Placenta (2003), 24, 123-130

doi:10.1053/plac.2002.0887

CURRENT TOPIC

Genes, Development and Evolution of the Placenta

J. C. Cross™?, D. Baczyk®, N. Dobric?, M. Hemberger®, M. Hughes®, D. G. Simmons?,

H. Yamamoto?, and J. C. P. Kingdom"*
2 Genes & Development Research Group, Departments of Biochemistry & Molecular Biology, and Obstetrics & Gynaecology,
Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada, ® Samuel Lunenfeld Research Institute, Mount Sinai
Hospital and © Departments of Obstetrics & Gynaecology, University of Toronto, Toronto, Ontario, Canada

Through studies of transgenic and mutant mice, it is possible to describe molecular pathways that control the development of all
major trophoblast cell subtypes and structures of the placenta. For example, the proliferation of trophoblast stem cells is dependent
on FGF signalling and downstream transcription factors Cdx2, Eomes and Err2. Several bHLH transcription factors regulate the
progression from trophoblast stem cells to spongiotrophoblast and to trophoblast giant cells (Id1/2, Mash2, Hand1, Stral3).
Intercellular actions critical for maintaining stable precursor cell populations are dependent on the gap junction protein Cx31 and
the growth factor Nodal. Differentiation towards syncytiotrophoblast as well as the initiation of chorioallantoic (villous)
morphogenesis is regulated by the Gem1 transcription factor, and subsequent labyrinth development is dependent on Wnt, HGF
and FGF signalling. These insights suggest that most of the genes that evolved to regulate placental development are either
identical to ones used in other organ systems (e.g., FGF and epithelial branching morphogenesis), were co-opted to take on new
functions (e.g., AP-2y, DIx3, Handl), or arose via gene duplication to take on a specialized placental function (e.g., Geml, Mash2).
Many of the human orthologues of these critical genes show restricted expression patterns that are consistent with a conserved
function. Such information is aiding the comparison of the human and mouse placenta. In addition, the prospect of a conserved
function clearly suggests potential mechanisms for explaining complications of human placental development.
Placenta (2003), 24, 123-130 © 2002 Elsevier Science Ltd. All rights reserved.

