~ 1.9:.”

v

s . J . ‘ flu... 5.
I, inn" Ma .gnuuhrwutt

a? a... pthwamm

.3
a

1

.fi. . ‘ .Hfibu . ,..

>

3‘?“

TA 1: I.)
t. .4. r a!

 

4 . 1 ‘ ‘3
iw- 13...... k 4..

pl :‘ldll

. .wt‘ . . 6r... .- .lc\ i ..‘ .
wwWfiflwmamfifi7 ; , A _ ‘ . , h «E: 35w ,

..

 

fihmsms

, f0 6 9

This is to certify that the

thesis entitled

RETINOIC ACID REGULATION OF CARDIAC INFLOW TRACT
DEVELOPMENT IN THE AVIAN EMBRYO IS MEDIATED VIA
A TGFbetaZ SIGNALING PATHWAY.

presented by
CHRISTOPHER S. CARLSON

has been accepted towards fulfillment
of the requirements for

MS , HUMAN NUTRITION
degree in

///2/7 // k/k/

/ Major fifessor

 

 

 

Date 8/20/2002

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

MAR 1 7 2005 I.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DaIeDue.p65~p.15

RETINOIC ACID REGULATION OF CARDIAC INFLOW TRACT
DEVELOPMENT IN THE AVIAN EMBRYO IS MEDIATED VIA A TGFBZ
SIGNALING PATHWAY
By

Christopher S. Carlson

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Food Science and Human Nutrition

2002

ABSTRACT
RETINOIC ACID REGULATION OF CARDIAC INFLOW TRACT
DEVELOPMENT IN THE AVIAN EMBRYO IS MEDIATED VIA A TGFBZ
SIGNALING PATHWAY
By

Christopher S. Carlson

Vitamin A is essential for normal embryonic cardiac morphogenesis. The
vitamin A-deficient (VAD) phenotype in the avian embryo includes an abnormal
heart-tube closed at the inflow tracts and the absence of omphalomesenteric
veins that normally connect the embryo to the extraembryonic circulatory
system. In VAD embryos, there were decreased presumptive endothelial cells in
the heart-forming region and in the area pellucida from the beginning of
differentiation (1 58), through the initiation of heart beating (10/11 ss). TGFBZ,
a multi-functional growth regulator and TGFBZ mRNA were increased in the
posterior heart-forming region of VAD embryos. Significantly, TGFBZ
transcripts were increased in the inflow tracts and the adjacent head mesenchyme
prior to inflow tract closure in the VAD embryo. The administration of TGFBZ
antisense oligonucleotides to the VAD embryo, specifically decreased TGFBZ
and TBRII transcripts in the inflow tracts, which remained open, and increased
presumptive endothelial cells were observed in the inflow tracts and the
surrounding area pellucida. Our results indicate that TGFBZ is a component of

the retinoid-mediated cardiogenic pathway, specific for inflow tract

development.

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Maija Zile for her constant
encouragement and support. I want to thank her not only for her guidance, but
also for her willingness to let me think independently. In addition, I would like
to thank my committee members Dr. Maurice Bennink, Dr. Norman Hord, and
Dr. Joseph Schroeder for their valuable wisdom and precious time. I want to
express my gratitude for my lab mates, Dr. Jian Cui and Dr. Mahmoud Romeih
for their patience, insight, and friendship.

To my parents who supported me and allowed me to be the person that I
am, I give you my deepest respect and admiration. I would especially like to
thank my wife Tausha for her constant love and devotion and for showing me the
best that the world has to offer.

This research was supported by an NIH grant.

iii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................. Vi
LIST OF FIGURES ........................................................................... vii
KEY TO ABBREVIATIONS ................................................................. x
CHAPTER 1: INTRODUCTION ............................................................ 1
1.1. Vitamin A: historical background ............................................. 1
1.2. Vitamin A and human nutrition ................................................ 3
1.3. Vitamin A deficiency ............................................................. 7
1.4. Models for studying Vitamin A function in embryonic
development ....................................................................... 10
1.4.1. Mammalian models .................................................. 11
1.4.2. Avian embryo “retinoid ligand knockout” model .............. 13
1.5. Avian cardiovascular development .......................................... 14
1.5.1. Avian cardiogenesis .................................................. 15
1.5.2. Avian vascular development ....................................... 20
1.5.3. Avian inflow tract development ................................... 22
1.5.4. Retinoids and cardiovascular development ...................... 23
1.6. Retinoid-regulated molecular mechanisms of cardiovascular
development ....................................................................... 25
1.6.1. Cardiogenic genes regulated by retinoic acid ................... 25
1.6.2. Vasculogenic genes and Vitamin A status ........................ 27
1.6.3. Transforming growth factor beta .................................. 29
CHAPTER 2: RATIONALE AND OBJECTIVES ...................................... 33
CHAPTER 3: MATERIALS AND METHODS ......................................... 38
3.1 Animals and diets ................................................................ 38
3.2 The quail embryo retinoid ligand knockout model ....................... 40
3.3 Dissection ......................................................................... 40
3.4 Sectioning ........................................................................ 41
3.5 In ovo microinjection .......................................................... 42
3.5.1 Rescue of VAD embryos with Vitamin A-active
compounds ............................................................. 42
3.5.2 Blocking of development with anti-all-trans-retinoic
acid monoclonal antibody ........................................... 44
3.6 Immunohistochemical localization ofendothelial cells ................. 45
3.7 Immunohistochemical localization of TGFBZ............ ................. 46
3.8 Analysis ofTGFBZ mRNA expression by in situ hybridization ....... 47
3.8.1 Recovery of plasmid ................................................. 47

iv

3.8.2 Transfection of E. coli with plasmid containing

TGFBZ cDNA .......................................................... 47
3.8.3 Purification of plasmid .............................................. 48
3.8.4 Sub-cloning of TGFBZ cDNA into pBSK' ...................... 49
3.8.5 Synthesis of TGFBZ riboprobe ..................................... 51
3.8.6 Analysis ofTGFB2 expression by whole-mount
in situ hybridization ................................................. 53
3.8.7 Cellular analysis of TGFBZ expression by cross-
sectional examination ................................................ 54
3.9 Analysis of TBRII expression by in situ hybridization .................. 55
3.9.1 Synthesis of TBRII riboprobes .................................... 55
3.9.2 Analysis of TBRII expression by whole-mount in situ
hybridization .......................................................... 56
3.9.3 Cellular analysis of TBRII expression by cross-
sectional examination ............................................... 56
3.10 Antisense oligonucleotide blocking of TGFBZ ............................ 57
CHAPTER 4: RESULTS ..................................................................... 58
4.1 VAD embryos display abnormal IFT development during early
cardiac morphogenesis ......................................................... 58
4.2 QH-l monoclonal antibody is a marker for presumptive
endothelial cells in normal and VAD embryos ........................... 64
4.3 Enhanced TGFBZ transcripts and protein levels precede
abnormal IFT development in VAD embryos ............................. 69
4.4 TBRII transcripts are elevated prior to abnormal IFT
morphogenesis in VAD embryos ............................................ 75
4.5 TGFB2 mediates IFT development in quail embryos ..................... 78
CHAPTER 5: DISCUSSION ................................................................ 82
CHAPTER 6: FUTURE WORK ............................................................ 92
APPENDIX ...................................................................................... 94

Appendix 1. Additional procedures and composition of solutions.........95

REFRENCES .................................................................................... 97

LIST OF TABLES

Table 1. Comparative chronology of cardiovascular development ............... 18

Table 2. Composition of semi-purified diets ......................................... 39

Vi

LIST OF FIGURES

Figure 1. Schematic ofthe location of cardiogenic progenitors in the

avian embryo. The relative anterior-to-posterior positions of precardiac

cells in the primitive streak (HH stage 3) are retained in the lateral plate

(as shown in HH stage 5 and 7), and in the heart tube at HH stage 10

(Fishman and Chien, 1997). Cardiogenic progenitors are shown in black.

All Views are ventral ............................................................................ 17

Figure 2. Cross-sections through the avian embryonic HFR. Cardiogenic
progenitors in the mesoderm respond to inductive signals from the endo-

derm. By HH stage 9-10 the endocardial tubes fuse, forming one heart

tube, surrounded by epimyocardial and epicardial cells. Adapted from the

Atlas of Descriptive Embryology, 4‘h Edition ............................................ 19

Figure 3. Schematic diagram ofthe avian IFT at HHlO. The vascular
endocardium (EN), surrounded by the epimyocardium (EP), connects to

the extraembryonal vascular networks at the posterior region of the IFT

through the omphalomesenteric veins (OMV). The EP forms the lateral
boundary of the IFT and the anterior intestinal portal (AIP) forms the

medial boundary. Adapted from Atlas of Descriptive Embryology, Fourth
Edition, 1986 .................................................................................... 22

Figure 4. Signaling by TGFB through its serine/threonine kinase receptors

and Smad proteins. The active pro-peptide dimer, cleaved from LAP and

LTBP by proteolysis, binds to a heterotetramer of the type II and type I

TGFB receptors R-lI and R-I, respectively. Although not essential for

signal transduction, the type III TGFB receptor (R-II) can facilitate the

binding of active ligands to R-II. I—Smads inhibit signaling by
R-SmadszCo-Smad complexes by competitive inhibition.

R-SmadzCo-Smad complexes bind to Smad responsive elements (SRE),

DNA binding proteins, and coactivators to elicit transcriptional activity

Adapted from Miyazono, 2000 ............................................................. 31

Figure 5. Heart development in VAD quail embryos. There are no
morphological differences observed in VAD and normal quail embryos at
4/5 ss. The bilateral heart primordia in the heart-forming region (HFR)

line the anterior intestinal portal (AIP) and extend caudally as development
continues. The heart-tubes fuse at the midline at 9 ss in the normal embryo,
delayed to 10/1 1 ss in the VAD embryo. Heart (Ht) orientation is random-
ized in the VAD embryo (shown on left) and the inflow tracts (IFT) are
closed by 12 ss. The Ht is a ballooned single-chamber at 2.5 d, in contrast
to the two-chambered heart of normal embryos. VAD embryos lack
omphalomesenteric veins (OMV) and omphalomesenteric arteries (OMA)

vii

to connect the embryo to the extraembryonic circulation. Injection of VAD
embryos by 4/5 55 with biologically active retinoids, completely rescues
cardiovascular development. If RA function is blocked in the normal

embryo by injecting anti-atRA MoAb, cardiovascular abnormalities are

observed consistent with the VAD phenotype ........................................... 62

Figure 6. Composite pictures of normal and VAD embryos from 4/5 35 to
10/11 ss exposed to QH-l antibody (ventral views). Endothelial cells as
well as endothelial precursors are fluorescent (yellow). Angioblasts
differentiate from mesoderm in the heart-forming region (HFR) and conger-
gate along the lateral edge of the anterior intestinal portal (AIP).
Simultaneously, angioblasts in the extraembryonic area differentiate from
extraembryonic mesoderm and form vascular networks (arrow). Extra-
embryonic vascular networks connect with the heart (Ht) at the
omphalomesenteric veins (OMV) and with the dorsal aorta lining the
somites at the omphalomesenteric arteries (OMA). VAD embryos have
fewer angioblasts, most notable in the IFT lining the AIP; the angioblasts
are poorly organized and do not form OMV or OMA. The administration
of atRA to VAD embryos completely rescues vascular development.
Treating normal embryos with anti-atRA MoAb interferes with vascular
patterning and results in a lack of connection of the heart to the extra-
embryonic circulation at the OMV ......................................................... 68

Figure 7. Immunolocalization of TGFBZ protein in normal and VAD quail
embryos. At 3/4 55 TGFBZ protein is ectopically expressed in the extra-
embryonic area. TGFBZ-specific immunofluorescence (red) is observed in
the neural tubes (NT) of both normal and VAD embryos at comparable
levels. TGFBZ-specific immunofluorescence is increased in the posterior
heart-forming region (HFR) and caudal to the Hensen’s node (Hn) in the
VAD embryo at 3/4 ss. At 10/11 ss TGFBZ is localized in the periphery of
the developing inflow tracts (IFT) in the normal embryo and the absence of
fluorescence at the site of the OMVs indicates that the IFT are open, while
in the VAD embryo TGFBZ-specific fluorescence is increased in the IFT
and the IFT are narrowed at the posterior ends. All views are ventral ............. 72

Figure 8. Expression patterns of TGFBZ in whole-mounts of normal and
VAD quail embryos, as well as in transverse sections through the inflow
tract (IFT). Note the increased expression caudal to the Hensen’s node (Hn)
and in the heart-forming region (HFR) lining the anterior intestinal portal
(AIP) ofthe VAD embryo beginning at the 4/5 ss. Increased expression in
the IFT continues through 10/11 ss. The expression of TGFBZ in the
neural tubes (NT) of the VAD embryo preceded the observed expression in
the normal embryo. As noted by transverse sectioning, at the level noted
by the black horizontal lines bilaterally, TGFBZ is expressed in all cell
layers (6/7 35). In VAD embryos, increased TGFBZ trascripts were
observed in the head mesenchyme (HM); cells expressing TGFBZ filled the

viii

IFT between the epimyocardium (EP) and the endocardium (EN).

Administration of atRA to VAD embryos reduced expression ofTGFBZ in

the HM and the IFT, and the IFT were open. Treating normal embryos

with anti-atRA MoAb resulted in increased TGFBZ expression in the HM

and the IFT, and the IFT were closed. All views are ventral ......................... 74

Figure 9. Expression patterns of TBRII in whole-mounts of normal and
VAD quail embryos, and in transverse sections through the inflow tract
(IFT). TBRII expression patterns correspond with TGFBZ expression
patterns in both normal and VAD embryos. VAD embryos have increased
TBRII expression beginning at 4/5 55 and continuing through 10/11 ss.
TBRII expression is increased in the IFT of VAD embryos beginning at
7/8 ss, prior to IFT closure. Transverse sectioning, at the level noted by
the black horizontal lines bilaterally, show a narrowing of the lumen
between the EP and the EN in VAD embryos, and increased TBRII
expression in the HM. Administration of atRA to VAD embryos reduced
expression of TBRII in the HM and in the IFT, and the IFT were open.
Treating normal embryos with anti-atRA MoAb resulted in increased
TBRII expression in the HM and the IFT, and the IFT were closed. All
views ventral. Anterior intestinal portal, AIP; endocardium, EN;
epimyocardium, EP; head mesenchyme, HM; heart—forming region, HFR;
Hensen’s node, Hn; neural tubes, NT. .................................................... 77

Figure 10. Blocking TGFBZ transcripts in the VAD embryo by injecting
antisense oligonucleotides to TGFBZ. The heart (Ht) of VAD injected with
TGFBZ antisense oligonucleotides was unaltered. Gross morphological
examination as well as transverse sectioning at the level ofthe black lines
located bilaterally, showed that the inflow tracts (IFT) remained open.
VAD embryos injected with TGFBZ antisense oligonucleotides,
immunolabeled with QH-l, showed increased presumptive endothelial
cells, specifically in the IFT, and the presence of omphalomesenteric veins
(OMV). Vascular networks (arrow) in the extraembryonic area were better
organized than in the VAD embryo; there were thin strands of presumptive
endothelial cells extending from the extraembryonic area to the OMV;
however, the vascular networks were not as organized, or robust as in
normal embryos. Expression of TGFBZ transcripts was specifically
decreased in the IFT when VAD embryos were treated with TGFBZ anti-
sense oligonucleotides. Furthermore, treatment of VAD embryos with
TGFBZ antisense oligonucleotides had the secondary effect of decreasing
TBRII transcripts in the IFT to levels comparable with the normal embryo.
All views ventral ............................................................................... 80

ix

Ab

atRA

atROH

BLAST

CRABP

ECM

EN

EP

FITC

FN

HFR

HH

HM

Hn

IFT

LAP

LRAT

LTBP

MoAb

NT

PFA

KEY TO ABBREVIATIONS

anfibody

all-trans-retinoic acid
all-trans-retinol

Basic Local Alignment Search Tool
cellular retinoic acid binding protein
extracellular matrix

endocardium

epimyocardium

fluorescein isothiocyanate
fibronectin

heart-forming region

Hamburger and Hamilton (stages of avian embryonic development)
head mesenchyme

Hensen’s node

inflow tract

latency associated protein (TGFB)
lecithinzretinol acyltransferase
latent TGFB binding protein
monoclonal antibody

neuraltube

paraformaldehyde

PBS
RA
RAE
RAR
RARE
RE
RBP
RXR

SS

TBR
TGFB
TRITC
TTR
VAD

VEGF

phosphate buffered saline
retinoic acid

retinol activity equivalents
retinoic acid receptor

retinoic acid responsive element
retinyl ester

retinol binding protein

retinoid X-receptor

somite stage

TGFB receptor

transforming growth factor beta
tetramethylrhodamine isothiocyanate
transthyretin

Vitamin A-deficient

vascular endothelial growth factor

xi

CHAPTER 1

INTRODUCTION

1.1 Vitamin A: historical background

Night blindness was a recognized disease entity in ancient Egypt and it
was suggested that a component in food was essential for night vision (McLaren,
1980). McCollum and Davis (1913) discovered a fat-soluble component of food
that was capable of promoting growth in rats. This substance was initially
termed “Fat Soluble A” and was later named Vitamin A. Early researchers
focused on physiology and metabolism of vitamin A. Wald (1934) and Morton
(1944) determined that the chromophore of the Visual pigment is retinaldehyde.
These early studies elucidated the role of Vitamin A in vision (Wald, 1968) and
confirmed the ancient Egyptian beliefs about essential components in food.

In 1937 Holmes and Corbet (1937) crystallized pure retinol from fish
liver. Later, Arens and van Dorp (1946) and Isler and his associates (Isler et al.,
1947) succeeded in achieving the chemical synthesis of pure retinoic acid and
retinol. These discoveries provided the groundwork on which further research
was based to study the function, metabolism, regulation, and pharmacological
use of vitamin A.

The importance of vitamin A throughout the life cycle has been well
established. The requirement for vitamin A begins early in normal embryonic
development (Wolf, 1984; Hofmann and Eichele, 1994) and continues in the

adult organism (Moore, 1957; DeLuca, 1991; Gudas et al., 1994; DeLuca et al.,

1997). Although it was clear that vitamin A was essential, the mechanism of
action used in Vision could not explain the vast biological processes that were
regulated by Vitamin A. Many researchers focused on the molecular mechanism
of action of Vitamin A and these mechanisms are still under investigation. The
regulation at the gene level attributed to vitamin A is due to all-trans-retinoic
acid (atRA), the transcriptionally active form of vitamin A and a recognized
regulator of cell division and differentiation in tissues of ectodermal,
endodermal, and mesodermal origin (Roberts and Sporn, 1984; Wolf, 1984;
Gudas et al., 1994). Recent advances in the understanding ofthe mechanisms of
action of Vitamin A can be attributed to the discovery of nuclear receptors for
retinoic acid (RARs) and their nuclear receptor partners, the retinoid X-receptors
(RXRs)(Mangelsdorf et al., 1995; Pfahl and Chytil, 1996). In light ofthe
discoveries about the molecular mechanism, the term “retinoids” has recently
been redefined to refer to any substance that exerts Vitamin A function via
retinoic acid nuclear receptors, eliciting a biological response (Sporn and
Roberts, 1994).

Vitamin A and more generally, the retinoids, are capable of producing
alterations in the expression ofa diverse group of genes (DeLuca, 1991; Gudas et
al., 1994). The discovery ofthe retinoid specific nuclear receptors provided an
explanation as to how vitamin A controls such a large network of gene activation
processes. Genes that have retinoic acid response elements (RARE) in their

promoter region are affected by Vitamin A status and are involved in diverse

biological functions; however, their functions are interconnected (DeLuca, 1991;
Mangelsdorf et al., 1995; Mangelsdorf et al., 1996).

Inadequate maternal Vitamin A during pregnancy results in fetal death and
congenital malformations (Mason, 1935; Hale, 1937). Vitamin A deficiency
affects several target tissues including the heart, the ocular tissues, and the
circulatory, urogenital, respiratory and central nervous systems (Wilson et al.,
1953; Thompson, 1969; Zile, 1998; Gale etal., 1999; Maden, 1999; Zile, 1999;
Ross et al., 2000). At the other end of the spectrum, excess Vitamin A causes
congenital malformations during embryogenesis and is hepatotoxic in adult
animals (Nau et al., 1998). The targets ofthe teratogenic effects of excess
retinoids have been demonstrated to be the heart, skull, skeleton, limbs, central
nervous system, brain, eyes, and craniofacial structures (Moore, 1957; Morriss
and Steele, 1972; Shenefelt, 1972; Brockes, 1989; Kochhar et al., 1993; Rosa,
1993; Kochhar and Christian, 1997; Nau et al., 1998;). There is considerable
overlap in the targets and the malformation associated with vitamin A deficiency
and with exposure to excess retinoids. These findings suggest that there are
common targets in deficiency and in excess and that retinoid regulation is

required for proper development and health.

1.2 Vitamin A and human nutrition
As elucidated in the previous section, Vitamin A is an essential nutrient.
Vitamin A enters the body through consumption of foods that contain preformed

dietary Vitamin A, or provitamin A carotenoids. The absorption, metabolism and

transport of retinoids has most recently been reviewed by Blaner et a1. (1999).
Retinoids are converted to retinol (ROH) and solubilized into mixed-micellar
solution before efficient absorption can occur at the brush boarder of the small
intestine. ROH is converted to retinyl ester (RE), through the action ofthe
intestinal lecithinzretinol acyltransferase (LRAT) for packaging in nascent
chylomicrons (Quick and Ong, 1990; Ong et al., 1994). Lipoprotein lipase
catalyzes lipolysis of nascent chylomicrons in the lymphatic system, giving rise
to chylomicron remnants, which enter the circulation and are transported to the
liver for storage. RE of chylomicron remnants are first taken up by parenchymal
cells ofthe liver and are then transferred to stellate cells for storage. The stellate
cells contain enzymes that can hydrolyze RE to ROH when ROH is needed.
Subsequently in hepatocytes, ROH is bound to the retinol binding protein (RBP)-
transthyretin (TTR) complex for transport in the plasma to the target tissues.
ROH is taken-up by the target cells. The exact mechanism for cellular uptake of
retinol is still unclear. It is possible that there is more than one mechanism for
cellular-uptake, which may be mediated by the cell type and retinoid status in the
target cells. Inside the cell retinol can be metabolized; specific proteins bind
retinol and direct it to retinoid-metabolizing enzymes (Napoli, 2000). These
proteins act as chaperones helping retinol to be metabolically converted to
retinoic acid (RA), the transcriptionally active form of Vitamin A (McCaffery
and Drager, 1994; Chen et al., 1995; Niederreither et al., 1997; Moss et al.,

1998). RA is transported to the nucleus where it acts via its nuclear receptors,

the RARs and RXRs, to control transcription of a diverse group of genes (Gudas
etaL,1994)

A small percentage of dietary provitamin A and preformed Vitamin A is
converted to RA in the intestine. RA is absorbed in the intestine Via the portal
system. In circulation it is bound to serum albumin (Blaner and Olson, 1994)
and is delivered to target cells, where uncharged RA rapidly and spontaneously
crosses the cell membrane (Noy, 1992). Once inside the cell, RA binds to either
cellular retinoic acid-binding protein I or II (CRABP) in the cytoplasm, which
acts as a chaperone, protecting the molecule until it is used in the nucleus (Chytil
and Ong, 1983; Napoli et al., 1991). While there is metabolic interconversion
between the reduced forms of Vitamin A, RA cannot be reduced and converted to
ROH.

