetic markers for the definitive endoderm we performed expression profiling on post-implantation mouse embryos using Affymetrix™ GeneChips®. From this analysis we identified and characterized a novel marker of the mouse definitive endoderm.
Examination of this, and other, novel endoderm markers in Foxh1 and Foxa2 deficient mouse
ii embryos revealed that contrary to current models of definitive endoderm formation, we find some definitive endoderm is formed in these mutants. Specifically, specification of the midgut and hindgut definitive endoderm is largely unaffected, while foregut formation is severely affected. These results suggest that the formation of the midgut and hindgut definitive endoderm populations is independent of the anterior primitive streak and separate from the foregut definitive endoderm. This represents a major insight into the mechanisms regulating endoderm formation and patterning.
iii TABLE OF CONTENTS
ABSTRACT...... ii
TABLE OF CONTENTS ...... iv
LIST OF TABLES ...... vii
LIST OF FIGURES ...... vii
ACKNOWLEDGEMENTS ...... ix
DEDICATION...... x
CO-AUTHORSHIP STATEMENT ...... xi
CHAPTER 1 INTRODUCTION...... 1 1.1 GENERAL OVERVIEW OF THE EMBRYONIC DEVELOPMENT OF MUS MUSCULUS...... 1 1.1.1 An introduction to the lineages constituting the mouse embryo ...... 1 1.1.1.1 The first lineages are specified during the pre-implantation stage...... 4 1.1.1.2 The primary germ layers are formed during the post-implantation stage ...... 5 1.2 THE MOUSE GASTRULA ORGANIZER ...... 6 1.2.1 The mouse gastrula organizer is located at the anterior end of the primitive streak and contributes to the midline tissues of the embryo ...... 7 1.2.2 Tissues derived from the anterior primitive streak contribute to the patterning of all three body axes ...... 9 1.2.3 Specification of the anterior primitive streak ...... 10 1.3 THE MOUSE DEFINITIVE ENDODERM...... 11 1.3.1 Formation and morphogenesis of the definitive endoderm...... 12 1.3.2 Regulation of definitive endoderm cell fate ...... 15 1.3.3 Regionalization of the definitive endoderm ...... 17 1.4 THE NODAL SIGNALING PATHWAY...... 18 1.4.1 Components of the Nodal signaling pathway...... 19 1.4.2 Regulation of cell fate decisions in the mouse embryo by the Nodal pathway...... 23 1.4.2.1 There is a graded response to Nodal signaling within the primitive streak of the mouse embryo ...... 23 1.5 THE FORKHEAD BOX PROTEINS FOXH1 AND FOXA2...... 28 1.5.1 Foxh1 and Foxa2 are winged helix transcription factors ...... 28 1.5.1.1 Foxh1...... 31 1.5.1.2 Foxa2...... 32 1.5.2 Roles of Foxh1 and Foxa2 in mouse embryogenesis...... 34 1.5.2.1 Expression of Foxh1 and Foxa2 in the mouse embryo...... 34 1.5.2.2 Foxh1 and Foxa2 are essential for specification of the anterior primitive streak and its derivative tissues ...... 35 1.5.2.3 Downstream targets of Foxh1 and Foxa2 in the mouse embryo...... 39 1.6 AIM OF THE STUDY...... 41 1.7 REFERENCES ...... 43
iv CHAPTER 2 A DIFFERENTIAL SCREEN FOR GENES INVOLVED IN SPECIFICATION OF THE ANTERIOR PRIMITIVE STREAK USING FOXH1 DEFICIENT EMBRYOS ...... 52 2.1 INTRODUCTION ...... 52 2.2 RESULTS ...... 54 2.2.1 Identification of genes differentially expressed between E6.5 wild-type and Foxh1 null embryos using Affymetrix™ GeneChips® ...... 54 2.2.2 Assessment of microarray sensitivity and data quality ...... 58 2.2.3 Validation of microarray results by RT-PCR and qRT-PCR ...... 60 2.2.4 Analysis of BG078657, an uncharacterized expressed sequence tag, in E6.5 mouse embryos ...... 64 2.2.5 Characterization of Trh, a regulator of the hypothalamic-pituitary-thyroid axis, in E6.5 mouse embryos...... 66 2.3 DISCUSSION...... 69 2.3.1 Multiple factors contributed to the discrepancies between the observed and expected results for control genes used to measure array quality and sensitivity ...... 69 2.3.2 The gene expression profiles of E6.5 wild-type and Foxh1 null embryos are largely similar ...... 71 2.3.3 Contrary to its primary role as a transcriptional activator, loss of Foxh1 results in an increase in gene expression ...... 73 2.3.4 EST BG078657 is likely the 3’ end of an extended mRNA of the Cer1 gene ...... 74 2.3.5 Thyrotropin-releasing hormone is a prospective marker of the pluripotent epiblast and definitive endoderm...... 75 2.4 EXPERIMENTAL PROCEDURES...... 77 2.4.1 Mouse strains, dissection, staging, and genotyping...... 77 2.4.2 RNA extraction, cDNA synthesis, amplification, and labeling...... 78 2.4.3 Affymetrix™ GeneChip® data acquisition and analysis ...... 78 2.4.4 Reverse Transcription-PCR and Quantitative Real Time PCR ...... 79 2.4.5 In situ hybridization probe preparation ...... 82 2.4.6 Whole mount in situ hybridization ...... 82 2.4.7 Imaging and histology ...... 83 2.5 REFERENCES ...... 84
CHAPTER 3 DYNAMIC EXPRESSION OF THYROTROPIN-RELEASING HORMONE IN THE MOUSE DEFINITIVE ENDODERM ...... 88 3.1 INTRODUCTION ...... 88 3.2 RESULTS ...... 92 3.2.1 Expression of components of the HPT axis during post-implantation stages in the mouse embryo...... 92 3.2.2 Transition of Trh expression from the epiblast to the definitive endoderm during gastrulation ...... 94 3.2.3 Trh expression in the definitive endoderm...... 97 3.2.4 Rapid shift in Trh expression from the definitive endoderm to neural tissues...... 97 3.3 DISCUSSION...... 100 3.4 EXPERIMENTAL PROCEDURES...... 103 3.4.1 Mouse strains, dissection and staging...... 103 3.4.2 Serial Analysis of Gene Expression library analysis...... 104 3.4.3 RNA extraction, cDNA synthesis, and RT-PCR...... 104 3.4.4 Probe preparation...... 105
v 3.4.5 Whole mount in situ hybridization ...... 105 3.4.6 Imaging and histology ...... 106 3.5 REFERENCES ...... 108
CHAPTER 4 MIDGUT AND HINDGUT DEFINITIVE ENDODERM FORMATION IN THE MOUSE EMBRYO IS INDEPENDENT OF THE ANTERIOR PRIMITIVE STREAK ...... 111 4.1 INTRODUCTION ...... 111 4.2 RESULTS ...... 115 4.2.1 Midgut and hindgut formation is unperturbed in Foxh1 and Foxa2 null embryos ...... 115 4.2.2 Formation of the foregut definitive endoderm is severely affected in Foxh1 and Foxa2 null embryos ...... 117 4.2.3 The visceral endoderm is incompletely displaced from the embryonic region of Foxh1 and Foxa2 null embryos during gastrulation...... 120 4.2.4 Visceral endoderm that is not displaced during gastrulation becomes incorporated into the foregut of Foxh1 and Foxa2 null embryos ...... 123 4.2.5 Analysis of the function and differentiation potential of the definitive endoderm in Foxh1-/- mutants...... 126 4.3 DISCUSSION...... 131 4.4 EXPERIMENTAL PROCEDURES...... 136 4.4.1 Mouse strains, dissection, and staging...... 136 4.4.2 Probe preparation...... 137 4.4.3 Whole mount in situ hybridization ...... 137 4.4.4. Imaging and histology ...... 138 4.4.5 Endoderm explant culture...... 139 4.4.6 Horseradish peroxidase uptake assay ...... 139 4.4.7 RNA extraction, cDNA synthesis, and RT-PCR...... 139 4.5 REFERENCES ...... 141
CHAPTER 5 SUMMARY, PERSPECTIVES, AND FUTURE DIRECTIONS...... 145 5.1 GENE EXPRESSION PROFILING REVEALS NOVEL ATTRIBUTES OF THE MOUSE DEFINITIVE ENDODERM...... 145 5.1.2 Tissue specific markers will facilitate the creation of essential molecular tools for the study of mouse embryogenesis...... 148 5.1.3 Revisiting previously characterized mouse mutants with novel markers is important to further refine our understanding of germ layer specification and patterning ...... 149 5.2 A THOROUGH UNDERSTANDING OF SPECIFICATION OF THE DEFINITIVE ENDODERM IS NECESSARY FOR CLINICAL APPLICATION OF REGENERATIVE MEDICINE...... 152 5.3 REFERENCES ...... 156
APPENDIX...... 158 ANIMAL CARE CERTIFICATE A02-1852 ...... 158 ANIMAL CARE CERTIFICATE A02-1854 ...... 160
vi LIST OF TABLES
CHAPTER 2
Table 2.1 Genes downregulated greater than 2 fold in Foxh1-/- embryos...... 55 Table 2.2 Genes upregulated between 2 and 3 fold in Foxh1-/- embryos ...... 56 Table 2.3 Genes upregulated greater than 3 fold in Foxh1-/- embryos...... 57 Table 2.4 Primers used for RT-PCR and qRT-PCR analysis ...... 81
vii LIST OF FIGURES
CHAPTER 1
Figure 1.1 The lineage hierarchy constituting the mouse embryo...... 2 Figure 1.2 Schematic diagrams of mouse embryos from pre-implantation, post-implantation, and early organogenesis stages...... 3 Figure 1.3 The mouse gastrula organizer is located at the anterior end of the primitive streak throughout gastrulation and gives rise to midline tissues of the embryo...... 8 Figure 1.4 Formation and morphogenesis of the mouse definitive endoderm ...... 13 Figure 1.5 The Nodal signaling pathway...... 20 Figure 1.6 Structural diagram of Foxh1 and Foxa2 proteins...... 30
CHAPTER 2
Figure 2.1 Expression intensities of control genes in E6.5 Foxh1+/+ and Foxh1-/- embryos……..59 Figure 2.2 RT-PCR validation of microarray data……………………………………………….61 Figure 2.3 qRT-PCR validation of differentially expressed genes...... 63 Figure 2.4 EST BG078657 is expressed in an identical pattern to that of Cer1 in E6.5, E7.5, and E8.5 mouse embryos...... 65 Figure 2.5 The expression level of Trh, but not its localization, is altered in Foxh1-/- embryos .. 68
CHAPTER 3
Figure 3.1 Expression of components of the HPT axis during mouse embryogenesis...... 93 Figure 3.2 Expression of Trh during gastrulation...... 95 Figure 3.3 Trh expression in the definitive endoderm...... 98 Figure 3.4 Shift in Trh expression from the definitive endoderm to neural tissues...... 99
CHAPTER 4
Figure 4.1 The midgut and hindgut definitive endoderm remains in Foxh1-/- and Foxa2-/- null embryos...... 116 Figure 4.2 There is a deficiency of foregut definitive endoderm in Foxh1-/- and Foxa2-/- embryos ...... 119 Figure 4.3 There is incomplete displacement of the visceral endoderm from the embryonic region of Foxh1-/- and Foxa2-/- embryos during gastrulation...... 121 Figure 4.4 Visceral endoderm not displaced during gastrulation becomes incorporated into the foregut of Foxh1-/- and Foxa2-/- embryos...... 124 Figure 4.5 The endoderm in the midgut and hindgut of Foxh1-/- does not display properties of the visceral endoderm ...... 127 Figure 4.6 The definitive endoderm in Foxh1-/- embryos retains further differentiation potential ...... 130
CHAPTER 5
Figure 5.1 Current model of definitive endoderm formation during gastrulation in the mouse embryo ...... 151 Figure 5.2 A novel model of definitive endoderm specification in the mouse embryo...... 153
viii ACKNOWLEDGEMENTS
First and foremost I would like to thank my family, my parents Ron and Alison
Armbruster and my sister Jacqueline Armbruster. Your support and encouragement throughout the years was invaluable.
To my research supervisor and mentor, Dr. Pamela Hoodless; thank you for giving the opportunity to work with you. I have learned so much from you over the years. Your enthusiasm and excitement for research was a source of motivation and inspiration. Thank you also to my thesis committee, Dr. Wan Lam, Dr. Gerry Krystal and Dr. Diana Juriloff.
To the members of the Hoodless lab whom I have had the pleasure to work with over the years: Rachel Montpetit, Rebecca Cullum and Mona Wu – three extraordinary research assistants who kept the lab together. Fellow students Lynn Mar, Pavle Vrljicak, Ali Saleem Hassan, and
Sam Lee. We’ve all been through so many ups and downs together on the journey that is graduate school. Drs. Jim Rupert, Robin Dickinson, Juan Hou and Elizabeth Wederell – I have learned so much from each of you. I will miss working with you all.
A special thank you to my friend and benchmate Rachel Montpetit for her friendship and support throughout the years. Also to Dr. Juan Hou who has taught me so much and contributed to Chapter 4 of this thesis. Also a thank you to Mr. David Lefelaar, my high school biology teacher who inspired me to pursue a career in academic science. Lastly, to my dear friends
Ayron, Lisa, Diana, Charlotte, Jocelyn, Melanie, Lindsay, Christy, Bari, Becky, Jillian, and
Wendy. Thank you for your love and support. You were always there when I needed you.
Financial support for my graduate career was provided by the Cordula and Gunter
Paetzold University of British Columbia University Graduate Fellowship and a Postgraduate
Scholarship (Doctoral) from the Natural Sciences and Engineering Council of Canada.
ix DEDICATION
To my family and friends who enrich my life and fill it with love and laughter, and to my mentors who have taught and inspired me
The most exciting phrase to hear in science, the one that heralds new discoveries, is not Eureka! (I found it!) but rather, "hmm.... that's funny...."
- Isaac Asimov -
x CO-AUTHORSHIP STATEMENT
CHAPTER 2
McKnight KD and Hoodless PA. (2009) A differential screen for genes involved in specification of the anterior primitive streak using Foxh1 deficient embryos. To be submitted for publication.
KDM and PAH designed the research program. KDM performed the research, analyzed the data, and prepared the manuscript.
CHAPTER 3
McKnight KD, Hou J, Hoodless PA. (2007) Dynamic expression of thyrotropin-releasing hormone in the mouse definitive endoderm. Developmental Dynamics. 236:2909-2917.
KDM and PAH designed the research program. KDM performed the research and prepared the manuscript. KDM, JH, and PAH analyzed the data. JH contributed reagents.
CHAPTER 4
McKnight KD, Hou J, Hoodless PA. (2008) The anterior primitive streak is not required for midgut and hindgut definitive endoderm formation in the mouse embryo. To be submitted to
Developmental Biology.
KDM, JH, and PAH designed the research program. KDM and JH performed the research and analyzed the data. KDM prepared the manuscript. JH performed the experiments presented in Figure 4.6.
xi Chapter 1
INTRODUCTION
1.1 GENERAL OVERVIEW OF THE EMBRYONIC DEVELOPMENT OF MUS
MUSCULUS
Embryogenesis results in the transformation of a single totipotent cell into a wide variety
of cell types with specialized functions that become organized into complex structures that
comprise the adult body form. This process requires the spatial and temporal regulation of cell
fate determination, differentiation, and migration. Embryonic development in the mouse can be
divided into the following stages – pre-implantation (cleavage and blastulation), post-
implantation (gastrulation), organogenesis, and fetal growth and development.
1.1.1 An introduction to the lineages constituting the mouse embryo
Formation of the different cell lineages of the mouse follows a very stereotypical pattern resulting in the transformation of a single cell, the fertilized egg, into the tissues that will form the adult organism as well as the supporting structures, such as the placenta, that are essential for embryogenesis. The lineages constituting the mouse embryo are summarized in Figure 1.1.
Diagrams of pre-implantation, post-implantation, and early organogenesis stage embryos can be found in Figure 1.2.
1
Figure 1.1 The lineage hierarchy constituting the mouse embryo
For a description see Section 1.1.1. Colored boxes indicate relevant lineages to this thesis, colored as follows: Grey – totipotent and pluripotent cell lineages, Green - extraembryonic endoderm lineages, Yellow – embryonic endoderm lineages, Red - mesoderm lineages, Blue – ectoderm lineages, Pink – germ cell lineage, White – extraembryonic lineages. Figure recreated from Manipulating the Mouse Embryo: A laboratory manual. Third Edition (Nagy, 2003).
2
Figure 1.2 Schematic diagrams of mouse embryos from pre-implantation, post- implantation, and early organogenesis stages
For detailed description see Section 1.1.1. Implantation occurs at E4.5. Gastrulation begins at E6.5. Grey – primitive ectoderm (inner cell mass/epiblast), Pink – mural trophectoderm and ectoplacental cone, Brown – extraembryonic ectoderm. Green- primitive endoderm (visceral endoderm), Blue – ectoderm, Yellow – definitive endoderm, Red – mesoderm. Dashed line indicates extent of primitive streak.
3 1.1.1.1 The first lineages are specified during the pre-implantation stage
The pre-implantation stage of embryogenesis begins when the single-cell fertilized egg undergoes 3 rounds of cleavage (cell division) to give rise to the 8-cell morula. Following formation of the morula, the embryo undergoes the process of compaction at approximately embryonic day (E) 2.5 to form the blastocyst. Following compaction, the cells flatten and acquire an apical-basal polarity. Cells that end up on the inside of the compacted embryo give rise to the inner cell mass whereas the outer cells give rise to the trophectoderm (reviewed in Pedersen,
1986). Thus, during the process of blastulation the first lineage commitment decision occurs in the embryo, resulting in the formation of two distinct cell lineages. The trophectoderm overlying the inner cell mass is known as the polar trophectoderm and the cells surrounding the blastocyst cavity constitute the mural trophectoderm. The trophectoderm contributes to the extraembryonic ectoderm, parietal yolk sac and the chorion whereas the inner cell mass generates all the tissues that will give rise to the embryo proper as well as the supporting extraembryonic tissues. The second lineage decision that occurs is marked by the appearance of the primitive endoderm
(hypoblast) on the surface of the inner cell mass cells. The hypoblast cells have been shown to contribute to the extraembryonic parietal and visceral endoderm of the yolk sac that surrounds the developing embryo (Gardner, 1997). The remaining cells of the inner cell mass become organized into a layer known as the epiblast (primitive ectoderm), which will give rise to the ectodermal, mesodermal, and endodermal tissues of the fetus, to the germ cells, and to the mesodermal components of the extraembryonic membranes and placenta. At E4.5 the blastocyst attaches to the uterine wall in the process of implantation.
4 1.1.1.2 The primary germ layers are formed during the post-implantation stage
At the time of implantation the mouse embryo is composed of three distinct tissue lineages: trophectoderm, hypoblast (primitive endoderm), and epiblast (primitive ectoderm). The cells of the epiblast continue to divide and expand down into the blastocyst cavity. Shortly after implantation the epiblast cells undergo cavitation, where the cells become organized into an epithelial layer surrounding a small central cavity (the proamniotic cavity).
Gastrulation is the process by which the three primary germ layers – the mesoderm, ectoderm, and definitive endoderm – are generated from the epiblast and organized into the fetal body plan (reviewed in Tam and Behringer, 1997). The ectoderm forms the skin and tissues of the central nervous system, the mesoderm forms the skeleton, muscles (including heart), connective tissues, blood, and kidneys while the definitive endoderm forms the liver, pancreas, lungs and epithelial lining of the digestive tract (stomach, duodenum, colon). Prior to the onset of gastrulation the embryo proper consists of two cell layers – the inner epiblast, that will give rise to all the tissues of the embryo, is surrounded by a layer of extraembryonic visceral endoderm that is derived from the primitive endoderm and will contribute to the yolk sac and other supporting tissues.
Gastrulation begins at E6.5 with the formation of a transient structure called the primitive streak in a localized region of the epiblast adjacent to the embryonic-extraembryonic junction on the prospective posterior side of the embryo (reviewed in Tam and Behringer, 1997). The epiblast cells in the region of the streak undergo an epithelial to mesenchymal transformation, delaminate, and ingress through the streak emerging between the epiblast and visceral endoderm to form the mesoderm, or intercalating into the outer visceral endoderm forming the definitive
5 endoderm (see Section 1.3). The epiblast cells that do not move through the primitive streak will give rise to the ectoderm. By E7.5 gastrulation is complete with the definitive endoderm as the outer layer of the embryonic cup overlying the mesoderm and ectoderm below. The mouse embryo changes substantially between E7.5 and E8.5, where the three primary germ layers begin to differentiate into their derivative tissues and are organized into the characteristic body plan.
1.2 THE MOUSE GASTRULA ORGANIZER
Gastrulation results not only in the generation of the three primary germ layers but also the establishment of the body plan of the embryo. The mammalian body plan consists of three axes: anterior-posterior, dorsal-ventral, and left-right. A key player in axis specification and germ layer patterning is a group of cells referred to as the embryonic organizer. The concept of an embryonic organizer was first introduced by the elegant transplantation studies of Spemann and Mangold using salamander embryos in the 1920’s (reviewed in Gilbert and Saxen, 1993).
They demonstrated that transplantation of the dorsal lip of the blastopore to an ectopic location on a host embryo induced the formation of a secondary body axis and induced de novo differentiation of the host tissue. Equivalent structures have been characterized in numerous model organisms, including the dorsal lip of the blastopore (Spemann organizer) in Xenopus laevis, Hensen’s node in Gallus gallus (chicken), the embryonic shield in Danio renio
(zebrafish), and the anterior primitive streak in Mus musculus (mouse) (Beddington, 1994;
Nieuwkoop, 1985; Shih and Fraser, 1996; Storey et al., 1995). Cells found within the organizer in all species show similar tissue fates and are the primary source of the axial mesoderm, which includes the prechordal plate and the notochord (Beddington, 1994; Kinder et al., 2001; Sulik et al., 1994). The axial mesoderm tissues play a critical role in patterning the embryonic axes, discussed in Section 1.2.2.
6 1.2.1 The mouse gastrula organizer is located at the anterior end of the primitive streak and contributes to the midline tissues of the embryo
The existence of the mouse embryonic organizer was first demonstrated by Rosa
Beddington in 1994 (Beddington, 1994). These experiments showed that transplantation of the cells located at the anterior end of the primitive streak to the lateral side of a similar staged embryo induced a secondary axis and differentiation of the host tissue. Since then, many studies have demonstrated and characterized the inductive capacity of this group of cells. The mouse organizer is specified at the onset of gastrulation and can be detected by the expression of the transcription factors Foxa2 and goosecoid at the anterior end of the primitive streak (APS) (Ang and Rossant, 1994; Blum et al., 1992; Monaghan et al., 1993). Cell populations that display organizer activity have been identified in early-streak, mid- to late-streak and early bud stage embryos. These organizer populations have been termed the early gastrula organizer (EGO), mid gastrula organizer (MGO), and node respectively (reviewed in Davidson and Tam, 2000). It is important to note that these different terms (EGO, MGO, node) do not reflect separate cell populations, but rather reflect the name of the organizer population at different embryonic stages.
This is diagramed in Figure 1.3. These cell populations differ slightly in their genetic activity, with the later organizer populations (the MGO and node) expressing Chordin, Noggin, Shh, and
Brachyury in addition to Foxa2 and goosecoid which are the only factors expressed in the EGO.
Further transplantation experiments have demonstrated that the organizer populations present at various stages of gastrulation differ slightly in their axis- inducing capabilities
7
Figure 1.3 The mouse gastrula organizer is located at the anterior end of the primitive streak throughout gastrulation and gives rise to midline tissues of the embryo
The cells at the anterior end of the primitive streak (orange) have been shown to have organizer properties by transplantation experiments. These experiments have also shown that these cells give rise to tissues the lie in the midline of the embryo – the axial mesendoderm (anterior definitive endoderm (yellow) and prechordal plate (red)), the notochord (red), and the floor plate of the neural tube (not shown).
8 (Kinder et al., 2001). The EGO and MGO have slightly different axis inducing capabilities.
Transplantation of the EGO and node generates a second body axis in host embryos, however,
the MGO has the ability to induce a complete secondary body axis with anterior and posterior molecular characteristics. Lineage tracing and cell ablation studies have clearly demonstrated that the cells of the anterior primitive streak give rise to tissues that lie in the midline of the embryo, specifically the prechordal plate, anterior definitive endoderm, node, notochord, and floor plate (reviewed in Camus and Tam, 1999). This is diagramed in Figure 1.3. The EGO,
MGO, and node have also been shown to differ slightly in their tissue fates (Kinder et al., 2001).
The EGO contributes primarily to the prechordal plate underlying the forebrain and the anterior part of the notochord. Cells derived from the MGO are found in the anterior axial mesendoderm, the foregut definitive endoderm and part of the trunk notochord. Lastly, cells of the node are found in the posterior region of the trunk notochord and the gut endoderm (Kinder et al., 2001;
Sulik et al., 1994).
1.2.2 Tissues derived from the anterior primitive streak contribute to the patterning of all three body axes
As discussed above, the mammalian body plan consists of three body axes that are specified in the following order: anterior-posterior, dorsal-ventral, and left-right. The tissues derived from the anterior primitive streak play key roles in the establishment and/or maintenance of each of these axes. The anterior-posterior axis is specified prior to the onset of gastrulation by the anterior visceral endoderm (also referred to as the “head organizer”). The anterior visceral endoderm is the source of antagonists of the Nodal signaling pathway, ultimately resulting in the restriction of Nodal activity to the presumptive posterior side of the embryo where the primitive streak will form (Perea-Gomez et al., 2002). The initial signals that establish the anterior region
9 of the embryo require a secondary reinforcing signal which is provided by the anterior definitive
endoderm and prechordal plate (anterior mesendoderm) which is a tissue derived from the
anterior primitive streak. In agreement with this, embryos that do not specify the anterior
primitive streak or lack the anterior mesendoderm have varying degrees of anterior truncations
(reviewed in Camus and Tam, 1999). The anterior primitive streak has also been shown to give
rise to the node (posterior notochordal plate), notochord, and the floor plate of the neural tube.
Signals from the notochord and floor plate act to provide dorsal-ventral pattern to the neural
tube, which is essential for the proper specification and localization of different neural cell types
(Wilson and Maden, 2005). In embryos lacking the notochord or floor plate this results in a loss
of ventral structures in the brain and spinal cord. Lastly, the node (which is derived from the
anterior primitive streak) plays a critical role in the initial breaking of left-right symmetry in the
embryo and the notochord acts to reinforce this asymmetry by acting as a midline “barrier”
preventing the spread of left-specific signals to the right side of the embryo (Yamamoto et al.,
2003). Loss of the node or notochord results in aberrant expression of left-sided markers, such as
Pitx2 (Davidson et al., 1999).
1.2.3 Specification of the anterior primitive streak
One of the major questions yet to be answered is how cells with organizer properties are
induced in the mouse embryo. In other vertebrate embryos, such as Xenopus laevis and Gallus gallus, the organizer is established by inductive interactions with non-embryonic tissues, however, an “organizer inducer” has not yet been discovered in the mammalian embryo
(Bachvarova et al., 1998; De Robertis et al., 2000; Harland and Gerhart, 1997). It has been established, however, that the Nodal signaling pathway plays an evolutionarily conserved role in specification of the organizer in numerous species, including mouse, frog, fish, and chick
10 (reviewed in Camus and Tam, 1999). Gene targeting experiments in the mouse have
demonstrated that complete loss of Nodal signaling results in a failure of the embryos to
gastrulate, while Nodal hypomorphs display varying defects based on the amount of residual
Nodal signaling remaining. Genetic manipulation of Nodal levels in the mouse embryo has
revealed that the highest levels of nodal signaling are involved in specification of the anterior
primitive streak and its derivative tissues while lower levels are involved in mesoderm
specification (Vincent et al., 2003). The Nodal signaling pathway will be discussed in further detail in Section 1.4. A role for the forkhead proteins Foxh1 and Foxa2 have also been shown to be essential for specification of the anterior primitive streak. These proteins are discussed further in Section 1.5.
1.3 THE MOUSE DEFINITIVE ENDODERM
The mouse embryo contains three distinct endoderm lineages – the parietal endoderm, the visceral endoderm, and the definitive endoderm. As discussed above, one of the first lineage divisions to occur in the mouse embryo is the formation of the epiblast (primitive ectoderm) and
hypoblast (primitive endoderm) from the inner cell mass of the blastocyst. The parietal and
visceral endoderm are derived from the hypoblast and contribute to the extraembryonic tissues,
including the yolk sac. The definitive endoderm, however, is derived from the epiblast during the
process of gastrulation and will generate the epithelial lining of the gastrotintestinal tract and
lungs, as well as form the pancreas, liver, and thyroid.
11 1.3.1 Formation and morphogenesis of the definitive endoderm
The formation and morphogenesis of the definitive endoderm is diagramed in Figure 1.4.
Prior to gastrulation the mouse embryo is a bilayered cup consisting of the inner epiblast layer and the outer visceral endoderm layer. As discussed above gastrulation is the process by which the three primary germ layers of the embryo – the ectoderm, mesoderm, and definitive endoderm
– are generated from the epiblast. During this process a transient structure called the primitive streak forms on the posterior of the embryo. Epiblast cells in the vicinity of the streak undergo an epithelial-to-mesenchymal transformation and move through the streak, emerging as mesoderm or definitive endoderm. Fate mapping studies of pre-streak and early-streak embryos demonstrated that the precursors of the definitive endoderm are found in the epiblast adjacent to the anterior domain of the primitive streak (Lawson et al., 1986). Studies of mid-streak embryos further demonstrated that the endoderm cells overlying the anterior primitive streak contributed to the anterior definitive endoderm (Lawson and Pedersen, 1987; Tam et al., 2007).
As the definitive endoderm cells emerge from the primitive streak they first intercalate into the visceral endoderm layer, and then subsequently expand to cover the entire embryonic region of the embryo, thereby displacing the visceral endoderm proximally to the extraembryonic region (Lawson et al., 1986; Lawson and Pedersen, 1987; Tam et al., 2007).
Thus at the completion of gastrulation, the mouse embryo consists of three tissue layers with the definitive endoderm overlying the mesoderm and ectoderm. Once formed, the single cell layered epithelial sheet of definitive endoderm undergoes morphogenesis to form the gut tube. Between
E7.5 and E8.0 the foregut and hindgut pockets form, generating the anterior and posterior intestinal portals (Franklin et al., 2008; Wells and Melton, 1999). The anterior and posterior pockets can be further subdivided into ventral and dorsal regions, such that the ventral foregut
12
Figure 1.4 Formation and morphogenesis of the mouse definitive endoderm
For detailed description see Section 1.3.1. Briefly, the definitive endoderm forms during gastrulation. As the newly formed definitive endoderm cells (yellow) emerge from the primitive streak they displace the visceral endoderm (green) towards the proximal region of the embryo. At the end of gastrulation the definitive endoderm is a single cell layered epithelium covering the embryonic region. Following gastrulation the endoderm undergoes several morphogenetic movements to form the gut tube, beginning at the anterior with the formation of the foregut and then in the posterior with the hindgut. The derivative organs of each of the regions of the gut tube are shown. Figure adapted from Wells and Melton, 1999.
13 endoderm is adjacent to the cardiogenic mesoderm and the dorsal foregut endoderm is adjacent
to the notochord. The anterior part of the gut tube, the foregut, forms first at E8.0 and originates from the endoderm in the most anterior portion of the embryo. The foregut contributes to a large number of organs, including the thyroid, thymus, esophagus, taste buds, trachea, lungs, liver, and ventral pancreas. The hindgut forms shortly after the foregut and is derived from the most posterior region of the endoderm overlying the primitive streak. The hindgut contributes to the small and large intestine as well as to the bladder and the urogenital tract. The open portion of the gut tube during these stages is referred to as the midgut. This is the last region of the endoderm to form a tube and contributes to the stomach, dorsal pancreas, and the duodenum
(reviewed in Wells and Melton, 1999).
The closure of the gut tube is linked to the change in body position of the embryo from the lordotic to the fetal position during the process of turning. When the embryo turns the endoderm layer is internalized, allowing the lateral walls of the endoderm sheet to move together and fuse, forming a tube. As the gut tube forms, the organ derivatives of the definitive endoderm are generated from organ buds that form at distinct locations. The domains that will generate the various derivative organs are specified by inductive interactions with the mesoderm or ectoderm.
Initially the gut tube is closed at both ends, but later in development it becomes perforated where the endoderm contacts the anterior and posterior ectoderm to form the mouth and anus, respectively. Subsequent stages of endoderm differentiation and organ development are beyond the scope of this thesis and will not be discussed here.
14 1.3.2 Regulation of definitive endoderm cell fate
As discussed in Section 1.4.2.1, the Nodal signaling pathway regulates cell fates in the
primitive streak of the mouse embryo. Specifically, the highest levels of Nodal signaling are
required for endoderm specification while lower levels specify mesoderm (Lowe et al., 2001). In all vertebrates analyzed, the Nodal signaling pathway is absolutely required, and in some cases sufficient, to initiate endoderm development. Studies in frog, zebrafish, chick, and mouse have revealed a conserved molecular pathway controlling vertebrate endoderm development
(reviewed in Zorn and Wells, 2007). Further studies of mouse mutants have demonstrated that graded Nodal signaling also governs endoderm cell fate decisions, with higher Nodal signaling specifying more anterior fates. Consistent with this, embryos harboring hypomorphic alleles of
Nodal or Smad2 display defects in the anterior (foregut) definitive endoderm, while posterior
(hindgut) endoderm formation is largely unaffected suggesting that the anterior definitive endoderm is more sensitive to reduction in Nodal signaling than the posterior definitive endoderm (Vincent et al., 2003). As discussed in Sections 1.4.2.1 and Section 1.5.2.2, embryos lacking Foxh1, a downstream target of Nodal, also fail to specify the anterior definitive endoderm (Hoodless et al., 2001; Yamamoto et al., 2001). Furthermore, chimera studies have demonstrated that embryonic stem cells lacking Smad2 or Nodal are unable to contribute to the definitive endoderm layer (Tremblay et al., 2000; Varlet et al., 1997). Foxh1 null embryonic stem cells are also deficient in their ability to contribute to the foregut definitive endoderm, however, they can still contribute to the hindgut definitive endoderm suggesting that other
Smad2 interacting proteins must regulate the Nodal-dependent specification of the posterior endoderm (Hoodless et al., 2001). This is discussed further in Chapter 4 of this thesis.
15 Aside from the Nodal pathway, several other signaling molecules and transcription
factors have been shown to be essential for endoderm specification in the mouse embryo.
Embryos lacking both Chordin and Noggin, secreted BMP antagonists expressed in the distal
primitive streak, node, and its derivatives, show a marked reduction in definitive endoderm at
early gastrulation resulting in a reduction in the pharyngeal endoderm at later stages (Bachiller et
al., 2000). Tcf1-/-;Tcf4-/- embryos have defects in hindgut expansion and an anterior
transformation of the duodenum (Gregorieff et al., 2004). Sox17 deficient embryos also have
severe hindgut defects that are apparent after gastrulation; however, the foregut is relatively
unaffected (Kanai-Azuma et al., 2002). Embryos lacking the homeodomain protein Mixl1 do not
have any overt endoderm specification defects; however, Mixl1 null embryonic stem cells are unable to contribute specifically to the hindgut definitive endoderm (Hart et al., 2002). As discussed in Section 1.5.2.2 the forkhead transcription factor Foxa2 plays an essential role in
endoderm specification. Chimera studies have demonstrated that Foxa2 is primarily required in the foregut and midgut definitive endoderm and not required for the formation of the hindgut definitive endoderm (Dufort et al., 1998). Embryos lacking both Tead1 and Tead2 are deficient in posterior endoderm (hindgut) (Sawada et al., 2008). Lastly, embryonic stem cells lacking
Danforth’s Short Tail (Sd) do not contribute to the definitive endoderm in chimera studies
(Maatman et al., 1997). Thus it appears that the Nodal pathway is at the top of a molecular hierarchy regulating definitive endoderm formation, and that this pathway regulates the expression of other factors, including those in the Sox, Mix, and Fox families, that are required for specification of different subsets of the definitive endoderm.
16 1.3.3 Regionalization of the definitive endoderm
At the end of gastrulation the definitive endoderm is a single-cell layered epithelium
overlying the mesoderm and ectoderm below that surrounds the entire embryonic region of the
embryo proper. Although morphologically homogeneous at this stage, several anterior-posterior
differences do exist. First, the cup of definitive endoderm can be divided based upon where in
the gut tube the cells will be located. As described in Section 1.3.1 the endoderm overlying the
anterior region of the cup will form the foregut and the endoderm overlying the primitive streak in the posterior region will form the hindgut. These regions are then further divided into domains that are defined by what organ derivatives it will give rise to. Second, the timing of endoderm formation in the primitive streak influences its final localization, where the first endoderm to exit the primitive streak during gastrulation migrates anteriorly and will contribute to the foregut definitive endoderm while the endoderm that forms later will contribute to the mid- and hindgut definitive endoderm (Franklin et al., 2008; Tam et al., 2007). Thus at the time of formation of the gut invaginations, the foregut endoderm is slightly older than the hindgut endoderm. However, the hindgut definitive endoderm has a higher cell division rate (Wells and Melton, 1999). Third, several differences exist in gene expression along the anterior posterior axis of the endoderm
cup. Genes such as Cer1, Otx2, and Hesx1 are expressed in the anterior region while Cdx2 and
Fabp2 are expressed in the posterior region (reviewed in Wells and Melton, 1999). Lastly, as
described in Section 1.3.2 Smad2, Foxh1, Foxa2, Sox17, and Mixl1 appear to regulate the
specification of different subpopulations of the definitive endoderm. Thus it appears that the
initial regionalization of the definitive endoderm occurs concurrently with its formation during
gastrulation as the endoderm cells exit the primitive streak.