INTRODUCTION after invading it) and produce growth factors, cytokines and
hormones which target maternal physiological systems, result-
Comparing the placenta from one species to another can be a
challenging task. On one hand, the gross anatomy of the ing in provision of more blood flow and nutrient delivery to the
placenta is at least superficially rather different among different
feto-placental unit (‘invasive and endocrine trophoblast’). Pre-
cisely how these structures and functions are orchestrated may
species (Wooding and Flint, 1994). To complicate matters,
even where analogous cell types exist, different names are often
vary among species, but the general principles are likely to
remain the same. Because of the explosion of new information
used to describe them in different species. The temptation is to
either give up or assume that the placenta evolved indepen-
emerging from analysis of mutant and transgenic mice con-
cerning the molecular control of placental development, it is an
dently into very different structures in different mammals.
important time to deal with the issue of how to extrapolate this
Although the differences are certainly interesting, a preoccu-
information to other species. Detailed descriptions and com-
pation with them obscures the conserved aspects of placental
parisons of the anatomy of the mouse and human placenta have
structure and function. At the most basic level, the placental
trophoblast-derived structure fulfils two distinct functions in
been published elsewhere (Adamson et al., 2002; Georgiades
et al, 2002). In this review, we present a framework for
every mammalian species. First, it generates a large surface
area for nutrient exchange, consisting of an epithelial barrier comparing the mouse and human placenta (Figure 1, Table 1)
and underlying foetal blood vessels (‘transport and barrier and provide molecular evidence in support (Table 2). Finally,
trophoblast’) that comprise the chorioallantoic placenta.
we discuss how the identification of genetic pathways in
mice has refined our thinking about the control of placental
Second, trophoblast cells interact closely with the uterus (often
development and how it may have evolved.
4To whom correspondence should be addressed at: Department of
Biochemistry & Molecular Biology, Faculty of Medicine, University EVOLUTIONARY ORIGINS OF THE PLACENTA
of Calgary, HSC Room 2279, 3330 Hospital Drive, N.W., Calgary, o . .
Alberta T2N 4N1 Canada. Fax: (403) 270-0727; E-mail The most primitive type of chorioallantoic placenta can be
[email protected] traced back to the egg-laying animals that, similar to eutherian
0143-4004/03/8-see front matter (© 2002 Elsevier Science Ltd. All rights reserved.
 Placenta (2003), Vol. 24
Endovascular trophoblast (Figure 1). The organization of the villi is one basis for
lant coll
classifying placental structures (Wooding and Flint, 1994).
Interstital glycogen
trophoblast cell The villi can be diffuse (pig, horse), or they can be tightly
Trophoblast grouped into a band (cat, dog) or into round structures called
giant cell cotyledons. Rodents have a single cotyledon (discoid placenta)
Spongiotrophoblast whereas humans have multiple cotyledons, though they are
Labyrinth
(smoyiotophobiast consolidated into a cluster resembling a single disc. In rumi-
cytoophosias ioma, Hood
vessel) nants (sheep, cow), multiple cotyledons are scattered across the
Fetus. placental surface. The structure of the trophoblast cell layer
that covers the villi and the type of maternal cells that it
Decidua Endovascular extravillous
trophoblast directly contacts vary between species. These differences are
another basis for placental classification (Wooding and Flint,
Interstitial extravilous
trophoblast 1994). In rodents and primates, the uterine epithelium
(endometrium) is eroded such that maternal blood comes into
Cytotrophobast direct contact with the trophoblast surface (haemochorial).
column
Chorfonic vill Whereas rodents have three trophoblast cell layers (tri-chorial)
(yneyiotophoblat vilous
yitophobian sioma, boss including two syncytial layers and a single ‘mononuclear’ cell
vessel)
type of unknown function, primates have a single syncytial
layer plus an underlying trophoblast stem cell layer. In
Fotus ruminants and horses, the endometrium essentially remains
intact and trophoblast cells lie in direct contact with uterine
Figure 1. Comparative anatomy of the human and mouse placenta.
epithelium (epithelio-chorial).
The ‘invasive and endocrine trophoblast’ subpopulation is
divergent across species in the extent of uterine invasion and
mammals, have two vascularized extraembryonic membranes the types of maternal-acting hormones that it can produce.
(Wooding and Flint, 1994; Gilbert, 1997). The yolk sac derives Ruminant bi-nucleate cells migrate essentially only the dis-
nutrients from the yolk and is essential throughout develop- tance of one cell diameter to enter the endometrium (Wooding
ment in reptiles and birds. Although it is often considered to and Flint, 1994). In contrast, rodent trophoblast giant cells and
be vestigial in cutherian mammals, the yolk sac is clearly glycogen trophoblast cells migrate several hundred microns
essential for nutrition of the early embryo prior to the time that into the uterus (Adamson et al., 2002), though still a relatively
the placenta has formed (Cross et al., 1994). The chorioallan- modest distance compared to the analogous cells in humans,
toic membrane that forms the placenta in higher mammals also the extravillous cytotrophoblast cells (Figure 1). The cell
exists in reptiles and birds albeit as a much simpler structure. adhesion molecules and matrix-degrading proteinases that
In birds, the chorioallantoic membrane underlies the eggshell mediate cell invasion in humans are expressed by trophoblast
and is essential for gas exchange and for transporting calcium giant cells in mice (Cross et al, 1994). The characteristic
from the shell to the developing fetus. Clearly, the mammalian feature of rodent trophoblast giant cells is that they are
placenta has evolved to take on many additional roles dealing polyploid (Zybina and Zybina, 1996), as a result of repeated
with intrauterine life, that are not undertaken by the chorioal- rounds of DNA replication in the absence of intervening
lantoic membrane in reptiles and birds. These include the mitoses, an unusual type of cell cycle term called endoredupli-
production of hormones that alter systemic maternal functions cation (MacAuley et al., 1998). Human extravillous cytotro-
and growth factors that alter the local uterine environment. In phoblast (Berezowsky et al., 1995) and bovine binucleate cell
addition, trophoblast cells, in some species, either fuse with nuclei (Klisch et al., 1999) are polyploid, though ploidy tends
endometrial cells or invade into the uterus to promote a more to be 4-8 N compared with up to 1024 N in rodents (Zybina
intimate contact between conceptus and mother. It is not clear and Zybina, 1996). This is different than syncytiotrophoblast
what to consider as the evolutionary precursor of the chorion cells, that form as the result of trophoblast cells exiting the cell
in lower vertebrates and invertebrates and thus there is no way cycle and fusing together to form a multi-nucleated syncytium
to identify candidate regulatory genes for the placenta by with each nucleus remaining diploid (Figure 2, Table 1).
analogy to genetically tractable organisms like Drosophila and Two general types of invasion are observed in the human
C. elegans. and rodent placenta (Pijnenborg et al, 1981) (Figure 1).
Trophoblast giant cells in rodents show invasion that is strictly
associated with maternal spiral arteries that bring blood to the
COMPARATIVE PLACENTAL STRUCTURE: A implantation site (Adamson et al., 2002). Glycogen trophoblast
GENERALIST'S VIEW cells, by contrast, invade into the interstitium of the maternal
decidua and show no close association with the spiral arteries
In most placental mammals, ‘transport and barrier trophoblast’ (Adamson et al., 2002). Differences in marker expression
is organized into highly branched villous, tree-like folds by these two cell types suggest that the endovascular and
Cross et al.: Genes, Development and Evolution of the Placenta

Table 1. Comparison of structures in the human and mouse placenta

Human Mouse

A. Invasive and endocrine trophoblast

Formal name Extravillous cytotrophoblast Trophoblast giant cell
Invasive Yes Yes
Proliferative No No
DNA content Mononuclear polyploid (4 N-16 N) Mononudlear polyploid (up to 1000 N)
B. Transport and barrier trophoblast
Formal name Chorionic villi Labyrinth
Haemochorial Yes Yes
Syncytiotrophoblast surface Yes Yes
Formed by cell fusion Yes Yes
Nuclear DNA content Diploid Diploid