Vitamin A in the diet comes from preformed retinoids as well as
provitamin A carotenoids. Preformed Vitamin A is present in meat and dairy
products, the best source being liver. Many fish liver oils may contain up to 300
ug of vitamin A per gram of oil (Moore, 1957; Lui and Roels, 1980). Unlike
preformed retinoids, provitamin A carotenoids are found in plant sources. There
are more than 600 carotenoids found in nature; however, only approximately 50
can be converted to vitamin A (Blaner and Olson, 1994). Carotenoids are widely
distributed in colored fruits and vegetables. Carrots are the major dietary source
ofa- and B-carotene (Chug-Ahuja et al., 1993). The Continuing Survey of Food
Intakes by Individuals (CFSII) in 1994-1996 reported that grains and vegetables

were the major source of vitamin A in the diet (approximately 55 percent),

followed by dairy and meat products (approximately 30 percent). The
bioconversion of carotenoids and the associated ROH activity is still under
scrutiny (Hume and Krebs, 1949; Torronen et al., 1996; Castenmiller and West,
1998; Rock et al., 1998; Van den Berg and van Vliet, 1998; Boileau et al., 1999).
Retinol Equivalent units were implemented to provide a common unit of
measurement for preformed retinoids and carotenoids based on observed ROH
activity (NRC, 1980; 1989). However, the ratios used to determine Retinol
Equivalents were not universally accepted, and Retinol Activity Equivalents
(RAE) replaced Retinol Equivalents in the 2001 Dietary Reference Intakes (DRI)
(NRC, 2001). RAE, like their predecessors, the Retinol Equivalents, provide
ratios to convert provitamin A carotenoids into a common unit of measure based
on ROH activity of carotenoids. Using RAE, 12 pg B-carotene is considered to
have the activity of 1 pg of ROH (Hume and Krebs, 1949; Chug-Ahuja et al.,
1993; Parker et al., 1999). Other dietary carotenoids have a conversion ratio of
24:1 (pg carotenoid: pg ROH) based on the observation that the activity of other
carotenoids is approximately halfthat of B-carotene (Deuel et al., 1949;
Bauernfeind, 1972). The vitamin A activity of provitamin A carotenoids using
RAE is half the Vitamin A activity when using Retinol Equivalents and can have
a substantial impact on assessing vitamin A nutriture since a large portion of
vitamin A in the diet is obtained through the consumption of carotenoid rich
fruits and vegetables.

Using RAE for conversion of carotenoids to Vitamin A, the DRI for

vitamin A is set at 900 pg for men and 700 pg for women (NRC, 2001).

Absorption of Vitamin A in the small intestine is affected by dietary fat,
infection, food matrix and processing, as well as nutrient-nutrient interactions
with iron, zinc, and alcohol (Dorea and Olson, 1986; Suhamo et al., 1993; Jalal
et al., 1998; Van het Hofet al., 1998; Boileau et al., 1999; Wang, 1999; Tang et
al., 2000).. These factors and Others need to be considered when making

individual recommendations and assessing nutriture based on consumption.

1.3 Vitamin A deficiency

There is an intense interest in Vitamin A in biology and in medicine, due
to the essentiality of vitamin A throughout life. It is well established that
maternal insufficiency of vitamin A results in death ofthe fetus as well as
congenital malformations, including those associated with the heart (Wolf, 1984;
Blomhoff, 1994; Zile, 1998; Ross, 2000). Vitamin A is also required throughout
life for maintenance of differentiation, proliferation, and apoptosis (Gudas et al.,
1994), as well as playing a critical role in reproduction and vision (Wald, 1968).

Xerophthalmia is synonymous with all ofthe clinical signs and symptoms
that affect the eye in vitamin A deficiency and is the most common clinical
effect of inadequate vitamin A intake (McLaren and Frigg, 2001). Some of the
common signs and symptoms of xerophthalmia are: night blindness, conjunctival
xerosis, and Bitot’s spot. There are an estimated 3 to 10 million children who
become xerophthalmic and 250,000 to 500,000 who become blind annually,
mostly occurring in developing countries (WHO, 1995; Sommer and West,

1996;). Xerophthalmia is the most common cause of blindness in the world. In

addition to the cases of clinical deficiency, it is estimated that there are 251
million children with subclinical Vitamin A deficiency (WHO, 1995).
Subclinical Vitamin A deficiency has been associated with increased morbidity
and mortality in children (Beaton et al., 1994; Sommer and West, 1996). Many
children with subclinical Vitamin A deficiency will go undiagnosed, increasing
their risk for morbidity and mortality, due to the inadequacy of many ofthe
health care systems in the developing countries that have a high prevalence of
vitamin A deficiency. This problem is complicated by the fact that the methods
used to assess vitamin A status are not adequate due to their lack of sensitivity
and practicality (Rosales and Ross, 1998; Filteau et al., 2000). It is important to
note that children are not the only ones subject to vitamin A deficiency;
however, children in developing countries are at the greatest risk.

The adverse affects associated with Vitamin A deficiency make
determining Vitamin A status essential, especially when dealing with high-risk
groups. Five states of vitamin A nutriture have been identified for defining
vitamin A status: deficient, marginal, satisfactory, excessive, and toxic (Olson,
1994). Vitamin A is stored in the liver as RE and the amount of vitamin A stored
in the liver is the key determinant when assessing Vitamin A status. Various
methods have been developed to assess Vitamin A status, each with strengths and
weaknesses; however, the methods currently available are insufficient to
properly assess Vitamin A status (NRC, 2001). Assessment of Vitamin A status
by plasma vitamin A and RBP levels is not a valid method, because plasma

levels of vitamin A, in the form of retinol bound to RBP and TTR, are

maintained under tight homeostatic control and are relatively unaffected by
vitamin A status, except when liver stores are very low or very high (McLaren
and Frigg, 2001). In addition to the lack of sensitivity of plasma values to
reflect actual liver reserves, plasma values are also negatively influenced by the
presence of infection and protein-energy malnutrition (Filteau et al., 2000),
rendering them ineffective for measuring Vitamin A status; however, they are
effective for population-based studies. Liver biopsy is the only accurate measure
of Vitamin A status over all five states of nutriture; however, the invasive nature
and the expense of the procedure makes it less than ideal for assessing vitamin A
status, especially with regards to population-based studies. The methods
currently available to test vitamin A status are inadequate, especially with
regards to marginal status. Researchers are working to improve the methods for
determining vitamin A status for all five states of nutriture (Sommer et al., 1983;
Sommer, 1994).

Vitamin A deficiency is most prevalent in developing countries (WHO,
1995) where staple crops make up the majority ofthe diet. Generally, the staple
crops contain a small concentration of carotenoids; however, rice is the main
exception and contains no carotenoids (McLaren and Frigg, 2001). Communities
that are rice-dependent are more vulnerable to Vitamin A deficiency. This is
complicated by the fact that vitamin A is important for the resistance to
infectious disease and the incidence of disease is often high in developing
countries (Beaton et al., 1994 and Sommer and West, 1996). Vitamin A

supplementation has been shown to reduce the risk of mortality associated with

these diseases in developing countries (Humphrey et al., 1996; West et al.,
1999). Controlling infections, providing food fortification, conducting dietary
interventions, and introducing genetically modified plants have been and
continue to be the focus for controlling Vitamin a deficiency. Children under the
age of six and pregnant and lactating women constitute the main vulnerable
groups in communities where Vitamin A deficiency has been identified as a
public health problem; supplementation should be considered for theSe

individuals wherever it is deemed appropriate (McLaren and Frigg, 2001).

1.4 Models for studying vitamin A function in embryonic development
The importance of vitamin A throughout the life cycle has been well
established (Moore, 1957; Wolf, 1984). As early as the 19305, maternal
insufficiency of vitamin A during pregnancy was recognized to result in
incomplete pregnancies, fetal death, and severe congenital malformations (Zile,
1999). These congenital abnormalities have been identified in various species;
however, in humans the effects of vitamin A deficiency are often disguised by
general malnutrition (Gerster, 1997). Multi-nutrient deficiency is a confounding
factor, making it difficult to isolate vitamin A deficiency as the sole cause of an
abnormality. Even though a direct link has not been made between Vitamin A
deficiency and congenital malformations in humans, circumstantial evidence is
available, i.e. premature neonates with low Vitamin A stores are at high risk of

respiratory abnormalities (Shenai et al., 1995), to suggest that abnormalities

10

experienced in humans share some similarity with those described in other
animals.

Similarly, excess retinoids can have a deleterious effect on embryonic
development. The malformations observed due to either vitamin A deficiency or
excess appear to overlap; however, there are examples of selective responses to
only deficiency or excess (Ross et al., 2000). Due to the strong evidence that
human embryonic exposure to excess vitamin A results in congenital
abnormalities (Kochhar et al., 1993; Rosa, 1993), treatment with retinoids is not
permitted during pregnancy (Chan et al., 1996).

The requirement for Vitamin A begins early in embryonic development
(Zile, 2001). As evidenced by the severe malformations associated with
deficiency and excess retinoids, developing embryos and fetuses require a
precisely regulated supply of Vitamin A. Recently, cell and molecular biology
research, along with developmental studies have provided a wealth of
information on the function of Vitamin A in development. However, it was the
discovery of the nuclear receptors for retinoic acid (RARs) and the retinoid X-
receptors (RXRs) that revealed how RA can exert an effect at the gene level

(Mangelsdorf et al., 1995; Pfahl and Chytil, 1996).

1.4.1 Mammalian models
Advances in molecular developmental biology, as well as the discovery of
retinoid nuclear receptors, the RARs and the RXRs, have provided researchers

with new methods to study the genomic role of Vitamin A in mechanisms in

11

development. RA via activating the RAR-RXR hetrodimers, which then binds to
the RARE in promoter regions of target genes, controls diverse, yet
interconnected biological processes (Ross et al., 2000). Transgenic mice with
alterations in retinoid receptor gene structure have been extensively used to
examine the role of vitamin A in development, focusing on the specific function
of different receptors during embryogenesis (Chambon, 1993; Kastner et al.,
1994; Sucov et al., 1994; Boylan et al., 1995; Giguere et al., 1996). Although
retinoid receptor isoforms have unique distribution patterns during embryonic
development (Durston et al., 1997), single isoform knockouts have failed to
produce phenotypic abnormalities (Kastner et al., 1995), due to retinoid receptor
redundancy (Luo et al., 1995). Developmental malformations can arise as a
result of knocking out the entire gene for a retinoid receptor; however,
embryogenesis is well protected by a system that allows most or all functions of
an absent receptor to be substituted by another (Ross et al., 2000). Receptor
knockouts have provided valuable information into the function of Vitamin A in
development; however, they alone cannot provide definitive answers.

Vitamin A-deficient (VAD) animals can be used to study the molecular
mechanisms of retinoid action by regulating the presence of retinoids and
examining gene expression. Rats can be used to target retinoid deficiency to
distinct gestational windows (White and Clagett-Dame, 1996; Wellik et al.,
1997; Smith et al., 1998). Rat embryos made deficient in retinoids during
gestational days 11.5-13.5 exhibit specific cardiac, limb, ocular, and nervous

system deficits, some of which are partially similar to those reported in retinoid

12

receptor knockout mice, while others are novel (Smith et al., 1998). The retinoid
deficient rodent model is useful; however, it is limited to studying later

gestational stages, and there is some uncertainty about the exact vitamin A status
of the embryo, since the embryo depends on the maternal supply for all nutrients,

and the mother must have Vitamin A for her own survival (Zile, 1999).

1.4.2 Avian embryo “retinoid ligand knockout” model

In contrast to the rodent model, the avian embryo model can be used to
study Vitamin A-dependent developmental events early in embryonic
development (Zile, 1998; Zile, 2001). The initial discovery by Thompson et al.
(1969) and subsequent work by Heine et al. (1985) has established that vitamin A
is necessary for the establishment ofthe early avian vasculature. Importantly,
RA fed to adult quail is not transported to the egg and thus completely VAD
embryos can be obtained from RA sufficient adult quails (Dong and Zile, 1995;
Chen et al., 1996). This avian embryo retinoid ligand knockout model provides
an ideal system to study the physiological role of Vitamin A in early
development. Work with this model has already made several contributions to
the understanding of the function of vitamin A in cardiovascular, hindbrain,
somite, and limb development (Zile, 1999; Zile, 2001).

Vitamin A is essential for early embryonic development and VAD
embryos die at day 3.5 of embryonic life (Dersch and Zile, 1993). Abnormalities
in the cardiovascular system, head, central nervous system, hematopoietic

organs, and the trunk are observed in VAD avian embryos (Dersch and Zile,

l3

1993; Twal et al., 1995; Maden et al., 1996, 1998; Zile, 1998; Zile, 2001).
Administration of vitamin A active compounds during the early stages of
development “rescues” the VAD embryo and results in normal development
(Dersch and Zile, 1993). Zile’s group has discovered a retinoid-requiring
developmental window, a stage in embryonic development concurrent with
initiation of morphogenesis and the early stages in the formation of the
cardiovascular system, during which there is an absolute requirement for vitamin
A (Zile, 1998; Zile, 2001).

The VAD quail model in our laboratory has been used mainly to study the
function of Vitamin A in early cardiovascular development. These studies are of
particular importance because of the obvious relevance to human cardiovascular
malformations, some of which may be diet-related during pregnancy (Ross et al.,
2000). The high incidence of vitamin A deficiency in developing countries may
account for the increased incidence of heart malformations in these populations
(Sommer et al., 1986). Very little is known about the etiology of congenital
heart malformations. Further research is needed to describe the etiological
factors, especially dietary ones during pregnancy, since diet directly contributes

to the physiology and pathology of the body.

1.5 Avian cardiovascular development
Cardiovascular development is a complicated process and involves the
simultaneous and interrelated development ofthe heart, the extra- and

extraembryonic vascular networks and the subsequent linking of the heart and

14

the circulation. Cardiovascular development is conserved across species,
especially in vertebrates, which have closely related developmental paradigms
(Fishman and Chien, 1997). Many cells and tissues are involved in
cardiovascular development; changes in cellular function and morphology are
essential for proper development. Several signaling pathways have been
identified to be involved in cardiovascular development; however, the whole
pathway remains to be elucidated. Development ofthe cardiovascular system
begins at an early developmental stage, with the simultaneous formation ofthe
cardiogenic precursors and the extra— and intraembryonic blood vessels. Two
basic types of processes are involved in cardiovascular development: those
processes that drive morphological development, such as migration,
proliferation, and apoptosis; and those processes that are necessary for functional
development, including specification, differentiation, and determination of

cardiogenic tissues (Farrell and Kirby, 2001).

1.5.1 Avian cardiogenesis

The heart is the first organ formed in the developing embryo and it serves
as the physiological center of the cardiovascular system. On the other end ofthe
life-spectrum, the cessation of a beating heart is often used as the definition of
embryonic death. Proper regulation of cardiogenesis is required for embryonic
development. Congenital heart malformations associated with developmental
abnormalities during cardiogenesis account for 20% of all spontaneous human

abortions (Hoffman, 1995). Cardiovascular development is fundamental to

15

vertebrate embryonic development and a large amount of work has been done to
elucidate the molecular and developmental mechanisms of heart development.
The term heart-forming region (HFR) is used to describe the area in which
cardiac progenitor cells arise. Figure l is a diagram ofthe relative positions of
cells that will form the heart, following their migrations from the primitive
streak, to the lateral plate, and finally to the heart tube during embryonic
development (Fishman and Chien, 1997; Redkar et al., 2001). Cells in the HFR

contribute to all layers ofthe mature heart (Redkar et al., 2001).

16

 

 

 

 

 

 

 

 

 

 

STAGE?

 

Figure 1. Schematic ofthe location of cardiogenic progenitors in the avian
embryo. The relative anterior-to-posterior positions of precardiac cells in the
primitive streak (HH stage 3) are retained in the lateral plate (as shown in HH
stage 5 and 7), and in the heart'tube at HH stage 10 (Fishman and Chien, 1997).
Cardiogenic progenitors are shown in black. All views are ventral.

Cardiac progenitors arise from the mesoderm; however, mesoderm alone
has been shown to be insufficient to generate heart tissue in explant cultures
from amphibians (Lough and Sugi, 2000). Epimyocardial and endocardial cells
are the first cardiogenic lineages that are specified. They arise from the
induction of mesoderm Via growth factors produced by the anterior endoderm

(Lough and Sugi, 2000; Sugi et al., 2001). Figure 2 shows cross-sections ofthe

HFR, depicting the cell layers that contribute to heart formation. The heart

17

begins as a paired tubular structure; however, the closure ofthe foregut draws
the cardiac precursors toward the midline where the endodermal components
fuse. Precursors of the endocardium cluster on the foregut side and are
distinguished from premyocardial cells by the expression of QH-l (Linask, 1992;
Linask and Lash, 1993) and flk, and they lack N-cadherin expression (Linask,
1992) and the muscle genes. By HH stage 9-10, the two-endocardial tubes have
fused to form one heart tube (Heine et al., 1985), which begins beating
simultaneously (Fishman and Chien, 1997). See Table 1 for comparative

chronology of heart development.

Table 1. Comparative Chronology of Cardiovascular Developmentl

 

 

QUAIL MOUSE HUMAN

Primitive streak HH 3 5-7 (I 15-16 (1
Assembly of lateral plate HH 5 7 d 18 d
Heart-tubes fuse HH 9 8 d 22 (1
Heart starts beating HH 10 8.5 d 23 d
Looping HH 11 8.5 d 23 (1
Birth HH 46 (17-18 d) 19.5 d 40 wk

 

 

1 Data from: Sissman, 1970 and Fishman and Chien, 1997

18

Ectodenn

 

HH Stage 8 —— Mesoderm

Foregut

l
Endodenn

Endocardial tube

'\....4‘
HH Stage 10 V

   

Figure 2. Cross-sections through the avian embryonic HFR. Cardiogenic
progenitors in the mesoderm respond to inductive signals from the endoderm.

By HH stage 9-10 the endocardial tubes fuse, forming one heart tube, surrounded
by epimyocardial and epicardial cells. Adapted from the Atlas of Descriptive
Embryology, 4th Edition.

Only the earliest developmental events in cardiogenesis have been
reviewed here. Reviews by Olson and Srivastava (1996), and Kirby and Waldo
(1990) describe in detail the later stages of cardiac morphogenesis. Briefly, after
the heart-tubes fuse, the heart orients on the right side of the embryo. As the
heart continues to develop and grow, the posterior and anterior ends remain in
fixed positions and the heart loops to the right in all vertebrates; however, the
exact mechanism for looping is not known. Recently, significant progress has
been made defining the positional and molecular cues that guide heart looping

and asymmetry (Levin, 1997; Beddington and Robertson, 1999; Brown and

19

Anderson, 1999; King and Brown, 1999; Majumder and Overbeek, 1999; Yost,
1999). It is clear that the asymmetry pathway is defined early in embryonic
development and is tightly regulated with stimulatory and inhibitory signals.
Neural crest cells and head mesenchyme also contribute to heart
development, specifically the formation of the outflow tract; these cells are
located outside of the HFR. This points out the fact that cells outside the HFR
need to be considered when examining heart development. Many factors
contribute to heart development and although great strides have been made into

the understanding of heart development, more research is still needed.

1.5.2 Avian vascular development

The vascular system arises early in embryogenesis to meet the nutritional
needs of the developing organism (Risau, 1995). Two mechanisms are involved
in embryonic vascular deve10pment: vasculogenesis and angiogenesis.
Vasculogenesis is defined as the differentiation of mesodermal cells into
endothelial precursors, which form the endothelial-lined blood vessels (Poole
and Coffin, 1989, 1991; Risau, 1991); angiogenesis is the sprouting and
branching of preexisting endothelial cells (Ausprunk and Folkman, 1977; Poole
and Coffin, 1989; Noden, 1991). These two mechanisms are interconnected and
together form the basis of vascular development.

The origin of endothelial cells and their assembly into the primary
vascular system have been extensively studied in avian embryos (Pardanaud et

al., 1987; Coffin and Poole, 1988; Poole and Coffin, 1989; 1991; DeRuiter eta1.,

20

1993). The precursor cells of the extraembryonal vascular network arise from
mesoderm in the area vasculosa. Precursor cells aggregate, forming blood
islands starting at the head-process stage (Pardanaud, 1987). Wilt (1966)
showed that differentiation of blood islands depends on an interaction between
the mesoderm and the underlying endoderm. Thus, like in heart development, an
inductive signal is needed for vascularization. In quail embryos this
differentiation begins at the head-process stage and by the one-somite stage
endothelial cells are detectable in the extraembryonic area (Coffin and Poole,
1988).

The blood islands consist of angioblasts, undifferentiated vascular cells
that will become endothelial cells and blood cells, which form in the area
vasculosa, and as the vascular system develops the blood island fuse to form the
primary vascular plexus (Patan, 2000). Angioblasts respond to stimulatory and
inhibitory signals, undergoing proliferation, differentiation, and apoptosis,
shaping the extraembryonicvascular plexus.

Intraembyonal vasculogenesis occurs simultaneously with the
development ofthe extraembryonal vascular plexus (Weinstein, 1999; Patan,
2000). Vascular development within the embryo is tightly correlated with heart
development, and some cells arise from the same precursors in the mesoderm of
the HFR (Inagaki et al., 1993). Endothelial cells can arise from the terminal
differentiation of endothelial cell precursors within the HFR (vasculogenesis
type I), or they can migrate to the HFR from the extraembryonal area pellucida

(vasculogenesis type II) (DeRuiter et al., 1993). Endothelial cells are first

21

apparent around HH stage 6-7 in the chicken embryo (Linask, 1992). These cells
assemble into a loose vascular plexus and coalesce into progressively larger
tubes, until eventually generating a single endocardial heart tube, by HH stage 9
(Sugi and Markwald, 1996). Endothelial cells, the main component ofthe

vascular system, line the heart as well as form the veins and arteries.

1.5.3 Avian inflow tract development

Although cardiogenic and vascular development are separate and distinct
processes, they require interaction for the development of a fully functional
cardiovascular system, evident at the inflow tract (IFT). A schematic diagram of

the IFT at early heart morphogenesis is presented in Figure 3.

 

 

Figure 3. Schematic diagram ofthe avian IFT at HHIO. The vascular
endocardium (EN), surrounded by the epimyocardium (EP), connects to the
extraembryonal vascular networks at the posterior region of the IFT through the
omphalomesenteric veins (OMV). The EP forms the lateral boundary of the IFT
and the anterior intestinal portal (AIP) forms the medial boundary. Adapted from
Atlas of Descriptive Embryology, Fourth Edition, 1986.