17 Despite the above lineage and gene expression evidence to suggest that the endoderm
cells are patterned at the end of gastrulation, tissue recombination experiments have
demonstrated that these cells are not yet committed to specific lineages (Horb, 2000; Kim et al.,
1997; Wells and Melton, 2000). These experiments also demonstrated that the endoderm acquires some positional identity and morphogenetic information from the adjacent mesodermal tissues. The anterior endoderm is in contact with the notochord while the posterior endoderm is in contact with the node and primitive streak, thus these endoderm regions are likely subject to different signaling influences from these different tissues. These recombination experiments demonstrated that endoderm co-cultured with different anterior or posterior populations of mesoderm will adopt the anterior or posterior character of the co-cultured mesoderm, suggesting that the endoderm cells at this stage still retain a high degree of developmental plasticity. These studies also identified Fgf4 as playing a role in the induction of posterior endoderm fates and repression of anterior fates. Thus it appears that the endoderm receives some degree of pattering information during gastrulation, but that it is not yet determined as it can change its fate upon receiving patterning cues.
1.4 THE NODAL SIGNALING PATHWAY
The Transforming Growth Factor Beta (TGFβ) signaling pathway is involved in many cellular processes in both the adult organism and the developing embryo, including cell growth, cell differentiation, apoptosis, cellular homeostasis and other cellular functions (reviewed in
Massague, 1998). There are two main branches of the TGFβ pathway, divided based upon the ligands, receptors, and intracellular effectors used. One branch utilizes ligands from the
TGFβ/Nodal/Activin family and the intracellular effectors Smad2 and Smad3, while the other branch utilizes ligands from the BMP family and the intracellular effectors Smad1, Smad5, and
18 Smad8. The basic mechanism of signal transduction of this pathway involves ligand binding to a
heteromeric receptor complex which then activates intracellular effector molecules which, upon
activation, move to the nucleus and interact with other DNA binding factors to regulate gene expression.
1.4.1 Components of the Nodal signaling pathway
The Nodal signaling pathway has been shown to play an important role in specification of the anterior-posterior and left-right body axes, formation of the mesoderm and endoderm during gastrulation, and patterning of the neural tissues (reviewed in Schier, 2003; Shen, 2007). A diagram of the key components of the Nodal pathway can be found in Figure 1.5. Nodal belongs to the TGFβ superfamily of signaling ligands and is only found in vertebrates. Like most TGFβ
superfamily ligands, Nodal is synthesized as a proprotein and is processed by the subtilisin-like
proprotein convertases Furin (Spc1, Pace) and Pace4 (Spc4, Pcsk6). It is thought that these
convertases may play an important regulatory role in regulating the spatial and temporal
distribution of active Nodal ligand as they have been found to process Nodal intracellularly and
extracellularly (Beck et al., 2002).
All ligands of the TGFβ superfamily bind as dimers to a complex consisting of type I and
type II serine/threonine kinase single-pass transmembrane receptors. Nodal ligands activate
intracellular events by assembling a complex consisting of two type I and two type II receptor
serine/threonine kinases and an EGF-CFC co-receptor. In the absence of ligand, the type II and
type I receptors exist as homodimers at the cell surface. Ligand initially binds to the type II
receptor and the type I receptor then recognizes the ligand-bound type II receptor. Ligand
binding causes the type II receptor to phosphorylate and activate the type I receptor in a cluster
19
Figure 1.5 The Nodal signaling pathway
See Section 1.4.1 for details. Adapted from Shen, 2007.
20 of five serine and threonine residues in the Gly-Ser (GS) domain of the type I receptor. Nodal
acts via the type I receptor ALK4 (ActRIB/Acvr1b) and the type II receptors ActRII (ActrIIA;
Acvr2a) or ActRIIB (AcvrIIb). Unlike other ligands in the TGFβ superfamily, Nodal requires the
presence of a co-factor to activate downstream signaling events. In the mouse, the Nodal co-
factors are encoded by the Cripto (Tdgf1) and Cyptic (Cfc1) genes, which are members of the
EGF-CFC family of extracellular, GPI-linked proteins (Yan et al., 2002; Yeo and Whitman,
2001). It is thought that regulation of the EGF-CFC proteins provides another mechanism by which to regulate Nodal activity (Gritsman et al., 1999).
Activation of Nodal signaling ultimately results in the regulation of gene transcription in the nucleus. Thus, the signal must be transferred from the receptors located at the plasma membrane to the nucleus. The intracellular effectors of the Nodal pathway are the receptor- associated Smads (R-Smad), Smad2 and Smad3, and the common mediator-Smad (co-Smad),
Smad4. The activated type I receptor phosphorylates and activates the R-Smads at an SSxS motif in the MH2 domain in the C-terminus. Phosphorylation of the Smads reduces their affinity for
SARA, a protein that recruits Smads to the membrane, and increases their affinity for Smad4.
The phosphorylated R-Smads then form a complex with the Co-Smad, Smad4 causing nuclear translocation. Activated Smad complexes then bind to additional tissue- and site-specific DNA binding co-factors which provide specificity in target gene expression. This is necessary as all
Smad proteins, with the exception of Smad2, bind DNA only weakly via their MH1 domain.
However, this affinity is too low to achieve unassisted binding to DNA. The MH2 domain of
Smads is important for transcriptional activation and has been shown to interact with either transcriptional coactivators or corepressors. The first Smad DNA binding co-factor identified was the forkhead transcription factor Foxh1 (Chen et al., 1996). Nodal has also been shown to regulate gene expression via the homeodomain protein Mixer (Germain et al., 2000). The cell-
21 type specific expression of the Smad-DNA binding cofactors confers cell type specificity to the
Nodal signal. Several lines of evidence indicate that Nodal can affect gene expression in a
Smad4 or Foxh1 independent manner, suggesting that other transcription factors involved in the regulation of Nodal downstream genes remain to be identified.
Aside from regulation of Nodal expression or activity via its processing or co-receptors,
Nodal is also regulated by the extracellular antagonists Lefty1, Lefty2, and Cerberus 1 homolog
(Cer1). Lefty molecules are related to the TGFβ superfamily of ligands and are thought to inhibit
Nodal signaling by binding to Nodal directly or to the EGF-CFC co-receptor (Chen and Shen,
2004; Chen and Schier, 2002). Leftys have been shown to act as feedback inhibitors of Nodal signaling as, in most tissues, Lefty expression is dependent upon Nodal signaling (Meno et al.,
1998). Cer1 belongs to the DAN family of cysteine knot proteins and acts to block Nodal signaling by binding Nodal directly (Belo et al., 1997; Shawlot et al., 1998). In addition to inhibiting Nodal signaling, Cer1 also blocks Bone Morphogenetic Protein (BMP) signaling
(Piccolo et al., 1999). Nodal signaling is also regulated intracellularly by the nuclear proteins
Dr1/Drap1 (NC2β/NC2α) and Arkadia. Dr1/Drap1 (NC2β/NC2α) is a general factor that represses basal transcription by RNA polymerase II, however, it can specifically block Nodal signaling by binding to Foxh1 and preventing its binding to DNA (Iratni et al., 2002). Arkadia is a RING-domain E3 ubiquitin ligase and has been shown to enhance Nodal signaling by promoting the ubiquitination and turnover of activated Smad2/Smad3 complexes (Episkopou et al., 2001; Levy et al., 2007; Mavrakis et al., 2007; Niederlander et al., 2001).
22 1.4.2 Regulation of cell fate decisions in the mouse embryo by the Nodal pathway
Nodal is dynamically expressed in the post-implantation mouse embryo (Conlon et al.,
1994; Varlet et al., 1997). Expression is first detected ubiquitously in the epiblast prior to gastrulation. Nodal is then restricted to the prospective posterior of the embryo where the primitive streak will form. At the completion of gastrulation (E7.0-E7.5), Nodal is expressed in the cells surrounding the node (posterior notochordal plate). At E8.5 Nodal is expressed asymmetrically in the lateral plate mesoderm on the left side of the embryo. In accordance with the localization of its expression, Nodal has been shown to regulate embryonic patterning prior to, during, and following gastrulation. Studies of Nodal signaling in various model organisms have described three properties of Nodals (reviewed in Schier, 2003; Shen, 2007). First, Nodals are produced locally and act as both long- and short-range signals to affect cell fate and tissue patterning. Second, different levels of Nodal signaling have been shown to induce different cell fates by acting as a morphogen. Third, both positive and negative regulatory loops act to regulate the activity of Nodal activity. The role of Nodal signaling in the mouse embryo has been elucidated using gene targeting strategies that have analyzed the role of Nodal as well as other components of the signaling pathway.
1.4.2.1 There is a graded response to Nodal signaling within the primitive streak of the mouse embryo
Nodal signaling has been shown to play a key role in several processes in the embryo, including axis specification, axial mesendoderm (prechordal plate, anterior definitive endoderm, node, and notochord) induction, neural patterning, and left-right development (reviewed in
Schier, 2003; Shen, 2007). Consistent with its expression in the early embryo, genetic ablation of
23 Nodal results in patterning defects and embryonic lethality. Nodal413.d embryos contain a
retroviral integration into the first intron of the Nodal gene, eliminating Nodal transcription and resulting in a null allele (Conlon et al., 1994). Homozygous Nodal413.d/413.d embryos arrest at the gastrulation stage of development, do not specify the anterior-posterior axis, and fail to form a primitive streak and thus lack definitive endoderm or mesoderm. Chimera studies further demonstrated that Nodal is required in the epiblast for primitive streak formation and gastrulation, and is required in the visceral endoderm for patterning of anterior neural tissues
(Varlet et al., 1997). Embryos harboring a hypomorphic allele of Nodal (Nodalfl1/Δ) display a range of phenotypes, including improper positioning of the anterior-posterior axis, patterning of the anterior neural tissues, formation of the axial mesendoderm (node, notochord, prechordal plate, and definitive endoderm), and specification of the left-right axis (Lowe et al., 2001).
Embryos lacking ActRIB, ActRIIA;ActRIIB, Smad2, Smad2;Smad3, Smad4, Cripto, or
Furin;Pace4 do not specify the anterior-posterior axis, fail to gastrulate and as a result lack mesoderm and definitive endoderm (Beck et al., 2002; Ding et al., 1998; Dunn et al., 2004; Gu et al., 1998; Nomura and Li, 1998; Sirard et al., 1998; Song et al., 1999; Waldrip et al., 1998;
Weinstein et al., 1998). Some degree of redundancy exists among some members of the Nodal pathway, as loss of Smad3, ActRIIA, ActRIIB, Lefty1, and Cer1 do not affect gastrulation (Belo et al., 1997; Datto et al., 1999; Matzuk et al., 1995; Meno et al., 1998; Oh and Li, 1997; Shawlot et al., 1998; Simpson et al., 1999; Stanley et al., 2000; Yang et al., 1999; Zhu et al., 1998).
Consistent with a role for Nodal signaling in mesoderm formation, loss of the Nodal pathway antagonists Drap1 and Lefty2 have an expanded primitive streak and excess mesoderm (Iratni et al., 2002; Meno et al., 1998). The phenotype of Drap1 null embryos is partially rescued by
Nodal heterozygosity. Compound mutants for Lefty1 and Cer1 have an expanded anterior primitive streak and fail to form middle streak derivatives or develop multiple primitive streaks
24 (Perea-Gomez et al., 2002). These phenotypes are also partially rescued in mice heterozygous for
a null allele of Nodal.
The role(s) of Nodal signaling in later stages of development (mesendoderm induction, neural patterning, and left-right development) has been demonstrated using hypomorphic alleles,
lineage-specific gene inactivation and chimera analysis. These studies demonstrated that there is
a dose dependent effect of Nodal signaling on cell fate decisions in the early embryo. Removal
of Smad4 specifically in the epiblast results in a loss of all anterior streak derivatives, including
the prechordal plate, anterior definitive endoderm, node, and notochord (Chu et al., 2004). Loss
of Foxh1 also results in a loss of all anterior streak derivatives, as these embryos lack the prechordal plate, anterior definitive endoderm, node and notochord (Hoodless et al., 2001;
Yamamoto et al., 2001). Removal of one copy of Smad3 in the context of a Smad2 deficient epiblast results in a failure to specify all axial midline tissues (the node, notochord, and prechordal plate) (Vincent et al., 2003). There are severe anterior truncations and the anterior gut is absent although some hindgut does form. No notochord is formed and as a result the somites are fused across the midline. Loss of Arkadia, which positively regulates Nodal signaling, results in a loss of the anterior primitive streak and its derivatives (Episkopou et al., 2001; Niederlander et al., 2001).
Various studies have characterized several regions of the Nodal promoter that regulate
Nodal expression in various regions of the embryo (Brennan et al., 2001; Norris and Robertson,
1999; Vincent et al., 2003). Based on these studies, Nodal expression was selectively removed from the proximal epiblast and primitive streak by deleting the proximal epiblast enhancer of
Nodal (Nodal∆PEE) (Vincent et al., 2003). In these embryos, there is a failure to specify the
anterior streak derivatives (the prechordal plate and anterior definitive endoderm), however, the
25 node, notochord, and floor plate still form. As a result of the failure to specify the prechordal
plate and anterior definitive endoderm these embryos have severe anterior truncations and the
gut tube is poorly elaborated in the anterior region; however, a distinct hindgut is present.
Mosaic deletion of Nodal in the epiblast using Mox2-Cre also results in a disruption of anterior
mesendoderm (prechordal plate and anterior definitive endoderm) formation resulting in varying
degrees of anterior truncations (Lu and Robertson, 2004).
Genetic manipulation of Smads in the mouse embryo results in a similar range of
phenotypes as described above in Nodal mutants. Smad4 chimeras (where extraembryonic
tissues are wild-type and embryonic derivatives lack Smad4) have anterior truncations, due to the
loss of anterior streak derivatives (Chu et al., 2004; Sirard et al., 1998; Sirard et al., 2000; Yang
et al., 1998). Removal of Smad2 specifically from the epiblast using Sox2-Cre results in a failure
to specify the anterior definitive endoderm and prechordal plate progenitors resulting in anterior
truncations; however, primitive streak formation and gastrulation appear normal and the node,
notochord, and floor plate still forms (Vincent et al., 2003). Similar to Nodal∆PEE embryos, a gut
tube forms but it is poorly elaborated in the anterior region while hindgut formation appears
normal. Embryos lacking one allele of both Smad2 and Smad3 display phenotypes ranging from
mild craniofacial and midline defects (such as holoprosencephaly) to severe midline defects that lead to embryonic lethality (Liu et al., 2004). It was postulated that these defects resulted from improper specification of the anterior definitive endoderm. However, other groups have reported no phenotype in Smad2+/-;Smad3+/-. This discrepancy is most likely due to differences in genetic
background (Dunn et al., 2004). AcrRIIA-/-;ActRIIB+/- embryos have variable defects in primitive
streak formation and various anterior truncations and patterning defects (Song et al., 1999). Loss
of Smad3 in the context of one wild-type copy of Smad2 results in impaired production of
26 anterior axial mesendoderm (prechordal plate and anterior definitive endoderm) while the node
and notochord are unaffected (Dunn et al., 2004).
Deletion of various components of the Nodal pathway, including Nodal itself, has
established a role for Nodal signaling in regulating specification of the left-right axis. Briefly,
Nodal has been shown to be required for the initial breaking of symmetry in the node, as well as transferring the initial asymmetric signal to the left side of the embryo (Brennan et al., 2002;
Saijoh et al., 2003). Furthermore, Nodal regulates the formation of midline tissues, such as the
notochord, which maintain the left-right asymmetry (Yamamoto et al., 2003). The role of Nodal in left-right patterning is beyond the scope of this thesis and will not be discussed further here.
Genetic manipulation of the level of Nodal signaling in the mouse embryo has demonstrated that the strength of Nodal signaling selectively regulates cell fate decisions in the primitive streak. Specifically, the highest levels of Nodal signal are required for specification of the prechordal plate and anterior definitive endoderm (axial mesendoderm), intermediate levels are involved in specification of the node and notochord (axial mesoderm), while low levels are involved in paraxial and lateral mesoderm specification. Although high levels of Nodal are required for specification of the anterior definitive endoderm, the role of Nodal in formation of the midgut and hindgut is currently unclear. The role of Nodal signaling in mesendoderm induction is evolutionarily conserved, as loss of function experiments have clearly demonstrated an essential role for Nodal in the induction of most mesodermal and endodermal cell types in other model organisms such as zebrafish (Danio renio), frog (Xenopus laevis), and chick (Gallus gallus) (reviewed in Schier, 2003).
27 1.5 THE FORKHEAD BOX PROTEINS FOXH1 AND FOXA2
Transcription factors are commonly divided into classes based upon their characteristic
DNA binding domains. One such class is the forkhead box (Fox) family of transcription factors which are named for the Drosophila melanogaster gene fork head (fkh), the founding member of this family (Weigel and Jackle, 1990). Mutation of the fork head gene in Drosophila results in
defects in terminal (end) segment differentiation, including gut invagination and head fold
involution, resulting in a characteristic spiked head appearance (Weigel et al., 1989). Over 100
Fox genes have since been identified and genetic analysis has shown many to have important
biological functions, such as regulation of morphogenesis and cell differentiation (reviewed in
Carlsson and Mahlapuu, 2002; Kaufmann and Knochel, 1996). Fox genes are found in organisms
ranging from yeast to humans and the number of forkhead genes appears to have increased
during evolution with greater numbers identified in vertebrates than invertebrates. To date 42
forkhead genes have been identified in humans, while Drosophila only has one member of this
gene family. The increase in diversity and number of Fox proteins correlates with the increasing
anatomical complexity of organisms and this increasing complexity in body plan may have been
a driving force behind the expansion of the forkhead gene family (discussed in Carlsson and
Mahlapuu, 2002).
1.5.1 Foxh1 and Foxa2 are winged helix transcription factors
Members of the forkhead protein family contain a highly conserved 110-amino acid
forkhead domain that functions as a DNA-binding domain (Weigel and Jackle, 1990). This
domain consists of 3 alpha helices and 2 beta sheets in a helix-turn-helix configuration which is
flanked by two loops or “wings” and is often referred to as a “winged helix” domain for this
28 reason (Clark et al., 1993; Lai et al., 1993). A schematic diagram of the protein domains of
Foxh1 and Foxa2 are shown in Figure 1.6. A large proportion of the amino acids in the forkhead
domain are invariant or highly conserved and there is limited variation in the 3D structure and
mode of DNA recognition within the forkhead family (Wijchers et al., 2006). Crystallography studies have provided several insights into the mechanism of DNA binding (Clark et al., 1993).
The forkhead domain makes 14 contacts with DNA ensuring binding specificity and these critical residues have been verified using mutational analysis (Overdier et al., 1994). Studies have also revealed that helix 3 (H3) binds to the major groove of DNA while wing 2 (W2) makes contacts in the minor groove (Clark et al., 1993; Marsden et al., 1997). These factors bind to
DNA as monomers on 15-17 base pair asymmetrical binding sites, with a 7 nucleotide core corresponding to where H3 binds the major groove. As expected from their similar structure, numerous known DNA target sites for different forkhead proteins share a common recognition motif. This 7 nucleotide core contains the consensus binding sequence of 5’ [(G/A) (T/C) (C/A)
A A (C/T) A] 3’ (Overdier et al., 1994; Pierrou et al., 1995; Weigelt et al., 2001). Further binding specificity also depends on flanking sequences on both sides of this core motif. Binding of the forkhead domain to a DNA target site appears to cause only minor structural changes in the forkhead domain whereas a substantial bend is induced in the DNA (Clark et al., 1993;
Pierrou et al., 1994). The forkhead domain also contains nuclear localization signals (located in helix H1 and the basic residues at the carboxyl-terminus) (Pani et al., 1992a). These factors are generally found to be nuclear localized.
The forkhead family of transcription factors act mostly as transcriptional activators, but not exclusively so. Recent studies have demonstrated that the forkhead domain can promote gene activation directly by opening up chromatin and not just by bringing in a separate transcriptional activation domain (Cirillo et al., 1998; Cirillo and Zaret, 1999). A winged helix fold similar to
29
Figure 1.6 Structural diagrams of Foxh1 and Foxa2 proteins
See text for details. FKHD – forkhead domain, SID – Smad interaction domain, TA – transactivation domain. Pink – nuclear localization signals.
30 that of forkhead proteins, except for the lack of a second wing, is found in the linker histones H1
and H5 (Clark et al., 1993). As a result forkhead proteins have been shown to open highly compacted chromatin in vitro and in a manner not requiring the SWI/SNF chromatin remodeling complex. In particular, the C-terminal domain of the Foxa subgroup has been shown to interact with core histones and thus Foxa proteins have been proposed to act as pioneer transcription factors, displacing linker histones from compacted chromatin and facilitating the binding of other transcription factors (Cirillo et al., 2002; Cirillo et al., 1998; Shim et al., 1998).
1.5.1.1 Foxh1
Foxh1 (Fast) was first identified in Xenopus based on its ability to mediate Activin- dependent induction of the Xenopus Mix.2 gene and was the first nuclear Smad binding partner to be identified (Chen et al., 1996). It was subsequently shown to mediate both activin and
TGFβ-dependent activation of the goosecoid promoter (Labbe et al., 1998). Foxh1 homologs
have since been identified in Xenopus laevis, Mus musculus, Danio renio, and Homo sapiens
(Boggetti et al., 2000; Chen et al., 1996; Liu et al., 1999; Weisberg et al., 1998; Yeo et al., 1999;
Zhou et al., 1998). The Foxh1 protein contains 401 amino acids and aside from the forkhead
domain contains a Smad Interaction Domain (SID) in the C-terminus. Two subdomains, the Fast
Motif (FM) and the Smad Interaction Motif (SIM) are located within the SID. The SID makes
direct contacts with the MH2 domain of Smad2/3 (Labbe et al., 1998). Foxh1 binds directly to
DNA in the promoters of target genes via the consensus sequence of 5’-TGT G/T T/G ATT-3’
which is present in all identified targets (Attisano et al., 2001). Although Foxh1 binds DNA
directly, this binding alone is not sufficient to activate transcription as it lacks a transcriptional
activation domain. Stimulation by Nodal/Activin/TGFβ ligands results in nuclear translocation of
R-Smad/Smad4 complexes which bind to Foxh1 and activate transcription (Chen et al., 1996;
31 Chen et al., 1997; Weisberg et al., 1998). Aside from the requirement of Smad2/3/4 binding for activity, little is known regarding the regulation of Foxh1 activity. There is a reported PKA phosphorylation site within the SID domain, however, its functional significance remains unknown (P.A.H unpublished). Additionally, there is an alternative splice variant of Foxh1 that lacks four amino acids in helix 2 (H2) which may affect DNA binding (Weisberg et al., 1998). It is inconclusive as to whether this form is expressed in vivo and/or has altered function (K.D.M, unpublished). Several proteins have been shown to bind to Foxh1 other than the Smad proteins that have been proposed to regulate the function of Foxh1. Foxh1, like several other forkhead proteins, has been shown to interact with homeodomain containing proteins. Binding of Foxh1 to
Goosecoid, a homeodomain protein expressed in the anterior primitive streak, causes Foxh1 to act as a transcriptional repressor (Izzi et al., 2008; Izzi et al., 2007). Foxh1 has also been reported to bind members of the groucho/TLE family of transcriptional repressors. The role of this interaction in vivo is beginning to be elucidated in Xenopus (Steiner, 2008). An important regulator of Foxh1 in the early embryo is Drap1, a general transcriptional repressor. Binding of
Drap1 to Foxh1 blocks the ability of Foxh1 to bind DNA, thus blocking Nodal signaling (Iratni et al., 2002). Lastly, another forkhead gene family member, Foxg1, has been shown to bind to the Smad Interaction Domain (SID) of Foxh1, preventing interaction of Foxh1 with the Smad proteins and thus blocking its activation (Dou et al., 2000). However, Foxg1 expression has not been reported in the early embryo and thus the functional significance of this interaction in vivo is unknown.
1.5.1.2 Foxa2
Foxa2 belongs to the Foxa subgroup of the forkhead family. This group consists of three genes, Foxa1 (HNF-3α), Foxa2 (HNF-3β), and Foxa3 (HNF-3γ) that were first identified in liver
32 nuclear extracts as transcription factors essential for the regulation of hepatocyte-specific
expression of several target genes (Lai et al., 1990; Lai et al., 1991). Biochemical and cell culture studies suggest that Foxa2 functions as a transcriptional activator. Furthermore, given the similarity between the structure of the forkhead domain and linker histones, as well as studies by
Cirillo et al (2002), it appears that Foxa2 can also act as a pioneer factor, binding to and opening compacted chromatin, as discussed in Section 1.5.1. The Foxa2 protein contains 459 amino acids, and in addition to the forkhead (DNA binding) domain, it also contains transactivation domains at its N- and C-terminus. The C-terminal domain has also been shown to interact with the core histones H3 and H4 (Cirillo et al., 1998). Furthermore, Foxa2 contains an AKT2/PKB phosphorylation site at the N terminus of the forkhead domain at position T156 (Wolfrum et al.,
2003). The function of this site is currently in dispute, but it is thought to regulate nuclear localization of Foxa2 (discussed in Friedman and Kaestner, 2006). All three members of the
Foxa subgroup are expressed exclusively in endoderm-derived tissues in the adult, such as the liver, lung, pancreas, and intestine (reviewed in Friedman and Kaestner, 2006). The embryonic expression of Foxa2 is discussed in depth in Section 1.5.2.1. Little is known regarding the regulation of Foxa2 activity in the early embryo. Similar to Foxh1, the primary mode of regulation may be via interaction with different protein binding partners. Foxa2 physically interacts with several homeodomain proteins that are expressed in the early embryo, such as
Goosecoid, Lim, and Otx2. Foxa2 has also been shown to interact with Pdx1, Engrailed-2 and the TLE/groucho family of co-repressors (Filosa et al., 1997; Foucher et al., 2003; Jin et al.,
2001; Marshak et al., 2000; Nakano et al., 2000; Perea-Gomez et al., 1999; Yaklichkin et al.,
2007). Collectively, this data suggests that regulation of Foxa2, and Foxh1, may be via the formation of various protein complexes.
33 1.5.2 Roles of Foxh1 and Foxa2 in mouse embryogenesis
1.5.2.1 Expression of Foxh1 and Foxa2 in the mouse embryo
Expression of both Foxh1 and Foxa2 is first detected during post-implantation stages in
the mouse embryo (Ang et al., 1993; Monaghan et al., 1993; Weisberg et al., 1998). Foxh1 is
expressed ubiquitously throughout the epiblast region of the embryo beginning at E5.5, and is
not expressed in the extraembryonic ectoderm. At E6.5 and E7.5 Foxh1 remains highly expressed throughout the epiblast. Expression of Foxh1 begins to decrease rapidly at the onset of
organogenesis. At E8.5 expression remains weakly in the heart and at E9.5 weak expression is
observed in the forebrain. Foxh1 expression is not detected in the mouse embryo following E9.5,
nor is it expressed in any adult mouse tissues. The factors upstream of Foxh1 that regulate its expression in the early embryo are unknown.
Expression of Foxa2 is first detected in the cells of the anterior primitive streak as well as the anterior visceral endoderm at E6.5. There is also weak expression throughout the entire visceral endoderm at this stage. At E7.5 Foxa2 expression is detected in the tissues derived from the anterior primitive streak - the node, notochord, and definitive endoderm. At E8.5 Foxa2 is also detected in the floor plate region of the neural tube. By E9.5 expression of Foxa2 begins in the liver primoridum. Extensive studies of the Foxa2 promoter have identified two distinct enhancers that mediate Foxa2 expression in the early mouse embryo (Nishizaki et al., 2001;
Sasaki and Hogan, 1996). Expression in the node and notochord is regulated by an upstream sequence while expression in the floor plate, posterior notochord, and posterior gut epithelium is regulated by a downstream sequence. There is evidence to suggest that the Gli family of transcription factors may regulate expression of Foxa2 in the floor plate (Sasaki et al., 1997).
34 Recently, the Tead family of proteins have been shown to regulate expression of the node/notochord enhancer of Foxa2 (Sawada et al., 2008). The promoter of Foxa2 contains Foxa2 consensus binding sites and it is thus thought to regulate its own expression (Pani et al., 1992b).
Consistent with this, the levels of β-galactosidase expression in Foxa2 null mice (where the
Foxa2 coding region has been replaced by the β-galactosidase gene) is significantly lower than in heterozygous animals (Weinstein et al., 1994). Clues to the factors upstream of Foxa2 have come from analysis of various mouse mutants. Expression of Foxa2 is lost in embryos lacking the RING domain protein Arkadia, as well as in embryos lacking Foxh1 (see Section 1.5.2.2)
(Episkopou et al., 2001; Hoodless et al., 2001; Niederlander et al., 2001; Yamamoto et al., 2001).
However, no Foxh1 consensus sites have been found in the Foxa2 promoter. Data from both frog and mouse embryos has suggested that the Nodal pathway plays a role in regulation of Foxa2 expression. It appears that Smad2 stimulates Foxa2 expression in the visceral endoderm in the mouse and this is supported by work in Xenopus which indicates that Foxa2 is a direct target of
Nodal/Smad2 (Howell and Hill, 1997). However, Foxa2 is expressed ectopically in
Nodal413.d/413.d mouse embryos, suggesting that in the epiblast Nodal may act to restrict Foxa2 expression (Brennan et al., 2001). In support of this, numerous Smad binding sites have been identified within the Foxa2 promoter, however, direct regulation of Foxa2 expression by Smad2 has not yet been demonstrated (Liu et al., 2004).
1.5.2.2 Foxh1 and Foxa2 are essential for specification of the anterior primitive streak and its derivative tissues
The temporally and spatially restricted expression patterns of both Foxh1 and Foxa2 suggest that they have a role in specifying developmental fates during early embryogenesis. The
35 function of these factors has been investigated using gene targeting. Consistent with their early
expression patterns, loss of Foxh1 or Foxa2 results in embryonic lethality.
Embryos homozygous for deletion of Foxa2 die at E9.5 (Ang and Rossant, 1994;
Weinstein et al., 1994). Mutant embryos are morphologically distinguishable from wild-type embryos beginning at E7.0 when a severe constriction at the boundary between the embryonic and extraembryonic regions appears. Gastrulation initiates in mutant embryos, however, the primitive streak fails to elongate, and expression of brachyury (T), a marker of the primitive streak, remains localized to a small cluster of cells in the posterior proximal region of the embryo. Consistent with failure of the primitive streak to elongate, cells expressing markers of the anterior primitive streak are absent or mislocalized, suggesting that this cell population is not specified in these embryos. Mutant embryos also lack a morphologically discernable node and its derivatives, the notochord and floor plate. As a result of the failure to form the prechordal plate and notochord, embryos lack anterior (forebrain) structures and the somites are fused in the midline. Contrary to the anterior primitive streak as the origin of the definitive endoderm (based on fate mapping and discussed in Section 1.3) and the expression of Foxa2 in the definitive endoderm it appears that some definitive endoderm, specifically the hindgut, forms in Foxa2 null embryos. This is discussed further in Chapter 4 of this thesis.
Foxa2 is expressed in both the extraembryonic visceral endoderm and the embryonic epiblast. To determine if Foxa2 function was required in the extraembryonic or embryonic tissues, Foxa2 was specifically removed from the embryonic lineage using a tetraploid aggregation strategy. In this method, tetraploid (4N) blastocysts are injected with Foxa2 null ES cells (Tam and Rossant, 2003). The tetraploid cells are unable to contribute to the embryonic lineages, however, they readily contribute to the extraembryonic lineages, thus all embryonic
36 tissues will be devoid of Foxa2 function. In chimeric embryos with wild-type extraembryonic
tissues and Foxa2 null embryonic tissues the deficiency in node, notochord, and definitive (gut)
endoderm remains, however, there is a rescue of both the anterior neural plate (forebrain)
specification and primitive streak elongation, suggesting that Foxa2 is required in the
extraembryonic tissues for these processes (Dufort et al., 1998). Furthermore, wild-type visceral
endoderm rescued the embryonic-extraembryonic constriction that occurs in Foxa2 null
embryos. Embryos also display severe heart defects, which are likely the cause of death at E9.5,
as well as laterality defects. The laterality defects are likely secondary to the loss of the node and notochord, which play a critical role in initiation and maintenance of left-right asymmetry
(reviewed in (Shiratori and Hamada, 2006). The heart defects may be explained by the lack of foregut definitive endoderm in the mutants, as endoderm-cardiac mesoderm interactions are
important for proper heart development. Tissue specific deletion of Foxa2 has also been
accomplished using Nestin-Cre to specifically remove Foxa2 from the embryonic tissues after
the onset of gastrulation thereby preserving Foxa2 function during the early stages of
embryogenesis in extraembryonic tissues and inactivating the gene in embryonic tissues only at
later stages (Hallonet et al., 2002). In these embryos the anterior neural plate and anterior
definitive endoderm are initially specified but fail to be maintained. Furthermore, the axial
mesendoderm fails to differentiate. These results support the knockout phenotype and highlight
the important role of Foxa2 in specification of midline tissues.
Genetic ablation of Foxh1 results in embryonic lethality at E9.5 and mutant embryos show a range of patterning defects resulting from a deficiency in Nodal signaling and failure to specify the anterior primitive streak (Hoodless et al., 2001; Yamamoto et al., 2001). The phenotype of Foxh1 null embryos is variable, with defects ranging from a lack of the node and notochord (axial mesoderm structures) and loss of anterior structures to a failure to specify the
37 anterior–posterior axis in a minority of embryos. The spectrum of phenotypes of Foxh1 null
embryos differs depending upon genetic background; however, there is still variability even on a constant genetic background. Similar to Foxh1 null mouse embryos, zebrafish mutants lacking
the Foxh1 homolog schmalspur display a loss of organizer and axial mesendoderm structures,
indicating an evolutionarily conserved role for Foxh1 (Pogoda et al., 2000; Sirotkin et al., 2000;
Watanabe and Whitman, 1999). Similar to Foxa2 null embryos, severely affected Foxh1 mutants
are first morphologically distinguishable from wild-type embryos at E7.0 by the presence of a
constriction between the embryonic and extraembryonic regions. In the most severe phenotype
the anterior-posterior axis is not specified and only extraembryonic tissues are generated during
gastrulation, resulting in no development of the embryo proper, similar to Nodal413.d/413.d or
Smad2Robm1/Robm1 embryos. In less severe mutants, gastrulation is initiated and a primitive streak
forms on the posterior of the embryo, but the cells of the anterior primitive streak fail to be
specified. Expression of the markers of the anterior primitive streak, Foxa2 and goosecoid, are
lost in Foxh1 null embryos and the tissues derived from the anterior streak, namely the node,
notochord, floor plate and anterior definitive endoderm do not form. As a result, the neural plate
remains unfolded and the somites are fused in the midline. Mutant embryos also display varying
degrees of anterior truncations due to the loss of the prechordal plate and anterior definitive
endoderm.
Analysis of mutant embryos at later stages revealed that Foxh1 mutant embryos also
display defects in heart morphogenesis and Foxh1 has subsequently been shown to be essential
for specification of the secondary heart field (von Both et al., 2004). Additionally, Foxh1 has
also been shown to regulate expression of members of the retinoic acid pathway in forebrain
specification (Silvestri et al., 2008). Conditional deletion of Foxh1 from the epiblast using
Lefty2-Cre rescues the anterior-posterior patterning defects, but not the anterior primitive streak
38 and midline defects (Yamamoto et al., 2001). Deletion of Foxh1 using Lefty2-Cre that is expressed specifically in the lateral plate mesoderm results in post-natal lethality with animals displaying right isomerism, indicating that Foxh1 also plays a role in left-right patterning of the embryo (Yamamoto et al., 2003). The role of Foxh1 has also been investigated using chimera analysis (Hoodless et al., 2001; Yamamoto et al., 2001). The results of these analyses demonstrated that Foxh1 is required in the visceral endoderm for anterior-posterior patterning and required in the epiblast for specification of the anterior primitive streak. These chimera studies also revealed that Foxh1 is required cell autonomously in the definitive endoderm, in agreement the phenotypes of Smad2 or Nodal deficient mutants. However, our recent analysis suggests that there may be subpopulations of the definitive endoderm that do not require Foxh1 activity. This is discussed further in Chapter 4 of this thesis.
1.5.2.3 Downstream targets of Foxh1 and Foxa2 in the mouse embryo
The direct transcriptional targets of Foxa2 in post-implantation mouse embryos are largely unknown. Foxa2 itself has a Foxa2 consensus site within its promoter and several lines of evidence support the claim that Foxa2 regulates its own expression (Pani et al., 1992b). Shh also appears to be regulated by Foxa2, as deletions in the Shh promoter that contain Foxa2 binding sites results in a loss of reporter gene expression in the notochord, suggesting that Foxa2 regulates expression of Shh in this tissue (Epstein et al., 1999). Lastly, the node specific enhancer in the promoter of the Nodal gene contains a predicted Foxa2 binding site. The absence of Nodal expression in embryos lacking Foxa2 specifically in the embryonic tissues suggests that Foxa2 may be a direct activator of Nodal expression in the mouse node (Dufort et al., 1998). Lastly,
Foxa2 has been shown to directly regulate the expression of Otx2 in the anterior visceral
39 endoderm (AVE) and thus plays a critical role in specification of the anterior-posterior axis
(Kimura-Yoshida et al., 2007).
Foxh1 targets include a variety of genes, all of which are primarily expressed during early stages of development. These include the homeodomain protein goosecoid and several members of the TGF-β superfamily of ligands, including Lefty1, Lefty2, and Nodal (Labbe et al.,
1998; Norris et al., 2002; Saijoh et al., 2000). Foxh1 has been shown to both activate and repress the goosecoid promoter, depending upon which Smad co-factor it binds (Labbe et al., 1998). In combination with Smad2 and Smad4 Foxh1 activates the goosecoid promoter, however, Foxh1
in combination with Smad3 and Smad4 has been demonstrated to repress the same Activin
response element (ARE) within the goosecoid promoter. Foxh1 has also been shown to both
activate and repress the expression of Mixl1 (Hart et al., 2005; Izzi et al., 2008; Izzi et al., 2007).