Table 2. Comparative gene expression in the human and mouse trophoblast cell lineage

Gene Human Mouse

Geml Syncytiotrophoblast and precursor Syncytiotrophoblast and precursor

Handl Trophectoderm Trophectoderm, trophoblast giant cells
12 Villous cytotrophoblast Chorionic trophoblast
Mash2 Column cytotrophoblast Spongiotrophoblast
Met Villous cytotrophoblast Chorionic trophoblast
Mmp9 Invasive cytotrophoblast ‘Trophoblast giant cells
Pecaml Endovascular cytotrophoblasts Endovascular trophoblast giant cells
Tef5 Syncytiotrophoblast Syncytiotrophoblast
uPA Invasive cytotrophoblast ‘Trophoblast giant cells

O
.ot 1) Cell Fusion
- syncytiotrophoblast
(Pijnenborg et al., 1981) but, unlike in rodents, molecular
markers have not been identified that can distinguish them.

MOLECULAR PATHWAYS REGULATING

PLACENTAL DEVELOPMENT
Insights from gene knockout mice

The mouse is an extremely powerful system for understand-

\. Endoreduplication ing the control of development because of the ability to alter
(polytene chromosomes)
- trophoblast giant cells and ablate the expression of genes through transgenic and
knockout mouse technologies. These experiments have pro-
duced an amazingly long list of genes that are essential for
placental development [reviewed in (Cross et al., 1994;
Rinkenberger et al., 1997; Cross, 2000; Hemberger and
Figure 2. Formation of multi-nucleated syncytiotrophoblast and mononu- Cross, 2001; Rossant and Cross, 2001)]. The detailed analysis
clear, polyploid trophoblast giant cells. of these genetically altered mice, combined with the simi-
larity of some of the mutant phenotypes, has allowed us to
describe these genes as part of genetic pathways, and to gain
interstitial cells represent distinct cell types (Adamson et al., precise understanding of the cellular and molecular functions
2002). Distinct endovascular and interstitial routes of tropho- of these pathways in tissue development (Cross, 2000;
blast invasion are observed in the placental bed of humans Hemberger and Cross, 2001; Rossant and Cross, 2001). Two
126 Placenta (2003), Vol. 24
general lessons have emerged from this work. First, develop- Nodal production from giant cells might negatively regulate
ment of the two major placental substructures—‘transport giant cell formation, it implies that different trophoblast
and Dbarrier trophoblast’ and ‘invasive and endocrine subtypes can interact with each other.
trophoblast'—is regulated by distinct genetic pathways
(Cross, 2000). Second, intercellular signalling events regu- Ectoplacental cone and spongiotrophoblast. At early post-
late many different aspects of trophoblast cell development implantation stages of development, trophoblast cells in the
(Rossant and Cross, 2001). ectoplacental cone lie between the chorion and the outer layer
of trophoblast giant cells. It is thought that ectoplacental cone
Trophoblast stem cells. Trophoblast stem cells are defined as cells later become the spongiotrophoblast, largely based on
those cells that have the potential to give rise to all differenti- gene expression patterns. For example, the genes Tpbp/4311
ated trophoblast cell subtypes. They can be isolated from the (Lescisin et al,, 1988) and Fit] (He et al., 1999) are expressed
blastocyst and early post-implantation stage mouse concep- in the ectoplacental cone and spongiotrophoblast, and not in
tuses by culturing them in the presence of fibroblast growth trophoblast giant cells, extraembryonic ectoderm or labyrinth.
factor4 (FGF4) and feeder cells (Tanaka et al, 1998). Tmportantly though, expression of Tpbp/4311 and Fitl is first
Removal of either factor results in cell proliferation arrest and detectable only in a few cells at the apex of the ectoplacental
differentiation to trophoblast giant cells. The transcription cone at E7.5 and 8.5. In contrast, Tphp/4311 and Fltl are
factors Cdx2 and Eomes are regulated by FGF signalling and expressed throughout the spongiotrophoblast layer at later
are in turn also essential for the maintenance of trophoblast times in development, albeit not at uniform levels in all cells.
stem cells (Chawengsaksophak et al., 1997; Tanaka et al., 1998; This shows that the ectoplacental cone is not uniform and is
Russ et al., 2000). Another transcription factor, AP-2y, is also composed of more than one cell type, though the functions of