22

Formation of the IFT requires cells from the HFR as well as vascular
precursors from the extraembryonal area to amalgamate, forming a cohesive
entity, consisting of an epimyocardial outer layer and an endocardial inner layer.
The IFT serves as the transition point between the extraembryonic vascular
plexus and the heart. As the intraembryonic vascular networks extend laterally,
the extraembryonic vascular networks extend medially and the two systems fuse
together by HH 9 at the IFT (Pardanaud et al., 1987; Coffin and Poole, 1988). In
the avian embryo, blood arises from the blood islands located in the
extraembryonal area vasculosa, and enters the atrium through the
omphalomesenteric veins (OMV) at the inflow tract (Patan, 2000). IFT
development is a complicated process, requiring proper mediation of cellular
differentiation, proliferation, migration, adhesion, apoptosis, cell-cell
recognition, cell-matrix interactions, and multiple cell lineages. Currently, there

is insufficient literature describing IFT development.

1.5.4 Retinoids and cardiovascular development

As early as the 19303 it was recognized that maternal insufficiency of
vitamin A during pregnancy results in fetal death or abnormalities in the
offspring that include abnormal heart development (Mason, 1935). Initial clues
into the function of retinoids in heart development came nearly 50 years ago,
with studies of the effects ofVAD diets on the pups of pregnant rats (Wilson et
al., 1953). Thompson et al. (1969) made the important discovery that vitamin A

is required for early avian embryonic cardiovascular development. Subsequent

23

studies by Heine et al. (1985) provided a descriptive analysis of the effects of
Vitamin A deficiency on avian cardiovascular development, in particular on the
requirement of vitamin A for development ofthe early vasculature, with
abnormalities in VAD embryos evident between the 5 and 15 somite stages.
These observations were confirmed and extended using the quail embryo ligand
knockout model (Dersch and Zile, 1993; Kostetskii et al., 1998; Zile, 2000). The
discovery of a critical retinoid-requiring developmental window by Kostetskii et
al. (1998) has made it possible to specifically address the role of Vitamin A in
early development. The abnormalities in heart development are the first
pronounced morphological changes due to vitamin A deficiency in the early
embryo (Zile, 1998, 2000, 2001); specifically, the earliest gross developmental
change evident at the time of formation of the cardiovascular system in the VAD
avian embryo is the abnormal IFT development (Zile, 2001). It is of interest to
note that RA synthesis in the mouse heart has been localized to the IFT during
early development (Moss et al., 1998), corresponding to the earliest
morphological defects in cardiogenesis observed in the VAD quail embryo.
Presently, it is unclear whether the abnormal IFT development in the VAD quail
embryo is the consequence of abnormal heart morphogenesis, or if IFT

development is a separate retinoid regulated process.

24

1.6 Retinoid-regulated molecular mechanisms of cardiovascular

development

The discovery of nuclear receptors for RA and their hetrodimer partners,
the RXRs has provided an explanation for the molecular mechanism by which
retinoids control many physiological processes, including cardiovascular
development (Mangelsdorf et al., 1995; Pfahl and Chytil, 1996). Vitamin A is
involved in early heart and vascular development, most likely Via the retinoid
receptors. Our laboratory is studying the function of retinoids in cardiovascular
development and has demonstrated that the early avian embryo expresses all the
genes for retinoid receptors, and their expression includes the HFR (Kostetskii et
al., 1998; Zile et al., 1997; Cui et al., in preparation). RAR132 expression in the
HFR is down regulated in the absence of vitamin A (Kostetskii et al., 1996,
1998). These findings in conjunction with the morphological studies on VAD
quail embryonic cardiovascular development make it logical to think that
retinoid receptors are involved in the transcriptional regulation of cardiovascular
development. However, the interaction among many factors is required for
cardiovascular development and the underlying mechanisms remain to be

elucidated.

1.6.1 Cardiogenic genes regulated by retinoic acid
There are no master regulators that control heart development; therefore,
the precise interaction among many genes is required for proper development

(Laverriere et al., 1994; Evans, 1997; Fishman and Chien, 1997). Our laboratory

25

has examined the role of vitamin A in the regulation of various cardiogenic
genes and has demonstrated that the heart specific gene, ka 2.5, as well as
cardiac muscle specifying genes are not affected by Vitamin A status (Kostetskii
et al., 1999). It was discovered, however that the cardiogenic gene GATA-4 is
altered in the VAD quail embryo (Kostetskii et al., 1999).

The GATA gene family encodes transcription factors involved in
hematopoetic lineage determination and in cardiovascular development
(Laverriere et al., 1994; Evans, 1997). GATA-4 expression has been reported to
be retinoic acid responsive (Arceci et al., 1993), and our studies with the quail
embryo retinoid-ligand knockout model have shown that GATA-4 is severely
down regulated in VAD quail embryos, especially in the area ofthe posterior
heart tube and subsequent inflow tract (Zile, 1998). It is possible that the
abnormal morphology associated with Vitamin A deficiency in the quail embryo
may be due in part to downstream cardiac-specific genes that are regulated by
GATA-4 via RA (Zile, 1999).

Vitamin A is also involved in determining cardiogenic asymmetry. In the
retinoid ligand knockout quail embryo heart position is randomized (Zile et al.,
2000). Heart sidedness can be rescued by administering vitamin A-active
compounds by HH stage 8 (Kostetskii et al., 1998; Zile etal., 1998, Zile et al.,
2000). These observations challenged the recently proposed molecular pathway
for determining cardiac L-R asymmetry, because the asymmetric expression of
all the early asymmetry genes including Shh in the quail Hensen’s node, is not

altered by vitamin A deficiency (Chen et al., 1996), and importantly, cardiac

26

asymmetry can be rescued by administering RA downstream of the Shh and other
early asymmetry signaling pathways (Zile et al., 2000). It is clear that vitamin A
is involved in the determination of cardiac asymmetry at the late stage of the
asymmetry pathway; however, more research is needed to determine the

molecular mechanisms of retinoid—regulated cardiac asymmetry aspects.

1.6.2 Vasculogenic genes and vitamin A status

Cell differentiation, proliferation, organization, and apoptosis are
maintained under strict control to complete the development of a single,
interconnected vascular system (Cleaver and Krieg, 1999). Recent work has
identified molecules involved in early differentiation of mesodermal cells into
endothelial cells, as well as molecules involved in endothelial cell proliferation,
migration, and remodeling (Cleaver and Krieg, 1999). Growth factors are
required for early differentiation of endothelial cell precursors, the angioblasts,
located in the mesoderm, into endothelial cells. Members ofthe fibroblast
growth factor (FGF) family play a critical role in the induction ofthe
mesodermal germ layer during the earliest stages of embryogenesis (Cleaver and
Krieg, 1999). Cell culture experiments have shown that FGF induces the
expression of flk-l, the receptor for vascular endothelial growth factor (VEGF),
which is required for angioblast differentiation (Eichmann et al., 1993; Cleaver
and Krieg, 1999). Once angioblasts differentiate into endothelial cells they

proliferate and migrate before assembling into blood vessels; both FGF and

27

VEGF are mitogens for endothelial cells (Keyt et al., 1996; Cleaver and Krieg,
1999)

The VAD embryo lacks a vascular link between the embryo and the
extraembryonal vascular system (Thompson, 1969; Heine et al., 1985; Dersch
and Zile, 1993; Twal et al., 1995; Zile, 1998; Zile, 1999; Zile, 2001). Thus, it is
very likely that Vitamin A is involved in the regulation of vasculogenesis at the
gene level.

Extracellular matrix (ECM) can modulate growth and differentiation of
endothelial cells and also influence their migration (Drake et al., 1992). These
developmental events require the coordinated recognition and response of cell
surface molecules to those in the ECM. Fibronectin (FN) is an ECM molecule
that has been linked to endothelial cell motility during vascular development
(Noden, 1991). Perturbations of normal avian embryogenesis with RA between
HH stages 5 and 8 show similarities to cardiovascular defects arising in embryos
treated with FN antibodies (Osmond et al., 1991) and in mouse embryos lacking
FN (George et al., 1993). Furthermore, preliminary work by Linask and our
laboratory has shown FN to be down regulated in VAD quail embryos (not
published). Cells interact with the ECM via cell adhesion molecules. Integrins
are the best characterized cell adhesion molecules and bind to collagen, laminin,
fibronectin, and thrombospondin ofthe ECM (Cleaver and Krieg, 1999). When
quail embryos are injected with CSAT, an antibody against Bl-integrin, vascular
development is arrested at the stage when slender cord-like assemblies of

angioblasts rearrange to form tubules (Drake et al., 1992). Some of the aspects

28

of retinoid-regulated vasculogenesis will be addressed in the present thesis

research.

1.6.3 Transforming growth factor beta

The transforming growth factor betas (TGFB) are a family of
multifunctional polypeptide growth factors capable of mediating cell growth,
differentiation, and morphogenesis (Koli et al., 2001) by modifying the
expression of a specific set oftarget genes (Massague and Chen, 2000). One of
the main functions of TGFB is the inhibition of cell proliferation (Piek et al.,
1999); however, the influence of TGFB is dependent on the cell lineage,
coexisting growth factors, and other biological and physiological variables
(Brand and Schneider, 1995). The pleiotropic actions of TGFB are best
illustrated by their ability to exert either positive or negative effects on cell
growth, illustrating the necessity to determine its biological role empirically in
any given context (Brand and Schneider, 1995).

Members ofthe TGFB family have been implicated in heart (Brand and
Schneider, 1995; McCormick, 2001) and vascular development (Brown et al.,
1999; Goumans et al., 1999; Dickson, et al., 1995). TGFBs have been shown to
induce the generation of myocardium (Muslin and Williams, 1991; Slager et al.,
1993) and to inhibit vascular development (Hayasaka et al., 1998; Zhao and
Overbeek, 2001) in tissue explants; however, the role of TGFB in early
cardiovascular development has not been elucidated. It is known that the TGFBS

are subject to regulation by retinoids (Mahmood et al., 1992; Salbert et al., 1993;

29

Mahmood et al., 1995; Han et al., 1997; Imai et al., 1997;Yoshizawa et al., 1998;
Koli et al., 2001); members ofthe TGFB family are candidate genes that mediate
the downstream effects of RA in the early heart tube (Kubalak and Sucov, 1999).
TGFB is typically secreted as part of a latent dimeric complex. The
complex consists of a homo- or hetrodimer pro-peptide domain, TGFB latency
associated protein (LAP), which is noncovalently associated with the mature
TGFB dimer, and latent TGFB binding protein (LTBP) covalently bound to LAP
(Brand and Schneider, 1995; Lawrence, 1995; Koli et al., 2001). The associated
proteins mediate proper folding and protect the mature peptide from degradation
(Lawrence, 1995). Once secreted, the latent TGFB complex associates with
interstitial ECM and basement membranes (Roberts and Sporn, 1996).
Mechanisms involving proteolysis are required to release and activate the pro-
peptide domain (Koli et al., 2001). The active TGFB binds to TGFB
serine/threonine protein kinase receptors, the TBRs, and elicits transcriptional

activity via an intracellular signal transduction pathway that is depicted in Figure

4.

3O

Precursor
Form

1"— Protcolysis

Active , ..
Ligand T @1911 +1 IBP

R- Ill

 

1?? R-111
'l/l/l/I/l/A’A I/IWIWW/ ////
t.‘ y to pl as I n

 

  
   

R- Smad

Co- Smad >\ \H
1 Smad

_ PP
‘5: PP Ubiquitin-dependent

de radation

 

Coactivator or
a .' Corepressor

protein

Figure 4. Signaling by TGFB through its serine/threonine kinase receptors and
Smad proteins. The active pro-peptide dimer, cleaved from LAP and LTBP by
proteolysis, binds to a heterotetramer of the type II and type I TGFB receptors R-
11 and R-I, respectively. Although not essential for signal transduction, the type
III TGFB receptor (R-II) can facilitate the binding of active ligands to R-II. I- ‘
Smads inhibit signaling by R-Smads:Co-Smad complexes by competitive
inhibition. R-SmadzCo-Smad complexes bind to Smad responsive elements
(SRE), DNA binding proteins, and coactivators to elicit transcriptional activity
Adapted from Miyazono, 2000.

The regulation of TGFB by retinoids is well recognized but is not
understood. The promoters of the TGFB genes do not contain RARE, indicating
that an indirect method of regulation is involved (Roberts and Sporn, 1992a).

Retinoids have been implicated in the activation of latent TGFB post-

31

transcriptionally (Kojima and Rifkin, 1993; Yoshizawa et al., 1998); however,
the mechanism remains unclear. Using both RA excess and VAD mouse

embryos, Mahmood et al. (1992, 1995) have shown that retinoids are capable of
differentially regulating TGFB isoforms through mechanisms involving different
stages in the process of TGFB synthesis and secretion. Most ofthe current
literature has focused on the increase in TGFB expression associated with the
addition of excess RA, either in cell culture (Kojima and Rifkin, 1993; Imai et
al., 1997; Yoshizawa et al., 1998), or in mouse embryonic development
(Mahmood et al., 1995). However, a recent study has shown that TGFBZ is
increased in RXRct knockout mice embryos, leading to increased apoptosis in the
outflow tract of the heart (Kubalak et al., 2002). Similarly, RARO. and RXRct
proteins haVe been shown to inhibit TGFB promoters via direct protein-protein
interactions in a concentration-dependent manner (Salbert et al., 1993). Our
findings that RARoc2 and RARBZ are severely down-regulated in VAD quail
embryos (Cui et al., in preparation), and our preliminary findings of altered
TGFB expression in these embryos (Kostetskii et al., 1993), suggest a role of
retinoids in the TGFB signaling pathway during quail embryonic development,

this forms the basis for the research described here.

32

CHAPTER 2

RATIONALE AND OBJECTIVES

The absolute requirement for vitamin A during embryonic development is
well established. Studies with the VAD avian embryo model in our laboratory
have provided important information regarding the cardiovascular phenotype
attributed solely to the absence of RA signaling, i.e. an abnormal, non-looped
heart with random left/right orientation, a thin-walled dilated heart cavity,
defects in the outflow tract, and an absence of the inflow tract (IFT) that links
the extraembryonal vasculature to the heart. Although these physiological
defects of the cardiovascular system have been attributed to VAD, the molecular
mechanisms remain to be elucidated. Our laboratory is focused on research
concerning the function of Vitamin A in cardiovascular development and has
identified several retinoid-regulated genes that are linked to heart morphogenesis
(Kostetskii et al., 1999; Zile et al., 2000; Zile, 2001). The lack of formation of
the IFT as well as the identification ofthe down regulation ofthe GATA-4 gene
in the posterior heart regions of the VAD embryo has led to hypothesize that the
initial effect of Vitamin A on heart development is at the site of cardiac IFT
formation in the posterior HFR, as this is the earliest event in cardiac
morphogenesis and the segmentation ofthe linear heart tube proceeds in a
posterior to anterior direction.

Development of the IFT is a complicated process requiring the mediation

of cell proliferation, differentiation, apoptosis, and migration and organization of

33

presumptive endothelial cells (angioblasts) from the extraembryonal area and
mesodermal cardiogenic cell in the HFR to give rise to the IFT. The IFT are
located bilaterally of the midline in the posterior HFR and contain the OMV that
link the embryo with the extraembryonal vasculature. Thus, it is clear that many
signaling pathways are involved in IFT development, and a likely candidate for

RA regulation for these events would be a growth factor that has multiple
regulatory functions. The TGFB family of multifunctional polypeptides are
possible candidate genes.

The TGFB family of cytokines that regulate cell proliferation,
differentiation, recognition, death, tissue recycling and repair (Massague, 1996),
have been linked with cardiovascular development. The primary receptor for
TGFB, transforming growth factor-[3 receptor type II (TBRII) is present in
endothelial cells and marks the extraembryonal vascular networks in quail and
chicken (Brown et al., 1999). Also, TGFB signaling is required for the
organization of extraembryonic endothelial cells into robust vessels (Dickson et
al., 1995; Goumans et al., 1999). The role of TGFB in cardiovascular
_ development is not limited to endothelial cells, as myocardial cells also express
TGFB during cardiogenesis (McCormick, 2001; Brand and Schneider, 1995;
Roberts and Sporn, 1992a), and TGFBZ knockout mice have congenital heart
defects, mainly in the myocardium ofthe outflow tract (Sanford et al., 1997). It
is well known that in many cells vitamin A regulates TGFB signaling (Mahmood
et al., 1992; Mahmood et al., 1995; Roberts and Sporn, 1992a). In vitro studies

using bovine endothelial cells have shown that RA up-regulates the expression of

34

TGFB receptors types I and II, resulting in enhancement of TGFB activity
(Yoshizawa et al., 1998). While our studies were in progress, work in mice has
shown that knocking out the retinoic X receptor or results in elevated TGFBZ,
which can contribute to abnormal outflow tract morphogenesis by enhancing
apoptosis in the endocardial cushions (Kubalak et al., 2002). However, the
regulation of TGFB by vitamin A has not been studied in the development ofthe
IFT. Furthermore, preliminary work in our laboratory demonstrated increased
TGFB mRNA in the early VAD quail embryo (Kostetskii and Zile, 1993). This
suggested a role of TGFBZ in embryonic development, possibly in cardiovascular
morphogenesis and IFT formation.

Presently, little is known about the development of the IFT and in
particular about the role vitamin A plays in its development. Previous work in
our laboratory has identified a critical retinoid-dependent developmental
window, the 4/5 somite stage of neurulation, at which time retinoids are
absolutely required for development to proceed normally and for the embryo to
survive (Kostetskii et al., 1999; Zile, 2001). The retinoid-dependent
developmental window is concurrent with the beginning of heart morphogenesis
and ofthe formation of the IFT. If biologically active retinoids are administered
before, or during the retinoid-dependent developmental window to VAD
embryos, then IFT development, as well as all other characteristics ofthe VAD
phenotype are rescued and the embryo survives (Kostetskii et al., 1999).

The overall goal of my research is to contribute to the understanding of

the function of Vitamin A in cardiovascular morphogenesis, in particular

35

understanding the molecular function of Vitamin A in IFT development focused

within the retinoid-regulated developmental window. The specific goals of my

research are as follows:

1)

2)

Compare normal and VAD cardiovascular development.

There has not been a systematic description ofcardiovascular
abnormalities associated with VAD during early embryogenesis. It is
imperative to determine the developmental stages when morphological
changes associated with vitamin A deficiency occur for subsequent
studies to address the molecular events that might be involved in these
developmental processes, especially in the formation ofthe IFT.
Normal and VAD cardiovascular development will be studied using
light microscopy to examine gross morphological development;
immunolocalization studies will be conducted‘to compare vascular
development using QH-l antibody as a biomarker for presumptive

endothelial cells.

Compare TGFBZ protein and mRNA and TBRII mRNA in normal and
VAD quail embryos during early embryogenesis.

Previously, we showed that TGFB2 expression is increased in the VAD
quail; however, spatio-temporal localization has not been examined in
early avian embryogenesis. We will examine normal and VAD TGFBZ

protein and mRNA levels and localization during early embryogenesis.

36

3)

If TGFBZ is involved in IFT development, we would expect it to be
expressed in the HFR. The expression of TBRII, the principal receptor
for TGFBZ, will also be studied during early development to determine
if its expression is altered in VAD embryo, and ifthe receptor

expression corresponds with ligand expression.

Examine the specific role of TGFBZ in cardiovascular IFT
morphogenesis.

If TGFBZ mRNA is increased in the HFR ofthe VAD quail embryo,
we will then elucidate the role of TGFBZ in development linked to the
VAD phenotype. The increased TGFBZ expression associated with
vitamin A deficiency will be blocked using antisense oligonucleotides
to TGFBZ. Morphology, vascular development (QH—l), TGFBZ
mRNA, and TBRII mRNA will be examined in VAD embryos injected

with antisense oligonucleotide to TGFBZ.

37

CHAPTER 3

MATERIALS AND METHODS

3.1 Animals and diets

Quail (Coturnix coturnix japonica) were housed at the Michigan State
University Poultry Research and Teaching Farm. Normal eggs were obtained
from hens fed game bird chow (Purina Mills Inc., St. Louis, MO).. Vitamin A-
deficient (VAD) eggs were obtained from hens fed a semi-purified diet adequate
in all nutrients, but with 10 mg of all-trans-retinoic acid added per kg of diet as
the only source of vitamin A, as previously described by Kostetskii et a1. (1996).
All birds were hatched from normal eggs obtained from the Michigan State
University Poultry Research and Teaching Farm. Birds that were designated to
be in the normal control group were fed Purina starter/grower diet from birth
until they began to lay eggs at 6-7 weeks of age at which time the diet was
changed to Purina breeder diet. Birds that were designated for the VAD
experimental group were fed a semi-purified starter diet, described in Table 2,
and switched to a semi-purified breeder diet at 6-7 weeks, when the birds began
to lay eggs. The eggs from the young birds were confirmed to be depleted of
vitamin A by collecting the VAD eggs from the young birds daily and examining
embryo morphology after three days of incubation. When all ofthe embryos
from eggs from the young birds were VAD according to our previously described

phenotype, the eggs were used in experiments. The birds were kept in cages with

38

a ratio of 2:1, females to males. The birds had continuous access to food and

water, which they received fresh daily.

Table 2. Composition of the Semi-Purified Diets

 

 

INGREDIENT (g/kg) STARTER BREEDER
Soybean meal (47.5% Protein) 505.00 464.00
Dextrose (monohydrate) 383.41 380.33
Mineral mixl 44.86 44.86
Soybean oil 40.00 40.00
Cellulose 10.00 0.00
Calcium phosphate (dibasic) 4.50 13.90
Choline dihydrogen citrate 4.20 3.20
DL-methionine 3.50 3.50
L-lysine hydrochloride 2.50 1.00
Calcium carbonate 1.50 48.70
all-trans-retinoic acid 0.01 0.01
Biotin 0.0006 0.0006
Vitamin B12 0.03 0.03
Vitamin D3 0.03 0.03
Folic acid 0.005 0.005
Menadione sodium bisulfite complex 0.0168 0.0168
Nicotinic acid 0.08 0.08
Calcium pantothenate 0.03 0.03
Pyridoxine hydrochloride 0.015 0.015
Riboflavin 0.015 0.015
Thiamin hydrochloride 0.025 0.025
DL~alpha-tocopheryl acetate 0.10 0.10
Butylated hydroxytoluene 0.10 0.10

 

 

I Custom Mineral Mix, DYET #290013 (Dyets Inc., Bethlehem, PA)

39

3.2 The quail embryo retinoid ligand knockout model

Quail eggs were collected daily, and the eggs from normal and VAD
groups were kept separately. The eggs were stored at 13°C, and used within one
week. Eggs were incubated at 385°C, with 98% humidity and with a 2-hr
rotation—cycle (Kostetskii et al., 1999; Zile et al., 2000). Embryos obtained from
eggs from hens fed normal chow diet were termed “normal”. Embryos from eggs
of hens fed the purified VAD diet were termed vitamin A deficient, because the
retinoic acid from the diet of hens cannot be transferred to the egg (Dersch and

Zile, 1993; Dong and Zile, 1995).