In vitro assays using the Mixl1 promoter have shown Foxh1 to positively regulate its expression.
However, in vivo Mixl1 is upregulated in Foxh1 null embryos, suggesting that Foxh1 plays a role
in repressing Mixl1 expression. This is supported by in vitro experiments that showed that
Foxh1, via an interaction with goosecoid, represses the Mixl1 promoter. Several studies have
demonstrated that expression of Nodal in the mouse epiblast is maintained by a positive
autoregulatory loop involving Foxh1 (Norris et al., 2002). Consistent with this, expression of
Nodal is downregulated, but not completely absent in Foxh1 null embryos. The expression of
Nodal in the lateral plate mesoderm also requires Foxh1 (Saijoh et al., 2000). Lastly, expression
of Lefty1 in the visceral endoderm is induced by Nodal/Foxh1, and consistent with this,
expression of this gene is lost in Foxh1 null embryos (Takaoka et al., 2006).
In later stages of embryogenesis Foxh1 has been shown to have a role in formation of the
secondary heart field, patterning of the neural tissue, and patterning of the left-right axis. In
40 cardiac specification Foxh1 interacts with Nkx2.5 to directly regulate the expression of Mef2c, a
MADS box transcription factor that is involved in morphogenesis of the right ventricle and
outflow tract of the developing heart (von Both et al., 2004). In the forebrain, Foxh1 has been
shown to regulate the expression of members of the aldehyde dehydrogenase “a” subfamily
Aldh1a1, Aldh1a2, and Aldh1a3 which are involved in the metabolism of Vitamin A to produce
retinoic acid, a potent mediator of neural specification and patterning, as well as Hesx1, Lmo1,
and Fgf8 (Silvestri et al., 2008). Expression of Nodal, Lefty2, and Pitx2 are restricted to the left lateral plate mesoderm following initial breaking of left-right asymmetry in the node. The left- specific enhancers of these genes have been shown to contain Foxh1 consensus sites and their expression is directly regulated by Foxh1 (Saijoh et al., 2000; Shiratori et al., 2001; Yamamoto et al., 2003).
1.6 AIM OF THE STUDY
The anterior primitive streak has been shown to be the location of the mouse gastrula
organizer, which plays an important role in mouse embryogenesis. The Nodal signaling pathway
and its downstream effector Foxh1 have been shown to play a critical role in specification of the
mouse gastrula organizer. However, the transcriptional targets of Foxh1, aside from Nodal, that
are involved in specification of the anterior primitive streak (mouse gastrula organizer) are
unknown. We hypothesized that genes downstream of Nodal/Foxh1 that are essential for
organizer specification would be expressed at significantly lower levels or not at all in Foxh1
null embryos. To investigate this we used Affymetrix™ GeneChip® arrays to compare the gene
expression profiles of wild-type and Foxh1 null embryos. We then sought to characterize the
expression patterns of genes predicted to be involved in organizer specification based on the
Affymetrix™ array data using whole mount in situ hybridization. Lastly, the expression patterns
41 of these candidate genes were to be analyzed in Foxh1 and Foxa2 null embryos, which do not specify the anterior primitive streak, to determine if the expression level, localization, or both were altered and to further investigate the regulation of the candidate genes.
42 1.7 REFERENCES
Ang, S. L., and Rossant, J. (1994). HNF-3 beta is essential for node and notochord formation in mouse development. Cell 78, 561-74. Ang, S. L., Wierda, A., Wong, D., Stevens, K. A., Cascio, S., Rossant, J., and Zaret, K. S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development 119, 1301-15. Attisano, L., Silvestri, C., Izzi, L., and Labbe, E. (2001). The transcriptional role of Smads and FAST (FoxH1) in TGFbeta and activin signalling. Mol Cell Endocrinol 180, 3-11. Bachiller, D., Klingensmith, J., Kemp, C., Belo, J. A., Anderson, R. M., May, S. R., McMahon, J. A., McMahon, A. P., Harland, R. M., Rossant, J., and De Robertis, E. M. (2000). The organizer factors Chordin and Noggin are required for mouse forebrain development. Nature 403, 658-61. Bachvarova, R. F., Skromne, I., and Stern, C. D. (1998). Induction of primitive streak and Hensen's node by the posterior marginal zone in the early chick embryo. Development 125, 3521-34. Beck, S., Le Good, J. A., Guzman, M., Ben Haim, N., Roy, K., Beermann, F., and Constam, D. B. (2002). Extraembryonic proteases regulate Nodal signalling during gastrulation. Nat Cell Biol 4, 981-5. Beddington, R. S. (1994). Induction of a second neural axis by the mouse node. Development 120, 613-20. Belo, J. A., Bouwmeester, T., Leyns, L., Kertesz, N., Gallo, M., Follettie, M., and De Robertis, E. M. (1997). Cerberus-like is a secreted factor with neutralizing activity expressed in the anterior primitive endoderm of the mouse gastrula. Mech Dev 68, 45-57. Blum, M., Gaunt, S. J., Cho, K. W., Steinbeisser, H., Blumberg, B., Bittner, D., and De Robertis, E. M. (1992). Gastrulation in the mouse: the role of the homeobox gene goosecoid. Cell 69, 1097-106. Boggetti, B., Argenton, F., Haffter, P., Bianchi, M. E., Cotelli, F., and Beltrame, M. (2000). Cloning and expression pattern of a zebrafish homolog of forkhead activin signal transducer (FAST), a transcription factor mediating Nodal-related signals. Mech Dev 99, 187-90. Brennan, J., Lu, C. C., Norris, D. P., Rodriguez, T. A., Beddington, R. S., and Robertson, E. J. (2001). Nodal signalling in the epiblast patterns the early mouse embryo. Nature 411, 965-9. Brennan, J., Norris, D. P., and Robertson, E. J. (2002). Nodal activity in the node governs left- right asymmetry. Genes Dev 16, 2339-44. Camus, A., and Tam, P. P. (1999). The organizer of the gastrulating mouse embryo. Curr Top Dev Biol 45, 117-53. Carlsson, P., and Mahlapuu, M. (2002). Forkhead transcription factors: key players in development and metabolism. Dev Biol 250, 1-23. Chen, C., and Shen, M. M. (2004). Two modes by which Lefty proteins inhibit nodal signaling. Curr Biol 14, 618-24. Chen, X., Rubock, M. J., and Whitman, M. (1996). A transcriptional partner for MAD proteins in TGF-beta signalling. Nature 383, 691-6. Chen, X., Weisberg, E., Fridmacher, V., Watanabe, M., Naco, G., and Whitman, M. (1997). Smad4 and FAST-1 in the assembly of activin-responsive factor. Nature 389, 85-9.
43 Chen, Y., and Schier, A. F. (2002). Lefty proteins are long-range inhibitors of squint-mediated nodal signaling. Curr Biol 12, 2124-8. Chu, G. C., Dunn, N. R., Anderson, D. C., Oxburgh, L., and Robertson, E. J. (2004). Differential requirements for Smad4 in TGFbeta-dependent patterning of the early mouse embryo. Development 131, 3501-12. Cirillo, L. A., Lin, F. R., Cuesta, I., Friedman, D., Jarnik, M., and Zaret, K. S. (2002). Opening of compacted chromatin by early developmental transcription factors HNF3 (FoxA) and GATA-4. Mol Cell 9, 279-89. Cirillo, L. A., McPherson, C. E., Bossard, P., Stevens, K., Cherian, S., Shim, E. Y., Clark, K. L., Burley, S. K., and Zaret, K. S. (1998). Binding of the winged-helix transcription factor HNF3 to a linker histone site on the nucleosome. Embo J 17, 244-54. Cirillo, L. A., and Zaret, K. S. (1999). An early developmental transcription factor complex that is more stable on nucleosome core particles than on free DNA. Mol Cell 4, 961-9. Clark, K. L., Halay, E. D., Lai, E., and Burley, S. K. (1993). Co-crystal structure of the HNF- 3/fork head DNA-recognition motif resembles histone H5. Nature 364, 412-20. Conlon, F. L., Lyons, K. M., Takaesu, N., Barth, K. S., Kispert, A., Herrmann, B., and Robertson, E. J. (1994). A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Development 120, 1919-28. Datto, M. B., Frederick, J. P., Pan, L., Borton, A. J., Zhuang, Y., and Wang, X. F. (1999). Targeted disruption of Smad3 reveals an essential role in transforming growth factor beta-mediated signal transduction. Mol Cell Biol 19, 2495-504. Davidson, B. P., Kinder, S. J., Steiner, K., Schoenwolf, G. C., and Tam, P. P. (1999). Impact of node ablation on the morphogenesis of the body axis and the lateral asymmetry of the mouse embryo during early organogenesis. Dev Biol 211, 11-26. Davidson, B. P., and Tam, P. P. (2000). The node of the mouse embryo. Curr Biol 10, R617-9. De Robertis, E. M., Larrain, J., Oelgeschlager, M., and Wessely, O. (2000). The establishment of Spemann's organizer and patterning of the vertebrate embryo. Nat Rev Genet 1, 171-81. Ding, J., Yang, L., Yan, Y. T., Chen, A., Desai, N., Wynshaw-Boris, A., and Shen, M. M. (1998). Cripto is required for correct orientation of the anterior-posterior axis in the mouse embryo. Nature 395, 702-7. Dou, C., Lee, J., Liu, B., Liu, F., Massague, J., Xuan, S., and Lai, E. (2000). BF-1 interferes with transforming growth factor beta signaling by associating with Smad partners. Mol Cell Biol 20, 6201-11. Dufort, D., Schwartz, L., Harpal, K., and Rossant, J. (1998). The transcription factor HNF3beta is required in visceral endoderm for normal primitive streak morphogenesis. Development 125, 3015-25. Dunn, N. R., Vincent, S. D., Oxburgh, L., Robertson, E. J., and Bikoff, E. K. (2004). Combinatorial activities of Smad2 and Smad3 regulate mesoderm formation and patterning in the mouse embryo. Development 131, 1717-28. Episkopou, V., Arkell, R., Timmons, P. M., Walsh, J. J., Andrew, R. L., and Swan, D. (2001). Induction of the mammalian node requires Arkadia function in the extraembryonic lineages. Nature 410, 825-30. Epstein, D. J., McMahon, A. P., and Joyner, A. L. (1999). Regionalization of Sonic hedgehog transcription along the anteroposterior axis of the mouse central nervous system is regulated by Hnf3-dependent and -independent mechanisms. Development 126, 281-92. Filosa, S., Rivera-Perez, J. A., Gomez, A. P., Gansmuller, A., Sasaki, H., Behringer, R. R., and Ang, S. L. (1997). Goosecoid and HNF-3beta genetically interact to regulate neural tube patterning during mouse embryogenesis. Development 124, 2843-54.
44 Foucher, I., Montesinos, M. L., Volovitch, M., Prochiantz, A., and Trembleau, A. (2003). Joint regulation of the MAP1B promoter by HNF3beta/Foxa2 and Engrailed is the result of a highly conserved mechanism for direct interaction of homeoproteins and Fox transcription factors. Development 130, 1867-76. Franklin, V., Khoo, P. L., Bildsoe, H., Wong, N., Lewis, S., and Tam, P. P. (2008). Regionalisation of the endoderm progenitors and morphogenesis of the gut portals of the mouse embryo. Mech Dev 125, 587-600. Friedman, J. R., and Kaestner, K. H. (2006). The Foxa family of transcription factors in development and metabolism. Cell Mol Life Sci 63, 2317-28. Gardner, R. L. (1997). The early blastocyst is bilaterally symmetrical and its axis of symmetry is aligned with the animal-vegetal axis of the zygote in the mouse. Development 124, 289- 301. Germain, S., Howell, M., Esslemont, G. M., and Hill, C. S. (2000). Homeodomain and winged- helix transcription factors recruit activated Smads to distinct promoter elements via a common Smad interaction motif. Genes Dev 14, 435-51. Gilbert, S. F., and Saxen, L. (1993). Spemann's organizer: models and molecules. Mech Dev 41, 73-89. Gregorieff, A., Grosschedl, R., and Clevers, H. (2004). Hindgut defects and transformation of the gastro-intestinal tract in Tcf4(-/-)/Tcf1(-/-) embryos. Embo J 23, 1825-33. Gritsman, K., Zhang, J., Cheng, S., Heckscher, E., Talbot, W. S., and Schier, A. F. (1999). The EGF-CFC protein one-eyed pinhead is essential for nodal signaling. Cell 97, 121-32. Gu, Z., Nomura, M., Simpson, B. B., Lei, H., Feijen, A., van den Eijnden-van Raaij, J., Donahoe, P. K., and Li, E. (1998). The type I activin receptor ActRIB is required for egg cylinder organization and gastrulation in the mouse. Genes Dev 12, 844-57. Hallonet, M., Kaestner, K. H., Martin-Parras, L., Sasaki, H., Betz, U. A., and Ang, S. L. (2002). Maintenance of the specification of the anterior definitive endoderm and forebrain depends on the axial mesendoderm: a study using HNF3beta/Foxa2 conditional mutants. Dev Biol 243, 20-33. Harland, R., and Gerhart, J. (1997). Formation and function of Spemann's organizer. Annu Rev Cell Dev Biol 13, 611-67. Hart, A. H., Hartley, L., Sourris, K., Stadler, E. S., Li, R., Stanley, E. G., Tam, P. P., Elefanty, A. G., and Robb, L. (2002). Mixl1 is required for axial mesendoderm morphogenesis and patterning in the murine embryo. Development 129, 3597-608. Hart, A. H., Willson, T. A., Wong, M., Parker, K., and Robb, L. (2005). Transcriptional regulation of the homeobox gene Mixl1 by TGF-beta and FoxH1. Biochem Biophys Res Commun 333, 1361-9. Hoodless, P. A., Pye, M., Chazaud, C., Labbe, E., Attisano, L., Rossant, J., and Wrana, J. L. (2001). FoxH1 (Fast) functions to specify the anterior primitive streak in the mouse. Genes Dev 15, 1257-71. Horb, M. E. (2000). Patterning the endoderm: the importance of neighbours. Bioessays 22, 599- 602. Howell, M., and Hill, C. S. (1997). XSmad2 directly activates the activin-inducible, dorsal mesoderm gene XFKH1 in Xenopus embryos. Embo J 16, 7411-21. Iratni, R., Yan, Y. T., Chen, C., Ding, J., Zhang, Y., Price, S. M., Reinberg, D., and Shen, M. M. (2002). Inhibition of excess nodal signaling during mouse gastrulation by the transcriptional corepressor DRAP1. Science 298, 1996-9. Izzi, L., Narimatsu, M., and Attisano, L. (2008). Sumoylation differentially regulates Goosecoid- mediated transcriptional repression. Exp Cell Res 314, 1585-94.
45 Izzi, L., Silvestri, C., von Both, I., Labbe, E., Zakin, L., Wrana, J. L., and Attisano, L. (2007). Foxh1 recruits Gsc to negatively regulate Mixl1 expression during early mouse development. Embo J 26, 3132-43. Jin, O., Harpal, K., Ang, S. L., and Rossant, J. (2001). Otx2 and HNF3beta genetically interact in anterior patterning. Int J Dev Biol 45, 357-65. Kanai-Azuma, M., Kanai, Y., Gad, J. M., Tajima, Y., Taya, C., Kurohmaru, M., Sanai, Y., Yonekawa, H., Yazaki, K., Tam, P. P., and Hayashi, Y. (2002). Depletion of definitive gut endoderm in Sox17-null mutant mice. Development 129, 2367-79. Kaufmann, E., and Knochel, W. (1996). Five years on the wings of fork head. Mech Dev 57, 3- 20. Kim, S. K., Hebrok, M., and Melton, D. A. (1997). Notochord to endoderm signaling is required for pancreas development. Development 124, 4243-52. Kimura-Yoshida, C., Tian, E., Nakano, H., Amazaki, S., Shimokawa, K., Rossant, J., Aizawa, S., and Matsuo, I. (2007). Crucial roles of Foxa2 in mouse anterior-posterior axis polarization via regulation of anterior visceral endoderm-specific genes. Proc Natl Acad Sci U S A 104, 5919-24. Kinder, S. J., Tsang, T. E., Wakamiya, M., Sasaki, H., Behringer, R. R., Nagy, A., and Tam, P. P. (2001). The organizer of the mouse gastrula is composed of a dynamic population of progenitor cells for the axial mesoderm. Development 128, 3623-34. Labbe, E., Silvestri, C., Hoodless, P. A., Wrana, J. L., and Attisano, L. (1998). Smad2 and Smad3 positively and negatively regulate TGF beta-dependent transcription through the forkhead DNA-binding protein FAST2. Mol Cell 2, 109-20. Lai, E., Clark, K. L., Burley, S. K., and Darnell, J. E., Jr. (1993). Hepatocyte nuclear factor 3/fork head or "winged helix" proteins: a family of transcription factors of diverse biologic function. Proc Natl Acad Sci U S A 90, 10421-3. Lai, E., Prezioso, V. R., Smith, E., Litvin, O., Costa, R. H., and Darnell, J. E., Jr. (1990). HNF- 3A, a hepatocyte-enriched transcription factor of novel structure is regulated transcriptionally. Genes Dev 4, 1427-36. Lai, E., Prezioso, V. R., Tao, W. F., Chen, W. S., and Darnell, J. E., Jr. (1991). Hepatocyte nuclear factor 3 alpha belongs to a gene family in mammals that is homologous to the Drosophila homeotic gene fork head. Genes Dev 5, 416-27. Lawson, K. A., Meneses, J. J., and Pedersen, R. A. (1986). Cell fate and cell lineage in the endoderm of the presomite mouse embryo, studied with an intracellular tracer. Dev Biol 115, 325-39. Lawson, K. A., and Pedersen, R. A. (1987). Cell fate, morphogenetic movement and population kinetics of embryonic endoderm at the time of germ layer formation in the mouse. Development 101, 627-52. Levy, L., Howell, M., Das, D., Harkin, S., Episkopou, V., and Hill, C. S. (2007). Arkadia activates Smad3/Smad4-dependent transcription by triggering signal-induced SnoN degradation. Mol Cell Biol 27, 6068-83. Liu, B., Dou, C. L., Prabhu, L., and Lai, E. (1999). FAST-2 is a mammalian winged-helix protein which mediates transforming growth factor beta signals. Mol Cell Biol 19, 424- 30. Liu, Y., Festing, M., Thompson, J. C., Hester, M., Rankin, S., El-Hodiri, H. M., Zorn, A. M., and Weinstein, M. (2004). Smad2 and Smad3 coordinately regulate craniofacial and endodermal development. Dev Biol 270, 411-26. Lowe, L. A., Yamada, S., and Kuehn, M. R. (2001). Genetic dissection of nodal function in patterning the mouse embryo. Development 128, 1831-43.
46 Lu, C. C., and Robertson, E. J. (2004). Multiple roles for Nodal in the epiblast of the mouse embryo in the establishment of anterior-posterior patterning. Dev Biol 273, 149-59. Maatman, R., Zachgo, J., and Gossler, A. (1997). The Danforth's short tail mutation acts cell autonomously in notochord cells and ventral hindgut endoderm. Development 124, 4019- 28. Marsden, I., Chen, Y., Jin, C., and Liao, X. (1997). Evidence that the DNA binding specificity of winged helix proteins is mediated by a structural change in the amino acid sequence adjacent to the principal DNA binding helix. Biochemistry 36, 13248-55. Marshak, S., Benshushan, E., Shoshkes, M., Havin, L., Cerasi, E., and Melloul, D. (2000). Functional conservation of regulatory elements in the pdx-1 gene: PDX-1 and hepatocyte nuclear factor 3beta transcription factors mediate beta-cell-specific expression. Mol Cell Biol 20, 7583-90. Massague, J. (1998). TGF-beta signal transduction. Annu Rev Biochem 67, 753-91. Matzuk, M. M., Kumar, T. R., and Bradley, A. (1995). Different phenotypes for mice deficient in either activins or activin receptor type II. Nature 374, 356-60. Mavrakis, K. J., Andrew, R. L., Lee, K. L., Petropoulou, C., Dixon, J. E., Navaratnam, N., Norris, D. P., and Episkopou, V. (2007). Arkadia enhances Nodal/TGF-beta signaling by coupling phospho-Smad2/3 activity and turnover. PLoS Biol 5, e67. Meno, C., Shimono, A., Saijoh, Y., Yashiro, K., Mochida, K., Ohishi, S., Noji, S., Kondoh, H., and Hamada, H. (1998). lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 94, 287-97. Monaghan, A. P., Kaestner, K. H., Grau, E., and Schutz, G. (1993). Postimplantation expression patterns indicate a role for the mouse forkhead/HNF-3 alpha, beta and gamma genes in determination of the definitive endoderm, chordamesoderm and neuroectoderm. Development 119, 567-78. Nagy, A., Gertsenstein, M., Vintersten, K., Behringer, R. (2003.). "Manipulating the Mouse Embryo: A laboratory manual. ." Cold Spring Harbor Laboratory Press., Cold Spring Harbor, New York. Nakano, T., Murata, T., Matsuo, I., and Aizawa, S. (2000). OTX2 directly interacts with LIM1 and HNF-3beta. Biochem Biophys Res Commun 267, 64-70. Niederlander, C., Walsh, J. J., Episkopou, V., and Jones, C. M. (2001). Arkadia enhances nodal- related signalling to induce mesendoderm. Nature 410, 830-4. Nieuwkoop, P. D. (1985). Inductive interactions in early amphibian development and their general nature. J Embryol Exp Morphol 89 Suppl, 333-47. Nishizaki, Y., Shimazu, K., Kondoh, H., and Sasaki, H. (2001). Identification of essential sequence motifs in the node/notochord enhancer of Foxa2 (Hnf3beta) gene that are conserved across vertebrate species. Mech Dev 102, 57-66. Nomura, M., and Li, E. (1998). Smad2 role in mesoderm formation, left-right patterning and craniofacial development. Nature 393, 786-90. Norris, D. P., Brennan, J., Bikoff, E. K., and Robertson, E. J. (2002). The Foxh1-dependent autoregulatory enhancer controls the level of Nodal signals in the mouse embryo. Development 129, 3455-68. Norris, D. P., and Robertson, E. J. (1999). Asymmetric and node-specific nodal expression patterns are controlled by two distinct cis-acting regulatory elements. Genes Dev 13, 1575-88. Oh, S. P., and Li, E. (1997). The signaling pathway mediated by the type IIB activin receptor controls axial patterning and lateral asymmetry in the mouse. Genes Dev 11, 1812-26.
47 Overdier, D. G., Porcella, A., and Costa, R. H. (1994). The DNA-binding specificity of the hepatocyte nuclear factor 3/forkhead domain is influenced by amino-acid residues adjacent to the recognition helix. Mol Cell Biol 14, 2755-66. Pani, L., Overdier, D. G., Porcella, A., Qian, X., Lai, E., and Costa, R. H. (1992a). 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. Pani, L., Quian, X. B., Clevidence, D., and Costa, R. H. (1992b). The restricted promoter activity of the liver transcription factor hepatocyte nuclear factor 3 beta involves a cell-specific factor and positive autoactivation. Mol Cell Biol 12, 552-62. Pedersen, R. A. (1986). "Potency, lineage, and allocation in pre-implantation mouse embryos." Cambridge University Press, New York. Perea-Gomez, A., Shawlot, W., Sasaki, H., Behringer, R. R., and Ang, S. (1999). HNF3beta and Lim1 interact in the visceral endoderm to regulate primitive streak formation and anterior-posterior polarity in the mouse embryo. Development 126, 4499-511. Perea-Gomez, A., Vella, F. D., Shawlot, W., Oulad-Abdelghani, M., Chazaud, C., Meno, C., Pfister, V., Chen, L., Robertson, E., Hamada, H., Behringer, R. R., and Ang, S. L. (2002). Nodal antagonists in the anterior visceral endoderm prevent the formation of multiple primitive streaks. Dev Cell 3, 745-56. Piccolo, S., Agius, E., Leyns, L., Bhattacharyya, S., Grunz, H., Bouwmeester, T., and De Robertis, E. M. (1999). The head inducer Cerberus is a multifunctional antagonist of Nodal, BMP and Wnt signals. Nature 397, 707-10. Pierrou, S., Enerback, S., and Carlsson, P. (1995). Selection of high-affinity binding sites for sequence-specific, DNA binding proteins from random sequence oligonucleotides. Anal Biochem 229, 99-105. Pierrou, S., Hellqvist, M., Samuelsson, L., Enerback, S., and Carlsson, P. (1994). Cloning and characterization of seven human forkhead proteins: binding site specificity and DNA bending. Embo J 13, 5002-12. Pogoda, H. M., Solnica-Krezel, L., Driever, W., and Meyer, D. (2000). The zebrafish forkhead transcription factor FoxH1/Fast1 is a modulator of nodal signaling required for organizer formation. Curr Biol 10, 1041-9. Saijoh, Y., Adachi, H., Sakuma, R., Yeo, C. Y., Yashiro, K., Watanabe, M., Hashiguchi, H., Mochida, K., Ohishi, S., Kawabata, M., Miyazono, K., Whitman, M., and Hamada, H. (2000). Left-right asymmetric expression of lefty2 and nodal is induced by a signaling pathway that includes the transcription factor FAST2. Mol Cell 5, 35-47. Saijoh, Y., Oki, S., Ohishi, S., and Hamada, H. (2003). Left-right patterning of the mouse lateral plate requires nodal produced in the node. Dev Biol 256, 160-72. Sasaki, H., and Hogan, B. L. (1996). Enhancer analysis of the mouse HNF-3 beta gene: regulatory elements for node/notochord and floor plate are independent and consist of multiple sub-elements. Genes Cells 1, 59-72. Sasaki, H., Hui, C., Nakafuku, M., and Kondoh, H. (1997). A binding site for Gli proteins is essential for HNF-3beta floor plate enhancer activity in transgenics and can respond to Shh in vitro. Development 124, 1313-22. Sawada, A., Kiyonari, H., Ukita, K., Nishioka, N., Imuta, Y., and Sasaki, H. (2008). Redundant roles of Tead1 and Tead2 in notochord development and the regulation of cell proliferation and survival. Mol Cell Biol 28, 3177-89. Schier, A. F. (2003). Nodal signaling in vertebrate development. Annu Rev Cell Dev Biol 19, 589-621.
48 Shawlot, W., Deng, J. M., and Behringer, R. R. (1998). Expression of the mouse cerberus-related gene, Cerr1, suggests a role in anterior neural induction and somitogenesis. Proc Natl Acad Sci U S A 95, 6198-203. Shen, M. M. (2007). Nodal signaling: developmental roles and regulation. Development 134, 1023-34. Shih, J., and Fraser, S. E. (1996). Characterizing the zebrafish organizer: microsurgical analysis at the early-shield stage. Development 122, 1313-22. Shim, E. Y., Woodcock, C., and Zaret, K. S. (1998). Nucleosome positioning by the winged helix transcription factor HNF3. Genes Dev 12, 5-10. Shiratori, H., and Hamada, H. (2006). The left-right axis in the mouse: from origin to morphology. Development 133, 2095-104. Shiratori, H., Sakuma, R., Watanabe, M., Hashiguchi, H., Mochida, K., Sakai, Y., Nishino, J., Saijoh, Y., Whitman, M., and Hamada, H. (2001). Two-step regulation of left-right asymmetric expression of Pitx2: initiation by nodal signaling and maintenance by Nkx2. Mol Cell 7, 137-49. Silvestri, C., Narimatsu, M., von Both, I., Liu, Y., Tan, N. B., Izzi, L., McCaffery, P., Wrana, J. L., and Attisano, L. (2008). Genome-wide identification of Smad/Foxh1 targets reveals a role for Foxh1 in retinoic acid regulation and forebrain development. Dev Cell 14, 411- 23. Simpson, E. H., Johnson, D. K., Hunsicker, P., Suffolk, R., Jordan, S. A., and Jackson, I. J. (1999). The mouse Cer1 (Cerberus related or homologue) gene is not required for anterior pattern formation. Dev Biol 213, 202-6. Sirard, C., de la Pompa, J. L., Elia, A., Itie, A., Mirtsos, C., Cheung, A., Hahn, S., Wakeham, A., Schwartz, L., Kern, S. E., Rossant, J., and Mak, T. W. (1998). The tumor suppressor gene Smad4/Dpc4 is required for gastrulation and later for anterior development of the mouse embryo. Genes Dev 12, 107-19. Sirard, C., Kim, S., Mirtsos, C., Tadich, P., Hoodless, P. A., Itie, A., Maxson, R., Wrana, J. L., and Mak, T. W. (2000). Targeted disruption in murine cells reveals variable requirement for Smad4 in transforming growth factor beta-related signaling. J Biol Chem 275, 2063- 70. Sirotkin, H. I., Gates, M. A., Kelly, P. D., Schier, A. F., and Talbot, W. S. (2000). Fast1 is required for the development of dorsal axial structures in zebrafish. Curr Biol 10, 1051-4. Song, J., Oh, S. P., Schrewe, H., Nomura, M., Lei, H., Okano, M., Gridley, T., and Li, E. (1999). The type II activin receptors are essential for egg cylinder growth, gastrulation, and rostral head development in mice. Dev Biol 213, 157-69. Stanley, E. G., Biben, C., Allison, J., Hartley, L., Wicks, I. P., Campbell, I. K., McKinley, M., Barnett, L., Koentgen, F., Robb, L., and Harvey, R. P. (2000). Targeted insertion of a lacZ reporter gene into the mouse Cer1 locus reveals complex and dynamic expression during embryogenesis. Genesis 26, 259-64. Steiner, A., Reid, C., Yaklichkin, S., Lu, G., Wang, S., Hennessey, M., Kessler, DS. (2008). Fast1 functions as a transcriptional switch for nodal signaling in Xenopus mesodermal development. In "Society for Developmental Biology 67th Annual Meeting." Vol. Abstract # 193. Storey, K. G., Selleck, M. A., and Stern, C. D. (1995). Neural induction and regionalisation by different subpopulations of cells in Hensen's node. Development 121, 417-28. Sulik, K., Dehart, D. B., Iangaki, T., Carson, J. L., Vrablic, T., Gesteland, K., and Schoenwolf, G. C. (1994). Morphogenesis of the murine node and notochordal plate. Dev Dyn 201, 260-78.
49 Takaoka, K., Yamamoto, M., Shiratori, H., Meno, C., Rossant, J., Saijoh, Y., and Hamada, H. (2006). The mouse embryo autonomously acquires anterior-posterior polarity at implantation. Dev Cell 10, 451-9. Tam, P. P., and Behringer, R. R. (1997). Mouse gastrulation: the formation of a mammalian body plan. Mech Dev 68, 3-25. Tam, P. P., Khoo, P. L., Lewis, S. L., Bildsoe, H., Wong, N., Tsang, T. E., Gad, J. M., and Robb, L. (2007). Sequential allocation and global pattern of movement of the definitive endoderm in the mouse embryo during gastrulation. Development 134, 251-60. Tam, P. P., and Rossant, J. (2003). Mouse embryonic chimeras: tools for studying mammalian development. Development 130, 6155-63. Tremblay, K. D., Hoodless, P. A., Bikoff, E. K., and Robertson, E. J. (2000). Formation of the definitive endoderm in mouse is a Smad2-dependent process. Development 127, 3079-90. Varlet, I., Collignon, J., and Robertson, E. J. (1997). nodal expression in the primitive endoderm is required for specification of the anterior axis during mouse gastrulation. Development 124, 1033-44. Vincent, S. D., Dunn, N. R., Hayashi, S., Norris, D. P., and Robertson, E. J. (2003). Cell fate decisions within the mouse organizer are governed by graded Nodal signals. Genes Dev 17, 1646-62. von Both, I., Silvestri, C., Erdemir, T., Lickert, H., Walls, J. R., Henkelman, R. M., Rossant, J., Harvey, R. P., Attisano, L., and Wrana, J. L. (2004). Foxh1 is essential for development of the anterior heart field. Dev Cell 7, 331-45. Waldrip, W. R., Bikoff, E. K., Hoodless, P. A., Wrana, J. L., and Robertson, E. J. (1998). Smad2 signaling in extraembryonic tissues determines anterior-posterior polarity of the early mouse embryo. Cell 92, 797-808. Watanabe, M., and Whitman, M. (1999). FAST-1 is a key maternal effector of mesoderm inducers in the early Xenopus embryo. Development 126, 5621-34. Weigel, D., and Jackle, H. (1990). The fork head domain: a novel DNA binding motif of eukaryotic transcription factors? Cell 63, 455-6. Weigel, D., Jurgens, G., Kuttner, F., Seifert, E., and Jackle, H. (1989). The homeotic gene fork head encodes a nuclear protein and is expressed in the terminal regions of the Drosophila embryo. Cell 57, 645-58. Weigelt, J., Climent, I., Dahlman-Wright, K., and Wikstrom, M. (2001). Solution structure of the DNA binding domain of the human forkhead transcription factor AFX (FOXO4). Biochemistry 40, 5861-9. Weinstein, D. C., Ruiz i Altaba, A., Chen, W. S., Hoodless, P., Prezioso, V. R., Jessell, T. M., and Darnell, J. E., Jr. (1994). The winged-helix transcription factor HNF-3 beta is required for notochord development in the mouse embryo. Cell 78, 575-88. Weinstein, M., Yang, X., Li, C., Xu, X., Gotay, J., and Deng, C. X. (1998). Failure of egg cylinder elongation and mesoderm induction in mouse embryos lacking the tumor suppressor smad2. Proc Natl Acad Sci U S A 95, 9378-83. Weisberg, E., Winnier, G. E., Chen, X., Farnsworth, C. L., Hogan, B. L., and Whitman, M. (1998). A mouse homologue of FAST-1 transduces TGF beta superfamily signals and is expressed during early embryogenesis. Mech Dev 79, 17-27. Wells, J. M., and Melton, D. A. (1999). Vertebrate endoderm development. Annu Rev Cell Dev Biol 15, 393-410. Wells, J. M., and Melton, D. A. (2000). Early mouse endoderm is patterned by soluble factors from adjacent germ layers. Development 127, 1563-72. Wijchers, P. J., Burbach, J. P., and Smidt, M. P. (2006). In control of biology: of mice, men and Foxes. Biochem J 397, 233-46.
50 Wilson, L., and Maden, M. (2005). The mechanisms of dorsoventral patterning in the vertebrate neural tube. Dev Biol 282, 1-13. Wolfrum, C., Besser, D., Luca, E., and Stoffel, M. (2003). Insulin regulates the activity of forkhead transcription factor Hnf-3beta/Foxa-2 by Akt-mediated phosphorylation and nuclear/cytosolic localization. Proc Natl Acad Sci U S A 100, 11624-9. Yaklichkin, S., Vekker, A., Stayrook, S., Lewis, M., and Kessler, D. S. (2007). Prevalence of the EH1 Groucho interaction motif in the metazoan Fox family of transcriptional regulators. BMC Genomics 8, 201. Yamamoto, M., Meno, C., Sakai, Y., Shiratori, H., Mochida, K., Ikawa, Y., Saijoh, Y., and Hamada, H. (2001). The transcription factor FoxH1 (FAST) mediates Nodal signaling during anterior-posterior patterning and node formation in the mouse. Genes Dev 15, 1242-56. Yamamoto, M., Mine, N., Mochida, K., Sakai, Y., Saijoh, Y., Meno, C., and Hamada, H. (2003). Nodal signaling induces the midline barrier by activating Nodal expression in the lateral plate. Development 130, 1795-804. Yan, Y. T., Liu, J. J., Luo, Y., E, C., Haltiwanger, R. S., Abate-Shen, C., and Shen, M. M. (2002). Dual roles of Cripto as a ligand and coreceptor in the nodal signaling pathway. Mol Cell Biol 22, 4439-49. Yang, X., Letterio, J. J., Lechleider, R. J., Chen, L., Hayman, R., Gu, H., Roberts, A. B., and Deng, C. (1999). Targeted disruption of SMAD3 results in impaired mucosal immunity and diminished T cell responsiveness to TGF-beta. Embo J 18, 1280-91. Yang, X., Li, C., Xu, X., and Deng, C. (1998). The tumor suppressor SMAD4/DPC4 is essential for epiblast proliferation and mesoderm induction in mice. Proc Natl Acad Sci U S A 95, 3667-72. Yeo, C., and Whitman, M. (2001). Nodal signals to Smads through Cripto-dependent and Cripto- independent mechanisms. Mol Cell 7, 949-57. Yeo, C. Y., Chen, X., and Whitman, M. (1999). The role of FAST-1 and Smads in transcriptional regulation by activin during early Xenopus embryogenesis. J Biol Chem 274, 26584-90. Zhou, S., Zawel, L., Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1998). Characterization of human FAST-1, a TGF beta and activin signal transducer. Mol Cell 2, 121-7. Zhu, Y., Richardson, J. A., Parada, L. F., and Graff, J. M. (1998). Smad3 mutant mice develop metastatic colorectal cancer. Cell 94, 703-14. Zorn, A. M., and Wells, J. M. (2007). Molecular basis of vertebrate endoderm development. Int Rev Cytol 259, 49-111.