essential for trophoblast stem cell maintenance (Auman et al., the two populations is not clear.
2002; Werling and Schorle, 2002), but whether it is regulated The basic helix-loop-helix (bHLH) transcription factor
by FGF is unclear. FGF4 is expressed by the inner cell mass Mash2 is expressed in the chorion, ectoplacental cone and later
at the blastocyst stage and subsequently in the epiblast (embry- the spongiotrophoblast (Guillemot et al., 1994; Nakayama
onic ectoderm) whereas the FGF receptor, FGFR2, is et al., 1996). Mash2-deficient embryos show an absence of
expressed in trophoblast cells immediately adjacent to the spongiotrophoblast by E10.5 as well as increased number of
epiblast (chorion or extraembryonic ectoderm) (Rossant and giant cells and a secondary failure of the labyrinth to form
Cross, 2001). Because FGF's are matrix associated, the effect of (Guillemot et al., 1994; Tanaka et al., 1997). The mutant
FGF4 is restricted to this close paracrine effect. After gastru- phenotype is interesting because, despite early expression,
lation, when the epiblast cells lose their immediate contact with Mash2 mutants have a normal ectoplacental cone and chorion
the chorionic trophoblast, it is unclear whether the FGF- up to at least E8.5, implying that Mash2 gene is only required
dependent trophoblast stem cell population continues to per- for the later maintenance of the ectoplacental cone-derived
sist. It is certainly possible that another FGF is produced to cells and not for their initial formation. While Mash2 mRNA is
sustain the cells. FGF-dependent TS cell lines have not been expressed in both the outer part of the ectoplacental cone,
established from mouse conceptuses after E7.5, however, coincident with Tpbp/4311 and Fit1, it is also expressed in the
implying that trophoblast stem cells may not persist (Rossant inner part (Guillemot et al., 1994; Nakayama et al., 1996).
and Cross, 2001). Other members of the bHLH transcription factor family that
Two other signalling systems appear to regulate the main- are obligate DNA binding partners for Mash2, Alfl/HEB and
tenance of trophoblast stem cells and suppress differentiation, Itf2, are only expressed in the inner part of the ectoplacental
at least at later times during development. The nuclear cone (Scott et al., 2000). This pattern is informative because it
receptor Err2/ErrB (encoded by the Esrrb gene) is expressed in implies that Mash2 function is actually limited to the inner
the extraembryonic ectoderm after E7.5 and Esrrb-deficient cells and, therefore, that these cells are somehow linked to the
embryos show premature loss of the extraembryonic ectoderm maintenance of the spongiotrophoblast. This is likely due to
and ectoplacental cone, with an associated increase in tropho- Mash2 promoting trophoblast proliferation (N.D. and J.C.C.,
blast giant cells (Luo et al., 1997). Err2/Errf appears to be a unpublished observations). It is notable that in addition to
constitutively active nuclear receptor, but upon binding to the Mash2, its potential partners Alfl/HEB and Itf2 are also
estrogen analogue diethylstilbestrol (DES), transcriptional expressed also in the chorion (Scott et al., 2000). Whether they
activity is repressed (Tremblay et al., 2001). Consistent with are able to dimerize with Mash2 is unclear, however, since cells
this, DES treatment of a cultured TS cell line promotes the of the chorion also express Id1 and 1d2 (Jen et al., 1997). The
loss of stem cell character and increases the formation of Id proteins are dominant-negative HLH factors that can
trophoblast giant cells (Tremblay et al., 2001). A TGFB family dimerize with Alfl/HEB, Itf2 as well as the related E12/47
member, Nodal, also has a role in regulating trophoblast proteins (Norton, 2000).
development in that trophoblast giant cell differentiation is In addition to intrinsic factors, it is clear that trophoblast
enhanced in Nodal-deficient embryos (Ma et al., 2001). Curi- development is regulated by cell extrinsic factors. Cell—cell
ously, trophoblast giant cells themselves express Nodal (Ma et interactions are important as embryos that are deficient for the
al., 2001). Although there is no direct evidence to suggest how gap junction protein Cx31 (expressed in the spongiotropho-
Cross et al.: Genes, Development and Evolution of the Placenta 127