3.3 Dissection

After incubation embryos were dissected from the yolk using
microsurgical scissors and forceps (Fine Science Tools Inc., Foster City, CA), in
a weigh boat containing phosphate-buffered saline (PBS) at RT (Kostetskii et al.,
1999; Zile et al., 2000). Embryos were then placed in a petri dish containing
PBS at RT under a Nikon SMZ-2T (Nikon Corporation, Japan) microscope, and
the embryonal membrane was removed using forceps. Embryos were staged
according to Hamburger and Hamilton (1951). Pictures for morphological
studies were taken of the embryos in PBS using a Nikon 6006 camera attached to
the microscope. Images in this thesis are presented in color. The embryos were

then fixed in 4% paraformaldehyde (PFA) in PBS at 4°C overnight.

40

3.4 Sectioning

To examine morphology at the cellular level, fixed embryos were stained
with a Toluidine Blue solution (Kostetskii et al., 1999) and cross-sectioned. The
methods used to embed the embryos in paraffin and to section the embryo have
been previously described (Kostetskii et al., 1996). Embryos were washed with
PBS twice for 10 min, and then dehydrated in an ascending EtOH series, 25%,
50%, 75%, and 100% for 15 min each. Embryos were stained with a Toluidine
Blue solution overnight at RT, and then washed 3 times with EtOH, 10 min each.
Embryos were then treated with isopropanol for 10 min at RT and then moved to ‘
a 1:1 mixture of xylene and paraffin at 65-70°C for an additional 10 min,
followed by treatment with paraffin for 10 min at 65-70°C. Embryos were
embedded vertically in paraffin blocks and allowed to harden overnight at 4°C.
Paraffin blocks were removed from their molds the next day, and the embryos
sectioned using a Spencer 820 Microtome (American Optical) at 10 pm using
standard methods (Kaufman, 1992). The sections were collected in a continuous
strand and placed in a water bath at 40°C. Sections were then placed on gelatin-
coated slides, examined under a light microscope (Ernst Leitz Wetzlar,
Germany), and the slides with desired sections were incubated in a 37°C oven
overnight to remove all moisture. Next, slides were dipped in xylene to remove
paraffin, then dipped in MeOH and placed in a 37°C oven for ten minutes to dry.
The sections were examined using a Nikon Optophot-2 compound microscope
that was equipped with a Nikon 6006 camera and pictures were taken of

representative sections.

41

3.5 In ovo microinjection

Exogenous compounds were administered to the developing embryo in
ovo Via microinjection, as described previously (Kostetskii et al., 1999;
Kostetskii et al., 1998). Eggs were incubated for 22-28 hr at 385°C at 98%
humidity with the eggs placed horizontally. The technique described by Selleck
(1996) was used for windowing of eggs and for in ovo injection. The eggs were
removed from the incubator and washed with 70% EtOH. A one m1 syringe was
used to remove 0.8 ml of egg albumin from the egg and the needle hole was
sealed with masking tape. A piece of masking tape was placed on the side ofthe
egg that was facing up during incubation and cut to lay flat. The tape stabilized
the eggshell, and a window was cut inside the edges ofthe tape, leaving one side-
ofthe window intact to act as a hinge. Air bubbles were removed from the egg
with a 1 m1 syringe. Embryos were sub-blastodermally microinjected, using a 10
pl Hamilton syringe (Hamilton Company, Reno, NV). The embryos visible
above the ink background were staged according to Hamburger and Hamilton
(1951). The eggshells were sealed with masking tape. Sealed eggs were
incubated horizontally for an additional 10 hr to 2 days in an incubator at 385°C

with 98% humidity.

3.5.1 Rescue of the VAD embryos with vitamin A-active compounds
Kostetskii et al. (1998) were able to rescue VAD embryos by injecting
biologically active retinoids. Previously, our lab discovered a critical retinoid-

dependent developmental window at the 4-5 somite stage of development (mid-

42

neurulation) (Zile, 2001). The presence of Vitamin A-active compound at this
stage is essential for further development to take place normally. The VAD
embryo can be rescued ifvitamin A-active compounds are provided at or prior to
this developmental window. In the present work, VAD embryos were treated
during this deve10pmental window so as to determine iftheir development could
be rescued.

Yellow lighting in the room was used when handling retinoid compounds.
Retinoids were kept under nitrogen and dissolved in EtOH. Approximately 3 mg
of either ROH or RA was weighed out and dissolved in 1 ml of EtOH. The
retinoid solution was diluted 1:1000 in EtOH and the concentration and spectra
of retinoids in the solution was determined using a CARY 3E spectrophotometer
(Varian, Inc., Palo Alto, CA). Absorbance and spectra were recorded from 400-
200 nm. A second peak, or an alteration in the main peak characteristic for the
retinoid signified degradation of the retinoid compounds, in which case the
retinoids were discarded. All-trans-ROH has a major peak at 326 nm and a
minor peak at 230 nm; all—trans-RA has a broad absorption with a maximum at
350 nm. Using a Hamilton syringe embryos were sub-blastodermally injected
with 10pl of Tyrode’s solution containing 10% VAD egg homogenate, 5% filter-
sterilized Pelikan Fount India ink, and 10 ng of biologically active retinoids,
either all-trans-ROH or all-trans-RA (Sigma Aldrich, St. Louis, MO); control
solution contained the vehicle alone. Embryos at 3-5 somite stages were selected

for experiments. The eggs were then sealed, and incubation was continued for an

43

additional 10 hr to 2 days. Normal and VAD control embryos were injected with

the vehicle.

3.5.2 Blocking of development with anti-all-trans-retinoic acid monoclonal
antibody

RA is the transcriptionally active form of Vitamin A. Twal et al. (1995)
showed that a monoclonal antibody against all-trans-RA blocked normal quail
development when injected under the blastoderm before the retinoid-dependent
developmental window. In the present study, anti-all-trans-RA MoAb was used
to block Vitamin A mediated development by sequestering all-trans-RA. The
antibody producing cells were grown in culture in our lab and the MoAb secreted
by the cells into mediums was purified using an affinity column (Twal et al.,
1995). Antibody concentration was determined using a Spectronic 21D
spectrophotometer (Spectronic Instruments, Rochester, NY).

Normal and VAD embryos were sub-blastodermally injected, using a 10p]
Hamilton syringe, with 10 p1 of Tyrode’s solution containing 10% VAD egg
homogenate, 5% filter-sterilized Pelikan Fount India ink, and 5 pg of anti-RA
MoAb. Normal and VAD control embryos were injected with the vehicle alone,
or mouse IgG (Sigma). Embryos at 1-3 somite stage were used for blocking

studies. The eggs were sealed and incubated for an additional 10 hr to 2 days.

44

3.6 Immunohistochemical localization of endothelial cells

Monoclonal antibodies against QH—l, isolated from cells obtained from
the Developmental Studies Hybridoma Bank at the University orowa and grown
in our laboratory, were used to identify endothelial cells in quail embryos (Twal
and Zile, 1997). Embryos were dissected from extraembryonal membranes and
staged according to Hamburger and Hamilton (1951). Embryos were fixed
overnight in 4% PFA in PBS and then rinsed 3 times in PBS, 30 min each. The
embryos were then treated for 15 min with PBS containing 1% Triton X-100
(PBST) and next treated for an additional 15 min with hyaluronidase, 2 mg/ml in
PBS. Blocking of nonspecific binding was performed by first incubating the
embryos for 5 min with PBST containing 0.01 M glycine (PBST/G) and then for
1 hr with PBST/G containing 10% normal goat serum. Embryos were next
incubated overnight at 4°C with, undiluted cell supernatant containing QH-.1
antibody. Normal and VAD control embryos were incubated in Dulbecco’s
Modified Eeagle’s Medium without the QH-l antibody. Embryos were rinsed 3
times (1 hr each) with PBST and incubated for 2 hr with fluorescein
isothiocyanate (FITC)-labeled goat anti-mouse IgG (Molecular Probes; Eugene,
OR) at a 1:50 dilution in PBS. Prior to incubation, the goat IgG was
preincubated for 1 hr with quail embryo acetone powder to reduce nonspecific
binding. Embryos were then rinsed 2 times (30 min each) with PBS and once for
5 min with PBS containing 0.003% Evans blue. After 2 brief rinses with PBS the
embryos were mounted in 40% PBS in glycerol and examined with an

epifluorescence equipped Nikon Optiphot-2 compound microscope. Pictures

45

were taken with a Nikon 6006 camera mounted on the microscope. All steps

were performed at RT except where indicated.

3.7 Immunohistochemical localization of TGFBZ

Rabbit polyclonal antibodies against TGFBZ were generously provided by
Dr. Anita Roberts at the National Cancer Institute (Bethesda, MD). Embryos
were fixed in 4% PFA, then rinsed 3 times in PBS, 30 min each, and then 3 times
with PBS containing 1% Triton-X (PBST), 10 min each. The embryos were then
treated 2 times, 1 hr each, with PBST containing 2% instant skim milk (PBSMT)
to solublize membranes and block nonspecific binding. The embryos were then
incubated for 15 min with 2 mg/ml hyaluronidase in PBS to further block
nonspecific binding. Embryos were next incubated overnight at 4°C with the
TGFBZ antibody diluted to a working concentration of 1 pl/ml in PBSMT.
Embryos were rinsed 5 times with PBSMT, one hr each, and incubated overnight
at 4°C with tetramethylrhodamine isothiocyanate (TRITC) —labeled donkey anti-
rabbit IgG (Sigma Aldrich) at a 1:100 dilution in PBSMT. Prior to incubation,
the donkey IgG was preincubated for 1 hr with quail embryo acetone powder.
Embryos were rinsed 5 times, 1 hr each, with PBSMT, then for 20 min with
PBST, and mounted in 50% PBS in glycerol, and examined with an
epifluorescence equipped Nikon Optiphot-Z compound microscope. Pictures
were taken with a Nikon 6006 camera mounted on the microscope. Normal and
VAD control embryos were incubated in PBSMT with non-specific mouse IgG

instead of the TGFBZ antibody.

46

3.8 Analysis of TGFBZ mRNA expression by in situ hybridization
3.8.1 Recovery of plasmid

cDNA for TGFB2 (1.9 kb) was a gift from Dr. Anita Roberts at the
National Cancer Institute. cDNA was inserted into the polyclonal region of
puc19 vector. The obtained plasmid was on filter paper. To reconstitute the
plasmid, the filter paper was placed in a sterile microfuge tube with 50 pl of
sterile ddeO and vortexed 3 times for 5 min, and then stored at 4°C overnight.
The filter paper was removed from the supernatant and the supernatant
containing the plasmid was stored at -20°C. To verify the presence ofthe
plasmid in the supernatant, 2 p1 ofthe supernatant was mixed with 3 pl of DNA

loading buffer, subjected to electrophoresis on 1% agarose gel, and stained with

ethidium bromide.

3.8.2 Transfection of E. coli with plasmid containing TGFBZ cDNA
Invitrogen One Shot TOP10 transformation kit (Invitrogen, Carlsbad, CA)
was used to transfect E. coli, competent cells, with the plasmid containing the
TGFBZ cDNA. All components used in the reaction were obtained from the kit,
except the supernatant containing the plasmid from above. The reaction was
carried out on ice. The vial of competent cells was thawed on ice, then 2 pl of
0.5M B-mercaptoethanol was added and the cells allowed to stand on ice for 20
min, then 10 pl of the above supernatant containing the plasmid was added to the
mixture. The competent cells were subjected to heat-shock in a 42°C water bath

for 35 sec and then returned directly to the ice bath for 2 min. Regular LB

47

media, 250 pl, was added to the reaction mixture and incubated at 37°C in a
water bath for 1 hr with vigorous shaking. The mixture was plated on LB plates
containing ampicillin (LB/Amp) and incubated at 37°C overnight and then stored

at 4°C until use. Colonies that grew contained the transfected plasmid.

3.8.3 Purification of plasmid

Single colonies that grew on the LB/Amp plate were inoculated in 3 ml
LB/Amp media in sterile test tubes and incubated at 37°C in a water bath
overnight with Vigorous shaking. Promega Wizard Plus Minipreps DNA
purification system (Promega, Madison, WI) was used to isolate the plasmid. All
solutions were from the kit unless specified. The bacteria were centrifuged at
10,000 g for 10 min and the supernatant poured off. Resuspension solution, 300
p1, was added to the test tube, mixed by vortexing, and then transferred to a
microfuge tube. Lysis solution, 300 pl, was added, the tube closed and mixed by
inverting until the solution was clear, then 300 pl of neutralization solution was
added and the sample centrifuged at 10,000 g for 5 min. An affinity
microcolumn provided in the kit for the purpose of isolating the plasmid was
placed on a 5 ml syringe and the plunger removed. Resin solution, 1 ml, was
added to the barrel of the syringe, then the supernatant was added to the syringe
and the solution was pushed through, trapping the plasmid in the microcolumn.
The microcolumn was removed from the syringe, then the plunger, and then the
microcolumn was placed back onto the syringe. Two ml of wash solution

containing EtOH was placed in the syringe and pushed through precipitating the

48

plasmid. The microcolumn was removed from the syringe and placed in a sterile
microfuge tube and subsequently centrifuged at 10,000 g for 2 min. The
microcolumn was removed, and placed in a sterile microfuge tube, then 50 pl
ddeO was added to the microcolumn and allowed to stand for l min, followed
by centrifugation at 10,000 g for 25 sec. To verify the presence ofthe plasmid, 2
pl ofthe collected solution, hereafter referred to as purified plasmid, was mixed
with 3 pl of DNA loading buffer, subjected to electrophoresis on 1% agarose gel,

and stained with ethidium bromide. The purified plasmid was stored at -20°C.

3.8.4 Sub-cloning of TGFBZ cDNA into pBSK'

The cDNA for TGFBZ was obtained in pUC19 vector, which does not
contain an mRNA promoter. In order to make an mRNA riboprobe the cDNA
was sub-cloned into a vector that contains an mRNA promoter, pBSK‘
(Stratagene, LaJolla, CA). Basic Local Alignment Search Tool (BLAST) was
used to check for homology between TGFBZ and other known genes in the NIH
database. A unique sequence was chosen that had no known homology with any
other avian genes. The pBSK' vector and the purified pUC19 plasmid containing
the TGFBZ cDNA were digested with Kpnl and EcoRV (New England Biolabs,
Beverly, MA) at 37°C for 3 hr. Digestion with Kpnl and EcoRV cut a 40 bp
section out of the polyclonal region ofthe pBSK' plasmid, and excised a 500 bp
section ofthe cDNA unique to TGFBZ from the pUCl9 plasmid. Complete

digestion was confirmed via electrophoresis on 1% agarose.

49

The entire sample from digestion was subjected to electrophoresis on 1%
agarose at a 1:1 ratio with DNA loading buffer. QIAEX 11 Gel Extraction Kit
(Qiagen, Inc., Valencia, CA) was used to isolate the digested pBSK' and the
TGFBZ segment. All solutions used for extraction were those provided in the kit,
unless specified. The corresponding DNA bands were cut out from the agarose
gel and placed in clear, sterilized microfuge tubes. The gel was weighed and 3
volumes of Buffer QXl was added to every volume of gel containing the DNA
fragments. QlEAX II was resuspended by vortexing and 10 pl of QIAEX II was
added to the sample. The sample was incubated at 50°C for 10 min to solubilize
the agarose and bind the DNA to QlEAX II. The sample was mixed by vortexing
every 2 min during the incubation to keep the QIAEX II in suspension. The
sample was centrifuged for 30 sec at 10,000g and the supernatant was discarded.
The pellet was washed with 500 pl of Buffer QXl, resuspended by vortexing,
centrifuged at 10,000g for 30 sec, and then the supernatant was discarded. The
pellet was washed twice with 500 pl of Buffer PE as described above. After
removing the supernatant from the second washing, the pellet was allowed to air-
dry until it was white and dry. The DNA was eluted by adding 20 pl of sterilized
ddH20 to the pellet and incubating at RT for 5 min. The sample was centrifuged
for 30 sec at 10,000g and the supernatant, which contains the DNA, was pipetted
into a sterile microfuge tube. Elution was repeated a second time to recover any
remaining DNA and to increase yield.

The isolated TGFBZ DNA fragment was then ligated into the isolated

pBSK' that had been digested, using T4 DNA ligase (Boehringer Mannheim,

50

Indianapolis, IN). pBSK', 0.1 pg, and an equimolar amount of the TGFBZ DNA
fragment were incubated in a sterile microfuge tube with 7.5 pl of sterile ddeO,
0.1 Weiss unit of T4 DNA ligase, 1 pl 10X ligation buffer (supplied with the
enzyme), and 1 pl of 5 mM ATP. Two control reactions were conducted: one
using the pBSK' vector alone and another using only the TGFBZ DNA fragments.
The samples were incubated at 16°C overnight. The ligated samples were
transfected into E. coli following the protocol described in Section 3.8.2, with
modifications. LB/Amp plates, used to culture the bacteria, were treated with 40
p1 of 20 mg/ml 5-Bromo-4-chloro-3-idolyl-B-D-galactopyranoside (X-
gal)(Boehringer Mannheim) and 4 pl of 250 mg/ml Isopropyl-B-D-
thiogalactoside (IPTG) (Boehringer Mannheim) for Blue/White screening. The
recombinant colonies (white) were selected for further analysis. The plasmid

was purified following the protocol outlined in Section 3.8.3.

3.8.5 Synthesis of TGFBZ riboprobe

The purified plasmid was digested with Kpn I (New England Biolabs) for
sense probe and Eco RV (New England Biolabs) for antisense probe synthesis as
follows: 5 p1 of purified plasmid, 3 pl of 10X buffer (supplied with the enzyme),
1 pl ofthe enzyme, 0.2 pl of bovine serum albumin (BSA) and 11.8 pl of diethyl
pyrocarbonate (DEPC) treated ddH20 was placed in a DEPC treated microfuge
tube and incubated in a water bath at 37°C for 2.5 hr. To verify complete

digestion ofthe plasmid, 2 pl ofthe digested plasmid was mixed with 3 pl of

51

DNA loading buffer, subjected to electrophoresis on 1% agarose gel, and stained
with ethidium bromide. The digested plasmid was stored at -20°C until use.

In vitro transcription was carried out using Ambion T3 and T7
transcription kits (Ambion Inc., Austin, TX), for the preparation of sense and
antisense probes, respectively. Two p1 of digested plasmid, 2 pl of 10X
transcription buffer, 2 pl ATP, 2 pl CTP, 2 pl GTP, 2 pl DigUTP2UTP (1 part
digoxigenin-l l-UTP (Roche Molecular Biochemicals, Indianapolis, IN):2 parts
UTP), 6 pl DEPC treated H20, 1 p1 RNase inhibitor, and 1 pl oftranscription
enzyme was added to DEPC treated microfuge tubes and incubated in a water
bath for 3 hr at 37°C. The probes were subjected to polyacrylamide gel
electrophoresis (PAGE) stained with ethidium bromide to determine the size and
concentration of the final products.

The samples were then treated with 1 pl of DNase (Ambion Inc.) at 37°C
for 20 min and purified by EtOH precipitation as follows: 3 volumes of cold
EtOH, 10 pg tRNA (Boehringer Mannheim), and 1/10‘h volume of sodium
acetate, 3M, pH 5.2 was added to the sample and stored at -20°C overnight. The
sample was then centrifuged for 20 min at 4°C and 10,000g and the supernatant
decanted. The sample was washed with 70% EtOH, dried in a Cenco Hyvac 14
speedvac (Central Scientific Company, Chicago, IL), and the pellet dissolved in
100 pl DEPC-H20. The probes were subjected to PAGE and stained with

ethidium bromide to determine the size and concentration ofthe final product.

52

3.8.6 Analysis of TGFBZ expression by whole-mount in situ hybridization
Embryos were dissected from extraembryonal membranes and fixed in 4%
PFA in PBS overnight at 4°C, then dehydrated through an ascending MeOH
series (25%, 50%, 75%, 100%) and stored at -20°C. Embryos were rehydrated
through a descending MeOH series (75%, 50%, 25%) at 30 min intervals and
then washed 2 times with 1% polyoxyethylene sorbitan monolaurate (Tween 20)
in PBS (PBT), 10 min each. The embryos were postfixed in 4% PFA/PBS for 30
min and then washed with PBS 3 times, 10 min each. Hybridization solution
[50% Formamide (Roche Molecular Biochemicals; Indianapolis, IN), 5X SSC
(20X: 3M NaCl, 0.3M sodium citrate, pH 7.0), 2X Denhardt’s (50X: 0.5g
Ficoll, 0.5g polyvinyl pyrrolidone, 0.5g BSA in 50 ml ddHZO), 1% sodium
dodecyl sulfate (SDS), 100 pg/ml DNA, 100 pg/ml RNA, and 50 pg/ml heparin]
was used to wash the embryos for 10 min, to prehybridize the embryos for 4 hr at
55°C, and to incubate the embryos with the specific riboprobe at a 1:1000
dilution overnight at 59°C. Embryos were washed 2 times at 60°C with 2X SSC,
1% SDS in PBT, 1 hr each. Then, 0.5X SSC, 1% SDS in PBT was used to wash
the embryos first at 65°C and then at 70°C, 90 min each. TBST (137mM NaCl,
3mM KCl, 25mM TrisCl pH 7.5 and 0.1% Tween 20) containing 10% heat-
inactivated goat serum and 4% dry nonfat milk was used to block non-specific
binding by incubating for 3 hrs at RT. Embryos were then incubated with anti-
digoxigenin alkaline phosphatase (Roche Molecular Biochemicals; Indianapolis,
IN) at a 1:2000 dilution in TBSTcontaining 0.5 mg/ml levamisole overnight at

4°C. The antibody was preabsorbed with 1% embryo acetone powder. Embryos

53

were washed 8 times, 30 min each, with TBST containing 0.5 mg/ml levamisole
at RT and then overnight at 4°C. Embryos were washed with TBST containing
0.5% levamisole 2 times for 10 min and then 2 times with NTMT (100mM NaCl,
50mM MgClz, 100mM TrisCl pH 9.5,0.1% Tween 20, and 0.5 mg/ml levamisole)
for 10 min each. Embryos were incubated with 4.5pl 4-nitroblue tetrazolium
chloride (Roche Molecular Biochemicals; Indianapolis, IN) and 2.3pl 5-bromo-4-
chloro-3-indolyl-phosphate (Roche Molecular Biochemicals; Indianapolis, IN)
per ml of NTMT at RT in a dark place until color had developed. The embryos
were washed 3 times with PBT at RT, 30 min each, and then stored in 60%
glycerol in PBS at 4°C. Pictures were taken ofthe embryos using a Nikon 6006
camera attached to a Nikon SMZ-2T light microscope. Experimental embryos
were incubated with antisense probes, while control embryos were incubated

with sense probes.

3.8.7 Cellular analysis of TGFBZ expression by cross-sectional examination
The embryos used for whole-mount in situ hybridization were sectioned
according to the protocol outlined in Section 3.4 with modifications. The
embryos were not stained with the Toluidine Blue solution; instead they were
placed directly in isopropanol after dehydration in EtOH. The embryos were
embedded in paraffin, sectioned, and deparaffinized. The sections were examined
using a Nikon Optophot-2 compound microscope. The microscope was equipped

with a Nikon 6006 camera, and pictures were taken of representative sections.