51 Chapter 2
A DIFFERENTIAL SCREEN FOR GENES INVOLVED IN
SPECIFICATION OF THE ANTERIOR PRIMITIVE STREAK USING
FOXH1 DEFICIENT EMBRYOS1
2.1 INTRODUCTION
Gastrulation is the process by which the three embryonic germ layers – the ectoderm,
endoderm, and mesoderm – are formed and the body plan of the embryo is established (reviewed
in Tam and Behringer, 1997). Gastrulation in the mouse embryo begins at approximately
embryonic day 6.5 (E6.5) with the formation of a transient structure called the primitive streak
on the posterior side of the embryo. Cells of the epiblast undergo an epithelial-to-mesenchymal
transformation then move through the primitive streak and emerge as either mesoderm or
definitive endoderm. The newly formed germ layers are then patterned in response to inductive
cues from the surrounding tissues and are organized into the three body axes – anterior-posterior,
dorsal-ventral, and left-right. The group of cells located at the anterior end of the primitive streak
(APS) in the mouse gastrula has been shown to play a critical role in embryogenesis (reviewed in
Camus and Tam, 1999; Robb and Tam, 2004). Transplantation experiments have demonstrated that these cells have the ability to induce a secondary axis, and thus this cell population is thought to function as the mouse embryonic organizer (Beddington, 1994). Fate mapping experiments have shown that these cells give rise to axial tissues, such as the anterior definitive
endoderm, prechordal plate, node and notochord, all which have an important role in patterning
surrounding embryonic tissues (Kinder et al., 2001). The anterior primitive streak cells can first
. . 1 A version of this chapter is to be submitted for publication. McKnight and Hoodless. (2009) 52 be detected at E6.5 by the expression of the transcription factors goosecoid homeobox (Gsc) and forkhead box A2 (Foxa2) (reviewed in Beddington and Robertson, 1999; Camus and Tam, 1999;
Davidson and Tam, 2000; Robb and Tam, 2004).
Gene targeting experiments have demonstrated that the Nodal signaling pathway plays a
critical role in the specification of the anterior primitive streak cell population; however, the
genes downstream of Nodal involved in this process are largely unknown (reviewed in
Beddington and Robertson, 1999). The transcription factor Foxh1 is a key transducer of Nodal signals and plays a critical role in the autoregulation of Nodal expression. Foxh1 has been shown to be essential for the specification of the anterior primitive streak cells (Hoodless et al., 2001;
Yamamoto et al., 2001). Foxh1 null embryos do not express the two markers of the anterior primitive streak, Foxa2 and Gsc. In agreement with a failure to specify the anterior primitive streak, embryos lacking Foxh1 do not form the tissues that are derived from this cell population, the prechordal plate, anterior definitive endoderm, node, and notochord (Hoodless et al., 2001;
Yamamoto et al., 2001). The transcriptional targets of Foxh1, other than Nodal, that are involved in this process remain unknown.
In this study the transcriptional regulation of the formation of the anterior primitive streak
will be examined. To investigate this we used Foxh1 null embryos in a differential screen with
the goal of identifying novel genes which may play a role in specification of the anterior
primitive streak. From this analysis we identified the EST BG078657 to be the putative 3’ end of
an alternative extended transcript of the Cerberus homolog 1 (Cer1) mRNA. We also identified
thyrotropin-releasing hormone (Trh) as being significantly downregulated in Foxh1 null embryos. Expression analysis revealed that Trh is likely a novel marker of the pluripotent epiblast and the definitive endoderm.
53 2.2 RESULTS
2.2.1 Identification of genes differentially expressed between E6.5 wild-type and Foxh1 null
embryos using Affymetrix™ GeneChips®
To identify genes involved in specification of the anterior primitive streak and novel
transcriptional targets of Foxh1, the gene expression profiles of wild-type and Foxh1 null embryos were compared. Embryos from Foxh1 heterozygous intercrosses were dissected at E6.5
and RNA from pooled wild-type and Foxh1 null embryos was amplified using the Affymetrix™
GeneChip® Eukaryotic Small Sample Target Labelling Assay and examined using Affymetrix™
GeneChips®. The analysis was performed at E6.5 to capture the specification of the anterior
primitive streak cells at the onset of gastrulation. The Affymetrix™ MOE 430 chip set used
contains probesets to examine the expression of 39,015 transcripts and variants from over 34,323
characterized mouse genes. Three biological replicates were performed and the array data were
analyzed using the GeneSpring software. The GC-RMA algorithm was used to normalize the
expression data as this approach has been shown to be best suited for experiments containing
multiple replicates and takes into account the GC content of probes when calculating expression
intensities (Bolstad et al., 2003; Wu, 2003). Following normalization, a paired t-test was used to
identify probesets with a statistically significant difference in expression between the wild-type
and Foxh1 null samples. This analysis identified 851 probesets with a p-value of less than the cutoff of 0.05. Probesets with a change in expression greater than 2-fold were chosen for further analysis and validation. This reduced the list of predicted differentially expressed probesets to
73, which represented 62 genes. Interestingly, only 10 genes were downregulated greater than 2- fold while 28 were upregulated between 2 and 3 fold and 24 were upregulated greater than 3-fold
in Foxh1 null embryos. The 73 differentially expressed probesets and their corresponding
annotations, fold changes, and p-values can be found in Tables 2.1, 2.2, and 2.3.
54
Table 2.1 Genes downregulated greater than 2 fold in Foxh1-/- embryos
Probeset IDs and corresponding gene annotations (NCBI accession numbers, gene symbols and gene titles) for 10 genes significantly (> 2-fold, p < 0.05) downregulated in E6.5 Foxh1-/- embryos. Fold changes and p-values were calculated using the GeneSpring software.
55
Table 2.2 Genes upregulated between 2 and 3 fold in Foxh1-/- embryos
Probeset IDs and corresponding gene annotations (NCBI accession numbers, gene symbols and gene titles) for 28 genes significantly (between 2-fold and 3-fold, p < 0.05) upregulated in E6.5 Foxh1-/- embryos. Fold changes and p-values were calculated using the GeneSpring software.
56
Table 2.3 Genes upregulated greater than 3 fold in Foxh1-/- embryos
Probeset IDs and corresponding gene annotations (NCBI accession numbers, gene symbols and gene titles) for 24 genes significantly (> 3-fold, p < 0.05) upregulated in E6.5 Foxh1-/- embryos. Fold changes and p-values were calculated using the GeneSpring software.
57 2.2.2 Assessment of microarray sensitivity and data quality
To assess the quality of the array data, the expression intensities of several genes known to be misexpressed in Foxh1 null embryos by in situ hybridization, as well as the expression of
Foxh1 itself, were examined. The normalized expression intensities of the control genes used in the analysis are shown in Figure 2.1 As the Foxh1 knockout is a null allele a significant downregulation of the probeset(s) for this gene was predicted (Hoodless et al., 2001; Yamamoto et al., 2001). A significant (2.1-fold, p < 0.05) downregulation between the wild-type and Foxh1 null sample of the Foxh1 probeset 1422213_s_at was observed, while there was no change (1.03- fold) in expression for probeset 1422212_at. Expression intensities of two markers of the anterior primitive streak cell population Gsc and Foxa2, whose expression in the APS is lost in Foxh1 null embryos, was also examined. Interestingly, the probesets for Gsc (1421412_at) and Foxa2
(1422833_at) showed only a slight (1.2-1.3 fold) decrease in expression intensity between the wild-type and Foxh1 null sample. Analysis of left right determination factor 1 (Lefty1)
(1417638_at) expression, a marker of the anterior visceral endoderm and a known transcriptional target of Foxh1, demonstrated a significant (2.01-fold, p < 0.05) downregulation in the Foxh1 null sample, consistent with previously reported data (Hoodless et al., 2001; Takaoka et al.,
2006; Yamamoto et al., 2001; Yamamoto et al., 2004). Cerberus 1 homolog (Cer1), a marker of the anterior visceral endoderm and the anterior definitive endoderm, has also been shown to have reduced expression in Foxh1 null embryos by in situ hybridization (Hoodless et al., 2001;
Yamamoto et al., 2001; Yamamoto et al., 2004). Analysis of the expression intensities of the two probesets for Cer1 in the wild-type and the Foxh1 null sample found a 1.89-fold decrease
(1450256_at) and a 1.06-fold decrease (1450257_at), respectively. In situ hybridization and qRT-PCR analysis has previously revealed that expression of Mix1 homeobox-like 1 (Mixl1) is upregulated in Foxh1 null embryos at E7.0 (Izzi et al., 2007). Thus, expression of the probeset
58 Figure 2.1 Expression intensities of control genes in E6.5 Foxh1+/+ and Foxh1-/- embryos
Normalized expression intensities from the Affymetrix™ GeneChips® for genes whose expression is known to be altered in Foxh1-/- embryos by RT-PCR or in situ hybridization. Blue bars indicate the expression intensity in the E6.5 wild-type sample while red bars represent the E6.5 Foxh1-/- sample. Error bars represent the standard deviation from the three biological replicates of the arrays. The gene and its corresponding probeset(s) are listed along the X-axis. Asterices indicate probesets with a fold-change > 2 and a p-value < 0.05.
59 for Mixl1 was expected to be higher in the Foxh1 null sample. When Mixl1 probeset 1421505_at
was examined, a 1.16-fold decrease in expression in the Foxh1 null sample was observed.
Finally, expression of Nodal was examined, as Foxh1 has been shown to play a critical role in
the autoregulation of Nodal expression (Brennan et al., 2001; Norris et al., 2002; Norris and
Robertson, 1999). Both probesets for Nodal (1422057_at and 1422058_at) showed slight
decreases (1.23-1.66 fold) in the Foxh1 null sample.
Collectively, analysis of a set of genes whose expression has been shown to be increased
or decreased in Foxh1 null embryos by in situ hybridization, as well as known transcriptional
targets of Foxh1, revealed that the expression level of several of these genes as measured by the
Affymetrix™ arrays was not significantly different between the wild-type and Foxh1 null
samples. However, in most cases while the change in expression was not significant (> 2-fold, p
< 0.05) the change in expression was in the correct direction, that is, for genes where expression
is downregulated in the Foxh1 null embryos the corresponding probesets showed a decrease in expression.
2.2.3 Validation of microarray results by RT-PCR and qRT-PCR
Confirmation of the upregulation or downregulation of the 62 genes predicted to be
differentially expressed was performed using reverse transcription polymerase chain reaction
(RT-PCR) and cDNA prepared from E6.5 wild-type and Foxh1 null embryos. The results of this
analysis are shown in Figure 2.2. Using this approach 16 genes were found to have a difference
in expression level between the wild-type and Foxh1 null samples (asterices in Figure 2.2). Of
the 16 genes that showed a difference in expression level between the two samples one gene,
Ly6c, showed the opposite change as to what was observed in the Affymetrix™ array data.
60 Figure 2.2 RT-PCR validation of microarray data
Reverse Transcription Polymerase Chain Reaction (RT-PCR) validation results for the 62 genes significantly (> 2-fold, p < 0.05) downregulated or upregulated in E6.5 Foxh1-/- embryos. The 10 downregulated genes are shown in the far left panel, the 52 upregulated genes are shown in the three right panels. Hypoxanthine guanine phosphoribosyl transferase (HPRT) was used as a loading control. Water (not shown) and cDNA prepared without reverse transcriptase (-RT) were used as negative controls. Genes that showed a downregulation or upregulation in the RT-PCR analysis are marked with a red or green asterisk respectively.
61 These data are the earliest reported expression for 11 of the 16 genes (Itgb3, Prl6a1,
Fmo1, Pcsk5, Pgr, Ipp, Hoxa11, Pdgfc, Ly6c, RIKEN 2310014D11, and Trh) while there is
previously reported evidence for expression in pre- or post-implantation stage mouse embryos
for the remaining five genes (Rgs2, EST BG078657, Lefty1, Tdo2, and Sox17) (MGI). The
function of nine of the genes (Itgb3, Pcsk5, Pgr, Rgs2, Hoxa11, Pdgfc, Trh, Lefty1, and Sox17) has previously been investigated using gene targeting. Of the nine knockouts, Lefty1, Sox17, and
Pcsk5 are embryonic lethal, while Pdgfc is neonatal lethal (Essalmani et al., 2006; Kanai-Azuma et al., 2002; Meno et al., 1998; Perea-Gomez et al., 2002). The phenotypes of the Hoxa11, Trh,
Itgb3, Pgr and Rgs2 knockouts manifest during adulthood and range from disruptions in blood pressure regulation, to impaired platelet function, sterility, and impaired organ formation or function (Holmback et al., 1996; Lydon et al., 1995; Small and Potter, 1993; Tang et al., 2003;
Yamada et al., 1997).
Quantitative real-time polymerase chain reaction (qRT-PCR) was used to further analyze the expression of the 16 genes which showed a difference by RT-PCR and to obtain a more accurate measurement of the amount of upregulation or downregulation. The results of this analysis are shown in Figure 2.3. Of the 15 genes that showed a difference in expression with
RT-PCR, 3 showed a > 2-fold difference in expression between the wild-type and Foxh1 null sample with qRT-PCR – EST BG078657, Trh, and RIKEN 2310012D11. Two additional genes
showed a difference in expression - Fmo1, upregulated in Foxh null embryos and Itgb3
(downregulated in Foxh1 null embryos), although the change in expression was small.
Interestingly, Ly6c, which was predicted to be upregulated in Foxh1 null embryos by the
Affymetrix™ arrays was found to be downregulated in the Foxh1 null embryos by qRT-PCR, consistent with the RT-PCR data. To address the aim of finding genes with a necessary role in specification of the anterior primitive streak we focused our analysis on the two genes that
62
Figure 2.3 qRT-PCR validation of differentially expressed genes
Quantitative Real Time Polymerase Chain Reaction (qRT-PCR) validation results for 15 genes which were found to be differentially expressed between E6.5 wild-type and Foxh1-/- embryos by RT-PCR. The three downregulated genes are shown on the left, the 12 upregulated genes are shown on the right. Blue bars indicate the expression intensity in the E6.5 wild-type sample while red bars represent the E6.5 Foxh1-/- sample. Expression was normalized to Hypoxanthine guanine phosphoribosyl transferase (HPRT) and plotted relative to the wild-type sample (set to 1). Data plotted represents the mean and standard deviation of results from three independent RNA samples. Genes that showed a downregulation or upregulation greater than 2-fold in the qRT-PCR analysis are marked with a red or green asterisk respectively.
63 showed a significant (> 2-fold) downregulation by qRT-PCR, the expressed sequence tag (EST)
BG078657 and thyrotropin-releasing hormone (Trh).
2.2.4 Analysis of BG078657, an uncharacterized expressed sequence tag, in E6.5 mouse embryos
One of the two transcripts that showed a significant downregulation in Foxh1 null embryos was the EST BG078657. This EST clone is among a set of clones from 11 cDNA libraries generated from pre-implantation stage (E0.5 to E4.5) embryos, E7.5 post-implantation stage embryos, the E12.5 female mesonephros/gonad, and newborn ovary cDNA library (Tam et al., 2006; Tremblay et al., 2000). Alignment of EST BG078657 to mRNA databases did not reveal any significant similarity to other annotated mRNA transcripts. It did however have 100%
similarity to the ESTs BI076493 and BG065374, which were cloned from the same set of tissues
as BG078657. A search for an open reading frame within the sequence of BG078657 revealed
several short (35-43 amino acid) open reading frames, however none of these ORFs contained
any known protein domains nor did these amino acid sequences show any similarity to other
protein sequences. The absence of a protein coding region within EST BG078657 could indicate
that this EST is part of an untranslated, functional RNA sequence. We surveyed the sequence of
EST BG078657 for the presence of functional RNA elements, including sequences encoding
microRNAs. This analysis revealed no elements encoding microRNAs or other functional RNA
elements.
The expression pattern of EST BG078657 in E6.5 embryos was determined using whole
mount in situ hybridization. The results of this analysis are shown in Figure 2.4. From E6.0 to
E7.0, EST BG078657 was specifically expressed in the cells of the anterior visceral endoderm
64
Figure 2.4 EST BG078657 is expressed in an identical pattern to that of Cer1 in E6.5, E7.5, and E8.5 mouse embryos
A-F: Expression of EST BG078657 in E6.0 to E9.0 wild-type mouse embryos. Lateral views (A- D and F) with anterior to the left or dorsal view (E) with anterior to the top. From E6.0 to E7.0 EST BG078657 is expressed in the anterior visceral endoderm (empty arrowheads in A-C). Expression is also observed in the nascent anterior definitive endoderm (solid arrowheads in B- D). At E8.5 and E9.0 EST BG078657 is expressed in the caudal-most somites (arrows in E and F). G-I: Expression of Cerberus 1 homolog in E6.5, E7.5, and E8.5 wild-type mouse embryos. At E6.5 and E7.5 Cer1 is expressed in the anterior visceral endoderm (empty arrowhead in G and H). At E7.5 expression is also observed in the nascent anterior definitive endoderm (solid arrowhead in H) while at E8.5 expression is only seen in the caudal-most somites (arrow in I).
65 (open arrowheads in Figure 2.4A,B,C). At E7.5, EST BG078657 was expressed in the nascent anterior definitive endoderm (solid arrowheads in Figure 2.4B,C,D). At E8.5 and E9.0 EST
BG078657 is expressed in the caudal somites (arrows in Figure 2.4E,F). Interestingly, this expression pattern is the same as that of Cer1, which at E6.5 marks the anterior visceral endoderm (open arrowhead in Figure 2.4G,H), then is expressed in the anterior definitive endoderm at E7.5 (solid arrowhead in Figure 2.4H) and finally in the caudal somites at E8.5
(arrow in Figure 2.4I) (Belo et al., 1997; Biben et al., 1998; Shawlot et al., 1998). Thus, EST
BG078657 appears to be expressed in an identical pattern to that of Cer1 in E6.5, E7.5, and E8.5 mouse embryos. Additionally, expression of Cer1 has been shown to be downregulated in E6.5
Foxh1 null embryos, as demonstrated by in situ hybridization, which is similar to the RT-PCR and qRT-PCR data for EST BG078657 in E6.5 Foxh1 null embryos (Hoodless et al., 2001;
Yamamoto et al., 2001). Interestingly, when the location EST BG078657 in the genome was examined it was found on chromosome 4 located 2.7 kilobases (kb) downstream of the 3’ end of the Cer1 gene. Given the lack of an open reading frame with functional protein domains or functional RNA motifs within EST BG078657, its downregulation in E6.5 Foxh1 null embryos, its close proximity to the Cer1 gene locus, and the identical tissue localization to Cer1 in the
E6.5-E8.5 mouse embryo, we hypothesize that EST BG078657 may be an alternate 3’ end of a novel extended variant of the Cer1 mRNA.
2.2.5 Characterization of Trh, a regulator of the hypothalamic-pituitary-thyroid axis, in
E6.5 mouse embryos
qRT-PCR validation of the Affymetrix™ array data revealed thyrotropin-releasing hormone (Trh) to be significantly downregulated in E6.5 Foxh1 null embryos. TRH is a tripeptide hormone that plays a critical role in regulation of the hypothalamic-pituitary-thyroid
66 axis via regulation of the synthesis and secretion of thyroid-stimulating hormone (TSH) from the
anterior pituitary (reviewed in Nillni and Sevarino, 1999; O'Leary and O'Connor, 1995). In the
adult, mRNA for Trh is detected only in the hypothalamus and testis, while in the developing
embryo mRNA for Trh has been detected as early as E8.0 in the developing neural folds and is
also expressed at the midbrain-hindbrain junction from E8.5 to E10.5 and in the developing
hypothalamus from E11.5 to E18.5 (Backman et al., 2003; Caqueret et al., 2006; Goshu et al.,
2004; Jukkola et al., 2006; Keith et al., 2001; Michaud et al., 2000; Michaud et al., 1998;
Schonemann et al., 1995; Wang and Lufkin, 2000). The detection of Trh expression at E6.5,
prior to the formation of the tissues previously shown to express Trh, was surprising and we
further characterized the expression pattern of Trh in the E6.5 mouse embryo using whole mount in situ hybridization. The results of this analysis are shown in Figure 2.5. At E6.5 Trh is
expressed ubiquitously in the epiblast, with no expression observed in the extraembryonic
visceral endoderm or the extraembryonic ectoderm (Figure 2.5A). We also analyzed the
expression pattern of Trh in E6.5 Foxh1 null embryos to determine if the level and/or
localization of Trh expression were altered. In agreement with the Affymetrix™ and qRT-PCR
validation data, whole mount in situ hybridization showed decreased expression of Trh mRNA in
the E6.5 Foxh1 null embryos; however, the extent of downregulation differed between embryos
with some embryos showing only a small decrease while other embryos showed a complete loss of Trh expression (Figure 2.5B,C,D,E,F). At E7.5 we found Trh to be expressed in the outer endoderm layer of the embryo which corresponds to the location of the definitive endoderm at this embryonic stage (Figure 2.5G,H). Furthermore, no expression was observed in the extraembryonic tissues or in the epiblast. Interestingly, expression of Trh remained in the outer endoderm layer of E7.5 Foxh1 null embryos (Figure 2.5I,J,K,L). This result was unexpected, as
Foxh1 null embryos are thought to be deficient in definitive endoderm specification. We investigated definitive endoderm specification in Foxh1 mutant embryos further by performing a
67
Figure 2.5 The expression level of Trh, but not its localization, is altered in Foxh1-/- embryos
A-F: Expression of Trh in E6.5 Foxh1+/+ (A) and Foxh1-/- (B-F) embryos. Lateral views with anterior to the left. At E6.5 Trh is expressed in the epiblast (ep) and not in the visceral endoderm (ve) or extraembryonic (exem) tissues in Foxh1+/+ embryos (A). Trh expression is downregulated in Foxh1-/- embryos to varying degrees ranging from a small decrease (B) to complete absence of expression (E and F). The localization of Trh is not affected in Foxh1-/- embryos. G-L: Expression of Trh in E7.5 Foxh1+/+ (G,H) and Foxh1-/- (I-L) embryos. Lateral views (G,I,K) and transverse sections (H,J,L) with anterior to the left. Location of sections is indicated by dashed line. In E7.5 Foxh1+/+ embryos Trh is expressed in the outer endoderm layer (end) of the embryonic region of the embryo (solid arrowhead in H) and is not expressed in the extraembryonic tissues (exem), ectoderm (ect) or mesoderm (mes). Expression is also not observed in the primitive streak (empty arrowhead in H). Trh expression is downregulated in E7.5 Foxh1-/- embryos (I,K) but expression remains localized to the outer endoderm layer (solid arrowheads in J and L).
68 comprehensive analysis of the expression pattern of Trh and additional markers of the definitive
endoderm in wild-type and Foxh1 null embryos using whole mount in situ hybridization. The
results of this analysis are presented in Chapter 3 and Chapter 4 of this thesis.
2.3 DISCUSSION
In this study a differential screen using Foxh1 null embryos was performed with the aim
of identifying novel genes involved in specification of the anterior primitive streak cell
population, and potentially novel transcriptional targets of Foxh1. This was accomplished using
the Affymetrix™ GeneChip® platform as this technology has previously been used to
successfully identify differentially expressed genes in many different cell and tissue types,
including post-implantation stage mouse embryos (Lickert et al., 2005; Morkel et al., 2003).
2.3.1 Multiple factors contributed to the discrepancies between the observed and expected
results for control genes used to measure array quality and sensitivity
Our preliminary analysis of the array data involved examining the expression levels of
several genes known to be misexpressed in E6.5 Foxh1 null embryos. The expression intensities
of these control genes did not show the expected difference between the wild-type and Foxh1
null samples, which can be attributed to several factors. First, with hybridization based fluorescent expression arrays there is always some amount of background signal, even for genes that are not expressed. This may explain the presence of a fluorescent signal for Foxh1 and other
genes whose expression should be absent in the Foxh1 null sample. Second, as the E6.5 mouse
epiblast contains approximately 600 cells it was necessary to amplify the RNA extracted from the pooled wild-type and Foxh1 null embryos to reduce the amount of required tissue (Snow,
69 1977). An undesired effect of this is the biased amplification of transcripts, where the 3’ end of
transcripts can be amplified more efficiently and/or to a greater extent than the 5’ end of
transcripts. Thus, as a result of the amplification protocol it may be that the region of the mRNA
corresponding to the 3’ probeset was amplified more than the region corresponding to the 5’
probeset. This may explain the discrepancy between the expression intensities for the two
probesets for Foxh1 and Cerberus. Of the two probesets for the Foxh1 mRNA, one showed no
difference between the wild-type and Foxh1 null samples, while the other showed a significant
(> 2-fold, p < 0.05) difference. The probeset which showed the greater difference is located 3’ to
the probeset which showed no difference. The 3’ bias from the amplification process then may
also explain the discrepancy in the data for the probesets for Cerberus, where the 3’ probeset
showed a nearly 2-fold downregulation in the Foxh1 null sample while the 5’ probeset showed
no difference.
As the anterior primitive streak represents a limited cell population of the mouse gastrula
(approximately 40/600 or 7% of the total cell number) the expression of genes expressed solely
in this cell population, such as Gsc, may be below the detection limits of the Affymetrix™ array
(Camus and Tam, 1999). Another concern with the detection of Gsc is that its transcript has a
high percentage of G and C nucleotides, and therefore may not have been efficiently amplified.
Lastly, all of the control genes analyzed are transcription factors or signaling molecules, which
are generally expressed at low levels and therefore may also be below the detection limits of the
Affymetrix™ array. An additional explanation for the lack of a significant difference of the
control genes is that their expression is not completely abolished in the E6.5 Foxh1 null embryos
or expression remains in another embryonic tissue despite the loss of expression in the anterior primitive streak. For example, Foxa2 is not only expressed in the anterior primitive streak but is also expressed in the extraembryonic visceral endoderm and this expression remains in Foxh1
70 null embryos (Hoodless et al., 2001; Yamamoto et al., 2001). Additionally, there is data suggesting that there is a very limited and transient expression of Foxa2 in the anterior primitive streak region of the least severe class of Foxh1 mutants, therefore it is possible that embryos were collected during the short time window in which Foxa2 is expressed in the less severe class of Foxh1 null embryos. Furthermore, while Foxh1 plays a key role in amplifying Nodal levels in the mouse gastrula, Nodal expression is also not completely abolished in E6.5 Foxh1 mutants
(Yamamoto et al., 2001; Yamamoto et al., 2004).
Collectively, the analysis of the control gene data suggests that the cRNA amplification process may have introduced a significant 3’ bias, resulting in probesets located in the central and/or 5’ locations in a transcript not giving an accurate measure of the expression level of the corresponding gene. It also suggests that the Affymetrix™ platform may not be sensitive enough to detect changes in gene expression in such a limited and heterogeneous cell population, changes in genes expressed at low levels such as transcription factors and signaling molecules, or subtle changes in expression levels.
2.3.2 The gene expression profiles of E6.5 wild-type and Foxh1 null embryos are largely similar
Application of statistical and fold change cutoffs to the Affymetrix™ dataset resulted in a list of 73 probesets (62 genes) with a significant change in expression between the wild-type and
Foxh1 null samples. With a significance cutoff of 0.05 approximately 5% of the array probesets
(2250/45,000) on the array would be expected to be called differentially expressed by chance.
Often when a statistical test involving multiple comparisons is performed a multiple testing correction is applied to re-calculate the p-values taking into account the multiple comparisons
71 performed. When a multiple testing correction was applied to this data it reduced the list of
predicted differentially expressed transcripts from 73 to zero. As a multiple testing correction
was not performed in this analysis there is a high possibility that many of the 73 probesets represent false positives. There are several possibilities which may explain the low number of genes predicted to be differentially expressed between the wild-type and Foxh1 null sample. It may be that the expression levels of the majority of genes expressed in the two samples are largely similar or that in fact there are very few genes differentially expressed between E6.5
wild-type and Foxh1 null embryos. Furthermore, it may be that there are genes whose expression
is different between the two samples but the difference was not large enough to satisfy the fold
change and p-value criteria, since many control genes did not meet these criteria, or that the
Affymetrix™ platform was not sensitive enough to detect the changes, as discussed above.
Additionally, the pooling of embryos may have averaged the gene expression levels in the wild-
type and Foxh1 null samples, particularly in the case of the Foxh1 null sample as there are
several classes of mutant phenotypes with different degrees of severity and presumably differing
levels of expression of some genes. This may also explain the large amount of variability seen in
the expression levels for the predicted differentially expressed genes. Lastly, the cRNA
amplification and/or normalization process may have artificially removed any differences
between the samples, causing any changes in expression level to no longer satisfy the fold
change and p-value criteria. Thus, despite the severe defects in embryonic patterning resulting
from the loss of Foxh1, the gene expression profiles of E6.5 wild-type and Foxh1 null embryos
are largely similar.
72 2.3.3 Contrary to its primary role as a transcriptional activator, loss of Foxh1 results in an increase in gene expression
An interesting feature of the list of genes predicted to be differentially expressed is the identification of relatively few genes predicted to be downregulated (10 genes) versus the number predicted to be upregulated in the Foxh1 null sample (52 genes). Several signaling antagonists, such as the BMP pathway inhibitors Chordin and Noggin, are expressed in the anterior primitive streak (Bachiller et al., 2000). In the absence of this cell population (as in
Foxh1 null embryos) these signaling antagonists are lost which may result in a general upregulation in gene expression. Furthermore, recent reports have demonstrated that Foxh1 can function as a transcriptional repressor (Chen et al., 2005; Izzi et al., 2008; Izzi et al., 2007; Labbe et al., 1998). In vivo Foxh1 has been shown to represses the expression of Mixl1, while in vitro
Foxh1 has also been shown to also act as a transcriptional repressor via an interaction with
Goosecoid. If Foxh1 functions as a transcriptional repressor in the mouse gastrula, loss of Foxh1 would result in an increase in expression of genes it normally represses, which may explain the higher number of upregulated genes in the E6.5 Foxh1 null embryos compared to wild-type embryos. In addition, Foxh1 null embryos have been shown to be slightly delayed in their development, which may also explain the higher number of upregulated than downregulated genes. For example, if a gene is highly expressed at a slightly earlier stage and then downregulated, this gene would appear upregulated in Foxh1 null embryos as they correspond to a slightly earlier embryonic stage. It is possible that the upregulation of genes observed in Foxh1 null embryos is due to the fact that some tissues are missing in these embryos and thus the proportion of remaining tissues is higher in wild-type embryos compared to E6.5 Foxh1 null embryos, however this is unlikely as the null embryos are only missing a very small cell population compared to wild-type.
73 Examination of data sets from other differential screens using post-implantation mouse embryos revealed that a greater number of upregulated genes than downregulated genes was observed when the opposite result was expected (Lickert et al., 2005; Morkel et al., 2003; Zakin et al., 2000). We also observed a higher number of genes upregulated in Foxh1 null embryos in a
Serial Analysis of Gene Expression (SAGE) library analysis of E6.5 and E7.5 wild-type embryos versus E6.5 and E7.5 Foxh1 and Foxa2 null embryos (performed independently of this study and data not shown). Thus it appears that loss of critical transcription factors in the early embryo may result in an overall deregulation of gene expression.
2.3.4 EST BG078657 is likely the 3’ end of an extended mRNA of the Cer1 gene
One of the two transcripts significantly downregulated in E6.5 Foxh1 null embryos was the EST BG078657. No open reading frames with functional protein domains or functional RNA motifs were detected within the EST sequence. Examination of the expression pattern of EST
BG078657 in E6.5 wild-type embryos showed that it was expressed in the anterior visceral endoderm, an important signaling centre in the embryo which plays a critical role in specification of anterior neural structures (Thomas and Beddington, 1996). At E7.5 EST BG078657 was found to be expressed in the anterior definitive endoderm and at E8.5 in the caudal-most somites.
Interestingly, this expression pattern exactly matches that of Cerberus 1 homolog (Cer1) (Belo et al., 1997; Biben et al., 1998; Shawlot et al., 1998). Examination of the genomic location of EST
BG078657 revealed that it was located 2.7 kb downstream from the Cer1 gene on Chromosome
4. Analysis of the expression of this extended Cer1 transcript in the Mouse Atlas of Gene
Expression SAGE libraries provided further evidence of expression of the mRNA for Cer1, the
EST BG078657, as well as the intervening sequence between the annotated 3’ end of the Cer1 gene and the 5’ end of EST BG078657 (data not shown). There was no evidence of expression
74 downstream of the 3’ end of EST BG078657. Furthermore, examination of the extended Cer1 mRNA revealed no consensus poly adenylation sites at the end of the predicted mRNA; however, a poly adenylation sequence was found at the 3’ end of EST BG078657. Collectively this data suggested that EST BG078657 is likely the 3’ end of an extended version of the Cer1 mRNA. The extended mRNA is predicted to have a total length of 6.7 kb with the 3’ untranslated region (3’-UTR) being 4.2 kb in length. The 3’UTR has been demonstrated to play a role in regulating transcript stability, localization, and translation (reviewed in de Moor et al.,
2005). Additionally, microRNAs have been found to target the 3’ UTR to exert their regulatory function (reviewed in de Moor et al., 2005). Thus this putative extended mRNA of Cer1 may serve an important role in regulating stability, localization, or translation of the transcript.
2.3.5 Thyrotropin-releasing hormone is a prospective marker of the pluripotent epiblast and definitive endoderm
The second transcript significantly downregulated in Foxh1 null embryos was thyrotropin-releasing hormone (Trh). TRH is a secreted tripeptide that is a key regulator of the hypothalamic-pituitary-thyroid (HPT) axis (reviewed in Nillni and Sevarino, 1999; O'Leary and
O'Connor, 1995). In the adult, TRH is secreted from the hypothalamus and binds to receptors on the pituitary resulting in the release of thyroid-stimulating hormone (TSH). Prior to this study the earliest reported expression of Trh in the mouse was in the developing headfolds at E8.0
(Jukkola et al., 2006). Our data is therefore the earliest reported expression of this gene. Despite the critical role of TRH in regulating the HPT axis and its expression in developing neural structures, mice deficient in Trh are viable and fertile (Yamada et al., 1997; Yamada et al.,
2003). Whole mount in situ hybridization was performed to determine the expression pattern of
Trh at E6.5. Trh was found to be highly and specifically expressed in the epiblast at E6.5. The
75 expression pattern of Trh overlaps with that of Foxh1, which is also ubiquitously expressed in
the epiblast at E6.5, suggesting that Trh may be a transcriptional target of Foxh1 (Weisberg et
al., 1998). In situ hybridization analysis of Trh in E6.5 Foxh1 null embryos embryos revealed varying degrees of downregulation, which could be the result of the different phenotypic classes of Foxh1. As some expression of Trh remained in mutant embryos (which was also observed in the RT-PCR analysis) this suggested that either Foxh1 does not directly regulate Trh expression or that it is not the sole transcription factor regulating Trh expression in the early embryo.
Interestingly, Trh has been identified as a putative transcriptional target of the pluripotency gene
Oct4 (Loh et al., 2006; Matoba et al., 2006). Oct4 is also ubiquitously expressed in the epiblast of E6.5 embryos (Rosner et al., 1990). Oct4 expression is downregulated in embryos with decreased Nodal signaling, such as Smad2Robm1/Robm1 or Smad2Robm1/CA;Smad3-/-, Furin-/-;Pace4-/- and Nodal413.d/413.d deficient embryos, suggesting that loss of Nodal signaling results in decreased
pluripotency of the epiblast (Beck et al., 2002; Brennan et al., 2001; Robertson et al., 2003;
Waldrip et al., 1998). If Trh is a transcriptional target of Oct4, this would explain the reduced
expression of this gene in Foxh1 null embryos, which are deficient in Nodal signaling. Analysis
of the expression pattern of Trh in E7.5 wild-type and Foxh1 null embryos revealed that Trh
expression had shifted from the inner ectoderm to the outer endoderm layer. The localization of
Trh expression at E7.5 strongly suggested that Trh was expressed in the definitive endoderm
cells of the embryo, making Trh a putative novel marker of this tissue. Surprisingly, cells expressing Trh were detected in the outer cell layer of E7.5 Foxh1 null embryos. This result was
particularly interesting, as Foxh1 null embryos phenocopy Nodal hypomorphs and thus are
thought to lack the definitive endoderm as the origin of this tissue, the anterior primitive streak, is not specified in Foxh1 null embryos (Hoodless et al., 2001; Yamamoto et al., 2001). A comprehensive analysis of the expression pattern of Trh in post-implantation embryos will be necessary to confirm whether Trh is a novel marker of the mouse definitive endoderm. If Trh is
76 in fact a novel marker of the mouse definitive endoderm the presence of Trh expressing cells in
E7.5 Foxh1 null embryos has interesting implications, as it suggests that, contrary to current endoderm formation models, that some definitive endoderm is formed in these mutants.
2.4 EXPERIMENTAL PROCEDURES
2.4.1 Mouse strains, dissection, staging, and genotyping
Embryos were obtained from crosses of inbred Foxh1 heterozygous mice and considered
0.5 days postcoitum at noon of the day of detection of the vaginal plug. The Foxh1 targeted allele was backcrossed from a mixed 129/ICR background onto a C57BL/6J background for 10 generations and subsequently maintained on the C57BL/6J background with intercrosses.
Embryos were dissected from the uterus in 1x Dulbecco’s Phosphate Buffered Saline
(Invitrogen) with 0.1% Fetal Bovine Serum (Invitrogen). Staging of embryos was carried out as previously described (Downs and Davies, 1993). Embryos were dissected between 2pm and
4pm. For in situ hybridization embryos were placed into individual tubes and fixed overnight at
4˚C in 4% paraformaldehyde, dehydrated through a graded methanol series and stored at -20˚C until use. For RNA collection embryos were placed into individual tubes of Trizol and stored at -
80˚C until use. Genotyping was carried out by removing the ectoplacental cone (epc) at the time of dissection and culturing it for 4-5 days in DMEM + 10% FBS on gelatin coated tissue culture dishes. Epc lysis and PCR genotyping was carried out as previously described (Hoodless et al.,
2001). Following genotyping embryos were pooled according to genotype (Foxh1+/+, Foxh1+/-,
Foxh1-/-) and processed for RNA extraction or in situ hybridization as described below.
77 2.4.2 RNA extraction, cDNA synthesis, amplification, and labeling
Embryos were placed into Trizol (Invitrogen) immediately following dissection and stored at -80˚C. Following genotyping, wild-type and Foxh1 null embryos were pooled (10 per replicate). RNA was extracted using MaXtract high density phase-lock gels (Qiagen) following manufacturer’s instructions. cRNA amplification and labeling for hybridization to the
Affymetrix™ oligonucleotide arrays was carried out at the BCCA Michael Smith Genome
Sciences Centre according to standard Affymetrix™ protocols. 100 ng of total RNA was amplified and labeled using the Affymetrix™ GeneChip® Eukaryotic Small Sample Target
Labeling Assay. cDNA synthesis for RT-PCR and qRT-PCR was performed using standard procedures, as described in McKnight et al., 2007.