blast) show a transient alteration in the spongiotrophoblast important for normal trophoblast development because mu-
layer (Plum et al., 2001). The proliferation of human tropho- tations in retinoic acid receptors have placental phenotypes
blast cells in vitro is highly regulated by oxygen tension. (Wendling et al., 1999; Sapin et al., 2000). In addition, the
Primary cells that are cultured in low oxygen proliferate, as retinoic acid-regulated gene Stral3 is expressed in trophoblast
they would be expected to do in vivo, whereas cells cultured in giant cells in vivo as a part of normal development (Boudjelal
an atmosphere with 20 per cent oxygen do not proliferate and et al, 1997), and Stral3 expression increases during the
instead differentiate into the invasive trophoblast subtype differentiation of TS cells in vitro (M. Hughes and J.C.C.,
(Genbacev et al.,, 1996; Genbacev et al., 1997). There are two unpublished observations).
ways to interpret these data. On one hand one could conclude
that low oxygen suppresses differentiation. Alternatively, one Glycogen trophoblast cells. The cell lineage origins of glycogen
could argue that exposure of early trophoblast cells to a high trophoblast cells are not exactly clear. However, they first
oxygen level might be toxic and that the response is to arrest appear within the spongiotrophoblast layer before they begin
cell proliferation and to differentiate. A precedent for tropho- to migrate into the decidual tissue of the mother and they
blast cells responding to a toxic insult by differentiating is that express the Tpbp/4311 gene, a gene otherwise exclusively
proliferating trophoblast cells exposed to radiation either die or expressed in spongiotrophoblast (Adamson et al., 2002).
differentiate (MacAuley et al., 1998; Heyer et al., 2000). No Therefore, it is very likely that they are a specialized subtype of
matter the effect by which trophoblast cells respond to differ- spongiotrophoblast. The only molecular clues that we have
ent oxygen levels, the mechanism of low oxygen action s likely about the control of their development is that fewer glycogen
mediated by the transcriptional complex HIF1a/Arnt. Arni- trophoblast cells are present Igf2 in mutants (Lopez et al.,
deficient embryos show a placental phenotype in which the 1996), and that more glycogen cells is a common finding
labyrinth and spongiotrophoblast layers are reduced in size, in several single gene mutants (Li and Behringer, 1998;
and giant cell numbers are increased (Adelman et al., 2000). Takahashi et al., 2000; Frank et al., 2002).
Culture of TS cells in low oxygen promotes their differenti-
ation into spongiotrophoblast-type cells and suppresses giant Syncytiotrophoblast and morphogenesis of the labyrinth. The
cell formation (Adelman et al., 2000). labyrinth layer of the placenta forms starting at E8.5 when
allantoic mesoderm attaches to the basal surface of the chorion
Trophoblast giant cells. Trophoblast giant cells form from the layer (Cross, 2000). While the chorion continues to have
mural trophectoderm immediately after implantation (primary proliferating cells within it, some of trophoblast cells begin to
giant cells) and then more giant cells emerge from the differentiate early on E9 and form primary branchpoints for
differentiation of cells at the outer edge of the ectoplacental the villous structure. Syncytiotrophoblast cell differentiation
cone (secondary giant cells) (Cross et al., 1994; Hemberger and also begins early on E9 and the Gem/ gene is necessary for both
Cross, 2001; Rossant and Cross, 2001). The differentiation of the initiation of morphogenesis and syncytiotrophoblast differ-
both primary and secondary giant cells is dependent on the entiation (Anson-Cartwright et al., 2000). Geml encodes a
bHLH factor Hand1 (Riley et al., 1998; Scott et al., 2000). transcription factor that is homologous to a factor that controls
Hand1 and Mash2 have opposing functions and indeed in vitro an important cell fate decision in the nervous system in
they show mutually antagonistic activities (Scott et al., 2000). Drosophila (Wegner and Riethmacher, 2001). While Gem!
In Handl mutants, Mash2 mRNA expression fails to be shut expression in mice appears to be essentially restricted to the
down (Riley et al., 1998) which may be functionally important placenta, it probably has an analogous cellular function in that
because ectopic expression of Mash2 in vitro is sufficient to it regulates a cell fate decision between chorion trophoblast and
block giant cell differentiation (Cross et al., 1995; Kraut et al., the progenitors of villi and syncytiotrophoblast. Gem! is the
1998; Scott et al., 2000). Suppression of Mash2 DNA-binding only gene identified to date that has been shown to be critical
activity is also likely the mode of action of the I-mfa protein for syncytiotrophoblast differentiation.
(Kraut et al, 1998). I-mfa binds to Mash2 in vitro and Several other transcription factors are critical for develop-
embryos lacking I-mfa show fewer giant cells than normal mment of the labyrinth layer after initiation of morphogenesis,
(Kraut et al., 1998). This phenotype is not observed in some such as DIx3 (Morasso et al., 1999), Ppary (Barak et al., 1999)
genetic backgrounds, however, indicating that this function is and Esx1 (Li and Behringer, 1998). The cellular function of
modulatory but not essential. these factors, and how they are regulated, is not entirely clear.
‘The differentiation oftrophoblast giant cells is thought to be In contrast, genes encoding several components of known
a ‘default’ pathway in the sense that if primary trophoblast biochemical pathways show similar mutant phenotypes
cells are cultured (in the absence of FGF) they spontaneously (Hemberger and Cross, 2001; Rossant and Cross, 2001).
differentiate into giant cells (Rossant, 1986). Despite this, it is Because of the similarity of their phenotypes, and localized
clear that some extrinsic cues can regulate giant cell differen- sites of expression or function, it is possible to describe
tiation. Whereas low oxygen levels, at least in vitro, can block signalling pathways among different cell types at the chorioal-
giant cell formation (Adelman et al., 2000), treatment with lantoic interface (Hemberger and Cross, 2001; Rossant and
retinoic acid stimulates giant cell differentiation both in vitro Cross, 2001). One example is the Wnt signalling pathway.