54

3.9 Analysis of TBRII expression by in situ hybridization
3.9.1 Synthesis of TBRII riboprobes

cDNA for TBRII was a gracious gift from Dr. Barnett at Vanderbilt
University. The cDNA was obtained on filter paper and the plasmid was
recovered as described in Section 3.8.1. The plasmid was then transfected into
E. coli as outlined in Section 3.8.2. The cDNA for TBRII was cloned into the
xbal site of pcDNA3. The vector, pcDNA3, contains mRNA promoters, so it
was not necessary to sub-clone the cDNA into pBSK‘.

The protocol outlined in Section 3.8.5 was used to synthesize riboprobes,
with alterations: First, BLAST was used to check for homology between TBRII
and other known genes in the NIH database. A unique sequence was chosen that
had no known homology with any other avian genes. The plasmid was digested
with pvu I (New England Biolabs) for both sense- and antisense probe formation
at the same concentrations specified previously. In vitro transcription was
completed using Ambion T7 and Sp6 transcription kits for sense and antisense
probe formation, respectively, as described in Section 3.8.5.

The sense probe was treated with 1 pl of DNase at 37°C for 20 min and

then purified by EtOH precipitation. The resultant pellet was dissolved in 100 p1
DEPC-H20. The probe was subjected to PAGE stained with ethidium bromide to
determine the size and concentration ofthe final product.

The anti-sense probe, ~1.8kb, was subjected to alkaline hydrolysis to
reduce the average probe length (Cox et al., 1984). One pl of DNase and 1 pl of

RNase were added to the anti-sense probe solution formed during in vitro

55

transcription. The sample was diluted to 100 pl using DEPC-H20. Hydrolysis
buffer (80 mM sodium bicarbonate, 120 mM sodium carbonate, 20 mM [3-
mercaptoethanol), 100 pl, was added to the sample and incubated in a 60°C
water bath for 30 min. The sample was removed from the water bath and 200 pl
of Stop buffer (0.2 M sodium acetate pH 6.0, 1.0% glacial acetic acid, 10 mM
dithiothreteitol) was added to the microfuge tube to terminate the reaction. The
probe was purified by EtOH precipitation as previously described, except 1 pl of
tRNA was added to the reaction. The resultant pellet was dissolved in 100 pl of

DEPC-H20. The probes were subjected to PAGE and stained with ethidium

bromide to determine the final size and concentration.

3.9.2 Analysis of TBRII expression by whole-mount in situ hybridization
Experimental embryos were hybridized with antisense probes and control
embryos were hybridized with sense probes, described above. The protocol
outlined in Section 3.8.6 was followed for in situ hybridization. Pictures were
taken ofthe embryos using a NikOn 6006 camera attached to a Nikon SMZ-2T

light microscope.

3.9.3 Cellular analysis of TBRII expression by cross-sectional examination
Embryos used for whole-mount in situ hybridization were sectioned
according to the protocol outlined in Section 3.8.7. The sections were examined
using a Nikon Optophot-2 compound microscope. The microscope was equipped

with a Nikon 6006 camera and pictures were taken of representative sections.

56

3.10 Antisense oligonucleotide blocking of TGFBZ

A unique sequence for TGFB2 (5’-GCA CAG AAG TTG GCA TTG TAT
CCT TTG GGT TC-3’) previously described by Jakowlew et al. (1990) was
chosen for the synthesis of antisense oligonucleotides. The oligonucleotides
were synthesized at the Macromolecular Structure Facility at Michigan State
University. The lypholized oligonucleotides were reconstituted in 100 pl of
Tyrode’s solution. The concentration ofthe oligonucleotides was determined in
a Spectronic 21D spectrophotometer and the oligonucleotides were diluted to a
working concentration of 50 pmol in Tyrode’s solution. The oligonucleotide
solution was sub-blastodermally injected into 1-2 somite stage embryos as
described in Section 3.5. The eggs were sealed and incubated horizontally at
385°C with 98% humidity for 10 to 24 hr. Normal and VAD control embryos
were injected with either the vehicle alone, or a non-specific oligonucleotide.
The embryos were dissected as described in Section 3.3 and pictures were taken
using a Nikon 6006 camera attached to the microscope. Embryos were then

fixed overnight in 4% PFA and stained and sectioned as described in Section 3.4.

57

CHAPTER 4

RESULTS

4.1 VAD embryos display abnormal IFT development during early

cardiac morphogenesis

The VAD avian embryo dies at 3.5 d of embryonic development
(Kostetskii et al., 1999; Zile, 2001). The most pronounced developmental defect
observed in the VAD avian embryo is the abnormal cardiovascular development
(Thompson et al., 1969; Heine et al., 1985; Zile, 2001), which is likely the cause
ofthe early embryo lethality observed in the VAD avian embryo, due to the lack
of nutrients and growth factors that are normally provided to the developing
embryo via the circulation. Our laboratory is focused on studying the effects of
VAD on quail embryonic cardiovascular development and the molecular
pathways that are responsible for the characteristic phenotype of the VAD quail
embryo (Zile, 2001). Our lab has identified a critical retinoid-regulated
developmental window at 4/5 somite stage (35), at which time biologically active
Vitamin A is essential for normal embryonic development and survival
(Kostetskii et al., 1999). Currently, there has not been a systematic description
of cardiovascular abnormalities in the VAD embryo, in particular focusing on the
inflow tract which connects the embryonic and extraembryonic vascular
networks and is absent in the VAD avian embryo (Heine et al., 1985; Dersch and
Zile, 1993; Zile, 1999; Zile et al., 2000; Zile, 2001). It is imperative to

determine the developmental stages when morphological changes associated with

58

VAD occur, for subsequent studies to address the molecular events that might be
involved in these developmental processes. Therefore, we examined
morphological cardiovascular development of normal and VAD embryos starting
at the critical retinoid-regulated developmental window (4/5 55) through the time
of VAD embryo lethality (3/3.5 d), focusing on IFT development.

No abnormalities were observed in VAD embryos between 4 and 7 ss.
Cells in the HFR began to form a symmetrical, paired-tubular structure in both
the normal and VAD embryos at 4/5 ss (Figure 5). The anterior intestinal portal
(AIP) establishes the medial boundary of the HFR, as mesodermal cells
differentiate into epimyocardium, they from the lateral boundary of the HF R.
The fusion of the two heart-tubes at the mid-line (at 9 ss), determination of heart
orientation and initiation of heart beating (at 10 ss), and heart looping (at 11/12
55) were all affected in the VAD embryo (Figure 5). The fusion ofthe heart-
tubes was delayed in the VAD embryo to 10/11 ss, and the single chambered,
linear heart-tube became dilated. In the normal embryo the linear heart-tube
orients on the right side of the midline; however, the determination of heart
orientation was altered in the VAD embryo and heart orientation was randomized
(heart shown on left in Figure 5). Shortly after heart-tube fusion (11/12 ss), the
VAD heart began beating; however, the rate was slow compared to that ofthe
normal heart. The IFT connects the embryo to the extraembryonal circulation
and the nutrients stored in the yolk. In the VAD embryo, by 10 55, a narrowing

and a subsequent closure (by 11/12 ss) of the IFT was observed. The narrowing

59

ofthe IFT was the first observable morphological change in VAD cardiovascular
development.

In the normal embryo by 2.5 d of embryonic development the
extraembryonic vasculature forms a distinct network of veins and arteries for
delivering nutrients from the yolk to the embryo Via the omphalomesenteric
veins (OMV) at the IFT and via the omphalomesenteric arteries (OMA) at the
dorsal aorta to the posterior embryo (Figure 5). The contracting, two-chambered
heart, oriented on the right side of the embryo, continuously pumps blood
throughout the circulatory system. At this stage of development the entire VAD
embryo had severe defects, including stunted growth and compressed somites. A
single chambered heart with a dilated cavity and with randomized orientation
was observed in the VAD embryo. Blood islands were present at the periphery
ofthe area vasculosa: there was limited formation of blood vessel endothelium
and there was no organization of cells into vascular networks. The IFT had
closed, and there was a complete absence of OMV and OMA in the VAD embryo
(Figure 5). The VAD heart continues to beat until 3-3.5 d, but there are no
significant changes in embryo morphology during the last day of embryonic
development.

Administration of atRA or all-trans-retinol before or during the retinoid-
regulated developmental window (4/5 ss) completely rescued the VAD
phenotype (Figure 5). Heart looping in the rescued embryos was normal and the
heart was oriented asymmetrically on the right side of the embryo. Blood vessels

coalesced into a well-defined vasculature, which connected to the embryo via the

60

OMV at the cardiac IFT and the OMA at the dorsal aorta. The rescued embryos
were normal in size and in all other developmental aspects. Administration of
biologically active retinoids to normal embryos at the concentrations used for
rescuing VAD embryos, had no observable effects on development. To verify
the role of atRA as the regulatory molecule in avian embryonic development,
normal embryos were treated with anti-atRA MoAb. Microinjection of anti-atRA
MoAb into normal embryos before the retinoid-regulated developmental window
resulted in embryos displaying the complete VAD phenotype (Figure 5),
including formation of a ballooned, single chamber heart, with randomized
orientation, and an absence of vascular networks in the extraembryonal area.

The OMV at the IFT and the OMA at the dorsal aorta were absent (Figure 5).
Administration of anti-all-trans-RA MoAb to VAD embryos at similar stages had
no further deleterious effects on development. Normal and VAD control
embryos injected with vehicle had no effect on development. A small percentage
of normal and VAD control embryos injected with mouse IgG had non-specific
abnormalities, i.e. the heart was positioned above the head, or spontaneous death

occurred; however, the majority of embryos had no developmental defects.

61

”U
("I

62

. kart s 4&4.”- .-

    

Figure 5.

, 2:90:23

Q<> 2: 53» 283550 69,830 2m worm—«88:? SEongEmo £202 «£3-55 wEuoomE .3 oxSEo 35.8: 2: E
@8303 E accuse <m .: .EoEQEoEc 53355630 8382 3303500 .mEoEHoH 038m EREMEEQ 5:5 mm m)» .3
mobnEo Q<> mo :28an dEEEoEo oEobnEombxo of 8 obnEo 2: 32:80 3 A<EOV mototm otoEomoEEmEmEo
Ea A>§Ov mEo> otoucomoEEEQEo v32 mobnEo Q<> .mobnEo 1:50: mo :3: coponfimnoécfi 05 8

SEES E 6 Wm 8 Engaged—mas Becca—En a E am 2:. .mm 2 .3 @320 8m ELEV Eon: BoCE 2: EB 32 :o .3653
932:5 Q<> 05 E Bonito—GEE E GEEEES Cmv two: .0585 Q<> 2: E mm 3:: 8 wean—ow 6.0580 RES:

2: E mm m S BEES 2: um 83 323.23: 23. .mosEEoo “guano—gov mm 3:253 @538 can 333 Eton REESE
Sigma 2: on: 372$ comm»: mEESmfiao: of E EEOAEE two: .8825 05. .mm m? E 3.39:8 :95 388:

can Q<> E 32030 moocoaotmv EEmEonfioE 0: En 82E. .moinEo 23c Q<> E EoEQEoEt teem .m 35w:—

63

4.2 QH-l monoclonal antibody is a marker for presumptive endothelial

cells in normal and VAD embryos

The development of the vascular system requires differentiation of
mesodermal cells in the HFR to endocardium (Coffin and Poole, 1988), as well
as differentiation of mesodermal cells in the area vasculosa to angioblasts
(Pardanaud, 1987). Inductive signals, i.e. growth factors, from the endoderm are
required for the differentiation of mesodermal cells from both cell populations
into presumptive endothelial cells (Wilt, 1965; Sugi and Lough, 1994).
Mesodermal cells in the HFR region give rise to epimyocardial cells, as well as
the endocardium; presumptive endothelial cells are distinguishable from
premyocardial cells by the production of QH-l and the expression of flk (Linask,
1992; Linask and Lash, 1993). In quail embryos this differentiation begins at
the head-process stage and by the one-somite stage endothelial cells are
detectable in the extraembryonic area and in the embryo in the area lateral to the
somites (Coffin and Poole, 1988). Presumptive endothelial cells migrate from
the extraembryonic area and the HFR towards the IFT, connecting at the IFT and
then forming vessels. The VAD quail embryo lacks a vascular link between the
extraembryonal and intraembryonic vascular networks at the IFT and at the
dorsal aorta. We examined vascular development using QH-l
immunolocalization in normal and VAD embryos between 1 35 (early
neurulation), when the first presumptive endothelial cells are detectable, and at
10/12 55, when we observed the first morphological defects in the VAD

cardiovascular development, i.e. the narrowing and closure of the IFT.

64

The first presumptive endothelial cells were observed at 1 ss in normal
and VAD embryos in the extraembryonic area vasculosa and in the area
pellucida, adjacent to the embryo (data not shown). QH-l positive cells rapidly
increased in both normal and VAD embryos between 1 ss and 4/5 ss. In the area
vasculosa and the area pellucida presumptive endothelial cells assembled into
loose polygonal-shaped networks, with strands of presumptive endothelial cells
surrounded by large avascular areas (Figure 6). While no morphological
differences between normal and VAD embryos were observed at this stage, VAD
embryos had fewer QH-l positive cells, most noticeable in the area pellucida
(Figure 6). The polygonal networks of presumptive endothelial cells were
incomplete and unorganized, with thin strands surrounded by large avascular
areas. In normal and VAD embryos, presumptive endothelial cells lined the
lateral edges 0f the AIP in the HFR. There were fewer QH-l positive cells in the
VAD embryo, judged by the lower intensity of fluorescence (Figure 6). A
continuous strand of QH-l positive cells ofthe dorsal aorta lined the somites by
7/8 35 in normal and VAD embryos; however, VAD embryos had fewer
presumptive endothelial cells in the dorsal aorta and their organization was
sporadic (Figure 6). By 7/8 55 vascular networks throughout the extraembryonic
area had increased in normal and VAD embryos (Figure 6). However, the
networks in the VAD embryo were sparse and there were fewer presumptive
endothelial cells in the HFR and in the extraembryonal area compared with the
normal embryo. Vascular networks in the normal embryo continued to coalesce

into progressively thicker stands of presumptive endothelial cells. As

65

differentiation of mesodermal cells continues in the HFR, a rapid increase in
presumptive endothelial cells in the HFR of the normal embryo was observed;
strands of presumptive endothelial cells had formed to connect the intra- and
extraembryonic vascular systems at the site that will later become the OMV
(Figure 6).

In normal embryos by 10/11 ss presumptive endothelial cells coalesce into
progressively thicker networks, surrounding small avascular areas. The
endocardium of the heart is connected to the extraembryonic circulation via the
OMV at the IFT. The formation of the OMA at the dorsal aorta completes the
circulation (Figure 6). Prior to closure ofthe IFT in the VAD embryo only a few
presumptive endothelial cells were seen in the posterior portion ofthe HFR and
in the area pellucida (7/8 58; Figure 6). At 10/11 ss there were relatively few
QH-l positive cells in the IFT ofthe VAD embryo, and the thin vascular
networks present in the extraembryonal region at 7/8 53 had diminished even .
more. The OMV and OMA were absent in VAD embryos (Figure 6).

Administration of atRA to the VAD embryo before or during the retinoid-
regulated developmental window completely rescued the abnormal presumptive
endothelial cell patterning (Figure 6). In the rescued embryos extraembryonal
vascular networks, OMV and OMA developed as in the normal embryo.
Conversely, microinjection ofthe normal embryo with anti-atRA MoAb before
the retinoid-regulated developmental window disrupted vascular patterning and
the vascular pattern resembled that ofthe VAD embryo; the extraembryonic

vascular networks were sparse and the OMV and OMA were disrupted.

66

Microinjection of normal and VAD control embryos with mouse IgG or vehicle

had no effect on presumptive endothelial cell localization and vasculogenesis.

67

10/11 ss 10/11 ss

 

Figure 6. Composite pictures of normal and VAD embryos from 4/5 55 to 10/ 11 SS
exposed to QH-l antibody (ventral views). Endothelial cells as well as endothelial
precursors are fluorescent (yellow). Angioblasts differentiate from mesoderm in the
heart-forming region (HFR) and congergate along the lateral edge of the anterior
intestinal portal (AIP). Simultaneously, angioblasts in the extraembryonic area
differentiate from extraembryonic mesoderm and form vascular networks (arrow).
Extraembryonic vascular networks connect with the heart (Ht) at the omphalomesenteric
veins (OMV) and with the dorsal aorta lining the somites at the omphalomesenteric
arteries (OMA). VAD embryos have fewer angioblasts, most notable in the IFT lining
the AIP; the angioblasts are poorly organized and do not form OMV or OMA. The
administration of atRA to VAD embryos completely rescues vascular
development. Treating normal embryos with anti-atRA MoAb interferes with
vascular patterning and results in a lack of connection of the heart to the extra-
embryonic circulation at the OMV

68

4.3 Enhanced TGFBZ transcripts and protein levels precede abnormal

IFT development in VAD embryos

The TGFB family ofproteins have been linked to cardiovascular
development. Furthermore, in several model systems retinoids have been
implicated in TGFB signaling (Jakowlew et al., 1992; Kojima and Rifkin, 1993;
Mahmood et al., 1995; Imai et al., 1997; Yoshizawa et al., 1998; Tsuiki and
Kishi, 1999). Additionally, we observed earlier that TGFB expression was
altered in the VAD quail embryo. To gain insight into TGFBZ function in the
normal and VAD developing embryo, we examined TGFBZ protein and mRNA
localization in normal and VAD embryos between 2/3 55 (early neurulation) and
10/12 55, when we observed the first morphological defects in cardiovascular
development ofthe VAD embryo, i.e. the narrowing and subsequent closure of
the IFT.

In 2/3 33 normal and VAD embryos we found TGFBZ transcripts expressed
in the neural tubes, neural folds, Hensen’s node, extraembryonal area, and in all
cell layers ofthe HFR, most strongly in the mesoderm lining the AIP (Figure 8).
VAD embryos expressed TGFBZ transcripts more strongly in the Hensen’s node
and in the neural folds. TGFBZ protein localization corresponded with TGFBZ
mRNA in normal and VAD embryo between 1/2 55 and 10/11 53 (Figure 7). At
4/5 53, TGFBZ transcripts increased in the HFR lining the AIP of normal and
VAD embryos, and as neurulation continued the expression oftranscripts
continued to extend caudally through the neural folds (Figure 8). Expression of~

TGFBZ transcripts and protein was increased in the posterior portion ofthe HFR

69

at this stage. Mesodermal cells in the HFR undergo differentiation into
epimyocardial and endocardial cells, and heart-tubes begin to form in the normal
and VAD embryo at 6/7 ss. At this stage, TGFBZ transcripts in the HFR were
expressed in the endoderm lateral to the, AIP and the epimyocardium, which
forms the lateral boundary ofthe HFR, throughout the IFT (Figure 8). Cross-
sections through the HFR at the level ofthe IFT revealed TGFBZ transcripts in
all cell layers, including head mesenchyme, which in part forms embryonal blood
vessels (Figure 8). Endothelial cells in the HFR form the OMV in the lumen
between the endoderm and epimyocardium. TGFBZ transcripts were expressed in
the endoderm and the epimyocardium surrounding the lumen between the two
cell layers (Figure 8) in normal embryos. VAD embryos expressed increased
levels of TGFBZ transcripts in the epimyocardium of the HFR, particularly the
posterior portion ofthe IFT, as well as in the head mesenchyme (Figure 8). Prior
to IFT closure, TGFBZ expressing cells had almost completely filled the lumen
between the endoderm and the epimyocardium in the IFT. TGFBZ protein was
also increased in the posterior IFT at this time (data not shown). Morphological
closure ofthe IFT was observed at lO/12 ss in the VAD, at the same time when
the expression of TGFBZ transcripts and protein was increased in the IFT and
caudal to the Hensen’s node (Figure 7 & 8), as it was at 6/7 ss. Head
mesenchyme continued to express increased levels of TGFBZ mRNA. The lumen
between cells expressing TGFBZ in the epimyocardium and the endoderm in the
IFT was becoming very narrow, and was completely closed in the posterior IFT

region (Figure 8). Normal embryos expressed TGFBZ transcripts and protein in

70

the endoderm and the epimyocardium ofthe IFT. At this developmental stage
the OMV connect the embryo to the extraembryonic circulation and there are no
TGFBZ expressing cells in the lumen where the OMV form (Figure 8).
Administration of atRA to VAD embryos before or during the retinoid-
regulated developmental window rescued TGFBZ expression, i.e. TGFBZ
expression was down-regulated and resembled that of the normal embryo (Figure
8). TGFBZ transcripts in the head mesenchyme were reduced and there was a
large lumen between the epimyocardium and the endoderm. In contrast,
microinjection of normal embryos with anti-atRA MoAb before the retinoid-
regulated developmental window increased TGFBZ transcripts, especially in the
IFT (Figure 8). Blocking the RA with anti-atRA MoAb in these embryos
resulted not only in increased TGFBZ transcripts in the IFT, but also caused the
lumen between the TGFBZ expressing cells to narrow, and resemble the
morphology characteristic ofthe VAD embryo. When normal and VAD control
embryos were injected with mouse IgG or vehicle, TGFBZ expression was not

altered.

71

3/4 ss

10/11 ss

 

Figure 7.

72

Figure 7. Immunolocalization of TGFBZ protein in normal and VAD quail
embryos. At 3/4 55 TGFBZ protein is ectopically expressed in the
extraembryonic area. TGFBZ-specific immunofluorescence (red) is observed in
the neural tubes (NT) of both normal and VAD embryos at comparable levels. TGFBZ-
specific immunofluorescence is increased in the posterior heart-forming region (HFR)
and caudal to the Hensen’s node (Hn) in theVAD embryo at 3/4 ss. At 10/11 53 T GFBZ
is localized in the periphery of the developing inflow tracts (IFT) in the normal embryo
and the absence of fluorescence at the site of the OMVs indicates that the IFT are open,
while in the VAD embryo TGFBZ-specific fluorescence is increased in the [FT and the
IFT are narrowed at the posterior ends. All views are ventral

73

10/11 ss 8/1055
._ ‘r

 

 

7«' . ~ \

HM HM " ‘-
fik.,';\ \. x ' r" '
IL} -K/l‘ 1’ .. "

l "y l “-4,

l \ w ‘ ~ ‘

J I Q" ' " ‘.
\ -L F.\ \ L"/—IFI'
5” EP "

1 Fl‘

     

HM .—.- .

ml— / ,1, .7 . j «6"» pt“

. . _ Lil-3P”), 5("‘ ‘l‘?

lFT— ‘ ' \ : iTWP

_/—F.\' ‘_ ' ', l" ”a. _.
u’ W" EN

Figure 8. Expression patterns of TGFBZ in whole-mounts of normal and VAD quail
embryos, as well as in transverse sections through the inflow tract (IFT). Note the
increased expression caudal to the Hensen’s node (Hn) and in the heart-forming region
(HFR) lining the anterior intestinal portal (AIP) of the VAD embryo beginning at the 4/5
ss. Increased expression in the IFT continues through 10/11 ss. The expression of
TGFBZ in the neural tubes (NT) of the VAD embryo preceded the observed expression in
the normal embryo. As noted by transverse sectioning, at the level noted by the black
horizontal lines bilaterally, TGFBZ is expressed in all cell layers (6/7 85). In
VAD embryos, increased TGFBZ trascripts were observed in the head
mesenchyme (HM); cells expressing TGFBZ filled the IFT between the
epimyocardium (EP) and the endocardium (EN). Administration of atRA to VAD
embryos reduced expression of TGFBZ in the HM and the IFT, and the IFT were
open. Treating normal embryos with anti-atRA MoAb resulted in increased
TGFBZ expression in the HM and the IFT, and the IFT were closed. All views
are ventral.