2.4.3 Affymetrix™ GeneChip® data acquisition and analysis
The Affymetrix MOE430 Expression Set was used for all three biological replicates.
Hybridization and scanning of the chips was performed at the BCCA Michael Smith Genome
Sciences Centre according to standard Affymetrix™ protocols. Data analysis was performed using the GeneSpring software (Silicon Genetics). CEL files were imported and expression intensities were normalized using the GC-RMA plug-in algorithm. A paired t-test was then performed to find probesets that had a significant change in expression (p < 0.05) between the wild-type and Foxh1 null sample. From the list of probesets with a p-value < 0.05 those with a change in expression > 2-fold were chosen for further validation and analysis.
78 2.4.4 Reverse Transcription-PCR and Quantitative Real Time PCR
RT-PCR was performed using HotStar Taq Polymerase (Qiagen) as per manufacturer’s
instructions in a PTC-200 thermal cycler (MJ Research). Thermal cycler conditions were as
follows: 95˚C 15 minutes, followed by N cycles of 95˚C 30 seconds, 55˚C 30 seconds, 72˚C 30 seconds, then 72˚C 10 minutes, where N ranged from 25-40 and was optimized for each primer set. qRT-PCR was performed using SYBR® Green supermix (Applied Biosystems) according to manufacturer’s instructions in an ABI 7500 real-time PCR system (Applied Biosystems).
Cycling conditions were as follows: 50˚C 2 minutes, followed by 40 cycles of 95˚C 15 seconds,
60˚C 1 minute. qRT-PCR was carried out on three biological replicates and all reactions were performed in triplicate. Data was analyzed using the Applied Biosystems 7500 System Software.
Samples were first normalized to HPRT, and the change in expression relative to the wild-type samples was calculated using the ∆∆Ct method. Primers used for RT-PCR and/or qRT-PCR are described in Table 2.4.
79
80
Table 2.4 Primers used for RT-PCR and qRT-PCR analysis
81 2.4.5 In situ hybridization probe preparation
Digoxigenin-UTP labeled mRNA probes were synthesized as per manufacturer’s
instructions (Roche). Briefly, 10 μg of template DNA was linearized with 10 Units of an
appropriate enzyme and purified. RNA probes were in vitro transcribed using 1 μg of linearized template and 20 Units of SP6, T3 or T7 RNA polymerase (Roche). RNA probes were purified using G-50 Sephadex columns (Amersham) and stored at -80˚C. In situ probes used in this study
(Trh and Cerberus homolog 1) are described in (Hoodless et al., 2001; Hou et al., 2007). The in situ probe for EST BG078657 was generated as follows: EST BG078657 was amplified using
E7.5 cDNA as template and the following primers: Forward 5’-
TTTGGGCATTTGTTGACATGGGTC-3’, Reverse 5’-
CCAGAAGCTTGCTGTGATGGCAAA-3’. The PCR product was subcloned into the pCRII vector (Invitrogen) as per the manufacturer’s instructions. Orientation was verified by sequencing. To generate the antisense probe the vector was linearized with BamH1 and RNA was transcribed using T7 polymerase. For the sense control probe the vector was linearized with
NotI and RNA was transcribed using SP6 polymerase.
2.4.6 Whole mount in situ hybridization
Whole mount in situ hybridization (WISH) was carried out essentially as in Wilkinson
(Wilkinson and Nieto, 1993) with modifications as described in (McKnight et al., 2007).
82 2.4.7 Imaging and histology
Following WISH, embryos were processed through a graded PBT:Glycerol series to 50% glycerol. Images were taken using a Leica MZ9.5 microscope and a Leica DFC420 camera with the Leica Application Suite software (Leica). For histology, embryos were incubated overnight at
4˚C in a 1:1 mixture of Tissue-Tek OCT embedding medium (Sakura Finetek) and 60% sucrose.
Embryos were embedded in 100% OCT (in 10x10 mm plastic mold) and frozen with a dry ice:ethanol mixture. Cryosections (10 uM) were cut using a Microm cryostat and placed on
Superfrost Plus glass slides (VWR) and allowed to air dry. Slides were washed twice in PBS for
10 minutes per wash and mounted using 50% glycerol and glass coverslips (VWR). Sections were imaged using Zeiss Axioplan microscope and OpenLab software (Improvision).
83 2.5 REFERENCES
Bachiller, D., Klingensmith, J., Kemp, C., Belo, J. A., Anderson, R. M., May, S. R., McMahon, J. A., McMahon, A. P., Harland, R. M., Rossant, J., and De Robertis, E. M. (2000). The organizer factors Chordin and Noggin are required for mouse forebrain development. Nature 403, 658-61. Backman, M., Machon, O., Van Den Bout, C. J., and Krauss, S. (2003). Targeted disruption of mouse Dach1 results in postnatal lethality. Dev Dyn 226, 139-44. Beck, S., Le Good, J. A., Guzman, M., Ben Haim, N., Roy, K., Beermann, F., and Constam, D. B. (2002). Extraembryonic proteases regulate Nodal signalling during gastrulation. Nat Cell Biol 4, 981-5. Beddington, R. S. (1994). Induction of a second neural axis by the mouse node. Development 120, 613-20. Beddington, R. S., and Robertson, E. J. (1999). Axis development and early asymmetry in mammals. Cell 96, 195-209. Belo, J. A., Bouwmeester, T., Leyns, L., Kertesz, N., Gallo, M., Follettie, M., and De Robertis, E. M. (1997). Cerberus-like is a secreted factor with neutralizing activity expressed in the anterior primitive endoderm of the mouse gastrula. Mech Dev 68, 45-57. Biben, C., Stanley, E., Fabri, L., Kotecha, S., Rhinn, M., Drinkwater, C., Lah, M., Wang, C. C., Nash, A., Hilton, D., Ang, S. L., Mohun, T., and Harvey, R. P. (1998). Murine cerberus homologue mCer-1: a candidate anterior patterning molecule. Dev Biol 194, 135-51. Bolstad, B. M., Irizarry, R. A., Astrand, M., and Speed, T. P. (2003). A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics 19, 185-93. Brennan, J., Lu, C. C., Norris, D. P., Rodriguez, T. A., Beddington, R. S., and Robertson, E. J. (2001). Nodal signalling in the epiblast patterns the early mouse embryo. Nature 411, 965-9. Camus, A., and Tam, P. P. (1999). The organizer of the gastrulating mouse embryo. Curr Top Dev Biol 45, 117-53. Caqueret, A., Boucher, F., and Michaud, J. L. (2006). Laminar organization of the early developing anterior hypothalamus. Dev Biol 298, 95-106. Chen, G., Nomura, M., Morinaga, H., Matsubara, E., Okabe, T., Goto, K., Yanase, T., Zheng, H., Lu, J., and Nawata, H. (2005). Modulation of androgen receptor transactivation by FoxH1. A newly identified androgen receptor corepressor. J Biol Chem 280, 36355-63. Davidson, B. P., and Tam, P. P. (2000). The node of the mouse embryo. Curr Biol 10, R617-9. de Moor, C. H., Meijer, H., and Lissenden, S. (2005). Mechanisms of translational control by the 3' UTR in development and differentiation. Semin Cell Dev Biol 16, 49-58. Downs, K. M., and Davies, T. (1993). Staging of gastrulating mouse embryos by morphological landmarks in the dissecting microscope. Development 118, 1255-66. Essalmani, R., Hamelin, J., Marcinkiewicz, J., Chamberland, A., Mbikay, M., Chretien, M., Seidah, N. G., and Prat, A. (2006). Deletion of the gene encoding proprotein convertase 5/6 causes early embryonic lethality in the mouse. Mol Cell Biol 26, 354-61. Goshu, E., Jin, H., Lovejoy, J., Marion, J. F., Michaud, J. L., and Fan, C. M. (2004). Sim2 contributes to neuroendocrine hormone gene expression in the anterior hypothalamus. Mol Endocrinol 18, 1251-62.
84 Holmback, K., Danton, M. J., Suh, T. T., Daugherty, C. C., and Degen, J. L. (1996). Impaired platelet aggregation and sustained bleeding in mice lacking the fibrinogen motif bound by integrin alpha IIb beta 3. Embo J 15, 5760-71. Hoodless, P. A., Pye, M., Chazaud, C., Labbe, E., Attisano, L., Rossant, J., and Wrana, J. L. (2001). FoxH1 (Fast) functions to specify the anterior primitive streak in the mouse. Genes Dev 15, 1257-71. Hou, J., Charters, A. M., Lee, S. C., Zhao, Y., Wu, M. K., Jones, S. J., Marra, M. A., and Hoodless, P. A. (2007). A systematic screen for genes expressed in definitive endoderm by Serial Analysis of Gene Expression (SAGE). BMC Dev Biol 7, 92. Izzi, L., Narimatsu, M., and Attisano, L. (2008). Sumoylation differentially regulates Goosecoid- mediated transcriptional repression. Exp Cell Res 314, 1585-94. Izzi, L., Silvestri, C., von Both, I., Labbe, E., Zakin, L., Wrana, J. L., and Attisano, L. (2007). Foxh1 recruits Gsc to negatively regulate Mixl1 expression during early mouse development. Embo J 26, 3132-43. Jukkola, T., Lahti, L., Naserke, T., Wurst, W., and Partanen, J. (2006). FGF regulated gene- expression and neuronal differentiation in the developing midbrain-hindbrain region. Dev Biol 297, 141-57. Kanai-Azuma, M., Kanai, Y., Gad, J. M., Tajima, Y., Taya, C., Kurohmaru, M., Sanai, Y., Yonekawa, H., Yazaki, K., Tam, P. P., and Hayashi, Y. (2002). Depletion of definitive gut endoderm in Sox17-null mutant mice. Development 129, 2367-79. Keith, B., Adelman, D. M., and Simon, M. C. (2001). Targeted mutation of the murine arylhydrocarbon receptor nuclear translocator 2 (Arnt2) gene reveals partial redundancy with Arnt. Proc Natl Acad Sci U S A 98, 6692-7. Kinder, S. J., Tsang, T. E., Wakamiya, M., Sasaki, H., Behringer, R. R., Nagy, A., and Tam, P. P. (2001). The organizer of the mouse gastrula is composed of a dynamic population of progenitor cells for the axial mesoderm. Development 128, 3623-34. Labbe, E., Silvestri, C., Hoodless, P. A., Wrana, J. L., and Attisano, L. (1998). Smad2 and Smad3 positively and negatively regulate TGF beta-dependent transcription through the forkhead DNA-binding protein FAST2. Mol Cell 2, 109-20. Lickert, H., Cox, B., Wehrle, C., Taketo, M. M., Kemler, R., and Rossant, J. (2005). Dissecting Wnt/beta-catenin signaling during gastrulation using RNA interference in mouse embryos. Development 132, 2599-609. Loh, Y. H., Wu, Q., Chew, J. L., Vega, V. B., Zhang, W., Chen, X., Bourque, G., George, J., Leong, B., Liu, J., Wong, K. Y., Sung, K. W., Lee, C. W., Zhao, X. D., Chiu, K. P., Lipovich, L., Kuznetsov, V. A., Robson, P., Stanton, L. W., Wei, C. L., Ruan, Y., Lim, B., and Ng, H. H. (2006). The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nat Genet 38, 431-40. Lydon, J. P., DeMayo, F. J., Funk, C. R., Mani, S. K., Hughes, A. R., Montgomery, C. A., Jr., Shyamala, G., Conneely, O. M., and O'Malley, B. W. (1995). Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev 9, 2266-78. Matoba, R., Niwa, H., Masui, S., Ohtsuka, S., Carter, M. G., Sharov, A. A., and Ko, M. S. H. (2006). Dissecting Oct3/4-Regulated Gene Networks in Embryonic Stem Cells by Expression Profiling. PLOS One, 1-15. McKnight, K. D., Hou, J., and Hoodless, P. A. (2007). Dynamic expression of Thyrotropin- releasing hormone in the mouse definitive endoderm. Dev Dyn 236, 2909-17. Meno, C., Shimono, A., Saijoh, Y., Yashiro, K., Mochida, K., Ohishi, S., Noji, S., Kondoh, H., and Hamada, H. (1998). lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 94, 287-97. MGI. Mouse Genome Informatics Database. www.informatics.jax.org.
85 Michaud, J. L., DeRossi, C., May, N. R., Holdener, B. C., and Fan, C. M. (2000). ARNT2 acts as the dimerization partner of SIM1 for the development of the hypothalamus. Mech Dev 90, 253-61. Michaud, J. L., Rosenquist, T., May, N. R., and Fan, C. M. (1998). Development of neuroendocrine lineages requires the bHLH-PAS transcription factor SIM1. Genes Dev 12, 3264-75. Morkel, M., Huelsken, J., Wakamiya, M., Ding, J., van de Wetering, M., Clevers, H., Taketo, M. M., Behringer, R. R., Shen, M. M., and Birchmeier, W. (2003). Beta-catenin regulates Cripto- and Wnt3-dependent gene expression programs in mouse axis and mesoderm formation. Development 130, 6283-94. Nillni, E. A., and Sevarino, K. A. (1999). The biology of pro-thyrotropin-releasing hormone- derived peptides. Endocr Rev 20, 599-648. Norris, D. P., Brennan, J., Bikoff, E. K., and Robertson, E. J. (2002). The Foxh1-dependent autoregulatory enhancer controls the level of Nodal signals in the mouse embryo. Development 129, 3455-68. Norris, D. P., and Robertson, E. J. (1999). Asymmetric and node-specific nodal expression patterns are controlled by two distinct cis-acting regulatory elements. Genes Dev 13, 1575-88. O'Leary, R., and O'Connor, B. (1995). Thyrotropin-releasing hormone. J Neurochem 65, 953-63. Perea-Gomez, A., Vella, F. D., Shawlot, W., Oulad-Abdelghani, M., Chazaud, C., Meno, C., Pfister, V., Chen, L., Robertson, E., Hamada, H., Behringer, R. R., and Ang, S. L. (2002). Nodal antagonists in the anterior visceral endoderm prevent the formation of multiple primitive streaks. Dev Cell 3, 745-56. Robb, L., and Tam, P. P. (2004). Gastrula organiser and embryonic patterning in the mouse. Semin Cell Dev Biol 15, 543-54. Robertson, E. J., Norris, D. P., Brennan, J., and Bikoff, E. K. (2003). Control of early anterior- posterior patterning in the mouse embryo by TGF-beta signalling. Philos Trans R Soc Lond B Biol Sci 358, 1351-7; discussion 1357. Rosner, M. H., Vigano, M. A., Ozato, K., Timmons, P. M., Poirier, F., Rigby, P. W., and Staudt, L. M. (1990). A POU-domain transcription factor in early stem cells and germ cells of the mammalian embryo. Nature 345, 686-92. Schonemann, M. D., Ryan, A. K., McEvilly, R. J., O'Connell, S. M., Arias, C. A., Kalla, K. A., Li, P., Sawchenko, P. E., and Rosenfeld, M. G. (1995). Development and survival of the endocrine hypothalamus and posterior pituitary gland requires the neuronal POU domain factor Brn-2. Genes Dev 9, 3122-35. Shawlot, W., Deng, J. M., and Behringer, R. R. (1998). Expression of the mouse cerberus-related gene, Cerr1, suggests a role in anterior neural induction and somitogenesis. Proc Natl Acad Sci U S A 95, 6198-203. Small, K. M., and Potter, S. S. (1993). Homeotic transformations and limb defects in Hox A11 mutant mice. Genes Dev 7, 2318-28. Snow, M. H. L. (1977). Gastrulation in the mouse: Growth and regionalization of the epiblast. Journal of embryology and experimental morphology 42, 293-303. Takaoka, K., Yamamoto, M., Shiratori, H., Meno, C., Rossant, J., Saijoh, Y., and Hamada, H. (2006). The mouse embryo autonomously acquires anterior-posterior polarity at implantation. Dev Cell 10, 451-9. Tam, P. P., and Behringer, R. R. (1997). Mouse gastrulation: the formation of a mammalian body plan. Mech Dev 68, 3-25. Tam, P. P., Loebel, D. A., and Tanaka, S. S. (2006). Building the mouse gastrula: signals, asymmetry and lineages. Curr Opin Genet Dev 16, 419-25.
86 Tang, K. M., Wang, G. R., Lu, P., Karas, R. H., Aronovitz, M., Heximer, S. P., Kaltenbronn, K. M., Blumer, K. J., Siderovski, D. P., Zhu, Y., and Mendelsohn, M. E. (2003). Regulator of G-protein signaling-2 mediates vascular smooth muscle relaxation and blood pressure. Nat Med 9, 1506-12. Thomas, P., and Beddington, R. (1996). Anterior primitive endoderm may be responsible for patterning the anterior neural plate in the mouse embryo. Curr Biol 6, 1487-96. Tremblay, K. D., Hoodless, P. A., Bikoff, E. K., and Robertson, E. J. (2000). Formation of the definitive endoderm in mouse is a Smad2-dependent process. Development 127, 3079-90. Waldrip, W. R., Bikoff, E. K., Hoodless, P. A., Wrana, J. L., and Robertson, E. J. (1998). Smad2 signaling in extraembryonic tissues determines anterior-posterior polarity of the early mouse embryo. Cell 92, 797-808. Wang, W., and Lufkin, T. (2000). The murine Otp homeobox gene plays an essential role in the specification of neuronal cell lineages in the developing hypothalamus. Dev Biol 227, 432-49. Weisberg, E., Winnier, G. E., Chen, X., Farnsworth, C. L., Hogan, B. L., and Whitman, M. (1998). A mouse homologue of FAST-1 transduces TGF beta superfamily signals and is expressed during early embryogenesis. Mech Dev 79, 17-27. Wilkinson, D. G., and Nieto, M. A. (1993). Detection of messenger RNA by in situ hybridization to tissue sections and whole mounts. Methods Enzymol 225, 361-73. Wu, Z., Irizarry, R.A., Gentleman, R., Murillo, F.M. & Spencer, F. (2003). A Model Based Background Adjustment for Oligonucleotide Expression Arrays. Technical Report. John Hopkins University, Department of Biostatistics Working Papers, Baltimore, MD. Yamada, M., Saga, Y., Shibusawa, N., Hirato, J., Murakami, M., Iwasaki, T., Hashimoto, K., Satoh, T., Wakabayashi, K., Taketo, M. M., and Mori, M. (1997). Tertiary hypothyroidism and hyperglycemia in mice with targeted disruption of the thyrotropin- releasing hormone gene. Proc Natl Acad Sci U S A 94, 10862-7. Yamada, M., Satoh, T., and Mori, M. (2003). Mice lacking the thyrotropin-releasing hormone gene: what do they tell us? Thyroid 13, 1111-21. Yamamoto, M., Meno, C., Sakai, Y., Shiratori, H., Mochida, K., Ikawa, Y., Saijoh, Y., and Hamada, H. (2001). The transcription factor FoxH1 (FAST) mediates Nodal signaling during anterior-posterior patterning and node formation in the mouse. Genes Dev 15, 1242-56. Yamamoto, M., Saijoh, Y., Perea-Gomez, A., Shawlot, W., Behringer, R. R., Ang, S. L., Hamada, H., and Meno, C. (2004). Nodal antagonists regulate formation of the anteroposterior axis of the mouse embryo. Nature 428, 387-92. Zakin, L., Reversade, B., Virlon, B., Rusniok, C., Glaser, P., Elalouf, J. M., and Brulet, P. (2000). Gene expression profiles in normal and Otx2-/- early gastrulating mouse embryos. Proc Natl Acad Sci U S A 97, 14388-93.
87 Chapter 3
DYNAMIC EXPRESSION OF THYROTROPIN-RELEASING HORMONE IN
THE MOUSE DEFINITIVE ENDODERM1
3.1 INTRODUCTION
Gastrulation is the process where the three primary germ layers of the embryo, the
ectoderm, mesoderm and endoderm are formed (reviewed in Tam and Behringer, 1997). Prior to
gastrulation the cup-shaped mouse embryo consists of two cell layers: the inner epiblast which
gives rise to all the embryonic tissues and the outer visceral endoderm which contributes to the
extraembryonic yolk sac. Gastrulation in the mouse embryo begins at approximately embryonic
day 6.5 (E6.5) with the formation of a transient structure called the primitive streak on the
posterior side of the embryo. Cells of the epiblast undergo an epithelial-to-mesenchymal
transformation then move through the primitive streak and emerge as either mesoderm or definitive endoderm. As the newly-formed definitive endoderm emerges from the primitive streak it intercalates into the pre-existing visceral endoderm layer then moves anteriorly and laterally, thereby displacing the visceral endoderm towards the proximal part of the embryo
(Lawson et al., 1986; Lawson and Pedersen, 1987). Thus, when gastrulation is complete the definitive endoderm comprises a single cell layer which covers the entire embryonic region of the mouse embryo.
. . 1 A version of this chapter has been published. McKnight et al. (2007) Dynamic expression of thyrotropin-releasing hormone in the mouse definitive endoderm. Developmental Dynamics. 236:2909-2917.
88 Numerous fate mapping studies have detailed the movement of the definitive endoderm as it emerges from the primitive streak and displaces the visceral endoderm (Beddington, 1983;
Lawson et al., 1986; Lawson and Pedersen, 1987; Tam and Beddington, 1992; Tam et al., 2007;
Tam et al., 2004). It has recently been shown that there are two paths of migration for the foregut precursors (Tam et al., 2007). At the mid-streak stage definitive endoderm cells emerge from the primitive streak and move towards the anterior of the embryo by way of the lateral sides of the embryo. Lineage tracing has shown these cells contribute to the ventral foregut and have been termed the “lateral” endoderm. Several hours later at the mid- to late streak stage, the primitive streak extends nearly to the distal tip of the embryo. Definitive endoderm cells emerging from the streak at this stage move towards the anterior of the embryo along the midline and contribute to the dorsal foregut. These cells have been termed the “medial” endoderm. Less is known about the formation of the midgut and hindgut precursors. It is thought that these cells are recruited during the late allantoic bud to early head fold stage as the primitive streak is regressing towards the posterior of the embryo and then subsequently expand to line the midgut region and hindgut invagination.
After recruitment of the definitive endoderm is complete it undergoes several morphogenetic movements resulting in the formation of the foregut and hindgut invaginations at the anterior and posterior of the embryo respectively (reviewed in Wells and Melton, 1999). Soon afterwards, specification and differentiation of the newly formed gut begins ultimately leading to the formation of the endoderm derived organs - the liver, pancreas, lungs, thyroid, thymus and epithelial lining of the gastrointestinal tract.
Of the three germ layers, the least is known about the definitive endoderm. This is due, in part, to the relative lack of genetic markers of this tissue. Many genes which are expressed in the
89 definitive endoderm, such as Hex, Cer1, Foxa2, and Sox17, are also expressed in the
extraembryonic visceral endoderm (Ang et al., 1993; Belo et al., 1997; Biben et al., 1998;
Hallonet et al., 2002; Kanai-Azuma et al., 2002; Shawlot et al., 1998; Thomas et al., 1998). A
second limitation of these definitive endoderm markers is that they are not expressed in the entire definitive endoderm at any point during gastrulation but instead only mark specific regions of the definitive endoderm. Identification of novel endoderm markers is necessary to further our understanding of definitive endoderm formation. Additionally, genes expressed exclusively in the definitive endoderm will be of great use in characterizing cell populations that are produced from embryonic stem cell protocols as potential therapeutic agents.
We have previously described a systematic screen to identify genes expressed in the definitive endoderm by Serial Analysis of Gene Expression (SAGE) (Hou et al., 2007). In this study, we identified thyrotropin-releasing hormone (Trh) as being differentially expressed between the foregut and hindgut definitive endoderm at the 8-12 somite stage. Specifically, Trh expression is higher in the foregut as compared to the hindgut. Thyrotropin-releasing hormone
(TRH) is a weakly basic tripeptide (pyroglutamyl-histidyl-proline amide) and plays a critical role in regulation of the hypothalamic-pituitary-thyroid (HPT) axis via regulation of the synthesis and secretion of thyroid-stimulating hormone (TSH) in the anterior pituitary (reviewed in Nillni and
Sevarino, 1999; O'Leary and O'Connor, 1995). The active form of TRH arises from the post- translational cleavage of a large precursor protein via the proprotein convertases PC1 and PC2 involving 3 cleavage steps at 14 cleavage sites within the precursor peptide (Schaner et al., 1997;
Sevarino et al., 1989). TRH acts via two receptors, TRHR and TRHR2, resulting in the production of second messengers and the subsequent activation of several downstream kinases ultimately leading to TSH release (reviewed in Nillni and Sevarino, 1999; O'Leary and O'Connor,
1995).
90 TRH, as well as several of its processing intermediates, are widely distributed throughout
adult body tissues (reviewed in Nillni and Sevarino, 1999; O'Leary and O'Connor, 1995). Despite
this, the mRNA for mouse Trh has only been reported in the hypothalamus and testis, suggesting that endogenous TRH is produced only in these locations (Satoh et al., 1992). The presence of
TRH throughout the central nervous system as well as in numerous peripheral organs supports a diverse range of roles for this molecule outside of the traditional HPT axis, including a potential role as a neurotransmitter. During mouse embryogenesis Trh mRNA has been detected in the neural folds at E8.0, at the midbrain-hindbrain junction from E8.5 to E10.5, and in the developing hypothalamus from E11.5 to E18.5 (Acampora et al., 1999; Backman et al., 2003; Caqueret et al.,
2006; Goshu et al., 2004; Jukkola et al., 2006; Keith et al., 2001; Michaud et al., 2000; Michaud et al., 1998; Schonemann et al., 1995; Wang and Lufkin, 2000). Despite the critical role of TRH in regulating the HPT axis and its expression in developing neural structures, mice deficient in
Trh are viable and fertile (Yamada et al., 1997; Yamada et al., 2003).
We have characterized the expression pattern of Trh in the developing mouse embryo using whole mount in situ hybridization. Trh is expressed at low levels in the epiblast prior to gastrulation, however during gastrulation Trh is exclusively expressed in the definitive endoderm until E8.5. Trh expression then rapidly decreases such that all endoderm expression is lost by
E9.0 coinciding with the onset of expression in the developing neural tissues. Thus, Trh is a novel marker of the mouse definitive endoderm that is expressed specifically and dynamically in this tissue.
91 3.2 RESULTS
3.2.1 Expression of components of the HPT axis during post-implantation stages in the mouse embryo
Our previous analysis identified Trh as being enriched in the foregut definitive endoderm of 8-12 somite (Theiler Stage (TS) 13) embryos (Hou et al., 2007). To determine at what embryonic stage Trh expression begins, we analyzed Serial Analysis of Gene Expression (SAGE)
libraries generated for the Mouse Atlas of Gene Expression project (www.mouseatlas.org)
(Figure 3.1A) (Siddiqui et al., 2005). SAGE is a powerful method for the analysis of gene
expression patterns where the underlying principle is that gene transcripts are represented as short
tag sequences that correlate to the abundance of that transcript (Velculescu et al., 1995). Trh
expression is detected in only 21 of 151 embryonic SAGE libraries, suggesting a very specific
and restricted expression. The highest Trh expression is observed between TS 9 (E6.5) and TS 13
(E8.5), corresponding to post-implantation stages. Specifically, high levels of Trh transcripts are
detected in the embryonic region of the conceptus and the early definitive endoderm. Trh
expression is also observed in the developing neural tissues, ovaries, spleen, pancreas and heart,
although at much lower levels. Interestingly, the highest level of Trh is seen in mouse embryonic
stem cells. However, despite the high expression of Trh at E6.5 and in embryonic stem cells, we
do not see any tags for Trh in pre-implantation stages or at E5.5.
To confirm the expression of Trh in the mouse embryo we used RT-PCR. This analysis confirmed the expression of Trh at E6.5, E7.5 and E8.5 observed in the SAGE libraries and revealed that the levels are nearly equivalent to those in adult mouse brain (Figure 3.1B). The expression of Trh at early embryonic stages prompted us to determine whether other components
92
Figure 3.1 Expression of components of the HPT axis during mouse embryogenesis
A: Detection of transcripts for Trh in Mouse Atlas of Gene Expression SAGE libraries. The data is represented as the tag count for Trh normalized to library size and expressed as tags per 100,000 total tags. TS: Theiler stage. B: RT-PCR analysis of components of the HPT axis during post-implantation stages in the mouse embryo. Water (dH20) and cDNA prepared without reverse transcriptase (-RT) were used as negative controls.
93 of the HPT axis are also expressed during post-implantation stages in the mouse embryo (Figure
3.1B). Using RT-PCR we analyzed the expression of the two receptors for TRH. Whereas Trhr2 is detected at E6.5, E7.5 and E8.5, Trhr is only detected in the adult brain. Since we detected expression of Trh and Trhr2 in the early mouse embryo we also wanted to examine the expression of the downstream targets of TRH signaling, specifically TSH which is the primary target of TRH in the adult. We examined the expression of the two subunits of TSH and found both Tsha and Tshb are only detected in the adult brain sample. Collectively, this data indicates that Trh and one of its receptors, Trhr2, are expressed during post-implantation stages in the mouse embryo whereas its primary targets, Tsha and Tshb, are not.
3.2.2 Transition of Trh expression from the epiblast to the definitive endoderm during gastrulation
The expression of Trh during post-implantation mouse development prior to formation of the hypothalamus, pituitary and other target organs of Trh is surprising and suggests a role for
TRH outside of regulating the HPT endocrine axis or acting as a neurotransmitter. As the tissue localization of Trh mRNA has yet to be analyzed in post-implantation stage mouse embryos, we performed an in situ hybridization analysis. We focused our analysis from E6.0 to E8.5 as these embryonic stages showed the highest levels of Trh expression in the SAGE library analysis.
In agreement with the SAGE data, Trh mRNA is not detected in E5.5 embryos (data not shown) and expression is first observed in the epiblast of early streak (ES) stage (E6.25) embryos
(Figure 3.2A,B). Trh is specific to the epiblast and is not expressed in the extraembryonic ectoderm or the visceral endoderm (Figure 3.2A,B). Interestingly, Trh is not expressed in the epiblast cells located in the newly forming primitive streak (Figure 3.2B). As gastrulation
94
Figure 3.2 Expression of Trh during gastrulation
A,B: Trh expression in early streak (ES) stage embryo. Lateral view (A) and transverse section (B). Trh is expressed in the epiblast and is absent from the primitive streak (ps), visceral endoderm (ve), and extraembryonic tissues (exem). C-G: Lateral views (C,E) and transverse sections (D,F,G) of late streak (LS) stage embryos. Expression of Trh in the distal region of the epiblast and on the lateral sides of the embryo (arrowheads in D,F). H-K: Trh expression in early bud (EB) stage embryo. Lateral (H) and frontal (I) views and transverse sections (J,K). Trh is expressed in the entire definitive endoderm and is absent from the extraembryonic visceral endoderm and notochordal plate (arrowheads in I,K). L-O: Trh expression in late bud (LB) stage embryo. Lateral (L) and frontal (M) views and transverse sections (N,O). Trh expression is enriched in the anterior definitive endoderm (arrowhead in L) and no expression is seen in the notochordal plate (arrowheads in M,N,O).
95 progresses, cells expressing high levels of Trh are observed on the lateral sides of the embryo
(Figure 3.2C). Examination of embryo sections shows that these cells are located in the outer
endoderm layer and that the weaker epiblast expression is now restricted to the distal region of
the embryo (Figure 3.2D). At the late streak (LS) stage (E6.75) additional Trh expressing cells
accumulate on the lateral sides of the embryo and Trh expression remains absent from the primitive streak (Figure 3.2E,F,G). The absence of Trh from the visceral endoderm at the onset of gastrulation and the appearance of Trh expressing cells in the outer endoderm layer during late streak stages suggests that the cells expressing Trh could be the newly formed definitive endoderm cells emerging from the primitive streak and intercalating into the pre-existing visceral endoderm layer.
At early and late allantoic bud stages (EB and LB respectively, E7.5) Trh is strongly expressed in the outer layer of endoderm cells that surround the embryonic portion of the embryo, corresponding to the presumptive definitive endoderm (Figure 3.2H,L). Trh is only present at low levels in the remaining epiblast cells and is not expressed in the extraembryonic tissues. Expression of Trh is higher in the anterior region of the embryo, which is also observed in the SAGE data (arrowheads in Figure 3.2H,L and Figure 3.1A). Interestingly, there appears to be a lower level of Trh expression in the anterior midline of the embryo (arrowheads in Figure
3.2I,M). Sections of these embryos show that there is a specific group of cells in the anterior midline of the embryo that do not express Trh (Figure 3.2K,N,O). The location and morphology of these cells suggests that they belong to the notochordal plate which originates within the node and displaces the definitive endoderm before moving anteriorly to lie subjacent to the developing neural tissues (Jurand, 1974; Sulik et al., 1994). Thus, during gastrulation Trh expression shifts from the epiblast to the newly formed definitive endoderm cells that are emerging from the primitive streak. At the end of gastrulation Trh is specifically expressed in the definitive
96 endoderm layer and is not expressed in the extraembryonic tissues or the notochordal plate.
3.2.3 Trh expression in the definitive endoderm
At early and late head fold stages (EHF and LHF respectively, E7.75) Trh expression remains in the entire definitive endoderm layer (Figure 3.3A,E,L) with the strongest expression just beneath the developing head folds (arrows in Figure 3.3B,F). Trh expression is absent from the notochordal plate as well as the node at these stages (Figure 3.3C,D,I,K). The specificity of
Trh for the definitive endoderm is clearly seen in embryo sections which show a sharp boundary
between the Trh expressing definitive endoderm cells and the more cuboidal extraembryonic
visceral endoderm cells which do not express Trh (Figure 3.3H). As foregut morphogenesis
begins, Trh expressing cells accumulate at the lip of the foregut invagination (arrowhead in
Figure 3.3M). Sectioning confirms expression of Trh in the cells lining the foregut with slightly
higher expression in the ventral foregut and absence from the dorsal midline cells of the foregut
(arrowheads in Figure 3.3O,P) and the node (arrow in Figure 3.3P).
3.2.4 Rapid shift in Trh expression from the definitive endoderm to neural tissues
At early somite stages (4-6 somites), Trh is highly expressed in both the foregut pocket
(Figure 3.4A,B,D,G,H,I) and presumptive hindgut region (Figure 3.4A,F,G,K) and at the 6
somite stage Trh expression begins in the posterior neural tube (Figure 3.4J). Embryo sections
show a higher level of Trh expression in the ventral foregut (Figure 3.4B,H). Trh remains absent
from the notochordal plate as the notochord has not yet folded off from the dorsal midline of the
foregut at this stage (Figure 3.4C,E,H,I) (Jurand, 1974; Sulik et al., 1994). As development
proceeds Trh expression rapidly decreases such that by E9.0 Trh expression is completely lost in
97
Figure 3.3 Trh expression in the definitive endoderm
A-D: Expression of Trh in early head fold (EHF) stage embryo. Lateral (A) and frontal (B) views and transverse sections (C,D). Trh is expressed in the definitive endoderm with enrichment in the anterior near the headfolds (arrow in B) and is absent from the notochordal plate in the anterior midline (arrowheads in B, inset in C, D). E-K: Expression of Trh in late head fold (LHF) stage embryo. Lateral (E) and frontal (F) views and transverse sections (G-K). Trh expression remains in the definitive endoderm and is absent from the node and notochordal plate (G,J, arrowheads in I,K). L-P: Expression of Trh in 0 somite (0 so) stage embryo. Lateral (L) and frontal (M) views and transverse sections (N-P). Trh expression is enriched in the foregut invagination (arrowhead in M) and absent from the dorsal midline cells of the foregut (arrowheads in O,P) and the node (arrow in P).
98
Figure 3.4 Shift in Trh expression from the definitive endoderm to neural tissues
A-F: Expression of Trh in 4 somite (4 so) embryo. Lateral (A) view and transverse sections (B- F). Trh is expressed in the entire definitive endoderm and is absent from the cells in the dorsal midline of the foregut (arrowheads in insets C,E). G-K: Expression of Trh in 6 somite (6 so) embryo. Lateral (G) view and transverse sections (H-K). Trh expression is observed in the foregut (H,I), hindgut (K), posterior neural tube (J), and is absent from the dorsal midline cells of the foregut (arrowheads in H,I). L-P: Expression of Trh in neural tissues at E9.0. Lateral (L) view and transverse sections (M-P). Trh is expressed at the midbrain-hindbrain junction (arrowhead in L and O), in the posterior neural tube (arrowhead in P) and is no longer detected in the definitive endoderm (de) (M,N). nc – notochord.
99 the definitive endoderm (Figure 3.4L,M,N). At this stage Trh is expressed at the midbrain- hindbrain junction (arrowhead in Figure 3.4L and Figure 3.4O) and remains in the posterior neural tube (arrowhead in Figure 3.4P). Thus, Trh is expressed in the definitive endoderm following gastrulation and continues to be expressed in the definitive endoderm during early morphogenesis of the gut tube. Trh expression then rapidly decreases concomitantly with the onset of expression in the neural tissues.
3.3 DISCUSSION
Recently, several groups have identified Trh as being highly expressed in undifferentiated mouse embryonic stem cells with its expression becoming downregulated upon differentiation
(Hailesellasse Sene et al., 2007; Yamamoto et al., 2005). Due to the lack of specific data for Trh in the early embryo, we analyzed its expression in a wide variety of mouse embryonic SAGE libraries. In addition to observing high expression in mouse embryonic stem cells, we also find it to be highly expressed during post-implantation stages in the mouse embryo. Surprisingly we do not see expression in earlier stages, including the blastocyst and E5.5 whole embryo. To confirm the SAGE data, we characterized Trh expression using whole mount in situ hybridization where we see expression at E6.5 but not at E5.5 (data not shown). Interestingly, it has recently been shown that Trh may be a direct transcriptional target of Oct4, a well characterized marker of pluripotency in embryonic stem cells and the epiblast (Loh et al., 2006; Matoba et al., 2006).