and in vivo (Yan et al., 2001). Retinoid signalling is likely to be Deficiency for 775 (Parr et al., 2001) or genes encoding the
128 Placenta (2003), Vol. 24
Whnt-activated transcription factors, T¢f/ Lefl (Galceran et al., MMP9 (Librach et al., 1991; Alexander et al., 1996), the cell
1999), is associated with failures in chorioallantoic attachment. adhesion molecule o,B, integrin (Sutherland et al., 1993;
This may reflect a requirement for Wnt signalling in either Damsky et al., 1994) and the transcription factor Stral3
polarity of the trophoblast cells in the chorion or induction of (Boudjelal et al., 1997; Janatpour et al., 1999). HANDI1
expression of the cell adhesion molecule a4 integrin in the expression in human trophoblast has been detected in troph-
mesothelial cells underlying the basal surface of the chorion ectoderm of blastocysts (Knofler et al., 2002), similar to mouse
(Parr et al,, 2001). Wnt signalling is also required later in (Cross et al., 1995), and in choriocarcinoma cells (Knofler et
labyrinth development, as shown by mutant phenotypes for al., 1998) but not as yet in the mature placenta. Gem1/GCMI
the WWnt2 (Monkley et al, 1996) and Fzd5 (encoding a Wnt is expressed in a defined subset of cells within the labyrinth in
receptor) (Ishikawa et al., 2001) genes. The V2 gene is mice (Anson-Cartwright et al., 2000) and the chorionic villi in
expressed by the allantois and Fzd5 appears to be expressed in humans (Janatpour et al., 1999; Nait-Oumesmar et al., 2000;
the trophoblast component of the labyrinth. Growth factor DB, J.C.C. and J.C.P.K., manuscript submitted). The syn-
signalling through receptor tyrosine kinases is also required for cytiotrophoblast layers of the murine labyrinth in mice and the
chorioallantoic morphogenesis (Rossant and Cross, 2001). floating chorionic villi in humans are characterized by Tef5/
‘Whereas a null mutation in Fgfr2 is associated with early TEFS expression (Jacquemin et al., 1998).
blastocyst-stage phenotype, consistent with a role in tropho- Given the considerable conservation in trophoblast subtype-
blast stem cell proliferation (Arman et al., 1998), a hypomor- specific expression between mice and humans, it seems likely
phic mutation results in a chorioallantoic branching defect (Xu that the molecular functions are conserved as well. While
et al., 1998). A similar phenotype is observed in mutants for human systems are not amenable to genetic manipulation in
the Hgf (Schmidt et al., 1995; Uehara et al., 1995) and ¢-Met the same way as the mouse, a limited amount of data is
(encoding the HGF receptor) (Bladt et al., 1995) genes. available that supports the hypothesis of conserved trophoblast
Likewise mutation of genes encoding signalling adaptor pro- cell subtype-specific functions. GCM1 and TEF5 have been
teins (Grb2, Sos1, Mek1, p38) and transcription factors (Fral, shown to transactivate the syncytiotrophoblast-expressed genes
JunB) known to be downstream of receptor tyrosine kinases all encoding aromatase (CYP19) (Yamada et al., 1999) and chori-
produce similar phenotypes (Rossant and Cross, 2001). Similar onic somatomammotropin (Jiang et al., 1999), respectively.
to Wnt2/Fzd5, the major downstream responses of the recep- Mis-expression of ID2 in cultured human trophoblast reduces
tor tyrosine kinases are in the trophoblast component of the their invasion and supports the maintenance of markers typical
labyrinth (Rossant and Cross, 2001). The major impact of of stem cells (Janatpour et al., 2000). The more complete
these findings is that it focuses attention on signalling inter- functional data in mice provided by knockout mice for a
actions between allantoic mesoderm and trophoblast as broader range of genes should clearly help to suggest molecular
important factors in labyrinth development. mechanisms underlying the developmental defects that are
associated with human diseases such as spontaneous abortions,
preeclampsia and intrauterine growth restriction.
Conservation of molecular pathways between
mice and humans
CONCLUSIONS: MOLECULAR EVOLUTION OF
Because many of the genes that are essential for trophoblast THE PLACENTA
development in mice show patterns of expression and func-
tions that are restricted to one specific trophoblast subtype When molecular biology investigations began into the control
(Table 2), a demonstration of where the human orthologues of placental development and regulation of placental-specific
are expressed helps to make explicit functional comparisons genes, the conventional wisdom was to go looking for genes
between the human and mouse placenta. Although this has that were uniquely expressed there. The logic was that the
only been done for a limited number of genes, the information most important developmental regulators would be
is quite revealing (Table 2). Mouse chorionic trophoblast and trophoblast-specific factors. Today though, it is clear that most
human villous cytotrophoblast cells both act as stem cell of the genes that have been shown to be essential for placental
populations and are identified by expression of 142/ID2 (Jen development are also involved in the development of other
et al., 1997; Janatpour et al., 2000). Mash2 is expressed in the organs and that expression of only a very limited number of
‘intermediate’ type of trophoblasts in the EPC/ genes is restricted to the placenta. This implies that evolution
spongiotrophoblast in mice (Guillemot et al., 1994), and its of the placenta among vertebrates did not involve invention of
human orthologue (MASH2 or HASH2) is detected in an entirely new set of genes. Rather, nature has re-used
cytotrophoblast cell columns (Alders et al., 1997; Janatpour et existing pathways for functions common to other systems (e.g.,
al, 1999). Tn both organisms, this gene is down regulated FGF signalling and branching morphogenesis in the lung as
during trophoblast differentiation (Guillemot et al., 1994; well as the placenta), re-cycled old genes to take on new
Janatpour et al., 1999; Scott et al., 2000). Invasive, extravillous functions (e.g., Hand] functions in placenta, heart and neural
cytotrophoblast cells in humans and trophoblast giant cells in crest), and duplicated and diverged existing multi-gene famil-
rodents share expression of the matrix-degrading enzyme ies to provide a pool of ‘genetic talent’ dedicated to placental
Cross et al.: Genes, Development and Evolution of the Placenta 129