74

4.4 TBRII transcripts are elevated prior to abnormal IFT morphogenesis

in VAD embryos

TBRII is one the transmembrane receptors for the TGFBS and is the initial
binding site for the mature ligand (Massague and Chen, 2000). As shown above,
TGFBZ transcripts and protein were increased in the IFT of VAD embryo. We
examined TBRII, the principal receptor for TGFBZ, at the same developmental
stages that we analyzed TGFBZ expression (2/3 ss-lO/ll 55), to determine if
receptor expression was affected in VAD embryos, and if the receptors
localization corresponded to TGFBZ expression patterns. In both normal and
VAD embryos expression of TBRII transcripts mimicked the expression pattern
of TGFBZ transcripts at all stages examined (Figure 9). At 2/3 ss TBRII was
expressed in the neural tubes, neural folds, Hensen’s node, and in all cell. layers
ofthe HFR, with increased expression in VAD embryo caudal to the Hensen’s
node at all stages examined. Expression of TBRII in the neural tubes extended
caudally, as neurulation progressed. At 6/7 55 in the normal embryo TBRII was
expressed in the epimyocardium and the endoderm of the IFT, surrounding the
developing OMV; by 10/11 ss TBRII transcripts were observed in the
epimyocardium and the endoderm surrounding the large lumen ofthe developing
IFT (Figure 9). In the VAD embryo, TBRII transcripts were increased in the
HFR at 4/5 55 and continued through 11 53 (Figure 9). Similarly, increased
expression in the head mesenchyme and in the epimyocardium, as well as a

narrowed lumen ofthe IFT, where the OMV typically connect to the heart, was

75

observed in the VAD embryo beginning at 6/7 35 and continued through the time
oleT closure at 10/11 ss.

Administration of atRA to VAD embryos before or during the retinoid-
regulated developmental window completely rescued TBRII expression and the
the IFT were open (Figure 9). Microinjection of anti-atRA MoAb into the
normal embryo before the retinoid-regulated developmental window, increased
TBRII expression, in particular in the IFT and resulted in a narrow lumen ofthe
IFT, and an expression pattern characteristic of the VAD embryo (Figure 9).
Administration of mouse IgG or vehicle to normal and VAD control embryos had

not effect on TBRII expression.

76

2/3 ss 4/5 ss 7/8 55 10/11 ss 9/10 ss

  

   

I ‘ or ,. ».j‘.\

+3;

, 4,; J
IFT E?

 

—. ‘ \.~
.4“ l
IFT

El’ 5‘

./;’\ \

    

,J/ ‘ RH.“
. '. ' ‘ EP
- L. .

t

Figure 9. Expression patterns of TBRII in whole-mounts of normal and VAD quail
embryos, and in transverse sections through the inflow tract (IFT). TBRII expression
patterns correspond with TGFBZ expression patterns in both normal and VAD embryos.
VAD embryos have increased TBRII expression beginning at 4/5 ss and continuing
through 10/11 ss. TBRII expression is increased in the IFT of VAD embryos beginning at
7/ 8 ss, prior to lFT closure. Transverse sectioning, at the level noted by the lack
horizontal lines bilaterally, show a narrowing of the lumen between the EP and the EN in
VAD embryos, and increased TBRII expression in the HM. Administration of atRA
to VAD embryos reduced expression of TBRII in the HM and in the IFT, and the
IFT were open. Treating normal embryos with anti-atRA MoAb resulted in
increased TBRII expression in the HM and the IFT, and the IFT were closed. All
views ventral. Anterior intestinal portal, AIP; endocardium, EN;
epimyocardium, EP; head mesenchyme, HM; heart-forming region, HFR;
Hensen’s node, Hn; neural tubes, NT.

77

4.5 TGFBZ mediates IFT development in quail embryos

As shown above, the initial morphological defect in VAD cardiovascular
deve10pment is the narrowing and closure ofthe IFT at 10/12 ss. This effect was
more apparent when vascular development was examined, noting that the lack of
a vascular link at the IFT preceded the morphological closure ofthe IFT. We
have also shown that VAD embryos have increased expression of TGFBZ protein
and mRNA, and TBRII mRNA in the IFT, prior to IFT closure. To determine if
the elevated level of TGFBZ was specifically involved in IFT closure in the VAD
embryo, we injected normal and VAD embryos at 2/3 85 with antisense
oligonucleotides specific for TGFBZ and allowed the embryos to develop until 9-
ll ss. VAD embryos injected with TGFBZ antisense oligonucleotides developed
normal, i.e. open, IFT (Figure 10). Treatment with TGFBZ antisense
oligonucleotides; however, did not normalize the other abnormalities of the VAD
heart. TGFBZ antisense oligonucleotides injection increased the level of
presumptive endothelial cells in the IFT of VAD embryos resulted in formation
of OMV that connected the heart to extraembryonic vascular networks; however,
the networks were sparse and poorly organized (Figure 10, QH-l). TBRII
expression was down-regulated in VAD embryos injected with TGFBZ antisense
oligonucleotides, specifically in the IFT (Figure 10). Similarly, TGFBZ
transcripts were decreased especially in the IFT, which resembled the open IFT
of normal embryos with TGFBZ expressing cells surrounding the OMV and the
IFT developing normally, unlike the narrowed IFT of untreated VAD embryos

(Figure 10).

78

Normal and VAD control embryos were injected with a random, non-
specific antisense oligonucleotide. A small percentage (<5%) of both normal and
VAD control embryos had abnormalities not associated with the VAD phenotype,
possibly due to the toxic nature of degraded nucleotides; in general there was no
_ effect on development. Embryos that exhibited these characteristic!abnormalities
were excluded from the sample group. Normal embryos injected with TGFBZ
antisense oligonuleotides at the concentration used in the current study

developed normally.

79

VAD + TGFBZ As

in. ‘1.

 

TBRII

 

Figure 10.

80

Figure 10. Blocking TGFBZ transcripts in the VAD embryo by injecting antisense
oligonucleotides to TGFBZ. The heart (Ht) of VAD injected with TGFB2 antisense
oligonucleotides was unaltered. Gross morphological examination as well as transverse
sectioning at the level of the black lines located bilaterally, showed that the inflow tracts
(IFT) remained open. VAD embryos injected with TGFBZ antisense oligonucleotides,
immunolabeled with QH-l, showed increased presumptive endothelial cells,
specifically in the IFT, and the presence of omphalomesenteric veins (OMV).
Vascular networks (arrow) in the extraembryonic area were better organized than
in the VAD embryo; there were thin strands of presumptive endothelial cells
extending from the extraembryonic area to the OMV; however, the vascular
networks were not as organized, or robust as in normal embryos. Expression of
TGFBZ transcripts was specifically decreased in the IFT when VAD embryos
were treated with TGFBZ antisense oligonucleotides. Furthermore, treatment of
VAD embryos with TGFBZ antisense oligonucleotides had the secondary effect
of decreasing TBRII transcripts in the IFT to levels comparable with the normal
embryo. All views ventral

81

CHAPTER 5

DISCUSSION

It is well known from work in our laboratory and earlier studies that
vitamin A is required for early avian development. In the VAD avian embryo,
the earliest gross morphological abnormality is observed in the heart, which is a
ballooned, non-compartmentalized single chamber structure randomly oriented
and lacking IFT; therefore, without a link to the extraembryonal circulation.

The role of vitamin A in cardiovascular development is not understood.
Our laboratory has addressed this question by examining the impact of VAD on
cardiogenic genes, including heart asymmetry genes (Zile et al., 2000).
However, no studies have been conducted on the role of vitamin A in IFT
formation in the posterior heart region. The work in this thesis is the first to
address specifically the question of how vitamin A regulates IFT formation, and
to identify a gene, i.e. TGFBZ, as a downstream effector in the retinoic acid
regulated pathway for IFT formation.

An association between RA signaling and the establishment of the
posterior heart regions during early development is supported by work in our
laboratory (Dersch and Zile, 1993; Twal et al., 1995; Kostetskii et al., 1999) and
the observation that excess RA in the chick causes the domains of atrial myosin
heavy chain (MHCl) expression to expand anteriorly, at the expense of
ventricular MHCl expression (Yutzey et al., 1994). Furthermore, retinaldehyde

dehydrogenase, RALDHZ, the earliest known RA synthesizing enzyme in the

82

mouse embryo, as well as RA response are initially restricted to the posterior
regions of the developing mouse heart, suggesting a role of RA in the
specification of cardiac inflow structures (Moss et al., 1998).

IFT development is a critical event in early embryogenesis and is a
complicated process that requires differentiation of mesodermal cells in the HFR
to epimyocardium and endocardium, migration of cells from the extraembryonic
area towards the HFR and of cells from the HFR towards the extraembryonic
vascular networks, which fuse together to form the basis of the circulatory
system; proliferation and apoptosis are also a part ofthese processes. Clearly,
many regulatory processes are involved, and it would be logical that the RA
regulated pathway would involve multifunctional growth regulators. Therefore,
TGFBs are likely candidates to be involved in RA-regulated cardiovascular
development; furthermore, RA regulates TGFB in many cell types.

Previous work in our laboratory using Northern blot, showed that TGFB
mRNA was up-regulated in VAD quail embryos (Kostetskii and Zile, 1993);
however, these findings gave us no indications of where TGFB was located in the
developing embryo. If TGFBZ were to be involved in heart development, we
would expect it to be present in the HFR. There is only limited data regarding
TGFBZ expression in the heart during early cardiac morphogenesis. At later
stages of cardiogenesis, after the heart has begun looping, TGFBZ mRNA is
expressed in the myocardial layer of the heart tube (Mahmood et al., 1995) and

TGFBZ protein is localized in all layers ofthe heart (Mahmood et al.,l992). In

chicken embryos at stage 14, both TGFBZ mRNA and protein were localized in

83

the myocardium and endothelium of the heart (Boyer et al., 1999; Barnett et al.,
1994). Chicken embryos exhibit similar expression patterns to those in mice
suggesting that TGFBZ expression, and possibly function is conserved across
species. Studies in mice have provided the only descriptive analysis of TGFBZ
expression during early cardiac morphogenesis showing that TGFBZ mRNA is
expressed in the promyocardium of 1-5 55 mice (Dickson et al., 1993). This
result is consistent with our observation in both normal and VAD avian embryos,
which express TGFBZ mRNA in the heart splanchnic mesoderm during early
neurulation. In mice embryos at 5-7 ss, TGFBZ mRNA is expressed in the
promyocardium of the vitelline veins and sinus venosus ofthe IFT, as well as in
the foregut endoderm (Dickson et al., 1993). Similarly, we observed TGFBZ
mRNA in the epimyocardium ofthe IFT and in the foregut endoderm in stage-
matched embryos. This expression pattern is conserved in the heart through the
initiation of the heart beating at HH 10 or 8.5 d in the quail and mouse,
respectively. Despite the early appearance of TGFBZ mRNA, TGFBZ protein is
not detected until 8.5 d in the mouse and then it is expressed mainly in the
cardiomyocytes (Dickson et al., 1993). In our current work we detect TGFBZ
early in neurulation, by 1/2 53 and continuing throughout the stages we examined
(until 10/11 55); however, recent work with chicken embryos using antibodies
specific for active TGFBZ protein did not find active TGFBZ in the HFR at 10/11
ss (McCormick, 2001). However, in the same study, using ELISA, TGFBZ was
observed in whole-heart homogenates of 10/11 ss chicken embryos with 70% of

the TGFBZ in the active form. The levels of protein were shown to increase with

84

development, but the percentage of active TGFBZ decreased with increased
production of protein. Using whole mount embryos, we were able to show for
the first time that TGFBZ is located in the HFR and IFT of the deve10ping normal
avian embryo during the early stages of cardiac morphogenesis. Normally
TGFBZ is expressed at low levels, but the absence of vitamin A caused an over-
expression of TGFBZ throughout the embryo, but especially in the IFT. The
concentration of antisense oligonucleotides used to block TGFBZ in VAD
embryos was sufficient to rescue IFT development, but it did not rescue other
abnormalities associated with the VAD phenotype, suggesting a specific function
in IFT development. Normal embryos injected with identical concentrations of
antisense oligonucleotides exhibited no alterations in IFT development. This
suggests that low levels of TGFBZ specifically control normal, RA-regulated IFT
development. Our findings are in agreement with the data showing that active
TGFBZ is present in the chick heart at the 10/11 ss (McCormick, 2001), and add
support for the role of TGFBZ in early cardiac morphogenesis.

Due to the complex nature of TGFBZ signaling under physiological
conditions, only limited data is available regarding the interaction of RA and
TGFBZ pathways in the heart. RA deficiency in mice results in down-regulation
ofTGFBZ protein, while mRNA levels are unaltered (Mahmood et al., 1995),
whereas treatment with excess RA decreases TGFBZ protein in all embryonic
tissues (Mahmood et al., 1992). In support of our present observations is the
finding that knocking out RXRor results in increased TGFBZ protein in the

cardiac outflow tract of developing mice (Kubalak et al., 2002). The ability of

85

-<..~._.—.-...

RA to decrease and normalize TGFBZ mRNA expression in VAD quail embryos,
as well as the alteration of normal expression of TGFBZ when normal quail
embryos are injected with anti-atRA MoAb, indicates that RA, the active form of
vitamin A, regulates TGFBZ expression during avian embryogenesis. However,
the mechanism remains to be elucidated. Our overall findings are in line with
those of others, indicating that RA interacts with TGFBZ during embryogenesis
suggesting that RA interaction with TGFBZ is preserved across species.

TGFBZ signaling is initiated via the binding ofthe ligand to TBRII, the
primary receptor. TBRII mRNA is present in the developing chicken heart and
its expression increases with developmental time (Barnett et al., 1994).
However, TBRII localization in the developing heart has not been thoroughly
studied. Our current work with the quail embryo is in agreement with the above
studies in chicken as we find TBRII mRNA expressed in the embryonic heart and
find this expression increased with advancing developmental stages. Our work
here is the first to analyze the cellular expression of TBRII in early heart
morphogenesis, and to examine it in relation to TGFBZ. TBRII mRNA is
expressed in all layers ofthe developing quail heart and in the IFT, and is
expressed in the same cells that express TGFBZ. The colocalization of TBRII
with TGFBZ is in agreement with the current understanding that TGFBZ works
through either autocrine or paracrine signaling mechanisms (Dickson et al.,
1993; Boyer et al., 1999). As was the case with TGFBZ transcripts,
administration of RA to VAD embryos resulted in decreased TBRII transcripts in

the IFT, while injection of normal embryos with anti-atRA MoAb increased

86

TBRII, supporting the role of RA in TBRII regulation. The ability of TGFBZ
antisense oligonucleotides to decrease the expression of TBRII transcripts in
VAD embryos to normal levels, suggests that RA regulation of TBRII is a
secondary effect of RA regulation of TGFBZ. In addition to decreasing TBRII
mRNA expression, blocking TGFBZ mRNA with antisense oligonucleotides
enabled the IFT of VAD embryos to develop normally and to remain open, and
also increased the number of presumptive endothelial cells in the IFT and the
area pellucida of the VAD embryo. The specific role of TGFBZ in RA-regulated
cardiogenesis is restricted to the IFT at the developmental stages we examined,
and is supported by the finding that the other phenotypical cardiac abnormalities
ofthe VAD embryo were not affected by TGFBZ antisense oligonucleotide
blocking in VAD embryos. In the VAD embryo we see a decrease in
presumptive endothelial cells (QH-l positive) in the IFT, but generally there are
more and likely, different cells that fill the openings. These results have led us
to hypothesize that TGFBZ inhibits endothelial cells in the inflow tracts during
early cardiovascular development.

There are several possible explanations for the decreased number of presumptive
endothelial cells in the VAD embryo: abnormal differentiation, decreased proliferation,
increased apoptosis, and defective organization. We have shown that there are fewer
presumptive endothelial cells in the VAD embryo and in the extraembryonic area from
the beginning of vascularization and throughout the development of the cardiovascular
system. We observed that presumptive endothelial cell numbers in VAD embryos are

always below the level of that in normal embryos and that the addition of RA can rapidly

87

and drastically increased the number of presumptive endothelial cells; therefore, it is
likely that RA mediates differentiation of mesodermal cells into presumptive endothelial
cells, as well as mediating the proliferation of the newly generated presumptive
endothelial cells. Our observation of fewer presumptive endothelial cells in the VAD
embryo is supported by current work by LaRue and others indicating that vitamin A is
required for the generation of the proper number of endothelial cells (LaRue et al., in
preparation). In our studies we find that TGF132 expression is increased in the IFT of the
VAD avian embryo; the increase of presumptive endothelial cells in VAD embryos when
blocking TGFBZ expression with antisense oligonucleotides, suggests that during normal
embryonic development TGFBZ restricts the generation of presumptive endothelial cells.
As presumptive endothelial cells differentiate from mesodermal cells they
undergo rapid'proliferation. Several lines of evidence support the idea that
TGFB influences cell proliferation. Some studies show that TGFBI is a potent
inhibitor of epithelial cell growth (Howe et al., 1993), while in others TGFBl has
been shown to induce FGF-2, which results in fibroblast proliferation (Strutz et
al., 2001). It is clear that TGFB plays a role in cell proliferation; however, the
direction and mechanism of regulation on cell proliferation needs to be
determined empirically under different physiological conditions. It is possible
that endothelial cell proliferation is decreased in VAD avian embryos, supported
by evidence that has shown RA to stimulate proliferation of endothelial cell
numbers (Melnykovych, 1981; Junquero, 1990; Lansink, 1998). In agreement with

the role of TGFBZ in both differentiation and proliferation of the vascular networks, is the

observation that increased expression of TGFBZ can inhibit vascular development at the

88

level of differentiation and proliferation in ocular tissue (Hayasaka et al., 1998; Zhao and
Overbeek, 2001). All of the above observations support our hypothesis of a role of
TGFBZ as an growth regulator, limiting presumptive endothelial cell production in the
RA-regulated formation of the IFT, possibly via decreased differentiation of mesodermal
cells into presumptive endothelial cells and a subsequent decrease in presumptive
endothelial cell proliferation.

TGFBZ has been implicated as an inducer of cardiac muscle formation and
differentiation. The spatio-temporal localization of TGFBZ, as well as other
TGFB isoforms, in the developing mouse embryo shows characteristics expected
of a paracrine factor for cardiac muscle induction (Dickson et al., 1993).
Specifically, TGFBZ mRNA is present in cardiac precursor cells and in cells of
the foregut epithelium, which is thought to have cardiac-inducing activity
(Dickson et al., 1993), an expression pattern that we observed in both normal and
VAD quail embryos. In vitro, TGFBZ enhances cardiac muscle formation in
cultured mouse embryonic stem cells, both in the rate of induction and in the
number of muscle cells produced (Slager et al., 1993). While the above
observations suggest a role of TGFBZ in vitamin A-regulated cardiogenesis, there
is no evidence that early cardiac muscle cell differentiation is regulated via a
RA-TGFBZ signaling pathway. Morphological examination of the heart and the
IFT in VAD embryos does not reveal a thicker layer of epimyocardium
surrounding the heart-tube and the IFT, as would be expected ifthe increased
TGFBZ expression in the VAD embryo IFT were to enhance cardiac muscle cell

differentiation. Furthermore, examination of cardiac muscle cell-specific genes

89

in our laboratory has shown that expression of these genes is not affected in the
VAD embryo (Kostetskii et al., 1999). While it is possible that TGFBZ is
involved in cardiac muscle induction, it seems unlikely that the retinoid-
regulated physiological function of TGFBZ in the early avian embryo is linked to
this function.

Apoptosis is a critical process in vascular development needed for
controlling cell populations and for remodeling. TGFBZ and vitamin A have
both been implicated in apoptosis; however, the mechanisms are unclear. Recent
work has shown that reducing endogenous TGFB prevents apoptosis of neurons
in the developing retina of chick embryos (Dunker et al., 2001), while elevated
TGFBZ in RXROL knockouts enhances apoptosis in the cardiac outflow tract of
developing mice (Kubalak et al., 2002). These findings suggest that TGFB is
capable of promoting apoptosis. A closer examination of the IFT of VAD
embryos by sectioning the IFT of 10/11 ss embryos, stained to examine all cells,
revealed that the lumen of the IFT in the VAD embryos were filled with cells.
However, there were fewer presumptive endothelial cells in these IFT; thus
TGFBZ mediated apoptosis in the IFT would haveto be a cell-specific signal for
presumptive endothelial cells. Therefore, one cannot rule out an increased
apoptosis of presumptive endothelial cells in VAD embryos, possibly mediated
via excessive TGFBZ.

Another mechanism to explain the RA-TGFBZ linked regulation of
endothelial cell incorporation into the IFT would be due to an indirect effect of

lack of RA on fibronectin. As mesodermal cells differentiate into presumptive

9O

endothelial cells they migrate into strands before they form vessels (Pardanaud et
al., 1987; Coffin and Poole, 1988; DeRuiter et al., 1993). Presumptive
endothelial cells have been shown to migrate following fibronectin depostion in
the extracellular matrix (Linask and Lash, 1986; 1990; 1992); moreover, it has
been observed that fibronectin deposition is altered in VAD quail embryos
(Linask and Zile, unpublished). Furthermore, antibody neutralization of TGFB
has been shown to reduce fibronectin deposition (Sanders et al., 1993), thus
suggesting a possible role of TGFB in endothelial cell migration. Since VAD
quail embryos exhibit poorly organized vascular networks, it is possible that RA,
via specific isoforms of TGFB, mediates endothelial cell migration and
organization.

In summary, RA-regulated cardiac IFT development is mediated via a
TGFBZ signaling pathway that most likely controls endothelial cell incorporation
into the IFT structure, so as to form the link between the extraembryonic

circulation and the heart. The specific target genes of TGFBZ, as well as the

mechanism of RA regulation of TGFBZ signaling, remain to be elucidated.