Oct4 expression in the epiblast overlaps with that of Trh, suggesting that Oct4 may in part be regulating Trh expression at this stage (Rosner et al., 1990).
During gastrulation we observe a transition in Trh expression from the epiblast to the newly formed definitive endoderm cells which have emerged from the primitive streak and
100 intercalated into the visceral endoderm on the lateral sides of the embryo. Interestingly, Trh
expression is not observed in the primitive streak itself but is re-expressed in the cells that have
committed to the definitive endoderm lineage. This suggests that Trh expression is very
dynamically regulated during gastrulation.
At the completion of gastrulation, Trh transiently marks the entire definitive endoderm.
This makes Trh unique among current definitive endoderm markers as many are only expressed
in a subset of definitive endoderm cells. Trh is not expressed in the visceral endoderm or the node
and notochordal plate. This is further evidence that Trh expression is specific to the definitive
endoderm. Although it is expressed in the entire definitive endoderm in head fold stage embryos, we find Trh to be enriched in the anterior, while at early somite stages Trh is highly expressed in the foregut invagination. It is possible that the enrichment of Trh expressing cells in the anterior of the embryo reflects the accumulation of foregut precursors in this region prior to foregut morphogenesis.
The dynamic expression pattern of Trh parallels a model of definitive endoderm formation recently proposed using lineage tracing and fate mapping experiments (Tam et al.,
2007). At early gastrulation stages Trh expression appears to mark the lateral definitive endoderm population which originates from the anterior end of the primitive streak at the mid-streak stage.
However, in head fold stage embryos we see Trh expression in the entire definitive endoderm and at later stages in the dorsal foregut. This suggests that Trh may also be expressed in the medial endoderm population which forms at the mid-streak stage and contributes to the dorsal foregut.
More in depth lineage tracing studies will be required to confirm this.
Trh expression is retained in the definitive endoderm during gut tube morphogenesis and
101 then rapidly decreases such that all expression in the definitive endoderm is lost by E9.0. At this
stage, we begin to see expression in the developing neural tissues, specifically at the midbrain- hindbrain junction and in the posterior neural tube. A recently published report has described Trh
expression from E8.0 to E10.5 (Jukkola et al., 2006). In agreement with our data this analysis
also observed Trh expression in the developing neural folds although at slightly earlier stages
than in our analysis. Their analysis did not detect Trh expression in the definitive endoderm,
which may reflect differences in staining protocols.
The expression of Trh during post-implantation stages in the mouse embryo raises several questions as to its possible function in these tissues. The role of TRH in regulating the HPT axis has been known for several years and is well characterized. TRH has also been postulated to act as a neurotransmitter in the adult (reviewed in Nillni and Sevarino, 1999; O'Leary and O'Connor,
1995). Additionally, several of the peptide intermediates produced during processing of the TRH precursor peptide have been shown to have biological functions independent of TRH (reviewed in Nillni and Sevarino, 1999; O'Leary and O'Connor, 1995). The absence of the main target of
TRH, TSH, from the early mouse embryo strongly suggests that TRH is acting in a novel manner, perhaps via TRHR2. The precise function of TRH in the early embryo is difficult to
determine as there are no obvious definitive endoderm formation or patterning defects in TRH-
deficient mice (Yamada et al., 1997; Yamada et al., 2003). However, the effect of loss of TRHR2
in mice has yet to be determined. Furthermore, it is not known if the TRH peptide is expressed
and processed in the definitive endoderm. If TRH has a yet uncharacterized function in the early
embryo the possibility exists that other unidentified factors may be compensating for the loss of
TRH in null embryos.
Understanding of definitive endoderm formation has lagged behind that of the
102 neurectoderm and mesoderm in part due to the lack of genetic markers for this tissue. Several
recent studies have used several approaches, including tissue micro-dissection and fluorescence
activated cell sorting (FACS), to identify endoderm-specific markers in early mouse embryos
(Sherwood et al., 2007). In agreement with our studies, Trh has recently been identified as being
enriched in E7.5 endoderm versus mesoderm and ectoderm using a microarray approach (Gu et
al., 2004). Novel genetic markers of the definitive endoderm will have several uses. Genes that
are expressed in all definitive endoderm cells will be useful to study the formation and movement
of this tissue during gastrulation. Genes that distinguish between extraembryonic and definitive
endoderm are also necessary to define the cell populations that are produced during in vitro
differentiation of embryonic stem cells. Lastly, identification of novel definitive endoderm genes
will be useful in analyzing gastrulation and definitive endoderm formation in mouse mutants. Trh
is thus a valuable marker of the definitive endoderm that can be used to address outstanding
questions regarding definitive endoderm formation in the mouse embryo.
3.4 EXPERIMENTAL PROCEDURES
3.4.1 Mouse strains, dissection and staging
Embryos were obtained from crosses of outbred ICR mice and considered 0.5 days postcoitum at noon of the day of detection of the vaginal plug. Embryos were dissected from the uterus in Dulbecco’s Phosphate Buffered Saline (Invitrogen) with 0.1% Fetal Bovine Serum
(Invitrogen). Staging of embryos was carried out as previously described (Downs and Davies,
1993). Embryos were fixed overnight at 4˚C in 4% paraformaldehyde, dehydrated through a graded methanol series and stored at -20˚C until use.
103 3.4.2 Serial Analysis of Gene Expression library analysis
Construction of the Mouse Atlas of Gene Expression project (www.mouseatlas.org)
SAGE libraries has been previously described (Siddiqui et al., 2005). Data was analyzed using
the DiscoverySpace 4 Software (http://www.bcgsc.ca/bioinfo/software/ds) (Robertson et al.,
2007).
3.4.3 RNA extraction, cDNA synthesis, and RT-PCR
Embryos were placed into Trizol (Invitrogen) immediately following dissection and
stored at -80˚C. RNA was extracted using MaXtract high density phase-lock gels (Qiagen) following manufacturer’s instructions. RT-PCR was performed as follows. 1 μg of total RNA was treated with DNAase I (Invitrogen) prior to cDNA synthesis. cDNA was synthesized using
Superscript II Reverse Transcriptase (Invitrogen) as per the manufacturer’s instructions with the
following modifications. 1/10 of the DNase I treated RNA was mixed with 6 μg random hexamer
primers (Invitrogen), 0.5 mM dNTPs (Invitrogen) and DEPC-treated water, and incubated for 5
minutes at 65˚C then placed on ice. 1X 1st strand synthesis buffer (Invitrogen), 0.02 mM DTT
(Invitrogen) and 50 Units of RNaseOUT (Invitrogen) were added and samples were incubated at
42˚C for 2 minutes. The reaction was divided into 2 equal volumes and 200 Units of SuperScript
II was added to one set of tubes while no enzyme was added to the minus reverse transcriptase (-
RT) negative control. Samples were incubated for 50 minutes at 42˚C followed by 15 minutes at
70˚C. Reactions were diluted to 50 μL with molecular biology grade water (Invitrogen) and 2 μL
was used per PCR reaction. PCR was performed using HotStar Taq Polymerase (Qiagen) as per
manufacturer’s instructions in a PTC-200 thermal cycler (MJ Research). Thermal cycler
conditions were as follows: 95˚C 15 minutes, 95˚C 30 seconds, 55˚C 30 seconds, 72˚C 30
104 seconds, 72˚C 10 minutes for 40 cycles. Primers used were as follows: Trh (NM_009426):
Forward: 5’- TTCTTGAGGAAAGACCTCCAGCGT-3’, Reverse: 5’-
TGGCTCTTTGAAGTTCCTGAAGTG-3’, Trhr (NM_013696) Forward: 5’-
AAGTCTGGCTGTGGCAGATCTCAT-3’, Reverse: 5’-
ATCCAGCAGGAAGAACCAGAGCAT-3’, Trhr2 (NM_133202) Forward: 5’-
GTACATCGCCATTTGCCACCCAAT-3’ , Reverse: 5’-
AAAGAGCAACACAATCACGGCCAG-3’, Tsha (NM_009889) Forward: 5’-
GCCCAGAATGTAAACTAAAGGAAA-3’, Reverse: 5’-
TAACCGTAAAGAGGCAGTGTGTAA-3’, , Tshb (NM_009432) Forward: 5’-
ATTGTATGACACGGGATATCAATG-5’, Reverse: 5’-
TTCGTTCTATTCCAGGTAAACACA-3’, HPRT (NM_013556) Forward: 5’-
AGCGCAAGTTGAATCTGC-3’, Reverse: 5’-AGCGACAATCTACCAGAG-3’
3.4.4 Probe preparation
Digoxigenin-UTP labeled mRNA probes were synthesized as per manufacturer’s
instructions (Roche). Briefly, 10 μg of template DNA was linearized with 10 Units an appropriate enzyme and purified. RNA probes were in vitro transcribed using 1 μg of linearized template and
20 Units of SP6, T3 or T7 RNA polymerase (Roche). RNA probes were purified using G-50
Sephadex columns (Amersham) and stored at -80˚C. Description of the Trh in situ probe can be
found in Hou et al., 2007.
3.4.5 Whole mount in situ hybridization
Whole mount in situ hybridization (WISH) was carried out essentially as in Wilkinson
(Wilkinson and Nieto, 1993) with the following modifications. Pre-hybridization and
105 hybridization steps were performed as described using the following hybridization solution: 50%
formamide, 1.3 X SSC (pH 5), 50 μg/mL yeast tRNA, 0.2% Tween-20, 0.5% CHAPS, 100
μg/mL Heparin). Post-hybridization washes were performed as follows: 3 rinses followed by 2
30-minute washes at 65˚C with hybridization solution, 10 minute wash with 1:1 mixture of
hybridization solution and MABT (100 mM maleic acid, 150 mM NaCl, pH 7.5, 0.1% Tween-
20), then 2 rinses followed by 1 wash with MABT. Blocking was performed for 1 hour at room
temperature with 2% Blocking Reagent (Roche) dissolved in MABT and an additional 2 hours at
room temperature with 2% Blocking Reagent (Roche) and 20% heat inactivated goat serum
(Invitrogen) in MABT. Embryos were incubated in anti-Digoxigenin antibody (1:2000) (Roche)
in 2% Blocking Reagent and 20% heat inactivated goat serum in MABT overnight at 4˚C.
Embryos were rinsed and washed 3 times for 1 hour with MABT followed by 2 10 minute
washes in NTMTL (100 mM, NaCl, 100 mM Tris-HCL pH9.5, 50 mM MgCl2, 1% Tween-20, 2
mM Levamisole). Embryos were incubated in BM purple alkaline phosphatase substrate (Roche)
at 4˚C or room temperature. Staining reactions were stopped by washes in PBS with 0.1%
Tween-20.
3.4.6 Imaging and histology
Following WISH, embryos were processed through a graded PBT:Glycerol series to 50%
glycerol. Images were taken using a Leica MZ9.5 microscope and a Leica DFC420 camera with the Leica Application Suite software (Leica). For histology, embryos were incubated overnight at
4˚C in a 1:1 mixture of Tissue-Tek OCT embedding medium (Sakura Finetek) and 60% sucrose.
Embryos were embedded in 100% OCT (in 10x10 mm plastic mold) and frozen with a dry
ice:ethanol mixture. Cryosections (10 uM) were cut using a Microm cryostat and placed on
Superfrost Plus glass slides (VWR) and allowed to air dry. Slides were washed twice in PBS for
106 10 minutes per wash and mounted using 50% glycerol and glass coverslips (VWR). Sections were imaged using Zeiss Axioplan microscope and OpenLab software (Improvision).
107 3.5 REFERENCES
Acampora, D., Postiglione, M. P., Avantaggiato, V., Di Bonito, M., Vaccarino, F. M., Michaud, J., and Simeone, A. (1999). Progressive impairment of developing neuroendocrine cell lineages in the hypothalamus of mice lacking the Orthopedia gene. Genes Dev 13, 2787- 800. Ang, S. L., Wierda, A., Wong, D., Stevens, K. A., Cascio, S., Rossant, J., and Zaret, K. S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development 119, 1301-15. Backman, M., Machon, O., Van Den Bout, C. J., and Krauss, S. (2003). Targeted disruption of mouse Dach1 results in postnatal lethality. Dev Dyn 226, 139-44. Beddington, R. S. (1983). Histogenetic and neoplastic potential of different regions of the mouse embryonic egg cylinder. J Embryol Exp Morphol 75, 189-204. Belo, J. A., Bouwmeester, T., Leyns, L., Kertesz, N., Gallo, M., Follettie, M., and De Robertis, E. M. (1997). Cerberus-like is a secreted factor with neutralizing activity expressed in the anterior primitive endoderm of the mouse gastrula. Mech Dev 68, 45-57. Biben, C., Stanley, E., Fabri, L., Kotecha, S., Rhinn, M., Drinkwater, C., Lah, M., Wang, C. C., Nash, A., Hilton, D., Ang, S. L., Mohun, T., and Harvey, R. P. (1998). Murine cerberus homologue mCer-1: a candidate anterior patterning molecule. Dev Biol 194, 135-51. Caqueret, A., Boucher, F., and Michaud, J. L. (2006). Laminar organization of the early developing anterior hypothalamus. Dev Biol 298, 95-106. Downs, K. M., and Davies, T. (1993). Staging of gastrulating mouse embryos by morphological landmarks in the dissecting microscope. Development 118, 1255-66. Goshu, E., Jin, H., Lovejoy, J., Marion, J. F., Michaud, J. L., and Fan, C. M. (2004). Sim2 contributes to neuroendocrine hormone gene expression in the anterior hypothalamus. Mol Endocrinol 18, 1251-62. Gu, G., Wells, J. M., Dombkowski, D., Preffer, F., Aronow, B., and Melton, D. A. (2004). Global expression analysis of gene regulatory pathways during endocrine pancreatic development. Development 131, 165-79. Hailesellasse Sene, K., Porter, C. J., Palidwor, G., Perez-Iratxeta, C., Muro, E. M., Campbell, P. A., Rudnicki, M. A., and Andrade-Navarro, M. A. (2007). Gene function in early mouse embryonic stem cell differentiation. BMC Genomics 8, 85. Hallonet, M., Kaestner, K. H., Martin-Parras, L., Sasaki, H., Betz, U. A., and Ang, S. L. (2002). Maintenance of the specification of the anterior definitive endoderm and forebrain depends on the axial mesendoderm: a study using HNF3beta/Foxa2 conditional mutants. Dev Biol 243, 20-33. Hou, J., Charters, A. M., Lee, S. C., Zhao, Y., Wu, M. K., Jones, S. J., Marra, M. A., and Hoodless, P. A. (2007). A systematic screen for genes expressed in definitive endoderm by Serial Analysis of Gene Expression (SAGE). BMC Dev Biol 7, 92. Jukkola, T., Lahti, L., Naserke, T., Wurst, W., and Partanen, J. (2006). FGF regulated gene- expression and neuronal differentiation in the developing midbrain-hindbrain region. Dev Biol 297, 141-57. Jurand, A. (1974). Some aspects of the development of the notochord in mouse embryos. J Embryol Exp Morphol 32, 1-33. Kanai-Azuma, M., Kanai, Y., Gad, J. M., Tajima, Y., Taya, C., Kurohmaru, M., Sanai, Y., Yonekawa, H., Yazaki, K., Tam, P. P., and Hayashi, Y. (2002). Depletion of definitive gut endoderm in Sox17-null mutant mice. Development 129, 2367-79.
108 Keith, B., Adelman, D. M., and Simon, M. C. (2001). Targeted mutation of the murine arylhydrocarbon receptor nuclear translocator 2 (Arnt2) gene reveals partial redundancy with Arnt. Proc Natl Acad Sci U S A 98, 6692-7. Lawson, K. A., Meneses, J. J., and Pedersen, R. A. (1986). Cell fate and cell lineage in the endoderm of the presomite mouse embryo, studied with an intracellular tracer. Dev Biol 115, 325-39. Lawson, K. A., and Pedersen, R. A. (1987). Cell fate, morphogenetic movement and population kinetics of embryonic endoderm at the time of germ layer formation in the mouse. Development 101, 627-52. Loh, Y. H., Wu, Q., Chew, J. L., Vega, V. B., Zhang, W., Chen, X., Bourque, G., George, J., Leong, B., Liu, J., Wong, K. Y., Sung, K. W., Lee, C. W., Zhao, X. D., Chiu, K. P., Lipovich, L., Kuznetsov, V. A., Robson, P., Stanton, L. W., Wei, C. L., Ruan, Y., Lim, B., and Ng, H. H. (2006). The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nat Genet 38, 431-40. Matoba, R., Niwa, H., Masui, S., Ohtsuka, S., Carter, M. G., Sharov, A. A., and Ko, M. S. H. (2006). Dissecting Oct3/4-Regulated Gene Networks in Embryonic Stem Cells by Expression Profiling. PLOS One, 1-15. Michaud, J. L., DeRossi, C., May, N. R., Holdener, B. C., and Fan, C. M. (2000). ARNT2 acts as the dimerization partner of SIM1 for the development of the hypothalamus. Mech Dev 90, 253-61. Michaud, J. L., Rosenquist, T., May, N. R., and Fan, C. M. (1998). Development of neuroendocrine lineages requires the bHLH-PAS transcription factor SIM1. Genes Dev 12, 3264-75. Nillni, E. A., and Sevarino, K. A. (1999). The biology of pro-thyrotropin-releasing hormone- derived peptides. Endocr Rev 20, 599-648. O'Leary, R., and O'Connor, B. (1995). Thyrotropin-releasing hormone. J Neurochem 65, 953-63. Robertson, N., Oveisi-Fordorei, M., Zuyderduyn, S. D., Varhol, R. J., Fjell, C., Marra, M., Jones, S., and Siddiqui, A. (2007). DiscoverySpace: an interactive data analysis application. Genome Biol 8, R6. Rosner, M. H., Vigano, M. A., Ozato, K., Timmons, P. M., Poirier, F., Rigby, P. W., and Staudt, L. M. (1990). A POU-domain transcription factor in early stem cells and germ cells of the mammalian embryo. Nature 345, 686-92. Satoh, T., Yamada, M., Monden, T., Iizuka, M., and Mori, M. (1992). Cloning of the mouse hypothalamic preprothyrotropin-releasing hormone (TRH) cDNA and tissue distribution of its mRNA. Brain Res Mol Brain Res 14, 131-5. Schaner, P., Todd, R. B., Seidah, N. G., and Nillni, E. A. (1997). Processing of prothyrotropin- releasing hormone by the family of prohormone convertases. J Biol Chem 272, 19958-68. Schonemann, M. D., Ryan, A. K., McEvilly, R. J., O'Connell, S. M., Arias, C. A., Kalla, K. A., Li, P., Sawchenko, P. E., and Rosenfeld, M. G. (1995). Development and survival of the endocrine hypothalamus and posterior pituitary gland requires the neuronal POU domain factor Brn-2. Genes Dev 9, 3122-35. Sevarino, K. A., Goodman, R. H., Spiess, J., Jackson, I. M., and Wu, P. (1989). Thyrotropin- releasing hormone (TRH) precursor processing. Characterization of mature TRH and non-TRH peptides synthesized by transfected mammalian cells. J Biol Chem 264, 21529- 35. Shawlot, W., Deng, J. M., and Behringer, R. R. (1998). Expression of the mouse cerberus-related gene, Cerr1, suggests a role in anterior neural induction and somitogenesis. Proc Natl Acad Sci U S A 95, 6198-203.
109 Sherwood, R. I., Jitianu, C., Cleaver, O., Shaywitz, D. A., Lamenzo, J. O., Chen, A. E., Golub, T. R., and Melton, D. A. (2007). Prospective isolation and global gene expression analysis of definitive and visceral endoderm. Dev Biol 304, 541-55. Siddiqui, A. S., Khattra, J., Delaney, A. D., Zhao, Y., Astell, C., Asano, J., Babakaiff, R., Barber, S., Beland, J., Bohacec, S., Brown-John, M., Chand, S., Charest, D., Charters, A. M., Cullum, R., Dhalla, N., Featherstone, R., Gerhard, D. S., Hoffman, B., Holt, R. A., Hou, J., Kuo, B. Y., Lee, L. L., Lee, S., Leung, D., Ma, K., Matsuo, C., Mayo, M., McDonald, H., Prabhu, A. L., Pandoh, P., Riggins, G. J., de Algara, T. R., Rupert, J. L., Smailus, D., Stott, J., Tsai, M., Varhol, R., Vrljicak, P., Wong, D., Wu, M. K., Xie, Y. Y., Yang, G., Zhang, I., Hirst, M., Jones, S. J., Helgason, C. D., Simpson, E. M., Hoodless, P. A., and Marra, M. A. (2005). A mouse atlas of gene expression: large-scale digital gene- expression profiles from precisely defined developing C57BL/6J mouse tissues and cells. Proc Natl Acad Sci U S A 102, 18485-90. Sulik, K., Dehart, D. B., Iangaki, T., Carson, J. L., Vrablic, T., Gesteland, K., and Schoenwolf, G. C. (1994). Morphogenesis of the murine node and notochordal plate. Dev Dyn 201, 260-78. Tam, P. P., and Beddington, R. S. (1992). Establishment and organization of germ layers in the gastrulating mouse embryo. Ciba Found Symp 165, 27-41; discussion 42-9. Tam, P. P., and Behringer, R. R. (1997). Mouse gastrulation: the formation of a mammalian body plan. Mech Dev 68, 3-25. Tam, P. P., Khoo, P. L., Lewis, S. L., Bildsoe, H., Wong, N., Tsang, T. E., Gad, J. M., and Robb, L. (2007). Sequential allocation and global pattern of movement of the definitive endoderm in the mouse embryo during gastrulation. Development 134, 251-60. Tam, P. P., Khoo, P. L., Wong, N., Tsang, T. E., and Behringer, R. R. (2004). Regionalization of cell fates and cell movement in the endoderm of the mouse gastrula and the impact of loss of Lhx1(Lim1) function. Dev Biol 274, 171-87. Thomas, P. Q., Brown, A., and Beddington, R. S. (1998). Hex: a homeobox gene revealing peri- implantation asymmetry in the mouse embryo and an early transient marker of endothelial cell precursors. Development 125, 85-94. Velculescu, V. E., Zhang, L., Vogelstein, B., and Kinzler, K. W. (1995). Serial analysis of gene expression. Science 270, 484-7. Wang, W., and Lufkin, T. (2000). The murine Otp homeobox gene plays an essential role in the specification of neuronal cell lineages in the developing hypothalamus. Dev Biol 227, 432-49. Wells, J. M., and Melton, D. A. (1999). Vertebrate endoderm development. Annu Rev Cell Dev Biol 15, 393-410. Wilkinson, D. G., and Nieto, M. A. (1993). Detection of messenger RNA by in situ hybridization to tissue sections and whole mounts. Methods Enzymol 225, 361-73. Yamada, M., Saga, Y., Shibusawa, N., Hirato, J., Murakami, M., Iwasaki, T., Hashimoto, K., Satoh, T., Wakabayashi, K., Taketo, M. M., and Mori, M. (1997). Tertiary hypothyroidism and hyperglycemia in mice with targeted disruption of the thyrotropin- releasing hormone gene. Proc Natl Acad Sci U S A 94, 10862-7. Yamada, M., Satoh, T., and Mori, M. (2003). Mice lacking the thyrotropin-releasing hormone gene: what do they tell us? Thyroid 13, 1111-21. Yamamoto, Y., Teratani, T., Yamamoto, H., Quinn, G., Murata, S., Ikeda, R., Kinoshita, K., Matsubara, K., Kato, T., and Ochiya, T. (2005). Recapitulation of in vivo gene expression during hepatic differentiation from murine embryonic stem cells. Hepatology 42, 558-67.
110 Chapter 4
MIDGUT AND HINDGUT DEFINITIVE ENDODERM FORMATION IN
THE MOUSE EMBRYO IS INDEPENDENT OF THE ANTERIOR
PRIMITIVE STREAK1
4.1 INTRODUCTION
The definitive endoderm is one of the three germ layers that are formed during gastrulation and will give rise to the inner lining of the gastrointestinal tract, as well as associated organs such as the liver and pancreas (reviewed in Wells and Melton, 1999). Prior to gastrulation the mouse embryo consists of two cell layers – the inner epiblast and the outer extraembryonic visceral endoderm. Gastrulation commences at embryonic day 6.5 (E6.5) and is marked by the formation of a transient structure called the primitive streak on the posterior side of the embryo
(reviewed in Tam and Behringer, 1997). During gastrulation, cells of the epiblast undergo an epithelial-to-mesenchymal transformation and move through the primitive streak and emerge as either mesoderm or definitive endoderm. As the cells of the definitive endoderm emerge from the primitive streak, they first intercalate into the pre-existing visceral endoderm layer and subsequently expand to form a single cell layer encompassing the entire embryonic region, thereby displacing the visceral endoderm proximally to the extraembryonic region (Lawson et al., 1986; Lawson and Pedersen, 1987; Tam and Beddington, 1992). Following gastrulation morphogenetic movements transform the epithelial sheet of definitive endoderm into a tube,
. . 1 A version of this chapter has been submitted for publication. McKnight et al. (2008) Midgut and hindgut definitive endoderm formation in the mouse embryo is independent of the anterior primitive streak.
111 beginning at the anterior with the foregut invagination and then at the posterior with the hindgut
invagination. The midgut region remains open until E9.0 when the embryo turns from the
lordotic to fetal position. Following formation of the gut tube the derivative organs of the
definitive endoderm begin to form, beginning as small buds of the endoderm epithelial layer which then further differentiate as embryogenesis proceeds.
Extensive fate mapping of the definitive endoderm has been performed to examine both the spatial and temporal origin of different regions of the definitive endoderm. Early fate mapping studies in pre-streak stage embryos by Lawson et al demonstrated that the majority of definitive endoderm arises from the anterior end of the primitive streak (Lawson et al., 1986;
Lawson and Pedersen, 1987). This region, marked by the expression of the transcription factors
Foxa2 and goosecoid (Gsc), is thought to be the location of the mouse embryonic organizer
(Camus and Tam, 1999; Robb and Tam, 2004). In addition to the definitive endoderm, cells at
the anterior primitive streak also contain the progenitors for the node, prechordal plate, and
notochord (Kinder et al., 2001b). More recent fate mapping studies performed by Tam et al have
refined the initial fate maps by examining the origin of the definitive endoderm in later stage
(mid-streak through early somite stage) embryos. This lineage tracing analysis demonstrated that
specific regions of the definitive endoderm, such as the ventral and dorsal foregut, midgut, and
hindgut are specified at different levels of the primitive streak at specific times during
gastrulation. These studies also demonstrated that these subpopulations of definitive endoderm
migrate independently from the mesoderm and in a stereotypical fashion.
The Nodal signaling pathway has been shown to play a critical and evolutionarily
conserved role in specification of the anterior primitive streak and its derivative tissues,
including the definitive endoderm (reviewed in Schier, 2003; Shen, 2007). Nodal belongs to a
112 family of secreted signaling ligands which also includes Activin and Transforming Growth
Factor beta (TGF-β). Nodal acts via a complex of the type I serine/threonine kinase
transmembrane receptor ActRIB (Acvr1b, ALK4) and the type II receptors ActRIIA (Acvr2a) or
ActRIIB (Acvr2b), with ligand binding resulting in the phosphorylation and activation of the
receptor regulated Smad proteins, Smad2 and Smad3. Upon activation, the Smads move to the
nucleus to form a complex with the common Smad, Smad4, and sequence-specific DNA binding
factors to regulate gene expression. The role of the Nodal pathway in specification of the anterior
primitive streak and its derivative tissues in the mouse has been demonstrated by genetic
manipulation of components of the Nodal pathway. These studies have revealed that there is a
graded response to Nodal/Smad2 signaling in the mouse embryo, where specification of the anterior definitive endoderm and prechordal plate require the highest levels of signaling,
intermediate levels of signaling specify the node, and notochord (axial mesoderm) and lower
levels of signaling are involved in specification of paraxial and lateral mesoderm.
In addition to the Nodal pathway, the forkhead proteins Foxh1 and Foxa2 have been shown to play an important role in specification of the anterior primitive streak and its derivative tissues (Ang and Rossant, 1994; Hoodless et al., 2001; Weinstein et al., 1994; Yamamoto et al.,
2001). Foxh1 is a Smad co-factor involved in regulating gene expression downstream of both
Nodal and Activin (Chen et al., 1996; Labbe et al., 1998; Liu et al., 1999; Weisberg et al., 1998).
Foxh1 regulates the expression of several genes involved in embryonic patterning, including
Nodal, Lefty1, Mix1, and Gsc (Hart et al., 2005; Izzi et al., 2007; Labbe et al., 1998; Norris et al.,
2002; Takaoka et al., 2006). Loss of Foxh1 in the mouse embryo results in a failure to specify the anterior primitive streak, as demonstrated by the loss of expression of the markers of this cell population, Foxa2 and Gsc, as well as the absence of the anterior primitive streak derived tissues, the node, notochord, and anterior definitive endoderm (Hoodless et al., 2001; Yamamoto et al.,
113 2001). The role of Foxh1 in specification of the definitive endoderm is supported by chimera
studies, which showed that Foxh1 null ES cells do not contribute to the definitive endoderm
layer, suggesting that Foxh1 is required cell autonomously for the specification of this tissue
(Hoodless et al., 2001). Interestingly, a small number of Foxh1 null ES cells were found in the
hindgut of chimeric embryos, suggesting that Foxh1 may not be required for formation of this
subpopulation of definitive endoderm. In accordance with its loss of expression in Foxh1 null
embryos, Foxa2 null embryos also fail to specify the anterior primitive streak and thus lack a
node, notochord, prechordal plate, floor plate, and fail to form the majority of definitive endoderm (Ang and Rossant, 1994; Weinstein et al., 1994). Analysis of the definitive endoderm markers Foxa1 and Shh revealed that loss of Foxa2 primarily affected specification of the foregut definitive endoderm and that the midgut and hindgut definitive endoderm was still present. Chimera analysis using Foxa2 null ES cells demonstrated that Gata4, a marker of the extraembryonic visceral endoderm, extended into the the foregut and midgut regions of mutant chimera embryos, suggesting that Foxa2 is required for specification of the foregut and midgut, but not the hindgut (Dufort et al., 1998).
Understanding of the definitive endoderm has lagged behind that of the other two germ layers, mainly due to the lack of both pan-endodermal and region specific genetic markers of this tissue. We recently performed gene expression profiling of the mouse definitive endoderm and identified a panel of novel definitive and visceral endoderm markers (Hou et al., 2007; McKnight et al., 2007). Here, we analyze these markers in Foxh1 and Foxa2 null embryos and conclusively demonstrate that formation of the midgut and hindgut definitive endoderm does not require the transcription factors Foxh1 or Foxa2 or the anterior primitive streak cell population.
Furthermore, we confirm previous observations that formation of the foregut definitive endoderm is severely affected in these mutants and demonstrate that the foregut invagination is
114 instead lined with visceral endoderm that was not displaced during gastrulation. Based on these
observations we propose a model where formation of the midgut and hindgut definitive
endoderm is independent of the anterior primitive streak and suggest that specification of the
midgut and hindgut definitive endoderm may be regulated by signaling pathways other than
Nodal/TGF-β.
4.2 RESULTS
4.2.1 Midgut and hindgut formation is unperturbed in Foxh1 and Foxa2 null embryos
Previous studies have demonstrated that high levels of Nodal signaling are required for specification of foregut and anterior axial mesoderm populations. However, analysis of the posterior endoderm populations has been hampered by the lack of region specific genetic markers of the definitive endoderm. We recently identified and characterized a novel marker of the midgut definitive endoderm, nephrocan (Nepn) (Hou et al., 2007). We analyzed the expression of Nepn in Foxh1 and Foxa2 null embryos to specifically analyze midgut formation in the absence of the anterior primitive streak. We also examined the expression of SRY-box containing gene 17 (Sox17), a well characterized transient marker of the hindgut definitive
endoderm at E8.5, whose expression has not previously been analyzed in Foxh1 or Foxa2
mutants.
In E8.5 wild-type embryos, Nepn is expressed specifically in the midgut definitive
endoderm (Figure 4.1A,B) (Hou et al., 2007). Expression does not extend into the foregut or
hindgut invaginations. Nepn expression is specific to the definitive endodermal layer, as it is not
expressed in the midline cells of the notochordal plate (arrowhead in Figure 4.1C). Interestingly,
115
Figure 4.1 The midgut and hindgut definitive endoderm remains in Foxh1-/- and Foxa2-/- null embryos
A-I: Expression of Nepn in E8.5 wild-type (A-C), Foxh1-/- (D-F), and Foxa2-/- (G-I) embryos; lateral (A,D,G) and ventral (B,E,H) views with anterior to the left and transverse sections (C,F,I). In wild-type embryos Nepn is expressed specifically in the midgut definitive endoderm; expression does not extend into the foregut or hindgut invaginations (A,B). Nepn is expressed specifically in the endoderm layer and is not expressed in the midline notochordal plate cells (arrowhead in C). Expression of Nepn remains in the midgut region of both Foxh1-/- and Foxa2-/- embryos (D,E,G,H). The notochordal plate cells which do not express Nepn are absent in both Foxh1-/- and Foxa2-/- embryos (arrowheads in F,I). J-U: Expression of Sox17 in E8.5 wild-type (J-M), Foxh1-/- (N-Q), and Foxa2-/- (R-U) embryos; lateral (J,N,R) and ventral (K,O,S) views with anterior to the left and transverse sections (L,M,P,Q,T,U). In wild-type embryos Sox17 is expressed strongly in the hindgut invagination (arrow in M) and the hemangioblasts (arrowhead in K); weak expression is also observed in the foregut invagination, midgut, and visceral endoderm (L). Sox17 is expressed in both the anterior foregut region (P) and the hindgut invagination (arrow in Q) in Foxh1-/- embryos. The bilateral hemangioblast expression domain is fused in the midline (arrowhead in O). Sox17 expression also remains in the hindgut of Foxa2-/- embryos (arrow in U). fg – foregut, hg- hindgut, mg – midgut, hb – hemangioblasts, nc – notochordal plate. Approximate locations of transverse sections are indicated by dashed line.
116 strong expression of Nepn in the midgut region is observed in both Foxh1 and Foxa2 null embryos (Figure 4.1D,E,G,H). Consistent with a loss of the notochordal plate in these mutants, the bilateral expression domain of Nepn is fused at the midline (arrowheads in Figure 4.1F,I).
The domain of Nepn expression is reduced in Foxa2 null embryos compared to Foxh1 null embryos, consistent with the more severe phenotype and smaller size of these mutants.
Sox17 is initially expressed in the nascent definitive endoderm at E7.5 and by E8.5 expression is restricted to the hindgut definitive endoderm (arrow in Figure 4.1M) (Kanai-
Azuma et al., 2002). Sox17 is also expressed strongly in the hemangioblasts (arrowheads in
Figure 4.1K) and weakly in the visceral endoderm. Sox17 expression is observed in both the hindgut and foregut of Foxh1 null embryos (Figure 4.1O,P, arrow in Q). The bilateral hemangioblast expression domain is fused in the midline of Foxh1 null embryos (arrowhead in
Figure 4.1O). Similar results were observed in Foxa2 null embryos, with weak expression of
Sox17 in the hindgut region and strong expression in the foregut invagination (Figure 4.1S,T, arrow in U). The expression of Sox17 in the foregut region of these mutants may reflect either extraembryonic visceral endoderm that remains in the foregut region or definitive endoderm that has not migrated properly. This analysis provides the first conclusive evidence that the midgut definitive endoderm forms in Foxh1 and Foxa2 null embryos and suggests that formation of these subpopulations of definitive endoderm is independent of the anterior primitive streak.
4.2.2 Formation of the foregut definitive endoderm is severely affected in Foxh1 and Foxa2 null embryos
We sought to further characterize definitive endoderm formation in anterior primitive streak mutants by examining the expression of the novel foregut markers peptide YY (Pyy) and
117 thyrotropin releasing hormone (Trh), both of which are strongly expressed in the foregut invagination (Hou et al., 2007; McKnight et al., 2007). Previous analysis of anterior (foregut) formation in these mutants have utilized the marker hematopoietically expressed homeobox
(Hhex); however, Hhex is only expressed in a small sub-population of cells in the foregut
definitive endoderm, as it is limited to the lip of the foregut invagination corresponding to the location of the liver progenitor cells (Keng et al., 1998). Using this marker, it has been
demonstrated that the anterior definitive endoderm (foregut) is more sensitive to reduced Nodal
signaling or loss of the anterior primitive streak, as in Foxh1 or Foxa2 null embryos.
At E8.0, Trh is expressed in the definitive endoderm layer with expression slightly higher
in the foregut compared to the hindgut (Figure 4.2A,B,C,D) (Hou et al., 2007; McKnight et al.,
2007). Similar to Nepn, Trh is not expressed in the midline cells of the notochordal plate
(arrowhead in Figure 4.2C). Expression of Trh remains in both the foregut and hindgut definitive
endoderm of Foxh1 null embryos (Figure 4.2E,F,G,H), albeit at significantly lower levels.
Consistent with the failure to specify the notochord in these mutants, the midline downregulation
of Trh is not observed in Foxh1 null embryos (arrowhead in Figure 4.2G). Cells expressing Trh
are also observed in the anterior region near the foregut (arrow in Figure 4.2K) and in the
hindgut region (arrow in Figure 4.2K) of Foxa2 null embryos.