development (e.g., Geml, Mash2). The variations in placental reflect slightly more or less elaborate use of the ‘re-use,
structures that we see among eutherian mammals may simply re-cycle and duplicate’ plan.

ACKNOWLEDGEMENTS
Work performed by the authors was supported by grants from the Canadian Institutes of Health Rescarch (CIHR), Alberta Heritage Foundation for Medical
Rescarch (AHFMR) and the Physicians Services Incorporated Foundation of Ontario. JCC is an Investigator of the CIHR and a Senior Scholar of the AHFMR.

REFERENCES Genbacev O, Joslin R, Damsky CH, Polliotti BM & Fisher SJ (1996)

Hypoxia alters carly gestation human cytotrophoblast ~differentation/
Adamson SL, Lu Y, Whiteley KJ, Holmyard D, Hemberger M, Pfarrer invasion in vitro and models the placental defects that occur in preeclamp-
C & Cross JC (2002) Interactions between trophoblast cells and the sia. 7 Clin Trvest, 97, 540-530.
maternal and fetal circulation in the mouse placenta. Dee Biol, in press. Genbacev O, Zhou Y, Ludlow JW & Fisher SJ (1997) Regulation of human
Adelman DM, Gertsenstein M, Nagy A, Simon MC & Maltepe E (2000) placental development by oxygen tension. Science, 277, 1669-1672.
Placental cell fates are reguiated in vivo by HIF-mediated hypoxia Georgiades P, Ferguson-Smith AC & Burton GJ (2002) Comparative
responses. Genes Dez, 14, 3191-3203. developmental anatomy of the murine and human definitive placentac.
Alders M, Hodges M, Hadjantonakis AK, Postmus J, van Wijk I, Blick Placenta, 23, 3-19.
J, de Meulemeester M, Westerveld A, Guillemot F, Oudejans C, Gilbert SF (1997) Decelopmental Biology. Sunderland, MA: Sinauer Associ-
Little P & Mannens M (1997) The human Achacte-Scute homologue 2 ates Inc.
(ASCL2.HASH2) maps to chromosome 11p15.5, close to IGF2 and is Guillemot F, Nagy A, Auerbach A, Rossant ] & Joyner AL (19%4)
expressd in extravillus trophoblasts. Hum Mol Genet, 6, 859-867. Essential role of Mash-2 in extraembryonic development. Nature, 371,
Alexander CM, Hansell EJ, Behrendtsen O, Flannery ML, Kishnani 333336,
NS, Hawkes SP & Werb Z (1996) Expression and function of matrix HeY, Smith SK, Day KA, Clark DE, Licence DR & Charnock-Jones DS
metalloproteinases and their inhibitors at the maternal-embryonic boundary (1999) Alternative splicing of vascular endothelial growth factor (VEGF)-R1
during mouse embryo imphintation. Decelopment, 122, 1723-1736. (FLT-1) pre-mRNA is important for the regulation of VEGF activity. Mol
Anson-Cartwright L, Dawson K, Holmyard D, Fisher SJ, Lazzarini RA Endocrinol, 13, 537-545.
& Cross JC (2000) The glial cells missing-1 protein is cssential for Hemberger M & Cross JC (2001) Genes governing placental development.
branching morphogenesis in the chorioallantoic placenta. Nat Genet, 25, Trends Endocrin Metab, 12, 162-168.
311-314. Heyer BS, MacAuley A, Behrendtsen O & Werb Z (2000) Hypersensitivity
Arman E, Haffner-Krausz R, Chen Y, Heath JK & Lonai P (1998) to DNA damage leads to increased apoptosis during early mouse develop-
Targeted disruption of fibroblast growth factor (FGF) receptor 2 suggests a ment. Genes Dev, 14, 2072-2084.
role for FGF signaling in pregastrulation mammalian development. Proc Ishikawa T, Tamai Y, Zorn AM, Yoshida H, Seldin MF, Nishikawa S &
Natl Acad Sci USA, 95, 5082-5087. Taketo MM (2001) Mouse Wt receptor gene Fzd is essential for yolk sac
Auman HJ, Nottoli T, Lakiza O, Winger Q, Donaldson S & Williams T and placental angiogenesis. Development, 128, 25-33.
(2002) Transcription factor AP-2gamma is essential in the extra-cmbryonic Jacquemin P, Sapin V, Alsat E, Evain-Brion D, Dolle P & Davidson T
lincages for carly postimplantation development. Development, 129, 2733— (1998) Differential expression of the TEF family of transcription factors in