91

CHAPTER 6

FUTURE WORK

In this study we have provided evidence that links TGFBZ to the
development of the IFT and suggest that TGFBZ inhibits endothelial cell
differentiation; however, more research is needed to elucidate the role TGFBZ in
IFT development. As mesodermal cells differentiate, endothelial precursors
express flk and myocardial precursors express N-cadherin. We have recently
observed increased N-cadherin in the HFR ofthe VAD embryos. Moreover, N-
cadherin is known to interact with TGFBZ in many cell systems. A proposed
study would be injecting VAD embryos at 1/2 53 with antisense oligonucleotides
to TGFBZ and collecting embryos sequentially from 30 min to 9 hr at which time
the embryos will be approximately 7/8 ss. Then in situ hybridization can be
done with biomarkers for endocardial and myocardial precursors and compare
the results with normal and VAD embryos. If TGFBZ signal initiated
differentiation into muscle, then we would observe increased N-cadherin in the
VAD compared to the normal and treatment with oligonucleotides would
decrease the appearance of N-cadherin transcripts. To confirm the increase in
presumptive endothelial cells we observed in VAD embryos injected with TGFBZ
antisense oligonucleotides by QH-l immunolocalization, flk expression could be
examined in a time course fashion. It would also be of interest to do double in

situ hybridization with N-cadherin and TGFBZ as well as flk and TGFBZ to see if

TGFBZ is colocalized with a specific cell type, in order to gain insight into the

92

mechanism of action. We have recently linked RARa to IFT development.
Blocking RAROL in the normal embryo results in closure ofthe IFT. Examining
the expression of TGFBZ in these embryos could link TGFBZ activity in IFT
deve10pment with a specific RAR.

As we noted previously, we are not able to rule out other possible
mechanisms of TGFBZ regulation of presumptive endothelial cells during IFT
development; however, there are several experiments that could be done to
examine these mechanisms. We are currently developing an embryo culture
system, which will simplify the study of migration and proliferation.
Presumptive endothelial cell migration can be examined using DiI labeling of
presumptive endothelial cells and examining their migration during IFT
development to determine if and how endothelial cell organization is affected by
vitamin A. BrdU staining of presumptive endothelial cells could give some
insight into RA mediation of angioblast proliferation. It is possible that
presumptive endothelial cells are differentiating from mesoderm and TGFBZ is
mediating apoptosis of these cells shortly after they differentiate. TUNNEL can
be used to examine apoptosis at these early developmental stages after
angioblasts differentiate.

TGFBS usually function as hetrotetramers. It will be important to
determine the other TGFB isoforms that function with TGFB2 in mediating IFT
development. The larger and more difficult question will be to try to determine

what is upstream from TGFBZ in the RA signaling pathway, as well as to

determine the target genes of TGFBZ in cardiac IFT development.

93

APPENDIX

94

Appendix 1. Additional procedures and composition of solutions

DNA loading buffer

0.25 % bromophenol blue
0.25 % xylene cyanol FF
15 % Ficoll in H20

Embryo acetone powder

Dissect Embryos

Grind in homogenizer

Suspend in PBS (1 ml per 1 g tissue)

Put tissue/salin on ice for 5 min

Add 8 ml cold acetone per 2 ml of suspension and mix
Incubate at 0°C for 30 min with occasional mixing
Centrifuge at 10,000 g for 10 min (refrigerated centrifuge)
Discard supernatant

Add fresh cold acetone and allow to sit at 0°C for 10 min
Centrifuge at 10,000 g for 10 min (regrigerated centrifuge)
Discard supernatant

Transfer pellet to filter paper and allow to dry

Store at -20°C

Gelatin coated slides

Place slides in 70% EtOH + 1 % HCl for 2 hr
Rinse 10 times in ddeO
Dip in Chromo-alum-gelatin solution

0 Heat 1L of ddeO to 65°C

0 Add 2 g gelatin

0 Dissolve and cool to 35-40°C
Allow to drain
Dip again
Allow to dry overnight at 50°C

LB media/plates

95 ml ddH20

1 g bacto-tryptone

0.5 g bacto-yeast

1 g NaCl

Dissolve, pH 7.0

Add ddHZO till 100 ml

95

Autoclave

Allow to cool

For LB/Amp add Amp after cooled to working concentration of 50 jig/ml
For LB plates add 1.5 g bacto-Agar before autoclaving

Allow to cool, add Amp, pour plates, allow to harden, flame surface

Phosphate buffered saline (PBS)
0 8.0 g NaCl

1.44 g NazHPO4

0.24 g KH2P04

0.2 g KCl

1 L ddHZO

pH 7.4

RNA loading buffer

50 % glycerol

1 mM EDTA (ethylenediamine tetraacetate)
0.25 % bromophenol blue

0.25 % xylene cyanol FF

Toluidine Blue solution
1 g Touiding Blue
20 ml 95 % EtOH

65 ml dioxane

Tyrode’s solution
0 126 mM NaCl
22 mM dextrose
1 mM Mng
4.4 mM KCl
20 mM taurine
5 mM creatine
5 mM sodium pyruvate
1 111M N3H2PO4
24.2 mM NaHCO3
- 1.08 mM CaClz

96

REFRENCES

97

Arceci, R. J., King, A. A., Simon, M. C., Orkin, S. H., and Wilson, D. B. (1993). Mouse
GATA-4: A retinoic acid-inducible GATA-binding transcription factor expressed
in endodermally derived tissues and heart. Mol. Cell Biol. 13, 2235-2246.

Ausprunk, D. H. and Folkman, J. (1977). Migration and proliferation of endothelial cells
in preformed and newly formed blood vessels during tumor angiogenesis.
Microvasc. Res. 14, 53-65.

Arens, J. F. and van Dorp, D. A. (1946). Synthesis of some compounds possessing
vitamin A activity. Nature 157, 190-191.

Barnett, J. V., Moustakas, A., Lin, W., Wang, X. F., Lin, H. Y., Galper, J. B., and Maas,
R. L. (1994). Cloning and developmental expression of the chick type II and
type III TGF beta receptors. Dev. Dyn. 199, 12-27.

Bauernfeind, J .C. (1972). Carotenoid vitamin A precursors and analogs in foods and
feeds. J. Agric. Food Chem. 20, 456-473.

Beaton, G. H., Martorell, R., Aronson, K. A., Edmonston, B., McCabe, G., Ross, A. C.,
and Harvey, B. (1994). Vitamin A supplementation and child morbidity and
mortality in developing countries. Food Nutr. Bull. 15, 282-289.

Beddington, R. S. P. and Robertson, E. J. (1999). Axis development and early
asymmetry in mammals. Cell 96, 195-209.

Blaner, W. S. and Olson, J. A. (1994). Retinol and retinoic acid metabolism. In “The
Retinoids: Biology, Chemistry, and Medicine, 2"d edition” (M. B. Spom, A. B.
Roberts, D. S. Goodman, Eds), pp. 229-255. Raven Press, New York.

Blaner, W. S., Piantedosi, R., Sykes, A., and Vogel, S. (1999). Retinoic acid synthesis
and metabolism. In “Retinoids: The Biochemical and Molecular Basis of Vitamin
A and Retinoid Action” (G. V. R. Born, P. Cuatrecasas, D. Ganten, H. Herken, K.
Starke, and P. Taylor, Eds), pp. 117-149. Springer-Verlag, Heidelberg, Germany

Blomhoff, R. (1994). Transport and metabolism of vitamin A. Nutr. Rev. 52(2),13S-
23S.

Boileau, T. W., Morre, A. C., and Erdman, J. W. Jr. (1999). Carotenoids and Vitamin A.
In “Papas AM” “Antioxidant Status, Diet, Nutrition and Health” pp. 133-158.
CRC Press, Boca Raton, FL.

Boyer, A. S., Ayerinskas, I. 1., Vincent, E. B., McKinney, L. A., Weeks, D. L., and
Runyan, R. B. (1999). TGFBZ and TGFB3 have separate and sequential activities

during epithelial-mesenchymal cell transformation in the embryonic heart. Dev.
Biol. 208, 530-545.

98

Boylan, J. F., Luftkin, T., Achkar, C. C., Taneja, R., Chambon, P., and Gudas, L. J.
(1995). Targeted disruption of retinoic acid receptor alpha (RAR alpha) and RAR
gamma results in receptor-specific alterations in retinoic acid-mediated
differentiation and retinoic acid metabolism. Mol. Cell Biol. 15, 843-851.

Brand, T. and Schneider, M. D. (1995). The TGFB superfamily in myocardium: ligands,
receptors, transduction, and function. J. Mol. Cell Cardiol. 27, 5-18.

Brockes,J.P. (1989). Retinoids, homeobox genes, and limb morphogenesis. Neuron 2,
1285-1294.

Brown, N. A. and Anderson, R. H. (1999). Symmetry and laterality in the human heart:
Developmental implication. In “Heart Development” (R. P. Harvey and N.
Rosenthal, Eds), pp. 447-461. Academic Press, San Diego.

Brown, C. B., Drake, C. J ., and Barnett, J. V. (1999). Antibodies directed against the

chicken type II TGFB receptor identify endothelial cells in the developing chicken
and quail. Dev. Dynamics 215, 79-85.

Castenmiller, J. J. and West, C. E. (1998). Bioavailability and bioconversion of
carotenoids. Ann. Rev. Nutr. 18, 19-38.

Chambon, P. (1993). The molecular and genetic dissection of the retinoid signaling
pathway. Gene 135, 223-228.

Chan, A., Hanna, M., Abbott, M., and Keane, R.J. (1996). Oral retinoids and pregnancy.
Med. J. Aust. 165, 164-167.

Chen, Y. P., Dong, D., Kostetskii, I., and Zile, M.H. (1996). Hensen’s node from
vitamin A-deficient quail embryo induces chick limb bud duplication and retains

its normal asymmetric expression of sonic hedgehog (shh). Dev. Biol. 173, 256-
264.

Chug-Ahuja, J. K., Holden, J. M., Forrnan, M. R., Mangels, A. R., Beecher, G. R., and
Lanza, E. (1993). The development and application of a carotenoid database for

fruits, vegetables, and selected multicomponent foods. J. Am. Diet Assoc. 93,
318-323.

Chytil, F. and Ong, DE. (1983). Cellular retinol-retinoic acid-binding proteins. Adv.
Nutr. Res. 5, 13-29.

Cleaver, O. and Krieg, P. A. (1999). Molecular mechanisms of vascular development.
In “Heart Development” (R. P. Harvey, Ed.), pp. 221-247. Academic Press, San
Diego.

99

Coffin, J. D. and Poole, T. J. (1988). Embryonic vascular development:
imunohistochemical identification of the origin and subsequent morphogenesis of
the major vessel primordial in quail embryos. Dev. 102, 735-748.

Cohen-Gould, L. and Mikawa, T. (1996). The fate diversity of mesodermal cells within
the heart field during chicken early embryogenesis. Dev. Biol. 177, 265-273.

Cox, K. H., Deleon, D. V., Angerer, L. M., and Angerer, R. C. (1984). Detection of

mRNAs in sea urchin embryos by in situ hybridization using asymmetric RNA
probes. Dev. Biol. 101, 485-502.

CSFII. Food Surveys Research Group (F SRG). Results from the 1994-96 continuing
survey of food intakes by individuals.

http://www.barc.usda.gov/bhnrc/foodsurvey/96result.htrnl. August 1998.

Cui, J ., Romeih, M, Michaille, J -J ., Weihong, J., and Zile, M. H. (in preparation).
Function of RARa and RARy2 at the initiation of retinoid signaling is essential
for avian embryo survival and for distinct events in cardiac morphogenesis.

DeLuca, L. M. (1991). Retinoids and their receptors in differentiation, embryogenesis,
and neoplasia. FASEB J. 5, 2924-2933.

DeLuca, L. M., Dosa, K., and Andreola, F. (1997). The role of vitamin A in
differentiation and skin carcinogenesis. J. Nutr. Biochem. 8, 426-437.

Dersch, H. and Zile, M. H. (1993). Induction of normal cardiovascular development in
the vitamin A-deprived quail embryo by natural retinoids. Dev. Biol. 160, 424-
433.

DeRuiter, M. C., Poelmann, R. E., Mentink, M. M. T., Vaniperen, L., and Gittenberger-

DeGroot, A. C. (1993). Early formation of the vascular system in quail embryos.
Anat. Rec. 235, 261-274.

Deuel, H. J ., Greenberg, S. M., Straub, E., F ukui, T., Chatterjee, A., and Zechmeister, L.
(1949). Stereochemical configuration and provitamin A activity. VII.
Neocryptoxanthin U. Arch. Biochem. 23, 239-240.

Dickson, M. .C., Lsager, H. G., Duffie, E., Mummery, C. L., and Akhurst, R. J. (1993).

RNA and protein localizations of TGFBZ in the early mouse embryo suggest an
involvement in cardiac development. Dev. 117, 625-639.

Dickson, M. C., Martin, J. S., Cousins, F. M., Kulkarni, A. B., Karlsson, S., and Akhurst,
R. J. (1995). Defective haematopoiesis and vasculogenesis in transforming
growth factor-Bl knock out mice. Dev. 12], 1845-1854.

100

Dong, D. and Zile, M. H. (1995). Endogenous retinoids in the early avian embryo.
Biochem. Biophys. Res. Commun. 217, 1026-1031.

Dorea, J. G. and Olson, J. A. (1986). The rate of rhodopsin regeneration in the bleached
eyes of zinc-deficient rats in the dark. J. Nutr. 116, 121-127.

Drake, C. J ., Davis, L. A., and Little, C. D. (1992). Antibodies to Bl-integrins cause
alterations of aortic vasculogenesis, in vivo. Dev. Dyn. 193, 83-91.

Durston, A. J ., Van Der Wees, J ., Pijnappel, W. W. M., Schilthuis, J. G., and Godsave, S.

F. (1997). Retinoid signalling and axial patterning during early vertebrate
embryogenesis. Cell Mol. Life Sci. 53, 339-349.

Dunker, N., Schuster, N., and Krieglstein, K. (2001). TGF-B modulates programmed
cell death in the retina of the developing chick embryo. Dev. 128, 1933-1942.

Eichmann, A., Marcelle, C., Breant, C., and Le Douarin, N. M. (1993). Two molecules
related to the VEGF receptor are expressed in early endothelial cells during avian
embryonic development. Mech. Dev. 42(1-2), 33-48.

Evans, T. (1997). Regulation of cardiac gene expression by GATA-4/5/6. Trends
Cardiovasc. Med. 7, 75-83.

Farrell, M. J. and Kirby, M. L. (2001). Cell biology of cardiac development. Int]. Rev.
Cytol. 202, 99-146.

Filteau, S. M., Willumsen, J. F., Sullivan, K., Simmank, K., and Gamble, M. (2000).
Use of the retinol-binding proteinztransthyretin ratio for assessment of vitamin A
status during the acute-phase response. Brit. J. Nutr. 83, 513-520.

Fishman, M. C. and Chien, K. R. (1997). Fashioning the vertebrate heart: earliest
embryonic decisions. Dev. 124, 2099-2117.

Gale, E., Zile, M., and Maden, M. (1999). Hindbrain respecification in the retinoid-
deficient quail. Mech. Dev. 89, 43-54.

George, E. L., Georges-Labouesse, E. N., Patel-King, R. S., Rayburn, H., and Hynes, R.
O. (1993). Defects in mesoderm, neural tube and vascular development in mouse

embryos lacking fibronectin. Dev. 119(4), 1079-1091. ~

Gerster, H. (1997). Vitamin A-functions, dietary requirements and safety in humans.
Int. J. Vit. Nutr. Res. 67, 71-90. '

Giguere, V., Fawcett, D., Luo, J ., Evans, R. M., and Sucov, H. H. (1996). Genetic
analysis of the retinoid signal. Ann. NY Acad. Sci. 785, 12-22.

101

Goumans, M-J., Zwijsen, A., Van Rooijen, M. A., Huylebroeck, D., Roelen, B. A. J ., and
Mummery, C. L. (1999). Transforming growth factor-[3 signalling in

extraembryonic mesoderm is required for yolk sac vasculogenesis in mice. Dev.
126, 3473-3483.

Gudas,L. J ., Spom, M. B., and Roberts, A. B. (1994). Cellular biology and biochemistry
of the retinoids. In “The Retinoids: Biology, Chemistry, and Medicine,” 2nd ed.
(M. B. Spom, A. B. Roberts, and D. S. Goodman, Eds), pp. 443-520. Raven
Press, New York, NY.

Hale, F. (1937). Relation of maternal vitamin A deficiency to microphthalmia in pigs.
Texas State J. Med. 33, 228-232.

Hamburger, V. and Hamilton, H. L. (1951). A series of normal stages in the
development of the chick embryo. J. Morphol. 88, 49-92.

Hayasaka, K., Oikawa, S., Hashizume, E., Kotake, H., Midorikawa, H., Sekikawa, A.,
Hoshi, K., Hara, 8., Ishigaki, Y., and Toyota, T. (1998). Anti-angiogenic effect
of TGFB in aqueous humor. Life Sci. 63(13), 1089-1096.

Heine, U. L, Roberts, A. B., Munoz, E. F., Roche, N. S., and Spom, M. B. (1985).
Effects of retinoid deficiency on the development of the heart and vascular system
of the quail embryo. Cell Pathol. 50, 135-152.

Hofmann, C. and Eichele, G. (1994). Retinoids in development. In “The Retinoids:
Biology, Chemistry and Medicine,” 2nd ed. (M. B. Sporn, A. B. Roberts, and D. S.
Goodman, Eds), pp. 443-464. Raven Press, New York, NY.

Hoffman, J. I. E. (1995). Incidence of congenital heart disease: 1. Postnatal incidence.
Pediatr. Cardiol. 16, 103-113.

Holmes, H. N. and Corbet, R. E. (1937). The isolation of crystalline vitamin A. J. Am.
Chem. Soc. 59, 2042-2047.

Howe, P. H., Dobrowolski, S. F ., Reddy, K. B., and Stacey, D. W. (1993). Release from

G growth arrest by transforming growth factor [31 requires cellular ras activity.
J. Biol. Chem. 268(28), 21488-21452.

Hume, E. M. and Krebs, H. A. (1949). Vitamin A requirement of human adults. An
experimental study of vitamin A deprivation in man. Medical Research Council

Special Report Series No. 264. London: His Majesty’s Stationery Office.

Humphrey, J. H., Agoestin, T., Wu, L. et al. (1996). Impact of neonatal vitamin A
supplementation on infant mortality and morbidity. J. Pediatr. 128, 489-496.

102

Ii._._-.“'
' or

m- -

‘I

Imai, S., Okuno, M., Moriwaki, H., Muto, Y., Murakami, K., Shudo, K., Suzuki, Y., and
Kojima, S. (1997). 9,13-di-cis-retinoic acid induces the production of tPA and

activation of latent TGF-B via RARor in a human liver stellate cell line, L190.
FEBS Letters 411, 102-106.

lnagaki, T., Garcia-Martinez, V., and Schoenwolf, G. C. (1993). Regulative ability of
the prospective cardiogenic and vasculogenic areas of the primitive streak during
avian gastrulation. Dev. Dyn. 197, 57-68.

Isler, O., Huber, W., Ronco, A. and Kofler, M. (1947). Synthese des vitamin A. Helv.
Chim. Acta. 30, 1911-1927.

Jakowlew, S. B., Dillard, P. J ., Sporn, M. B., and Roberts, AB. (1990). Complementary
deoxyribonucleic acid cloning of an mRNA encoding transforming growth factor-
BZ from chicken embryo chondrocytes. Growth Factors 2, 123-133.

J alal, F ., Nesheim, M. C., Agus, Z., Sanjur, D., and Habicht, J .P. (1998). Serum retinol
concentrations in children are affected by food sources of beta-carotene, fat
intake, and anthelmintic drug treatment. Am. J. Clin. Nutr. 68, 623-629.

Junquero, D., Modat, G., Coquelet, C., and Bonne, C. (1990). Retinoid-induced
potentiation of epidermal growth factor mitogenic effect on corneal endothelial
cells. Cornea 9(1), 41-44.

Kastner, P., Grondona, J. M., Mark, M., Gansmuller, A., Lemeur, M., Decimo, D.,
Vonesch, J. L., Dolle, P. and Chambon, P. (1994). Genetic analysis of RXRa
developmental function: co,nvergence of RXR and RAR signaling pathways in
heart and eye morphogenesis. Cell 78, 987-1003.

Kastner, R, Mark, M., and Chambon, P. (1995). Nonsteroid nuclear receptors: what are
genetic studies telling us about their role in real life? Cell 83, 859-869.

Keyt, B. A., Berleau, L. T., Nguyen, H. V., Chen, H., Heinsohn, H., Vandlen, R., and
Ferrara, N. (1996). The carboxyl-tenninal domain (111-165) of vascular

endothelial growth factor is critical for its mitogenic potency. J. Biol. Chem.
271(13), 7788-7795.

King, T. and Brown, N. A. (1999). Developmental biology: Antagonists on the left
flank. Nature 401, 222-223.

Kirby, M. L. and Waldo, D. L. (1990). Role of neural crest in congenital heart disease.
Circulation 82, 332-340.

Kochhar, D. M. and Christian, M. S. (1997). Tretinoin: a review of the nonclinical
developmental toxicology experience. J. Am. Acad. Dermatol. 36, $47-$59.

103

Kochhar, D. M., J iang, H., Soprano, D. R., and Hamish, DC. (1993). Early embryonic
cell response in retinoid-induced teratogenesis. In “Retinoids: Progress in
Research and Clinical Applications” (M. A. Livrea and L. Packer, Eds), pp. 383-
396. Marcel Dekker, New York, NY.

Kojima, S. and Rifkin, D. B. (1993). Mechanism of retinoid-induced activation of latent

transforming growth factor-13 in bovine endothelial cells. J. Cell. Physio. 155,
323-332.

Koli, K., Saharinen, J ., Hyytiainen, M., Penttinen, C., and Keski-Oja, J. (2001). Latency,

activation, and binding proteins of TGF-B. Microscopy Res. Technique 52, 354-
362.

Kostetskii, I., J iang, Y., Kostetskaia, E., Yuan, S., Evans, S., and Zile, M. (1999).
Retinoid signaling required for normal heart development regulates GATA-4 in a
pathway distinct from cardiomyocyte differentiation. Dev. Biol. 206, 206-218.

Kostetskii, I., Linask, K. L., and Zile, M.H. (1996). Vitamin A deficiency and the
expression of retinoic acid receptors during cardiogenesis in early quail embryo.
Roux '3 Arch. Dev. Biol. 205, 260-271.

Kostetskii, I., Yuan, S-Y., Kostetskaia, E., Linask, K. K., Blanchet, S., Seleiro, E.,
Michaille, J -J ., Brickell, P., and Zile, M. (1998). Initial retinoid requirement for
early avian development coincides with retinoid receptor coexpression in the

precardiac fields and induction of normal cardiovascular development. Dev.
Dynamics 213, 188-198.

Kostetskii, I. and Zile, M. H. (1993). Expression of genes for retinoic acid receptors and
TGFB during quail embryogenesis. FASEB J. 7, A522.

Kubalak, S. W., Hutson, D. R., Scott, K. K., and Shannon, R. A. (2002). Elevated
transforming growth factor [32 enhances apoptosis and contributes to abnormal

outflow tract and aortic sac development in retinoic X receptor a knockout
embryos. Dev. 129, 733-746.

f.“

Kubalak, S. W. and Sucov, H. M. (1999). Retinoids in heart development. In “Heart
Development” (R. P. Harvey, Ed.), pp.209-219, Academic Press, San Diego.