Pyy is highly expressed in the foregut invagination at E8.5, with no expression in the
midline cells of the notochordal plate (Figure 4.2L,M, and arrowhead in N) (Hou et al., 2007). In
E8.5 Foxh1 null embryos the expression domain of Pyy is significantly reduced compared to
wild-type embryos, with expression remaining in a small number of cells within the foregut
invagination (arrows in Figure 4.2P,Q). Pyy expression in Foxa2 null embryos is even more
limited than in Foxh1 null embryos with very few Pyy expressing cells observed in the anterior
118
Figure 4.2 There is a deficiency of foregut definitive endoderm in Foxh1-/- and Foxa2-/- embryos
A-K: Expression of Trh in E8.5 wild-type (A-D), Foxh1-/- (E-H), and Foxa2-/- (I-K) embryos; lateral (A,E,I) and ventral (B,F,J) views with anterior to the left and transverse sections (C,D,G,H,K). In wild-type embryos Trh is expressed in the foregut and hindgut definitive endoderm (A,B). Trh is expressed specifically in the endoderm layer and is not expressed in the midline notochordal plate cells (arrowhead in C). Expression of Trh remains in the foregut and hindgut regions of Foxh1-/- embryos (E,F,G,H). The notochordal plate cells which do not express Trh are absent in Foxh1-/- (arrowhead in G). Trh expression remains in the foregut and hindgut region of Foxa2-/- embryos (I,J,K). L-T: Expression of Pyy in E8.5 wild-type (L-N), Foxh1-/- (O- Q), and Foxa2-/- (R-T) embryos; lateral (L,O,R) and ventral (M,P,S) views with anterior to the left and transverse sections (N,Q,T). In wild-type embryos Pyy is expressed strongly throughout the foregut invagination and is absent from the midline notochordal plate cells (arrowhead in N). Expression of Pyy is observed in a small number of cells within the foregut invagination of Foxh1-/- embryos (arrows in P,Q). Cells expressing Pyy are found scattered in the anterior region of Foxa2-/- embryos (arrows in R,S,T). fg – foregut, hg- hindgut, mg – midgut, nc – notochordal plate. Approximate locations of transverse sections are indicated by dashed line.
119 region of embryos (arrows in Figure 4.2R,S,T). Thus, while formation of the midgut and hindgut
definitive endoderm is largely unaffected in Foxh1 and Foxa2 mutants, as demonstrated by the
expression of Nepn, Sox17, and Trh in the posterior endoderm of these mutants, there is a
significant deficiency in foregut definitive endoderm, demonstrated by the reduced expression
domains of Trh and Pyy in the anterior endoderm.
4.2.3 The visceral endoderm is incompletely displaced from the embryonic region of Foxh1
and Foxa2 null embryos during gastrulation
Analysis of region specific definitive endoderm markers in Foxh1 and Foxa2 null
embryos has demonstrated that there is a significant reduction in foregut definitive endoderm in
these mutants, while the midgut and hindgut definitive endoderm domains are not reduced.
During gastrulation, the newly formed definitive endoderm emerges from the primitive streak
and intercalates into the pre-existing visceral endoderm layer then expands to cover the entire
embryonic cup, displacing the visceral endoderm proximally (Lawson et al., 1986; Lawson and
Pedersen, 1987). We hypothesized that the deficiency of foregut (anterior) definitive endoderm
would result in the incomplete displacement of the visceral endoderm from the embryonic
region. To examine this we performed in situ hybridization for markers of both the nascent
definitive endoderm and the visceral endoderm in E7.5 wild-type and Foxh1 and Foxa2 null
embryos.
Prior to gastrulation Cerberus 1 homolog (Cer1) is expressed in the anterior visceral endoderm and is later expressed in the nascent definitive endoderm, becoming restricted to the anterior definitive endoderm by E7.5 (Figure 4.3A) (Belo et al., 1997; Biben et al., 1998;
Shawlot et al., 1998). In E7.5 Foxh1 null embryos, strong expression of Cer1 is found in the
120
Figure 4.3 There is incomplete displacement of the visceral endoderm from the embryonic region of Foxh1-/- and Foxa2-/- embryos during gastrulation
A-D: Expression of Cer1 in E7.5 wild-type (A), Foxh1-/- (B-C), and Foxa2-/- (D) embryos; lateral views with anterior to the left. In wild-type embryos Cer1 is expressed in the anterior definitive endoderm (A). Cer1 expression is observed in the distal (anterior) region of Foxh1-/- (arrows in B and C) and Foxa2-/-embryos (arrow in D). E-H: Expression of Sox17 in E7.5 wild-type (E), Foxh1-/- (F-G), and Foxa2-/- (H) embryos; lateral views with anterior to the left. In wild-type embryos Sox17 is expressed in the anterior definitive endoderm and the extraembryonic visceral endoderm (E). Sox17 expression is observed in the proximal (posterior) region of Foxh1-/- embryos (arrows in F and G). Expression of Sox17 is also detected weakly in the outer layers of Foxa2-/- embryos (H). I-L: Expression of Trh in E7.5 wild-type (I), Foxh1-/- (J-K), and Foxa2-/- (L) embryos; lateral views with anterior to the left. In wild-type embryos Trh is expressed throughout the definitive endoderm with higher expression in the anterior definitive endoderm (I). Trh expression is observed scattered throughout the distal (anterior) region of Foxh1-/- embryos (arrow in J), while expression is restricted to the proximal (posterior) region in severely affected embryos (arrow in K) Expression of Trh is also detected in the proximal (posterior) region of Foxa2-/- embryos (L). M-P: Expression of AFP in E7.5 wild-type (M), Foxh1-/- (N-O), and Foxa2-/- (P) embryos; lateral views with anterior to the left. In wild-type embryos AFP is expressed in the extraembryonic visceral endoderm (M). AFP expression extends into the embryonic region of Foxh1-/- embryos (arrows in N and O). Expression of AFP is also detected in the distal (anterior) region of Foxa2-/- embryos (P). ade – anterior definitive endoderm, exve- extraembryonic visceral endoderm.
121 outer endoderm layer with the expression domain localized to the anterior (distal) region of the
embryo (arrow in Figure 4.3B). The restriction of Cer1 to the anterior of the embryo is more
pronounced in the more severe class of mutant embryos (arrow in Figure 4.3C). In Foxa2 null
embryos, Cer1 expression is also observed in the outer endoderm layer in the anterior (distal) region (arrow in Figure 4.3D). At E7.5, Sox17 is expressed in the anterior definitive endoderm, as well as in the extraembryonic visceral endoderm (Figure 4.3E) (Kanai-Azuma et al., 2002).
Expression of Sox17 is also observed in the outer endoderm layer in the posterior (proximal) embryonic region of E7.5 Foxh1 null embryos (arrow in Figure 4.3F,G) and weakly in the outer layer of Foxa2 null embryos (arrow in Figure 4.3H). The weak expression of Cer1 and Sox17 in
the anterior region of Foxa2 null mutants is in agreement with previous reports (Kinder et al.,
2001a; Klingensmith et al., 1999). These studies concluded that due to the embryonic delay and
aberrant morphology of Foxa2 mutants and that Cer1 and Sox17 are expressed in both the
visceral and definitive endoderm it could not be determined if the expression observed in these
mutants was in the visceral or definitive endoderm. To distinguish between these possibilities we
examined the expression of Trh in Foxh1 and Foxa2 mutants. Unlike Cer1 and Sox17, Trh is not expressed in the visceral endoderm at any time point during gastrulation. Trh is expressed in the nascent definitive endoderm as it emerges from the primitive streak and is expressed in the entire definitive endoderm population at E7.5, with higher expression in the anterior region (Figure
4.3I) (McKnight et al., 2007). Strong expression of Trh remains in the outer endoderm layer of
E7.5 Foxh1 null embryos (Figure 4.3J,K). In the less affected embryos, expression of Trh is found throughout the distal tip (anterior) of the E7.5 embryo (arrow in Figure 4.3J) while in the most severe class of mutants expression is restricted to the posterior region (arrow in Figure
4.3K). In Foxa2 mutants, Trh expression is also observed in the proximal (posterior) region
(arrow in Figure 4.3L). Thus, markers of the nascent definitive endoderm are still expressed in
Foxh1 and Foxa2 null embryos, supporting the claim that some definitive endoderm is still
122 formed in the absence of the anterior primitive streak cell population. However, Cer1 and Sox17 are also expressed in the extraembryonic visceral endoderm, therefore it is possible that the observed expression in the embryonic region of Foxh1 null embryos may be visceral endoderm not displaced during gastrulation. To address this we examined the expression of AFP, a marker of the visceral endoderm in E7.5 Foxh1 and Foxa2 null embryos. At E7.5 AFP expression is restricted to the proximal extraembryonic region of embryos (Figure 4.3M) (Kwon et al., 2006).
The distal embryonic region that does not express AFP consists of newly formed definitive endoderm. Expression of AFP is observed in the distal embryonic region, in addition to the extraembryonic region, in both Foxh1 and Foxa2 null embryos (arrows in Figure 4.3N,O,P).
Collectively these results suggest that some nascent definitive endoderm is formed during gastrulation in Foxh1 and Foxa2 null embryos, but that there is a deficiency in the definitive endoderm resulting in an incomplete displacement of the visceral endoderm from the embryonic region.
4.2.4 Visceral endoderm that is not displaced during gastrulation becomes incorporated into the foregut of Foxh1 and Foxa2 null embryos
The deficiency in anterior definitive endoderm (foregut) in Foxh1 and Foxa2 null embryos results in an incomplete displacement of the visceral endoderm from the embryonic region. To examine the fate of this visceral endoderm we analyzed the expression of the visceral endoderm markers α-fetoprotein (AFP) and carboxypeptidase N polypeptide 1 (Cpn1) in E8.5 wild-type and Foxh1 and Foxa2 null embryos. At E8.5 AFP is expressed specifically in the visceral endoderm derived yolk sac (Figure 4.4A,B) (Kwon et al., 2006; Sellem et al., 1984).
AFP expression is absent from the domain corresponding to the definitive endoderm (Figure
4.4B, arrow in C and arrowheads in D). In Foxh1 null embryos, AFP expression extends into the
123
Figure 4.4 Visceral endoderm not displaced during gastrulation becomes incorporated into the foregut of Foxh1-/- and Foxa2-/- embryos
A-L: Expression of AFP in E8.5 wild-type (A-D), Foxh1-/- (E-H), and Foxa2-/- (I-L) embryos; lateral (A,E,I) and ventral (B,F,J) views with anterior to the left and transverse sections (C,D,G,H,K,L). In wild-type embryos AFP is expressed specifically in the visceral endoderm derived yolk sac; expression does not extend into the foregut or hindgut invaginations (A,B, arrow in C and arrowheads in D). Expression of AFP extends into the foregut invagination of Foxh1-/- embryos (E and arrow in G). Expression does not extend into the midgut and hindgut regions (F and arrowheads in H). AFP expressing cells are also observed in the foregut invagination of Foxa2-/- embryos (I and arrow in J and K) and are not observed in the midgut or hindgut (arrowheads in L). The expression domain of AFP does not extend into the midgut or hindgut regions (J and arrowheads in L). M-W: Expression of Cpn1 in E8.5 wild-type (M-P), Foxh1-/- (Q-T), and Foxa2-/- (U-W) embryos; lateral (M,Q,U) and ventral (N,R,V) views and transverse sections (O,P,S,T,W). In wild-type embryos Cpn1 is expressed in both the visceral yolk sac endoderm and the midgut and hindgut definitive endoderm (M,N and arrow in P). Expression does not extend into the foregut invagination (M and arrow in O). Cpn1 expression extends into the foregut invagination in Foxh1-/- embryos (Q and arrow in S). Expression of Cpn1 also remains in the midgut and hindgut regions (R and arrow in T). Cpn1 expression also remains in the hindgut of Foxa2-/- embryos (V and arrow in W). fg – foregut, hg- hindgut, ys – yolk sac. Approximate locations of transverse sections are indicated by dashed line.
124 foregut invagination (Figure 4.4E and arrow in G). Similarly in Foxa2 null embryos, AFP expressing cells are also found in the foregut invagination (arrow in Figure 4.4J). We also observed a downregulation of AFP expression in the visceral endoderm of both E7.5 and E8.5
Foxa2 null embryos, and this likely reflects the role of Foxa2 in regulation of AFP expression in this tissue. Interestingly, AFP expression is never observed in the midgut and hindgut regions of
Foxh1 and Foxa2 null embryos (Figure 4.4F and arrowheads in H), further suggesting that the endoderm in these regions is definitive endoderm.
Cpn1 is a marker of both the visceral and definitive endoderm (Hou et al., 2007). At E8.5
Cpn1 is expressed in the midgut and hindgut regions, as well as the visceral yolk sac endoderm and the neural tube (Figure 4.4M,N and arrow in P). Similar to AFP, Cpn1 is not expressed in the foregut invagination of wild-type embryos (Figure 4.4M and arrow in O), however, it is expressed throughout the hindgut (arrow in Figure 4.4P). In Foxh1 null embryos, Cpn1 expression is observed throughout the midgut and hindgut as well as in the foregut invagination
(Figure 4.4Q,R and arrows in S,T). As expression of Cpn1 does not extend into the foregut invagination of wild-type embryos, we concluded that the Cpn1 expressing cells in the foregut of
Foxh1 null embryos are visceral endoderm, whereas the Cpn1 expressing cells in the midgut and hindgut are definitive endoderm. Strong expression of Cpn1 is also observed in the midgut and hindgut of Foxa2 null embryos (arrow in Figure 4.4U,V and arrow W). Thus, as a result of the deficiency in foregut definitive endoderm in Foxh1 and Foxa2 mutants the visceral endoderm that is not displaced from the embryonic region becomes incorporated into the foregut invagination.
125 4.2.5 Analysis of the function and differentiation potential of the definitive endoderm in
Foxh1-/- mutants
Analysis of gene expression patterns of several visceral and definitive endoderm markers provides strong evidence that the midgut and hindgut definitive endoderm still form in Foxh1
and Foxa2 null embryos and suggests that formation of these endoderm populations is
independent of the anterior primitive streak. However, tissue recombination experiments have
demonstrated that the underlying mesoderm plays a critical role in patterning and inducing
differentiation of the definitive endoderm (Wells and Melton, 2000). To exclude the possibility
that the endoderm present in Foxh1 and Foxa2 mutants is visceral endoderm that has been
induced to express definitive endoderm markers, we analyzed the function and differentiation
potential of the endoderm in these mutants.
It has been demonstrated that visceral endoderm has the ability to actively endocytose
extracellular molecules, such as horseradish peroxidase (HRP), while definitive endoderm does
not have this ability (Balakier, 1984; Bielinska et al., 1999; Kadokawa et al., 1987). This
approach has been used previously to distinguish between visceral and definitive endoderm and
confirm an observed deficiency in definitive endoderm formation based on in situ hybridization
analysis (Kanai-Azuma et al., 2002; Maurer and Cooper, 2005; Sugimoto et al., 2003). We
compared the ability of the endoderm in wild-type and Foxh1 null embryos to take up exogenous
HRP at E7.5 and E8.5. In E7.5 wild-type embryos, the visceral endoderm in the extraembryonic
region has incorporated the HRP while the definitive endoderm in the embryonic region has not
(Figure 4.5A,B,C). Similar results are observed in E8.5 embryos, where the visceral yolk sac
endoderm has taken up the HRP and is strongly labelled, while the tissues of the embryo proper
remain unlabelled (Figure 4.5M,N,O,P). In E7.5 Foxh1 null embryos a discrete patch of HRP
126
Figure 4.5 The endoderm in the midgut and hindgut of Foxh1-/- embryos does not display properties of the visceral endoderm
A-I: HRP uptake assay in E7.5 wild-type (A-B) and Foxh1-/- (D-H) embryos; lateral (A,D,G) and anterior (B,E,H) views with anterior to the left and transverse sections (C,F,I). In wild-type embryos cells with the ability to take up extracellular HRP are localized to the proximal extraembryonic region (area above arrowhead in A,B and above arrowhead in inset in C). Cells with the ability to take up HRP are found in the embryonic region of Foxh1-/- embryos (arrows in D,E,F,G,H,I). J-L: Expression of AFP in E7.5 Foxh1-/- embryos, lateral (J) and anterior (K) views with anterior to the left and transverse section (L). AFP expressing cells are observed in the distal embryonic region of Foxh1-/- embryos (arrows in J,K,L). M-T: HRP uptake assay in E8.5 wild-type (M-P) and Foxh1-/- (Q-T) embryos, ventral (M,Q,R,S) and anterior (N) views and transverse sections (O,P,T). In wild-type embryos, cells with the ability to take up HRP are found in the yolk sac endoderm (M,N, and arrow in P) and are not observed in the definitive endoderm (M,N,O). Cells with the ability to take up HRP are found in the foregut of Foxh1-/- embryos (arrows in Q,R,S,T) but are not observed in the midgut or hindgut region (Q,R,S). fg – foregut, hg- hindgut, ys – yolk sac, exve – extraembryonic visceral endoderm, de – definitive endoderm. Approximate locations of transverse sections are indicated by dashed line.
127 labelled cells is observed in the outer layer of the embryonic region (arrows in Figure
4.5D,E,F,G,H,I). In E8.5 Foxh1 null embryos HRP positive cells are observed in the foregut invagination, as well as the visceral yolk sac (arrows in Figure 4.5Q,R,S,T). The HRP uptake assay results are in agreement with the expression of AFP in E7.5 (arrows in Figure 4.5J,K,L) and E8.5 (Figure 4.4E,F,G,H) Foxh1 null embryos and further supports the conclusion that the midgut and hindgut definitive endoderm remains in Foxh1 and Foxa2 null embryos. This results in an incomplete displacement of the visceral endoderm from the embryonic region and incorporation of these cells into the foregut invagination of the null embryos.
We further analyzed the differentiation potential of the definitive endoderm in Foxh1 null embryos by assaying for markers of definitive endoderm-derived organs. Since Foxh1 null embryos arrest at the beginning of the organogenesis stage (E9.5), we took advantage of an explant culture system (Gualdi et al., 1996) combined with RT-PCR to analyze the expression of the pancreas and duodenum marker, Pancreatic and duodenal homeobox 1 (Pdx1), and lung marker, NK2 homeobox 1 (Nkx2.1) (Ahlgren et al., 1996; Serls et al., 2005). The ectoderm was manually separated from the mesoderm and endoderm of early somite stage embryos and then cultured ex vivo (see experimental procedures for further details). After one week in culture, the explants were collected and RT-PCR was performed. Interestingly, Pdx1 and Nkx2.1 expression were found in the mesoderm and endoderm portion in both wild-type explants and the Foxh1 null explants (Figure 4.6A, lanes 1,3 and 5). No Pdx1 or Nkx2.1 expression was detected in the negative control of ectoderm explants (Figure 4.6A, lane 2 and 4). At E9.0 Pdx1 is known to be expressed in the dorsal pancreas bud and duodenum (which is derived from the midgut) and ventral pancreas bud (which is derived from the ventral foregut) (arrows in Figure 4.6B). To further discriminate whether the Pdx1 expression detected in Foxh1 null explants was in the ventral foregut or midgut, we cultured the mesoderm and endoderm with the ventral foregut
128 removed (mes + dfg-mg-hg), and the mesoderm and ventral foregut only with rest of the
endoderm removed (mes + vfg). As expected, Pdx1 expression was detected in the mes + dfg-
mg-hg mutant explants, but not in the mes + vfg mutant explants (Figure 4.6A, lanes 6-8). The
same result was observed for Nkx2.1 (Figure 4.6A, lanes 6-8). Overall, these results suggest that
the definitive endoderm formed in Foxh1-/- embryos possesses the potential to further
differentiate into lung, duodenum and/or dorsal pancreas, but not ventral pancreas due to the
failure of displacement of the visceral endoderm from the ventral foregut.
Next, we investigated Pdx1 expression in vivo using whole mount in situ hybridization.
Pdx1 is expressed in the developing pancreatic buds, as well as the duodenum (Jonsson et al.,
1994). In contrast with the dorsal and ventral pancreas expression of Pdx1 in WT embryos
(arrows in Figure 4.6B), analysis of the Pdx1 expression in E9.0 Foxh1 null embryos revealed
weak expression along the entire length of the midline (Figure 4.6C), however, the expression in
the ventral pancreas is absent. Sectioning confirmed that the expression of Pdx1 is in the dorsal
midline of the developing gut (arrowhead in inset of Figure 4.6D). Previous research has
suggested that expression of Pdx1 in the dorsal pancreas is negatively regulated by Sonic
Hedgehog (Shh) signaling from the notochord (Hebrok et al., 1998; Kim et al., 1997). Thus, the ectopic expression of Pdx1 in the entire gut midline could be due to the loss of notochord structure in Foxh1 null embryos and thus loss of Shh signaling. Since Pdx1 is expressed in both the developing pancreas and duodenum, it is still unclear as to whether the gut midline cells expressing Pdx1 in Foxh1 null embryos are pancreas or duodenum. Nevertheless, this result further supports the conclusion that the definitive endoderm formed in Foxh null embryos possesses a certain extent of further differentiation potential.
129
Figure 4.6 The definitive endoderm in Foxh1-/- embryos retains further differentiation potential
A: Different germ layers were dissected and cultured for a week, followed by RT-PCR analysis of the expression of Nkx2.1 (developing lung marker) and Pdx1 (developing pancreas and duodenum marker). Both Nkx2.1and Pdx1 are expressed in the mesoderm and endoderm samples, but not in the ectoderm samples, in either wild-type (WT) (lane1-2) or Foxh1-/- (lane3- 5). Analysis of mesoderm explants with ventral foregut (mes + vfg, lane 7) and the mesoderm and endoderm explants with ventral foregut removed (mes + dfg-mg-hg, lane 6 and 8) showed that Nkx2.1 and Pdx1 expression in the mes + dfg-mg-hg explants but not mes + vfg explants. Embryos were staged based on number of somites, which is indicated in each lane. B-D: Expression of Pdx1 in E9.0 wild-type (B) and Foxh1-/- (C) embryos; lateral views (B,C) with anterior to the left and transverse section (D). Pdx1 is expressed in the dorsal and ventral pancreatic bud in wild-type embryos (B). In Foxh1-/- embryos expression of Pdx1 is observed along the entire midline of the developing gut tube (C) and arrowhead in inset in D. The approximate location of the transverse section is indicated by dashed line. So – somite; mes - mesoderm; end - endoderm; ect - ectoderm; vfg: ventral foregut, mg – midgut, hg – hindgut, dp – dorsal pancreas, vp – ventral pancreas.
130 4.3 DISCUSSION
Using a panel of novel definitive and visceral endoderm markers, we have characterized
definitive endoderm formation in anterior primitive streak deficient mutants. Specifically, we
have performed an extensive analysis of markers specific to the foregut, midgut, and hindgut
endoderm populations. Our results provide evidence that loss of the anterior primitive streak cell
population specifically affects formation of the foregut definitive endoderm (anterior definitive
endoderm) and that the midgut and hindgut definitive endoderm (posterior definitive endoderm)
populations are largely unaffected by the loss of this cell population. The deficiency of foregut
definitive endoderm results in the incomplete displacement of the visceral endoderm to the
extraembryonic region during gastrulation and the subsequent incorporation of the visceral
endoderm into the foregut invagination. We further demonstrate that the definitive endoderm that
forms in the absence of the anterior primitive streak retains the ability to differentiate suggesting
it retains the ability to respond to patterning cues from the surrounding mesoderm.
The study of definitive endoderm formation in various mouse mutants has primarily
utilized genetic markers such as Shh, Foxa1, Foxa2, Hhex, and Cer1. A drawback of these markers is that expression of these genes is in all regions of the definitive endoderm (Foxa2,
Shh), at very low levels in the definitive endoderm (Shh), in a very limited population of the definitive endoderm (Hhex), or in tissues other than the definitive endoderm, such as the visceral endoderm or notochordal plate (Sox17, Cer1, Foxa2, Hhex) (Ang et al., 1993; Belo et al., 1997;
Biben et al., 1998; Echelard et al., 1993; Kanai-Azuma et al., 2002; Keng et al., 1998; Monaghan et al., 1993; Shawlot et al., 1998). A specific analysis of the midgut and hindgut definitive
endoderm has not been previously possible due to the lack of markers specific to these definitive
endoderm populations. We have previously shown Pyy, Nepn, and Trh to be markers of the
131 foregut, midgut, and whole definitive endoderm, respectively (Hou et al., 2007; McKnight et al.,
2007). Using these, and other, markers we have conclusively demonstrated that formation of the midgut and hindgut definitive endoderm does not require the transcription factors Foxh1 or
Foxa2. This finding is supported by several previous studies of these mouse mutants. Chimera studies using Foxh1 null ES cells demonstrated that these cells are unable to contribute to the foregut but retain the ability to populate the hindgut region at low levels (Hoodless et al., 2001).
Foxa2 null embryos are also thought to retain the ability to form definitive endoderm based on the presence of a closed gut-like structure in the posterior region of embryos and the expression of Foxa1, Shh, and Steel (Kitl) in the hindgut region (Ang and Rossant, 1994; Weinstein et al.,
1994). Furthermore, analysis of Gata4 expression in embryos lacking Foxa2 specifically in the embryonic tissues showed that formation of the hindgut was largely unaffected in these embryos
(Dufort et al., 1998). Furthermore, in embryos where the node (the organizer at late gastrulation stages and a derivative of the anterior primitive streak) was ablated, midgut and hindgut formation still occurred (Davidson et al., 1999). Thus, our analysis not only confirms the presence of the hindgut definitive endoderm in Foxh1 and Foxa2 null embryos but for the first time demonstrates that the midgut definitive endoderm also remains in these mutants.
Extensive analysis of the role of Nodal signaling in the mouse embryo has been performed by genetically manipulating the levels of Nodal using hypomorphic alleles or tissue specific ablation of other components of the pathway, such as Smad2 (Schier, 2003; Shen, 2007).
These studies have demonstrated that Nodal regulates cell fate during gastrulation in a dose- dependent manner with the highest levels of Nodal required for specification of the anterior streak derivatives such as the foregut and prechordal plate (anterior definitive endoderm and anterior axial mesoderm), while lower levels are required for specification of more middle and posterior streak derivatives such as the node, notochord and midgut/hindgut (posterior axial
132 mesoderm and posterior endoderm). Several studies have demonstrated that the hindgut
definitive endoderm still forms in the context of reduced Nodal signaling. Expression of Shh and
Foxa2, and a distinct hindgut invagination remains in the posterior region of
Smad2Robm1/+;Smad3-/-, Sox2CRE/+;Smad2CA/Robm1;Smad3+/-; Smad2ΔC/+;Smad3+/-, NodalΔfl/+,
NodalΔPEE/413.d, and Arkadia-/- embryos (Dunn et al., 2004; Liu et al., 2004; Lowe et al., 2001;
Niederlander et al., 2001; Vincent et al., 2003). However, in all of these mutants, in addition to
embryos lacking Foxh1 and Foxa2, there is a reduction or complete absence of the foregut
definitive endoderm. This data, in addition to findings presented here, suggests that specification
of the midgut and hindgut definitive endoderm requires lower levels of Nodal signaling than the
foregut definitive endoderm.
Foxh1 has been shown to play a key role in the autoregulation of Nodal levels in the
mouse embryo, and it is possible that the primary function of Foxh1 is to amplify Nodal levels to
reach a threshold that is sufficient to specify the anterior primitive streak and its derivatives
(Norris et al., 2002; Vincent et al., 2003). This is supported by the phenotype of Foxh1 null
embryos, which specifically lack anterior streak derivatives (Hoodless et al., 2001; Yamamoto et
al., 2001). As Nodal signaling is involved in specification of tissues in addition to the anterior
streak derivatives, this suggests that the role of Nodal and Smad2/3 signals in patterning mid-
and posterior streak derivatives is mediated independently of Foxh1. The data presented in this
study supports this model. Alternatively, specification of the midgut and hindgut (posterior
endoderm) lineages may be independent of either Nodal signaling or Foxh1/Foxa2. Tissue
recombination experiments have demonstrated that Fibroblast growth factor 4 (Fgf4) represses
anterior cell fate and promotes posterior (midgut and hindgut) endoderm fates. Thus FGF
signaling may act in a similar manner to the Nodal pathway where there is a graded response,
with high levels of FGF specifying the hindgut (posterior endoderm) and where no FGF
133 signaling or inhibition of FGF signaling is required for specifying the foregut (anterior
endoderm) (Wells and Melton, 2000).
Wnt signaling has also been implicated in endoderm specification. Embryos lacking both
Tcf1 and Tcf4 have defects in hindgut specification and display an anterior transformation of the
duodenum (Gregorieff et al., 2004). Other factors have also been implicated specifically in
specification of the posterior endoderm. Sox17 is a well conserved transcription factor and
homologs of this gene in zebrafish play a critical role in endoderm specification. In the mouse,
loss of Sox17 results in defects specifically in the hindgut definitive endoderm (Kanai-Azuma et al., 2002). Embryos lacking Tead1 and Tead2 (Tead1-/-;Tead2-/-) have reduced expression of
Sox17 in the hindgut, indicating defective hindgut formation (Sawada et al., 2008). Furthermore,
chimera studies using Mixl1 and Sd (Danforth’s short tail) null ES cells have demonstrated that
these cells contribute to all embryonic lineages but are specifically excluded from the hindgut
definitive endoderm (Hart et al., 2002; Maatman et al., 1997). Further analysis of endoderm
formation in these mutants using markers for specific endoderm populations, as well as lineage
tracing in these mutants will be necessary to define which signaling pathways are involved in
specification of the various subpopulations of definitive endoderm.
Lineage tracing studies have demonstrated that as the nascent definitive endoderm
emerges from the primitive streak it intercalates into the visceral endoderm layer and then
expands to cover the embryonic region (Lawson et al., 1986; Lawson and Pedersen, 1987). As a
result of the deficiency of foregut definitive endoderm in Foxh1 and Foxa2 null embryos there is
an incomplete displacement of the visceral endoderm from the embryonic region, demonstrated
by the presence of cells expressing Afp and having the ability to endocytose HRP remaining in
the embryonic region following gastrulation. This is also supported by the altered/reduced
134 expression domains of several markers of the nascent definitive endoderm in the embryonic
region of Foxh1 and Foxa2 null embryos. Analysis of the expression of visceral endoderm
markers in E8.5 embryos determined/demonstrated that the visceral endoderm that remains in the
embryonic region of Foxh1 and Foxa2 null embryos following gastrulation becomes
incorporated into the foregut invagination. Previous studies have also observed an expansion of
the visceral endoderm domain when there is a deficiency on definitive endoderm, demonstrated
by the expansion of Hnf4α into the embryonic region of Sox2Cre/+;Smad2CA/Robm1, expression of
Gata4 in the foregut (and midgut) of embryos with embryonic specific depletion of Foxa2, and
expansion of Gata4, Hnf4α, and Afp into the hindgut region of Sox17 null embryos. Additionally, cells with visceral endoderm morphology are found in the embryonic region of Foxh1-/- embryos
that have a severe constriction between the embryonic and extraembryonic regions (Yamamoto
et al., 2001). Several studies have demonstrated a role for the anterior definitive endoderm in
patterning the anterior neurectoderm in the embryo (Hallonet et al., 2002). The presence of
anterior truncations in Foxh1 null embryos where there is a foregut invagination that is lined
with visceral endoderm instead of definitive endoderm supports previous observations that the
visceral endoderm cannot functionally substitute/compensate for the definitive endoderm. As a
result it is unable to respond to signals from surrounding embryonic tissues and does not
differentiate into the organs that are derived from the foregut. In agreement with this Foxh1 null
embryos do not express markers of the liver (Hhex) or pancreatic (Pdx1) progenitors. However,
we have demonstrated that the midgut and hindgut definitive endoderm that forms in Foxh1 null
embryos retains its ability to differentiate into the organs that are derived from these regions of
the gut tube.
The emerging field of regenerative medicine has generated considerable interest in the
mouse definitive endoderm. Recent studies have involved the identification of novel markers
135 which can be used to monitor the differentiation of stem cells to therapeutically applicable cell types, as well as purify/isolate these cells following differentiation (Gu et al., 2004; Sherwood et al., 2007). In addition to potential clinical applications, extensive analysis of these novel markers during embryogenesis will aid greatly in our understanding of definitive endoderm specification and patterning. This in turn will further our efforts to efficiently differentiate cells for the treatment of diseases such as hepatitis and diabetes. A clear understanding of the signals involved in specification of the definitive endoderm, and in particular the various subpopulations of definitive endoderm and their organ primordia, will be essential to our understanding of how to differentiate stem cells or reprogram other cell types for use in organ replacement.
4.4 EXPERIMENTAL PROCEDURES
4.4.1 Mouse strains, dissection, and staging
Embryos were obtained from crosses of Foxh1 (maintained on a mixed 129-ICR background) or Foxa2 (maintained on an ICR background) heterozygous mice and considered
0.5 days postcoitum at noon of the day of detection of the vaginal plug. Embryos were dissected from the uterus in Dulbecco’s Phosphate Buffered Saline (Invitrogen) with 0.1% Fetal Bovine
Serum (Invitrogen). Staging of embryos was carried out as previously described (Downs and
Davies, 1993). For in situ hybridization embryos were placed into individual tubes and fixed overnight at 4˚C in 4% paraformaldehyde, dehydrated through a graded methanol series and stored at -20˚C until use. For explant cultures embryos were prepared as described in section
4.4.5. Genotyping was carried out by removing the ectoplacental cone (epc) at the time of dissection and culturing it for 4-5 days in DMEM + 10% FBS on gelatin coated tissue culture dishes. Epc lysis and PCR genotyping was carried out as previously described (Hoodless et al.,
136 2001; Weinstein et al., 1994). Following genotyping embryos were pooled according to genotype
and processed for in situ hybridization as described below.
4.4.2 Probe preparation
Digoxigenin-UTP labeled mRNA probes were synthesized as per manufacturer’s
instructions (Roche). Briefly, 10 μg of template DNA was linearized with 10 Units of an
appropriate enzyme and purified. RNA probes were in vitro transcribed using 1 μg of linearized template and 20 Units of SP6, T3 or T7 RNA polymerase (Roche). RNA probes were purified using G-50 Sephadex columns (Amersham) and stored at -80˚C. Probes used in this study are as follows: Cer1 (as described in Hoodless et al., 2001), Sox17 (kindly obtained from Y.Kanai, as described in Kanai-Azuma et al., 2002), Afp (kindly obtained from S. Duncan), Nepn, Trh, Pyy, and Cpn1 (as described in Hou et al., 2007), and Pdx1 (probe generated from PCR amplicon using primers described in section 4.4.7 and subcloned into the pCRII vector (Invitrogen).
4.4.3 Whole mount in situ hybridization
Whole mount in situ hybridization (WISH) was carried out essentially as in Wilkinson
(Wilkinson and Nieto, 1993) with the following modifications. Pre-hybridization and hybridization steps were performed as described using the following hybridization solution: 50% formamide, 1.3 X SSC (pH 5), 50 μg/mL yeast tRNA, 0.2% Tween-20, 0.5% CHAPS, 100
μg/mL Heparin). Post-hybridization washes were performed as follows: 3 rinses followed by 2
30-minute washes at 65˚C with hybridization solution, 10 minute wash with 1:1 mixture of hybridization solution and MABT (100 mM maleic acid, 150 mM NaCl, pH 7.5, 0.1% Tween-
20), then 2 rinses followed by 1 wash with MABT. Blocking was performed for 1 hour at room
137 temperature with 2% Blocking Reagent (Roche) dissolved in MABT and an additional 2 hours at
room temperature with 2% Blocking Reagent (Roche) and 20% heat inactivated goat serum
(Invitrogen) in MABT. Embryos were incubated in anti-Digoxigenin antibody (1:2000) (Roche)
in 2% Blocking Reagent and 20% heat inactivated goat serum in MABT overnight at 4˚C.
Embryos were rinsed and washed 3 times for 1 hour with MABT followed by 2 10 minute
washes in NTMTL (100 mM, NaCl, 100 mM Tris-HCL pH9.5, 50 mM MgCl2, 1% Tween-20, 2
mM Levamisole). Embryos were incubated in BM purple alkaline phosphatase substrate (Roche)
at 4˚C or room temperature. Staining reactions were stopped by washes in PBS with 0.1%
Tween-20.
4.4.4. Imaging and histology
Following WISH, embryos were processed through a graded PBT:Glycerol series to 50%
glycerol. Images were taken using a Leica MZ9.5 microscope and a Leica DFC420 camera with the Leica Application Suite software (Leica). For histology, embryos were incubated overnight at
4˚C in a 1:1 mixture of Tissue-Tek OCT embedding medium (Sakura Finetek) and 60% sucrose.
Embryos were embedded in 100% OCT (in 10x10 mm plastic mold) and frozen with a dry
ice:ethanol mixture. Cryosections (10 uM) were cut using a Microm cryostat and placed on
Superfrost Plus glass slides (VWR) and allowed to air dry. Slides were washed twice in PBS for
10 minutes per wash and mounted using 50% glycerol and glass coverslips (VWR). Sections
were imaged using Zeiss Axioplan microscope and OpenLab software (Improvision).
138 4.4.5 Endoderm explant culture
Early somite stage embryos were dissected in Dulbecco’s Phosphate Buffered Saline and
under the stereomicroscope the ectoderm was manually removed. The ectoderm or the
endoderm+mesoderm were cultured on a Collagen I coated 24-well tissue culture plate in
DMEM + 10 % Fetal Bovine Serum for 7 days. Following culture the explants were collected in
Trizol (Invitrogen) and RNA was extracted for RT-PCR analysis as described in section 4.4.7.
4.4.6 Horseradish peroxidase uptake assay
Embryos were dissected in pre-warmed (37°C) Dulbecco's Modified Eagle's Medium
(DMEM) + 10% Bovine Serum Albumin (BSA). Embryos were then incubated in 2 mg/mL of
Type IV Horseradish Peroxidase (HRP) freshly dissolved in DMEM + 10% BSA for 30 minutes at room temperature. Embryos were then washed 2 times for 10 minutes per wash in Phosphate
Buffered Saline (PBS). Embryos were then fixed for 30 minutes at room temperature in 4%
Paraformaldehyde (PFA) in PBS. Embryos were then washed 2 times for 10 minutes per wash with PBS. The 3, 3’-diaminobenzidine (DAB) HRP substrate was prepared as per manufacturers instructions (Vector Labs). Embryos were placed in the DAB substrate solution and the staining was monitored under a dissecting microscope. The staining reaction was stopped by washing the embryos 2 times for 10 minutes per wash in PBS. Embryos were processed through a glycerol:PBS series to a final concentration of 80% glycerol for imaging.
4.4.7 RNA extraction, cDNA synthesis, and RT-PCR
Embryos were placed into Trizol (Invitrogen) immediately following dissection and
stored at -80˚C. RNA was extracted using MaXtract high density phase-lock gels (Qiagen)
139 following manufacturer’s instructions. RT-PCR was performed as follows. 1 μg of total RNA
was treated with DNAase I (Invitrogen) prior to cDNA synthesis. cDNA was synthesized using
Superscript II Reverse Transcriptase (Invitrogen) as per the manufacturer’s instructions with the
following modifications. 1/10 of the DNase I treated RNA was mixed with 6 μg random hexamer
primers (Invitrogen), 0.5 mM dNTPs (Invitrogen) and DEPC-treated water, and incubated for 5
minutes at 65˚C then placed on ice. 1X 1st strand synthesis buffer (Invitrogen), 0.02 mM DTT
(Invitrogen) and 50 Units of RNaseOUT (Invitrogen) were added and samples were incubated at
42˚C for 2 minutes. The reaction was divided into 2 equal volumes and 200 Units of SuperScript
II was added to one set of tubes while no enzyme was added to the minus reverse transcriptase (-
RT) negative control. Samples were incubated for 50 minutes at 42˚C followed by 15 minutes at
70˚C. Reactions were diluted to 50 μL with molecular biology grade water (Invitrogen) and 2 μL
was used per PCR reaction. PCR was performed using HotStar Taq Polymerase (Qiagen) as per
manufacturer’s instructions in a PTC-200 thermal cycler (MJ Research). Thermal cycler
conditions were as follows: 95˚C 15 minutes, 95˚C 30 seconds, 55˚C 30 seconds, 72˚C 30
seconds, 72˚C 10 minutes for 40 cycles. Primers used were as follows: Gapdh (NM_008084)
Forward: 5’- -3’, Reverse: 5’- -3’; Nkx2.1 (NM_009385) Forward: 5’-
CCAGGACACCATGCGGAACA-3’, Reverse: 5’-GGCCATGTTCTTGCTCACGT-3’; Pdx1
(NM_008814) Forward: 5’-AGCAAGATTGTGCGGTGACC-3’, Reverse: 5’-
AGTTTGGAGCCCAGGTTGTC-3’.
140 4.5 REFERENCES
Ahlgren, U., Jonsson, J., and Edlund, H. (1996). The morphogenesis of the pancreatic mesenchyme is uncoupled from that of the pancreatic epithelium in IPF1/PDX1-deficient mice. Development 122, 1409-16. Ang, S. L., and Rossant, J. (1994). HNF-3 beta is essential for node and notochord formation in mouse development. Cell 78, 561-74. Ang, S. L., Wierda, A., Wong, D., Stevens, K. A., Cascio, S., Rossant, J., and Zaret, K. S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development 119, 1301-15. Balakier, H. (1984). Endogenous peroxidase in the visceral endoderm of early mouse embryos. J Exp Zool 231, 243-8. Belo, J. A., Bouwmeester, T., Leyns, L., Kertesz, N., Gallo, M., Follettie, M., and De Robertis, E. M. (1997). Cerberus-like is a secreted factor with neutralizing activity expressed in the anterior primitive endoderm of the mouse gastrula. Mech Dev 68, 45-57. Biben, C., Stanley, E., Fabri, L., Kotecha, S., Rhinn, M., Drinkwater, C., Lah, M., Wang, C. C., Nash, A., Hilton, D., Ang, S. L., Mohun, T., and Harvey, R. P. (1998). Murine cerberus homologue mCer-1: a candidate anterior patterning molecule. Dev Biol 194, 135-51. Bielinska, M., Narita, N., and Wilson, D. B. (1999). Distinct roles for visceral endoderm during embryonic mouse development. Int J Dev Biol 43, 183-205. Camus, A., and Tam, P. P. (1999). The organizer of the gastrulating mouse embryo. Curr Top Dev Biol 45, 117-53. Chen, X., Rubock, M. J., and Whitman, M. (1996). A transcriptional partner for MAD proteins in TGF-beta signalling. Nature 383, 691-6. Davidson, B. P., Kinder, S. J., Steiner, K., Schoenwolf, G. C., and Tam, P. P. (1999). Impact of node ablation on the morphogenesis of the body axis and the lateral asymmetry of the mouse embryo during early organogenesis. Dev Biol 211, 11-26. Downs, K. M., and Davies, T. (1993). Staging of gastrulating mouse embryos by morphological landmarks in the dissecting microscope. Development 118, 1255-66. Dufort, D., Schwartz, L., Harpal, K., and Rossant, J. (1998). The transcription factor HNF3beta is required in visceral endoderm for normal primitive streak morphogenesis. Development 125, 3015-25. Dunn, N. R., Vincent, S. D., Oxburgh, L., Robertson, E. J., and Bikoff, E. K. (2004). Combinatorial activities of Smad2 and Smad3 regulate mesoderm formation and patterning in the mouse embryo. Development 131, 1717-28. Echelard, Y., Epstein, D. J., St-Jacques, B., Shen, L., Mohler, J., McMahon, J. A., and McMahon, A. P. (1993). Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 75, 1417-30. Gregorieff, A., Grosschedl, R., and Clevers, H. (2004). Hindgut defects and transformation of the gastro-intestinal tract in Tcf4(-/-)/Tcf1(-/-) embryos. Embo J 23, 1825-33. Gu, G., Wells, J. M., Dombkowski, D., Preffer, F., Aronow, B., and Melton, D. A. (2004). Global expression analysis of gene regulatory pathways during endocrine pancreatic development. Development 131, 165-79. Gualdi, R., Bossard, P., Zheng, M., Hamada, Y., Coleman, J. R., and Zaret, K. S. (1996). Hepatic specification of the gut endoderm in vitro: cell signaling and transcriptional control. Genes Dev 10, 1670-82.
141 Hallonet, M., Kaestner, K. H., Martin-Parras, L., Sasaki, H., Betz, U. A., and Ang, S. L. (2002). Maintenance of the specification of the anterior definitive endoderm and forebrain depends on the axial mesendoderm: a study using HNF3beta/Foxa2 conditional mutants. Dev Biol 243, 20-33. Hart, A. H., Hartley, L., Sourris, K., Stadler, E. S., Li, R., Stanley, E. G., Tam, P. P., Elefanty, A. G., and Robb, L. (2002). Mixl1 is required for axial mesendoderm morphogenesis and patterning in the murine embryo. Development 129, 3597-608. Hart, A. H., Willson, T. A., Wong, M., Parker, K., and Robb, L. (2005). Transcriptional regulation of the homeobox gene Mixl1 by TGF-beta and FoxH1. Biochem Biophys Res Commun 333, 1361-9. Hebrok, M., Kim, S. K., and Melton, D. A. (1998). Notochord repression of endodermal Sonic hedgehog permits pancreas development. Genes Dev 12, 1705-13. Hoodless, P. A., Pye, M., Chazaud, C., Labbe, E., Attisano, L., Rossant, J., and Wrana, J. L. (2001). FoxH1 (Fast) functions to specify the anterior primitive streak in the mouse. Genes Dev 15, 1257-71. Hou, J., Charters, A. M., Lee, S. C., Zhao, Y., Wu, M. K., Jones, S. J., Marra, M. A., and Hoodless, P. A. (2007). A systematic screen for genes expressed in definitive endoderm by Serial Analysis of Gene Expression (SAGE). BMC Dev Biol 7, 92. Izzi, L., Silvestri, C., von Both, I., Labbe, E., Zakin, L., Wrana, J. L., and Attisano, L. (2007). Foxh1 recruits Gsc to negatively regulate Mixl1 expression during early mouse development. Embo J 26, 3132-43. Jonsson, J., Carlsson, L., Edlund, T., and Edlund, H. (1994). Insulin-promoter-factor 1 is required for pancreas development in mice. Nature 371, 606-9. Kadokawa, Y., Kato, Y., and Eguchi, G. (1987). Cell lineage analysis of the primitive and visceral endoderm of mouse embryos cultured in vitro. Cell Differ 21, 69-76. Kanai-Azuma, M., Kanai, Y., Gad, J. M., Tajima, Y., Taya, C., Kurohmaru, M., Sanai, Y., Yonekawa, H., Yazaki, K., Tam, P. P., and Hayashi, Y. (2002). Depletion of definitive gut endoderm in Sox17-null mutant mice. Development 129, 2367-79. Keng, V. W., Fujimori, K. E., Myint, Z., Tamamaki, N., Nojyo, Y., and Noguchi, T. (1998). Expression of Hex mRNA in early murine postimplantation embryo development. FEBS Lett 426, 183-6. Kim, S. K., Hebrok, M., and Melton, D. A. (1997). Notochord to endoderm signaling is required for pancreas development. Development 124, 4243-52. Kinder, S. J., Tsang, T. E., Ang, S. L., Behringer, R. R., and Tam, P. P. (2001a). Defects of the body plan of mutant embryos lacking Lim1, Otx2 or Hnf3beta activity. Int J Dev Biol 45, 347-55. Kinder, S. J., Tsang, T. E., Wakamiya, M., Sasaki, H., Behringer, R. R., Nagy, A., and Tam, P. P. (2001b). The organizer of the mouse gastrula is composed of a dynamic population of progenitor cells for the axial mesoderm. Development 128, 3623-34. Klingensmith, J., Ang, S. L., Bachiller, D., and Rossant, J. (1999). Neural induction and patterning in the mouse in the absence of the node and its derivatives. Dev Biol 216, 535- 49. Kwon, G. S., Fraser, S. T., Eakin, G. S., Mangano, M., Isern, J., Sahr, K. E., Hadjantonakis, A. K., and Baron, M. H. (2006). Tg(Afp-GFP) expression marks primitive and definitive endoderm lineages during mouse development. Dev Dyn 235, 2549-58. Labbe, E., Silvestri, C., Hoodless, P. A., Wrana, J. L., and Attisano, L. (1998). Smad2 and Smad3 positively and negatively regulate TGF beta-dependent transcription through the forkhead DNA-binding protein FAST2. Mol Cell 2, 109-20.
142 Lawson, K. A., Meneses, J. J., and Pedersen, R. A. (1986). Cell fate and cell lineage in the endoderm of the presomite mouse embryo, studied with an intracellular tracer. Dev Biol 115, 325-39. Lawson, K. A., and Pedersen, R. A. (1987). Cell fate, morphogenetic movement and population kinetics of embryonic endoderm at the time of germ layer formation in the mouse. Development 101, 627-52. Liu, B., Dou, C. L., Prabhu, L., and Lai, E. (1999). FAST-2 is a mammalian winged-helix protein which mediates transforming growth factor beta signals. Mol Cell Biol 19, 424- 30. Liu, Y., Festing, M., Thompson, J. C., Hester, M., Rankin, S., El-Hodiri, H. M., Zorn, A. M., and Weinstein, M. (2004). Smad2 and Smad3 coordinately regulate craniofacial and endodermal development. Dev Biol 270, 411-26. Lowe, L. A., Yamada, S., and Kuehn, M. R. (2001). Genetic dissection of nodal function in patterning the mouse embryo. Development 128, 1831-43. Maatman, R., Zachgo, J., and Gossler, A. (1997). The Danforth's short tail mutation acts cell autonomously in notochord cells and ventral hindgut endoderm. Development 124, 4019- 28. Maurer, M. E., and Cooper, J. A. (2005). Endocytosis of megalin by visceral endoderm cells requires the Dab2 adaptor protein. J Cell Sci 118, 5345-55. McKnight, K. D., Hou, J., and Hoodless, P. A. (2007). Dynamic expression of Thyrotropin- releasing hormone in the mouse definitive endoderm. Dev Dyn 236, 2909-17. Monaghan, A. P., Kaestner, K. H., Grau, E., and Schutz, G. (1993). Postimplantation expression patterns indicate a role for the mouse forkhead/HNF-3 alpha, beta and gamma genes in determination of the definitive endoderm, chordamesoderm and neuroectoderm. Development 119, 567-78. Niederlander, C., Walsh, J. J., Episkopou, V., and Jones, C. M. (2001). Arkadia enhances nodal- related signalling to induce mesendoderm. Nature 410, 830-4. Norris, D. P., Brennan, J., Bikoff, E. K., and Robertson, E. J. (2002). The Foxh1-dependent autoregulatory enhancer controls the level of Nodal signals in the mouse embryo. Development 129, 3455-68. Robb, L., and Tam, P. P. (2004). Gastrula organiser and embryonic patterning in the mouse. Semin Cell Dev Biol 15, 543-54. Sawada, A., Kiyonari, H., Ukita, K., Nishioka, N., Imuta, Y., and Sasaki, H. (2008). Redundant roles of Tead1 and Tead2 in notochord development and the regulation of cell proliferation and survival. Mol Cell Biol 28, 3177-89. Schier, A. F. (2003). Nodal signaling in vertebrate development. Annu Rev Cell Dev Biol 19, 589-621. Sellem, C. H., Frain, M., Erdos, T., and Sala-Trepat, J. M. (1984). Differential expression of albumin and alpha-fetoprotein genes in fetal tissues of mouse and rat. Dev Biol 102, 51- 60. Serls, A. E., Doherty, S., Parvatiyar, P., Wells, J. M., and Deutsch, G. H. (2005). Different thresholds of fibroblast growth factors pattern the ventral foregut into liver and lung. Development 132, 35-47. Shawlot, W., Deng, J. M., and Behringer, R. R. (1998). Expression of the mouse cerberus-related gene, Cerr1, suggests a role in anterior neural induction and somitogenesis. Proc Natl Acad Sci U S A 95, 6198-203. Shen, M. M. (2007). Nodal signaling: developmental roles and regulation. Development 134, 1023-34.
143 Sherwood, R. I., Jitianu, C., Cleaver, O., Shaywitz, D. A., Lamenzo, J. O., Chen, A. E., Golub, T. R., and Melton, D. A. (2007). Prospective isolation and global gene expression analysis of definitive and visceral endoderm. Dev Biol 304, 541-55. Sugimoto, M., Karashima, Y., Abe, K., Tan, S. S., and Takagi, N. (2003). Tetraploid embryos rescue the early defects of tw5/tw5 mouse embryos. Genesis 37, 162-71. Takaoka, K., Yamamoto, M., Shiratori, H., Meno, C., Rossant, J., Saijoh, Y., and Hamada, H. (2006). The mouse embryo autonomously acquires anterior-posterior polarity at implantation. Dev Cell 10, 451-9. Tam, P. P., and Beddington, R. S. (1992). Establishment and organization of germ layers in the gastrulating mouse embryo. Ciba Found Symp 165, 27-41; discussion 42-9. Tam, P. P., and Behringer, R. R. (1997). Mouse gastrulation: the formation of a mammalian body plan. Mech Dev 68, 3-25. Vincent, S. D., Dunn, N. R., Hayashi, S., Norris, D. P., and Robertson, E. J. (2003). Cell fate decisions within the mouse organizer are governed by graded Nodal signals. Genes Dev 17, 1646-62. Weinstein, D. C., Ruiz i Altaba, A., Chen, W. S., Hoodless, P., Prezioso, V. R., Jessell, T. M., and Darnell, J. E., Jr. (1994). The winged-helix transcription factor HNF-3 beta is required for notochord development in the mouse embryo. Cell 78, 575-88. Weisberg, E., Winnier, G. E., Chen, X., Farnsworth, C. L., Hogan, B. L., and Whitman, M. (1998). A mouse homologue of FAST-1 transduces TGF beta superfamily signals and is expressed during early embryogenesis. Mech Dev 79, 17-27. Wells, J. M., and Melton, D. A. (1999). Vertebrate endoderm development. Annu Rev Cell Dev Biol 15, 393-410. Wells, J. M., and Melton, D. A. (2000). Early mouse endoderm is patterned by soluble factors from adjacent germ layers. Development 127, 1563-72. Wilkinson, D. G., and Nieto, M. A. (1993). Detection of messenger RNA by in situ hybridization to tissue sections and whole mounts. Methods Enzymol 225, 361-73. Yamamoto, M., Meno, C., Sakai, Y., Shiratori, H., Mochida, K., Ikawa, Y., Saijoh, Y., and Hamada, H. (2001). The transcription factor FoxH1 (FAST) mediates Nodal signaling during anterior-posterior patterning and node formation in the mouse. Genes Dev 15, 1242-56.
144 Chapter 5
SUMMARY, PERSPECTIVES, AND FUTURE DIRECTIONS
5.1 GENE EXPRESSION PROFILING REVEALS NOVEL ATTRIBUTES OF THE
MOUSE DEFINITIVE ENDODERM
The anterior primitive streak is a group of approximately 40 cells that are specified at the
onset of gastrulation in the mouse embryo (Camus and Tam, 1999). This cell population has the
ability to induce a secondary axis upon heterologous transplantation, and as a result is referred to
as the mouse embryonic organizer (Beddington, 1994). In addition to its role as the embryonic
organizer, fate mapping studies have demonstrated that the anterior primitive streak contains
progenitors for midline tissues, including the anterior definitive endoderm, prechordal plate, notochord, and node (Kinder et al., 2001b). Data presented in this thesis has contributed significantly to our understanding of the role of the anterior primitive streak during mouse embryogenesis and specifically to its contribution to the formation of the mouse definitive endoderm. First, we performed gene expression profiling on wild-type and Foxh1 null embryos,
which lack the anterior primitive streak, to identify novel genes involved in specification of this
cell population. From this analysis we identified thyrotropin-releasing hormone (Trh) as being
downregulated in Foxh1 null embryos. Further characterization of the expression pattern of this
gene revealed it to be a novel marker of the mouse definitive endoderm (McKnight et al., 2007).
Trh is unique among markers of this germ layer as it marks the nascent definitive endoderm that
forms during gastrulation and goes on to mark the entire endodermal layer. Trh is not expressed in the extraembryonic visceral endoderm, unlike many other definitive endoderm markers.
145 Furthermore, Trh expression is rapidly downregulated as the definitive endoderm begins to differentiate, suggesting it may be a marker of primitive or undifferentiated cells. In accordance with this, Trh is highly expressed in the pluripotent epiblast as well as mouse embryonic stem cells (K.D.M, unpublished). Examination of the expression of Trh in Foxh1 null embryos revealed that formation of the definitive endoderm was not as severely compromised in these embryos as previously reported. This finding prompted us to examine the expression of a larger panel of novel endoderm markers we recently identified using Serial Analysis of Gene
Expression (SAGE) (Hou et al., 2007) in Foxh1 and Foxa2 null embryos, both of which lack the anterior primitive streak. This analysis revealed that while formation of the foregut definitive endoderm is severely compromised in anterior primitive streak deficient embryos, the majority of the midgut and hindgut definitive endoderm remains. This finding suggests that the midgut and hindgut definitive endoderm forms independently of the anterior primitive streak.
Furthermore, as the Nodal signaling pathway has been demonstrated to play a role in specification of the anterior primitive streak, these results suggest that formation of the posterior definitive endoderm (midgut and hindgut) is regulated by an alternate signaling pathway(s).
5.1.1 Gene expression profiling as a technique to study mouse developmental biology
Gene expression profiling provides a “snapshot” of all the genes that are expressed in a given cell or tissue type at a particular time. This allows one to analyze which signaling pathways are active and what cellular processes are occurring. This technique has been used for many years to rapidly characterize cell or tissue types, to identify novel markers, or to identify genes that are differentially expressed between various cell or tissue types. Expression profiling can also be used to predict transcriptional targets, identify co-expressed and co-regulated genes, and construct gene regulatory networks. Gene expression profiling, either array or sequence
146 based, have become popular methods for performing differential screens to identify differentially expressed genes. Recent advances in profiling technologies have made the application of this technique to the study of early mouse embryos more amenable, as they allow the use of a smaller amount of starting material. Additionally, these technologies can sequence to a depth previously unattainable due to financial constraints.
As described in Chapter 2 of this thesis, we used gene expression profiling to investigate the specification of the anterior primitive streak (mouse gastrula organizer). We used this technique with the aim of identifying genes that were differentially expressed between wild-type and Foxh1 null embryos, which do not specify this cell population (Hoodless et al., 2001;
Yamamoto et al., 2001). Upon analysis of the data, we concluded that the platform we chose
(Affymetrix™ GeneChips) was not sensitive enough to detect changes in such a limited cell population and of genes with low expression levels (such as transcription factors and signaling molecules). The limitations of the technology available to us at the time the study was performed required us to amplify the RNA from the wild-type and mutant samples, further reducing the likelihood of detecting differential expression. Alternatively, the lack of differential expression between the wild-type and Foxh1 null sample may reflect that the primary function of Foxh1 in specification of the anterior primitive streak is to amplify Nodal expression levels (Norris et al.,
2002). Recent advances in sequencing technologies and strategies to purify specific cell populations from mouse embryos will allow more sensitive and accurate expression profiling studies of the early mouse embryo to be performed and may be able to investigate the role of
Foxh1 in specification of the anterior primitive streak further. Nonetheless, we did identify
Thryotropin-releasing hormone (Trh) as being significantly downregulated in Foxh1 mutant embryos. Further studies of this gene in the mouse embryo have revealed several insights into the mouse definitive endoderm, as described in Chapter 3 and Chapter 4 of this thesis.
147 5.1.2 Tissue specific markers will facilitate the creation of essential molecular tools for the study of mouse embryogenesis
The field of developmental biology relies heavily on the use of tissue and cell-type specific genetic markers. These markers have several important functions, including characterizing defects in gene knockouts, monitoring in vitro differentiation of embryonic stem cells, creating (and verifying) fate maps of embryonic tissues, and generating molecular tools to analyze gene function in specific cell lineages.
As discussed in Section 5.1.1, gene expression profiling studies of post-implantation embryos have provided a large repertoire of genetic markers specific to various cell lineages.
Particular emphasis has been placed on identifying novel markers of the mouse definitive endoderm. Understanding of this germ layer has lagged behind that of the mesoderm and ectoderm primarily due to the lack of genetic markers specific to the definitive endoderm. As discussed in this thesis, many of the commonly used definitive endoderm markers, such as
Foxa2, Cer1, and Sox17, are also expressed in the extraembryonic visceral endoderm (Ang et al.,
1993; Belo et al., 1997; Biben et al., 1998; Kanai-Azuma et al., 2002; Monaghan et al., 1993).
Studies in several labs, including our own, have described several novel markers of both the definitive and visceral endoderm, such that there are now a host of genetic markers available to those in the field that can be used to study endoderm specification in the mouse embryo.
Attention should now turn to the careful characterization of the expression patterns of these markers in wild-type embryos and embryos deficient in various signaling pathways to better understand the regulation of these markers and their role in endoderm specification.
148 Additionally, these novel markers will serve as powerful genetic and molecular tools to
study endoderm formation in the mouse embryo. Identification of the regulatory elements that
control expression of genes expressed specifically in the definitive endoderm, such as Trh, will
be an important step in the generation of these molecular tools. Using the promoters of these
genes will allow the conditional (tissue and/or time specific) inactivation or overexpression
inactivation of genes specifically in the definitive endoderm lineage. Furthermore, expression of
a selectable marker, such as green fluorescent protein (GFP) will allow the purification of pure
populations of definitive endoderm using fluorescent activated cell sorting (FACS). This will
allow the in vitro manipulation of in vivo derived definitive endoderm cells to investigate the
signals involved in their differentiation. This system can also be used to investigate the regulation of novel definitive endoderm markers. Furthermore, it will allow the expression profiling of a homogeneous population of definitive endoderm cells. It will also aid in efforts to determine the transcriptional targets of the various transcription factors that are expressed in the definitive endoderm, such as Foxa2, using new technologies such as ChIP-Seq (Wederell et al.,
2008). The combination of ChIP-Seq and expression profiling will allow the construction of the gene regulatory networks that govern endoderm specification and patterning. Expression of an easily visualized marker, such as β-galactosidase or GFP will also allow genetic lineage tracing to be performed in both wild-type and mutant embryos. This technique would be valuable to study the origin of the midgut and hindgut endoderm that forms in Foxh1 and Foxa2 mutants.
5.1.3 Revisiting previously characterized mouse mutants with novel markers is important to further refine our understanding of germ layer specification and patterning
The application of gene targeting techniques has provided a wealth of knowledge regarding the genetic control of mouse embryogenesis over the past 20 years. Since the initial
149 characterization of these gene knockouts there have been many advances in the techniques
available to developmental biologists to study gene function in the mouse embryo. Generating
mouse embryos with conditional gene targeting, transgenic overexpression, or chimeric tissues
has shed further light on the signaling involved in tissue specification and embryo patterning. A
further approach is to re-examine these mouse phenotypes using the large number of lineage
specific markers that have been identified since the initial characterization of these phenotypes.
This will be important to further study and understand the signaling pathways regulating the
expression of these new tissue specific markers. This has been performed by several groups, and
the re-analysis of previously characterized mouse knockouts has contributed to our knowledge of
how cell fate and patterning is achieved in the mouse embryo (Kinder et al., 2001a).
Lineage tracing studies indicated that the mouse definitive endoderm originates from the
anterior region of the primitive streak, as shown in Figure 5.1 (Lawson et al., 1986; Lawson and
Pedersen, 1987). Furthermore, the study of mouse mutants lacking various components of the
Nodal signaling pathway has demonstrated an important role for this pathway in endoderm
specification (reviewed in Schier, 2003; Shen, 2007). Current models indicate that graded Nodal
signaling specifies different tissue derivatives in the primitive streak, and specifically, that the
highest levels of Nodal signaling are required for specification of the anterior primitive streak
derivatives – the prechordal plate, anterior definitive endoderm (foregut precursors), node, and
notochord (reviewed in Schier, 2003; Shen, 2007). These studies have also demonstrated that
Nodal plays an important role in specification of the definitive endoderm lineage, and in fact the role for Nodal signaling in definitive endoderm specification is evolutionarily conserved
(reviewed in Schier, 2003; Shen, 2007). However, the specific role of Nodal in regulating formation of the midgut and hindgut has not been possible due to the lack of genetic markers specific for these endoderm subpopulations. The analysis presented in Chapter 4 is the first
150
Figure 5.1 – Current model of definitive endoderm formation during gastrulation in the mouse embryo
During gastrulation cells of the epiblast in the posterior region undergo an epithelial to mesenchymal transformation and form a transient structure called the primitive streak. Epiblast cells move through the streak and emerge as either mesoderm or definitive endoderm. Fate mapping studies have suggested that the definitive endoderm forms from the anterior region of the primitive streak while the paraxial and lateral mesoderm forms from more posterior regions of the streak. The anterior primitive streak is also the location of the embryonic organizer and also contains precursors for the axial mesoderm. Red – mesodermal lineages, yellow – definitive endoderm lineages, green – primitive streak, orange – anterior primitive streak (axial mesoderm and definitive endoderm precursors), blue – ectoderm lineages, white – derivative organs and tissues.
151 evidence clearly demonstrating that the midgut and hindgut definitive endoderm forms in Foxh1
and Foxa2 null embryos, which have previously been shown to lack derivatives of the anterior
primitive streak. This suggests that formation of these definitive endoderm subpopulations (the midgut and hindgut) is independent of the anterior primitive streak. This new model of mouse
definitive endoderm specification is shown in Figure 5.2. Further study will be necessary to
determine if the specification of these subpopulations is completely independent of Nodal signaling and to identify the pathway(s) regulating the specification of these endodermal lineages. Understanding the signals that are involved in specification of the different subpopulations of definitive endoderm will aid in efforts to differentiate endodermal derivatives from embryonic stem cells for therapeutic purposes, as described in Section 5.2.
5.2 A THOROUGH UNDERSTANDING OF SPECIFICATION OF THE DEFINITIVE
ENDODERM IS NECESSARY FOR CLINICAL APPLICATION OF REGENERATIVE
MEDICINE
Differentiation of replacement cell types from embryonic stem cells (ESCs) is a popular area of research in regenerative medicine. A number of endoderm derived organs such as the liver and pancreas are potential targets for cell-based therapy, and so there is great interest in understanding the pathways that regulate the induction and specification of this germ layer
(Zaret, 2008). The most successful ESC differentiation strategies are those that recapitulate normal development. As described in this thesis, the Nodal signaling pathway plays a critical role in regulation of definitive endoderm formation in the mouse embryo. This knowledge has been applied to the differentiation of both human and mouse embryonic stem cells. Using
Activin, a TGFβ superfamily ligand that signals via the same receptors and intracellular effectors as Nodal, these groups generated cells from ESCs that expressed several markers of the
152
Figure 5.2 A novel model of definitive endoderm specification in the mouse embryo
Markers specific for the midgut and hindgut definitive endoderm lineages are still expressed in mouse mutants that lack the anterior primitive streak population, suggesting that formation of these subpopulations of the definitive endoderm is independent of the anterior primitive streak. Red – mesodermal lineages, yellow – definitive endoderm lineages, green – primitive streak, orange – anterior primitive streak (axial mesoderm and definitive endoderm precursors), blue – ectoderm lineages, white – derivative organs and tissues.
153 definitive endoderm (D'Amour et al., 2005; Yasunaga et al., 2005). More recent studies have
further differentiated these cells into insulin producing β-cells and hepatocytes (D'Amour et al.,
2006; Kroon et al., 2008; Shiraki et al., 2008; Yamamoto et al., 2005). Despite these successes,
much work has yet to be done before these cells will be of use in a clinical setting. One potential
issue is that many of the markers currently in use to identify definitive endoderm cells within a
differentiated culture are expressed in both the definitive and visceral endoderm in vivo.
Therefore, there could be contamination of these cultures with extraembryonic endoderm.
Additionally, while there has been great success in producing differentiated cell types, the
efficiency is often low, with relatively small numbers of cells being produced. Studies, such as
the one described in this thesis, are essential to the field of regenerative medicine for several
reasons. First, gene expression profiling of the endoderm (and characterization of novel endoderm markers such as described in Chapter 3) will allow more accurate monitoring of ESC derived definitive endoderm enabling investigators to distinguish between definitive and visceral
endoderm. Novel definitive endoderm markers expressed on the cell surface will also allow more
efficient purification of ESC derived definitive endoderm using techniques such as fluorescent activated cell sorting (FACS). Second, understanding the signaling pathway(s) regulating the differentiation of the various subpopulations/regions of the definitive endoderm may aid in the development of more directed differentiation strategies. For example, the work presented in
Chapter 4 demonstrates that formation of the midgut and hindgut definitive endoderm is independent of Nodal signaling. Therefore, Activin mediated differentiation strategies may not be the appropriate approach if the desired cell type is derived from the midgut or hindgut population. Lastly, recent work has demonstrated in the mouse that adult cells can be
“reprogrammed” to a new cell fate by overexpression of genes required for the differentiation of
that cell type in the embryo (Zhou et al., 2008). Therefore, it is crucial to understand the
154 signaling pathways and factors involved in specification of the various lineages of the definitive endoderm.
155 5.3 REFERENCES
Ang, S. L., Wierda, A., Wong, D., Stevens, K. A., Cascio, S., Rossant, J., and Zaret, K. S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development 119, 1301-15. Beddington, R. S. (1994). Induction of a second neural axis by the mouse node. Development 120, 613-20. Belo, J. A., Bouwmeester, T., Leyns, L., Kertesz, N., Gallo, M., Follettie, M., and De Robertis, E. M. (1997). Cerberus-like is a secreted factor with neutralizing activity expressed in the anterior primitive endoderm of the mouse gastrula. Mech Dev 68, 45-57. Biben, C., Stanley, E., Fabri, L., Kotecha, S., Rhinn, M., Drinkwater, C., Lah, M., Wang, C. C., Nash, A., Hilton, D., Ang, S. L., Mohun, T., and Harvey, R. P. (1998). Murine cerberus homologue mCer-1: a candidate anterior patterning molecule. Dev Biol 194, 135-51. Camus, A., and Tam, P. P. (1999). The organizer of the gastrulating mouse embryo. Curr Top Dev Biol 45, 117-53. D'Amour, K. A., Agulnick, A. D., Eliazer, S., Kelly, O. G., Kroon, E., and Baetge, E. E. (2005). Efficient differentiation of human embryonic stem cells to definitive endoderm. Nat Biotechnol 23, 1534-41. D'Amour, K. A., Bang, A. G., Eliazer, S., Kelly, O. G., Agulnick, A. D., Smart, N. G., Moorman, M. A., Kroon, E., Carpenter, M. K., and Baetge, E. E. (2006). Production of pancreatic hormone-expressing endocrine cells from human embryonic stem cells. Nat Biotechnol 24, 1392-401. Hoodless, P. A., Pye, M., Chazaud, C., Labbe, E., Attisano, L., Rossant, J., and Wrana, J. L. (2001). FoxH1 (Fast) functions to specify the anterior primitive streak in the mouse. Genes Dev 15, 1257-71. Hou, J., Charters, A. M., Lee, S. C., Zhao, Y., Wu, M. K., Jones, S. J., Marra, M. A., and Hoodless, P. A. (2007). A systematic screen for genes expressed in definitive endoderm by Serial Analysis of Gene Expression (SAGE). BMC Dev Biol 7, 92. Kanai-Azuma, M., Kanai, Y., Gad, J. M., Tajima, Y., Taya, C., Kurohmaru, M., Sanai, Y., Yonekawa, H., Yazaki, K., Tam, P. P., and Hayashi, Y. (2002). Depletion of definitive gut endoderm in Sox17-null mutant mice. Development 129, 2367-79. Kinder, S. J., Tsang, T. E., Ang, S. L., Behringer, R. R., and Tam, P. P. (2001a). Defects of the body plan of mutant embryos lacking Lim1, Otx2 or Hnf3beta activity. Int J Dev Biol 45, 347-55. Kinder, S. J., Tsang, T. E., Wakamiya, M., Sasaki, H., Behringer, R. R., Nagy, A., and Tam, P. P. (2001b). The organizer of the mouse gastrula is composed of a dynamic population of progenitor cells for the axial mesoderm. Development 128, 3623-34. Kroon, E., Martinson, L. A., Kadoya, K., Bang, A. G., Kelly, O. G., Eliazer, S., Young, H., Richardson, M., Smart, N. G., Cunningham, J., Agulnick, A. D., D'Amour, K. A., Carpenter, M. K., and Baetge, E. E. (2008). Pancreatic endoderm derived from human embryonic stem cells generates glucose-responsive insulin-secreting cells in vivo. Nat Biotechnol 26, 443-52. Lawson, K. A., Meneses, J. J., and Pedersen, R. A. (1986). Cell fate and cell lineage in the endoderm of the presomite mouse embryo, studied with an intracellular tracer. Dev Biol 115, 325-39.
156 Lawson, K. A., and Pedersen, R. A. (1987). Cell fate, morphogenetic movement and population kinetics of embryonic endoderm at the time of germ layer formation in the mouse. Development 101, 627-52. McKnight, K. D., Hou, J., and Hoodless, P. A. (2007). Dynamic expression of Thyrotropin- releasing hormone in the mouse definitive endoderm. Dev Dyn 236, 2909-17. Monaghan, A. P., Kaestner, K. H., Grau, E., and Schutz, G. (1993). Postimplantation expression patterns indicate a role for the mouse forkhead/HNF-3 alpha, beta and gamma genes in determination of the definitive endoderm, chordamesoderm and neuroectoderm. Development 119, 567-78. Norris, D. P., Brennan, J., Bikoff, E. K., and Robertson, E. J. (2002). The Foxh1-dependent autoregulatory enhancer controls the level of Nodal signals in the mouse embryo. Development 129, 3455-68. Schier, A. F. (2003). Nodal signaling in vertebrate development. Annu Rev Cell Dev Biol 19, 589-621. Shen, M. M. (2007). Nodal signaling: developmental roles and regulation. Development 134, 1023-34. Shiraki, N., Yoshida, T., Araki, K., Umezawa, A., Higuchi, Y., Goto, H., Kume, K., and Kume, S. (2008). Guided differentiation of embryonic stem cells into Pdx1-expressing regional- specific definitive endoderm. Stem Cells 26, 874-85. Wederell, E. D., Bilenky, M., Cullum, R., Thiessen, N., Dagpinar, M., Delaney, A., Varhol, R., Zhao, Y., Zeng, T., Bernier, B., Ingham, M., Hirst, M., Robertson, G., Marra, M. A., Jones, S., and Hoodless, P. A. (2008). Global analysis of in vivo Foxa2-binding sites in mouse adult liver using massively parallel sequencing. Nucleic Acids Res 36, 4549-64. Yamamoto, M., Meno, C., Sakai, Y., Shiratori, H., Mochida, K., Ikawa, Y., Saijoh, Y., and Hamada, H. (2001). The transcription factor FoxH1 (FAST) mediates Nodal signaling during anterior-posterior patterning and node formation in the mouse. Genes Dev 15, 1242-56. Yamamoto, Y., Teratani, T., Yamamoto, H., Quinn, G., Murata, S., Ikeda, R., Kinoshita, K., Matsubara, K., Kato, T., and Ochiya, T. (2005). Recapitulation of in vivo gene expression during hepatic differentiation from murine embryonic stem cells. Hepatology 42, 558-67. Yasunaga, M., Tada, S., Torikai-Nishikawa, S., Nakano, Y., Okada, M., Jakt, L. M., Nishikawa, S., Chiba, T., Era, T., and Nishikawa, S. (2005). Induction and monitoring of definitive and visceral endoderm differentiation of mouse ES cells. Nat Biotechnol 23, 1542-50. Zaret, K. S. (2008). Genetic programming of liver and pancreas progenitors: lessons for stem- cell differentiation. Nat Rev Genet 9, 329-40. Zhou, Q., Brown, J., Kanarek, A., Rajagopal, J., and Melton, D. A. (2008). In vivo reprogramming of adult pancreatic exocrine cells to beta-cells. Nature.
157 APPENDIX
The following are the animal care certificates required during this thesis.
ANIMAL CARE CERTIFICATE A02-1852
158
159 ANIMAL CARE CERTIFICATE A02-1854
160
161