Lansink, M., Koolwijk, P., van Hinsbergh, V., and Kooistra, T. (1998). Effect of steroid
hormones and retinoids on the formation of capillary-like tubular structures of

human microvascular endothelial cells in fibrin matrices is related to urokinase
expression. Blood 92(3), 927-938.

LaRue, A. C., Zile, M. H., Argraves, W. S., and Drake, C. J. (in preparation). Normal

vascular morphogenesis requires retinol-mediated regulation of endothelial cell
number.

104

Laverriere, A. C., McNeil, C., Mueller, C., Poelmann, R. E., Burch, J. B. E., and Evans,
T. (1994). GATA-4/5/6, a subfamily of three transcription factors transcribed in
developing heart and gut. J. Biol. Chem. 269, 23177-23184.

Lawrence, D. A. (1995). Transforming growth factor-B: an overview. Kidney Int].
47(S49), Sl9-S23.

Levin, M. (1997). Left-right asymmetry in vertebrate embryogenesis. BioEssays 19,
287-296.

Linask, K. K. (1992). N-cadherin localization in early heart development and polar
expression of Na-, K+-ATPase, and integrin during pericardial coelom formation
and epithelialization of the differentiating myocardium. Dev. Biol. 151, 213-224.

Linask, K. K. and Lash, J. W. (1986). Precardiac cell migration: Fibronectin localization
at mesoderm-endoderrn interface during directional movement. Dev. Biol. 114,
87-101.

Linask, K. K. and Lash, J. W. (1990). Fibronectin and integrin distribution on migrating
precardiac mesoderm cells. Ann. N. Y. Acad. Sci. 588, 417-420.

Linask, K. K. and Lash, J. W. (1993). Early heart development: dynamics of

endocardial cell sorting suggests a common origin with cardiomyocytes. Dev.
Dynamics 195, 62-69.

Lough, J. and Sugi, Y. (2000). Endoderm and heart development. Dev. Dynamics 217,
327-342.

Lui, N. S. T. and Roels, O. A. (1980). Vitamin A and carotene. In “Modern Nutrition in
Health and Disease” (R. S. Goodhart and M. E. Shils, Eds), pp. 142-159. Lea &
Febiger, Philadelphia, Pennsylvania.

Luo, J ., Pasceri, P., Conlon, R. A., Rosant, J ., and Giguere, V. (1995). Mice lacking all
isoforms of retinoic acid receptor [3 develop normally and are susceptible to the
teratogenic effects of retinoic acid. Mech. Dev. 53, 61-71.

Maden, M. (1999). Retinoids in neural development. In “Handbook of Experimental
Pharmacology,” vol. 139, “Retinoids, The Biochemical and Molecular Basis of
Vitamin A and Retinoid Action” (H. Nau and W. S. Blaner, Eds), pp. 399-442.
Springer-Verlag, Heidelberg, Germany.

Maden, M., Gale, E., Kostetskii, I., and Zile, M. (1996). Vitamin A-deficient quail
embryos have half a hindbrain and other neural defects. Curr. Biol. 6, 417-426.

105

Maden, M., Gale, E., and Zile, M.H. (1998). The role of vitamin A in the development
of the contral nervous system. J. Nutr. 128, 4718-475S.

Mahmood, R., Flanders, K. C., and Morriss-Kay, G. M. (1992). Interactions between
retinoids and TGF 03 in mouse morphogenesis. Dev. 115, 67-74.

Mahmood, R., Flanders, K. C., and Morriss-Kay, G. M. (1995). The effects of retinoid

status on TGF [3 expression during mouse embryogenesis. Anat. Embryol. 192,
21-33.

Majumder, K. and Overbeek, P. A. (1999). Left-right asymmetry and cardiac looping.
In “Heart Development” (R.P. Harvey and N. Rosenthal, Eds), pp. 391-402.
Academic Press, San Diego.

Mangelsdorf, D. J ., Thummel, C., Beato, M., Herrlich, P., Schutz, G., Umesono, K.,
Blumberg, B., Kastner, R, Mark, M., Chambon, P., and Evans, RM. (1995). The
nuclear receptor superfamily: the second decade. Cell 83, 835-839.

Mangelsdorf, D. J ., Umesono, K., and Evans, R. M. (1996). The retinoid receptors. In
“The Retinoids: Biology, Chemistry” (M. B. Spom, A. B. Roberts, and D. S.
Goodman, Eds), pp. 319-349. Raven Press, New York, NY.

Mason, K. E. (1935). Fetal death, prolonged gestation, and difficult parturition in the rat
as a result of vitamin A deficiency. Am. J. Anat. 57, 303-349.

Massague, J. (1996). TGFB signaling: receptors, transducers, and mad proteins. Cell
85, 947-950.

Massague, J. and Chen, Y-G. (2000). Controlling TGF-[3 signaling. Genes Dev. 14,
627-644.

Mathwes, w. w. (1986). Atlas ofDescriptive Embryology, 4th Edition. pp. 111-180.
Prentice-Hall, Englewood Cliffs, NJ.

McCaffery, P. and Drager, U. (1995'). Retinoic acid synthesizing enzymes in the
embryonic and adult vertebrate. Adv. Exp. Med. Biol. 372, 173-183.

McCollum, E. V. and Davis, M. (1913). The necessity of certain lipids in the diet during
growth. J. Biol. Chem. 15, 167-175.

McCormick, K. M. (2001). TGF 132 activation status during cardiac morphogenesis.
Dev. Dynamics 222, l7-25.

McLaren, D. S. ( 1980). Nutritional Ophthalmology. Academic Press, London.
McLaren, D. S. and Frigg, M. (2001). Sight and Life Manual on Vitamin A Deficiency
Disorders (VADD). 2nd Edition. Task Force Sight and Life, Basel, Switzerland.

106

Melnykovych, G. and Clowes, K. K. (1981). Growth stimulation of bovine endothelial
cells by vitamin A. J. Cell Physiol. 109(2), 265-270.

Miyazono, K. (2000). Positive and negative regulation of TGF- [3 signaling. J. Cell Sci.
113, 1101- 1109.

Moore, T. (1957). Vitamin A. Elsevier Publishing Company, Amsterdam.

Morriss, G. M. and Steele, C. E. (1972). Comparison of the effects of retinal and
retinoic acid on postimplantation rat embryos in vitro. T eratology 15, 109-119.

Morton, R. A. (1944). Chemical aspects of the visual process. Nature 153, 69-71.

Moss, J. B., Xavier-Neto, J ., Shapiro, M. D., Nayeem, S. M., McCaffery, P., Drager, U.
C., and Rosenthal, N. (1998). Dynamic patterns of retinoic acid synthesis and
response in the developing mammalian heart. Dev. Biol. 199, 55-71.

Napoli, J. L., Posch, K. P., Fiorella, P. D., and Boerrnan, M. H. (1991). Physiological
occurrence, biosynthesis and metabolism of retinoic acid: evidence for roles of
cellular retinol-binding protein (CRBP) and cellular retinoic acid-binding protein

(CRABP) in the pathway of retinoic acid homeostasis. Biomed. Pharmacother.
45(4-5), 131-143.

Napoli, J. L. (2000). A gene knockout corroborates the integral function of Cellular
Retinol-Binding Protein in retinoid metabolism. Nutr. Revs. 58, 230-241.

National Research Council. (1980) Recommended Dietary Allowances, 9th edition.
National Academy Press, Washington, DC.

National Research Council. (1989) Recommended Dietary Allowances, 10th edition.
National Academy Press, Washington, DC.

National Research Council. (2001) Dietary Reference Intakes for Vitamin A, Vitamin K,
Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum,
Nickel, Silicon, Vanadium, and Zinc. National Academy Press, Washington, DC.

Nau, H., Chahoud, 1., Dencker, L., Lammer, E. J ., and Scott, W.J. (1998).

Teratogenicity of vitamin A and retinoids. In “Vitamin A in Health and Disease”
(R. Blomhoff, Ed.), 615-663. Dekker, New York, NY.

Niederreither, K., McCaffery, P., Drager, U., Chambon, P., and Dolle, P. (1997).
Restricted expression and retinoic acid-induced downregulation of the
retinaldehyde dehydrogenase type 2 (RALDH-Z) gene during mouse
development. Mech. Dev. 62, 67-78.

107

Niederreither, K., Vermot, J., Messaddeq, N., Schuhbaur, B., Chambon, P., and Dolle, P.

(2001). Embryonic retinoic acid synthesis is essential for heart morphogenesis in
the mouse. Dev. 128, 1019-1031.

Noden, D. M. (1991). Development of craniofacial blood vessels. In “The Development
of the Vascular System” (R. N. Feinberg, G. K. Sherer, R. Auerbach, Eds), pp.
14, 124 Issues Biomed. Karger, Basel.

Olson, J. A. (1994). Needs and sources of carotenoids and vitamin A. Nutr. Rev. 52(2),
67S-73S.

Noy (1992). The ionization behavior of retinoic acid in lipid bilayers and in membranes.
Biochm. Biophys. Acta. 1106, 159-164.

Ong, D. E., Newcomer, M. E., and Chytil, F. (1994). Cellular retinoid-binding proteins.
In “The Retinoids” (M. B. Spom, A. B. Roberts, D. S. Goodman), pp. 283-317.
Raven Press, New York.

Olson, E. N. and Srivastava, D. (1996). Molecular pathways controlling heart
development. Science 272, 671-676.

Osmond, M. K., Butler, A. J ., Voon, F.C., and Bellairs, R. (1991). The effects of
retinoic acid on heart formation in the early chick embryo. Dev. 113(4), 1405-
1417.

Pardanaud, L., Altmann, C., Kitos, P., Dieterlen-Lievre, F ., and Buck, C. A. (1987).
Vasculogenesis in the early quail blastodisc as studied with a monoclonal
antibody recognizing endothelial cells. Dev. 100, 339-349.

Parker, R. S., Swanson, J. E., You, C. S., Edwards, A. J ., and Huang, T. (1999).
Bioavailability of carotenoids in human subjects. Proc. Nutr. Soc. 58, 155-162.

Patan, S. (2000). Vasculogenesis and angiogenesis as mechanisms of vascular network
formation, growth and remodeling. J. Neuro-Oncology 50, 1-15.

Pfahl, M. and Chytil, F. (1996). Regulation of metabolism by retinoic acid and its
nuclear receptors. Annu. Rev. Nutr. 16, 257-283.

Piek, E., Heldin, C-H, and ten Dijke, P. (1999). Specificity, diversity, and regulation in
TGF-B superfamily signaling. FASEB J. 13, 2105-2124.

Poole, T. J. and Coffin, J. D. (1989). Vasculogenesis and angiogenesis: Two distinct

morphogenetic mechanisms establish embryonic vascular patter. J. Exp. Zool.
251, 224-231.

108

Poole, T. J. and Coffin, J. D. (1991). Morphogenetic mechanisms in avian vascular
development. In “The Development of the Vascular System” (R. N. Feinberg, G.
K. Sherer, R. Auerbach, Eds), pp. 14, 25-36. Issues Biomed. Karger, Basel.

Quick, T. C. and Ong, D. E. (1990). Vitamin A metabolism in the human intestinal
Caco-2 cell line. Biochem. 29(50), 11116-11123.

Radkar, A., Montgomery, M., and Litvin, J. (2001). Fate map of early avian cardiac
progenitor cells. Dev. 128, 2269-2279.

Risau, W. (1991). Vasculogenesis, angiogenesis and endothelial cell differentiation
during embryonic development. In “The Development of the Vascular System”
(R. N. Feinberg, G. K. Sherer, R. Auerbach, Eds), pp. 14, 58-68. Issues Biomed.
Karger, Basel.

Risau, W., Flamme,1. (1995). Vasculogenesis. Annu. Rev. Cell Dev. Biol. 11, 73-91.

Roberts, A. B. and Spom, M. B. (1984) Cellular biology and biochemistry of retinoids.
In “The Retinoids” vol. 2 (M. B. Spom, A. B. Roberts, and D. S. Goodman,
Eds), pp. 209-286. Academic Press, Orlando, FL.

Roberts, A. B. and Spom, M. B. (1992). Differential expression of the TGF-[3 isoforms

in embryogenesis suggests specific roles in developing and adult tissues. M01.
Repro. Dev. 32, 91-98.

Roberts, A. B. and Spom, M. B. (1992a). Mechanistic interrelationships between two

superfamilies: the steroid/retinoid receptors and transforming growth factor-B.
Cancer Surveys 14, 205-220.

Roberts, A. and Spom, M. (1996). Transforming growth factor-B. In “The Molecular
and Cellular Biology of Wound Repair” (R. Clark, Ed.), pp. 275-308. Plenum
Press, New York.

Rock, C. L., Lovalvo, J. L., Emenhiser, C., Ruffin, M. T., Flatt, S. W., and Schwartz, S. J.

(1998). Bioavailability of B-carotene is lower in raw than processed carrots and
spinach in women. J. Nutr. 128, 913-916.

Rosa, R. W. (1993). Retinoid embryopathy in humans. In “Retinoids in Clinical
Practice. The Risk-Benefit Ratio.” (G. Koren, Ed.), pp. 77-109. Marcel Dekker,
New York, NY. "

Rosalas and Ross. (1998). A Low Molar Ratio of Retinol Binding Protein to
Transthyretin Indicates Vitamin A Deficiency during Inflammation: Studies in
Rats and A Posteriori Analysis of Vitamin A-Supplemented Children with
Measles. J. Nutr. 128, 1681-1687.

109

Ross, S. A., McCaffery, P. J ., Drager, U. C. and De Luca, L. M. (2000). Retinoids in
embryonal development. Physiol. Rev. 80, 1021-1054.

Salbert, G., Fanjul, A., Piedrafita, F. J., Lu, X. P., Kim, S. J., Tran, P., and Pfahl, M.
(1993). Retinoic acid receptors and retinoid X receptor-on down-regulate the

transforming growth factor-Bl promoter by antagonizing AP-l activity. Mol.
Endocrinol. 7(10), 1347-1356.

Sanders, E. J., Prasad, S., and Hu, N. (1993). The involvement of TGFBI in early avian
development: gastrulation and chondrogenesis. Anat. Embryo]. 187, 573-581.

Sanford, L. P., Ormsby, I. Gittenberger-de Groot, A. C., Sariola, H., Friedman, R.,
Boivin, G. P., Cardell, E. L., and Doetschman, T. (1997). TGFBZ knockout mice

have multiple developmental defects that are non-overlapping with other TGFB
knockout phenotypes. Dev. 124, 2659-2670.

Selleck, M. A. J. (1996). Culture and microsurical manipulation of the early avian
embryo. In “Methods in Cell Biology,” vol. 51, “Methods in Avian Embryology”
(M. Bronner-Fraser, ed.), pp. 1-21. Academic Press, New York.

Shenai, J. P., Rush, M. G., Parker, R. A., and Chytil, F. (1995). Sequential evaluation of
plasma retinol-binding protein response to vitamin A administration in very-low-
birth weight neonates. Biochem. Mo]. Med. 54, 67-74.

Shenefelt, R. E. (1972). Morphogenesis of malformations in hamsters caused by retinoic
acid: relation to dose and stage at treatment. T eratology 5, 103-118.

Slager, H. G., Freund, E., Buiting, A. M. J ., Feijen, A., and Mummery, CL. (1993).
Secretion of transforming grth factor-B isoforms by embryonic stem cells:

isoform and latency are dependent on direction of differentiation. J. Cellular
Physio. 156, 247-256.

Smith, S. M., Dickrnan, E. D., Power, S. C., and Lancman, J. (1998). Retinoids and their
receptors in vertebrate embryogenesis. J. Nutr. 128, 467S-470S.

Sommer, A. (1994). Vitamin A: its effects on childhood sight and life. Nutr. Rev.
52(2), 60S-66S.

Sommer, A., Hussaini, G., and Tarwotjo, I. (1983). Increased mortality in children with
mild vitamin A deficiency. Lancet. ii, 585-588.

Sommer, A., Tarwotjo, I., Djunaedi, E., West, K. P. Jr., and Loeden, A. A. (1986).

Impact of vitamin A supplementation on childhood mortality: a randomized
controlled community trial. Lancet. 1, 1169-1173.

110

Sommer, A. and West, K. P., Jr. (1996).; Vitamin A Deficiency: Health, Survival, and
Vision. Ox ford University Press, New York, NY.

Spom, M. B. and Roberts, A. B. (1994). In “The Retinoids: Biology, Chemistry, and
Medicine,” 2nd ed. (M. B. Spom, A. B. Roberts, and D. S. Goodman), Raven
Press, NY.

Strutz, F ., Zeisberg, M., Renziehausen, A., Raschke, 3., Becker, V., van Kooten, C., and

Muller, GA. (2001). TGF-Bl induces proliferation in human renal fibroblasts
via induction of basic fibroblast growth factor (FGF-2). Kidney Int]. 59, 579-
592.

Sucov, H. M., Dyson, E., Gumeringer, C. L., Price, J ., Chien, K. R., and Evans, RM.

(1994). RXRor mutant mice establish a genetic basis for vitamin A signaling in
heart morphogenesis. Genes Dev. 9, 1007-1018.

Sugi, Y. and Markwald, R. R. (1996). Formation and early morphogenesis of

endocardial endothelial precursor cells and the role of endoderm. Dev. Biol. 175,
66-83.

Sugi, Y., Myers, K. L., Brown, C. B., Barnett, J. V., and Markwald, RR. (2001).
Endodermal regulation of endocardial precursor cell formation from precardiac
mesoderm involves TGFB pathway. Unpublished.

Suhamo, D.,.West, C. E., Muhilal, Karyadi, D., and Hautvast, JG. (1993).
Supplementation with vitamin A and iron for nutritional anaemia in pregnant
women in West Java, Indonesia. Lancet. 342, 1325-1328.

Tang, G., Qin, J ., Dolnikowski, G. G., and Russell, R. M. (2000). Vitamin A

equivalence of B-carotene in a woman as determined by a stable isotope reference
method. Eur. J. Nutr. 39, 7-11.

Thompson, J. N. (1969). The role of vitamin A in reproduction. In “The Fat Soluble
Vitamins” (H. F. DeLuca and J. W. Suttie, Eds), pp. 267-281. University of
Wisconsin Press, Madison, WI.

Thompson, J. N., Howell, J ., Pitt, G. A. J ., and McLaughlin, CL (1969). The biological
activity of retinoic acid in the domestic fowl and the effects of vitamin A
deficiency on the chick embryo. Br. J. Nutr. 23, 471-490.

Torronen, R., Lehmusaho, M., Hakkinen, S., Hanninen, O., and Mykkanen, H. (1996).
Serum B-carotene response to supplementation with raw carrots, carrot juice or
purified B-carotene in healthy non-smoking women. Nutr. Res. 16, 565-575.

111

Twal, W. 0., Roze, L., and Zile, M.H. (1995). Anti-retinoic acid monoclonal antibody
localizes all-trans-retinoic acid in target cells and blocks normal development in
early quail embryo. Dev. Biol. 168, 225-234.

Twal, W. O. and Zile, M. H. (1997). Retinoic acid reverses ethanol-induced

cardiovascular abnormalities in quail embryos Alcohol Clin. Exp. Res. 21(6),
1 137-1 143.

Van den Berg, H. and van Vliet, T. (1998). Effect of simultaneous, single oral doses of

B-carotene with lutein or lycopene on the B-carotene and retinyl ester responses in
the triacylglycerol-rich lipoprotein fraction of men. Am. J. Clin. Nutr. 68, 82-89.

Van het Hof, K. H., Cartner, C., West, C. E., and Tijburg, LB. (1998). Potential of

vegetable processing to increase the delivery of carotenoids to man. Int. J. Vitam. 1,
Nutr. Res. 68, 366-370. ‘

Wald, G. (1934). Carotenoids and vitamin A cycle in vision. Nature 134, 65. ‘5 I
Wald, G. (1968). Molecular basis of visual excitation. Science 162, 230-239.

Wei, Y. and Mikawa, T. (2000). Fate diversity of primitive streak cells during heart
field formation in ovo. Dev. Dyn. 219, 505-513.

Weinstein, B. M. (1999). What guides early embryonic blood vessel formation? Dev.
Dynamics 215, 2-11.

Wellik, D. M., Norback, D. H., and DeLuca, H. F. (1997). Retinol is specifically
required during midgestation for neonatal survival. Am. J. Physiol. 272
(Endocrinol. Metab. 35: E25-E29).

West, K. P. Jr., Le Clerq, S. C., Khatry, S. K., et al. (1999). Double blind, cluster
randomized trial of low dose supplementation with vitamin A or B-carotene on
mortality related to pregnancy in Nepal. Brit. Med. J. 318, 570-575.

White, J. C. and Clagett-Dame, M. (1996). Abnormal development of the optic cup,
olfactory pit, cranial ganglia and anterior cardinal vein in vitamin A-deficient rat
embryos. Soc. Neurosci. Abstr. 22, 990 part 2.

 

Wilt, 'F. H. (1966). The concept of messenger RNA and cytodifferentiation. Am. Zoo].
6(1), 67-74. '

WHO (1995). Global prevalence of vitamin A deficiency. WHO/NUT/95.3, WHO,
Geneva.

112

 

Wilson, J. G., Roth, C. B. and Warkany, J. (1953). An analysis of the syndrome of
malformations induced by maternal vitamin A deficiency. Effects of restoration of
vitamin A at various times during gestation. Am. J. Anat. 92, 189-217.

Wolf, G. (1984). Multiple functions of vitamin A. Physiol. Rev. 64, 873-937.

Yoshizawa, M., Miyazaki, H., and Kojima, S. (1998). Retinoids potentiate transforming
growth factor-B activity in bovine endothelial cells through up-regulating the

expression of transforming growth factor-B receptors. J. Cellular Physio]. 176,
565-573.

Yutzey, K. E., Rhee, J. T., and Bader, D. (1994). Expression of the atrial-specific
myosin heavy chain AMHCl and the establishment of anteroposterior polarity in
the developing chicken heart. Dev. 120, 871-883.

Zhao, S. and Overbeek, P. A. (2001). Elevated TGFB signaling inhibits ocular vascular
development. Dev. Biol. 237, 45-53.

Zile, M. H. (1998). Vitamin A and embryonic development: an overview. J. Nutr. 128,
455S-458S.

Zile, M. H. (1999). Avian embryo as model for retinoid function in early development.
In “Handbook of Experimental Pharmacology,” vol. 139, Retinoids, “The
Biochemical and Molecular Basis of Vitamin A and Retinoid Action” (H. Nau
and W. S. Blaner, Eds), pp. 443-464. Springer-Verlag, Heidelberg, Germany.

Zile, M. H. (2001). Function of vitamin A in vertebrate embryonic development. J.
Nutr. 131, 705-708.

Zile, M. H., Kostetskii, I., Yuan, S., Kostetskaia, E., St. Arnand, T. R., Chen, Y., and

J iang, W. (2000). Retinoid signaling is required to complete the vertebrate
cardiac left/right asymmetry pathway. Dev. Biol. 223, 323-338.

113

 

1|.Iljllljjjjljljujjjjl