2747 the murine placenta and during differentiation of primary human tropho-
Barak Y, Nelson MC, Ong ES, Jones YZ, Ruiz-Lozano P, Chien KR, blasts in vitro. Dev Dyn, 212, 423-436.
Koder A & Evans RM (1999) PPAR gamma is required for placental, Janatpour M, McMaster M, Genbacev O, Zhou Y, Dong J, Cross JC,
cardiac, and adipose tissue development. Mol Cell, 4, 585-595. Isracl M & Fisher SJ (2000) 1d-2 reguiates critical aspects of human
Berczowsky J, Zbicranowski I, Demers ] & Murray D (1995) DNA ploidy cytotrophoblast differentiation, invasion and migration. Decelopment, 127,
of hydatidiform moles and nonmolar conceptuses: a study using flow and 549-538.
tissue section image cytometry. Mod Pathol, 8, 775-781. Janatpour MJ, Utset MF, Cross JC, Rossant J, Dong J, Tsracl MA &
Bladt F, Ricthmacher D, Isenmann S, Aguzzi A & Birchmeier C (1995) Fisher SJ (1999) A repertoire of differentialy expressed transcription
Essential role for the c-mret receptor in the migration of myogenic precursor factors that offers insight into mechanisms of human- cytotrophoblast
cells into the limb bud. Nature, 376, 768-771. differentiation. Dev Genet, 25, 146-157.
Boudjelal M, Tancja R, Matsubara S, Bouillet P, Dolle P & Chambon JenY, Manova K & Benezra R (1997) Each member of the Id gene family
exhibits a unique expression pattern in mouse gastrulation and neuro-
P (1997) Overexpression of Stral3, a novel retinoic acid-inducible gene of genesis. Dev Dyn, 208, 92-106.
the basic_helix-loop-helix family, inhibits mesodermal and promotes Jiang SW, Wu K & Eberhardt NL (1999) Human placental TEF-5
neuronal differentiation of P19 cells. Genes Dev, 11, 2052-2065. transactivates the human chorionic somatomammotropin gene enhancer.
Chawengsaksophak K, James R, Hammond VE, Kontgen F & Beck F Mol Endocrinol, 13, $79-889.
(1997) Homeosis and initestinal tumours in Cdx2 mutant mice. Natare, 386, Klisch K, Hecht W, Pfarrer C, Schuler G, Hoffmann B & Leiser R
84-87. (1999) DNA content and ploidy level of bovine placentomal trophoblast
Cross JC (2000) Genetic insights into trophoblast differentiation and placental giant cells. Placenta, 20, 451-458.
morphogenesis. Sem Cell Dev Biol, 11, 105-113. Knofler M, Meinhardt G, Bauer S, Loregger T, Vasicek R, Bloor DJ,
Cross JC, Flannery ML, Blanar MA, Steingrimsson E, Jenkins NA, Kimber SJ & Husslein P (2002) Human Handl basic helix-loop-helix
Copeland NG, Rutter W] & Werb Z (1995) Hxt encodes a basic (bHLH) protein: extracmbryonic expression pattern, interaction partners
helix-loop-helix transcription factor that regulates trophoblast cell develop- and identification of its transcriptional repressor domains. Biachem , 361,
ment. Development, 121, 2513-2523. 641651
Cross JC, Werb Z & Fisher S (1994) Implantation and the placenta: key Knofler M, Meinhardt G, Vasicek R, Husslein P & Egarter C (1998)
picces of the development puzzle. Science, 266, 1508-1518. Molecular cloning of the human Handl gene/cDNA and its. tissue-
Damsky CH, Librach C, Lim KH, Fitzgerald ML, McMaster MT, restricted expression in cytotrophoblastic cells and heart. Gene, 224, 77-86.
Janatpour M, Zhou Y, Logan SK & Fisher SJ (1994) Integrin switching Kraut N, Snider L, Chen C, Tapscott S] & Groudine M (1998)
regulates normal trophoblast invasion. Development, 120, 3657-3666. Requirement of the mouse I-mfa_gene for placental development and
Frank D, Fortino W, Clark L, Musalo R, Wang W, Saxena A, Li CM, skeletal patterning. EMBO 7, 17, 6276-6288.
Reik W, Ludwig T & Tycko B (2002) Placental overgrowth in mice Lescisin KR, Varmuza S & Rossant ] (1988) Isolation and characterization
lacking the imprinted gene Ipl. Proc Natl Acad Sci USA, 99, 7490-7495. of a novel trophoblast-specific cDNA in the mouse. Genes Dez, 2, 1639~
Galceran J, Farinas I, Depew MJ, Clevers H & Grosschedl R (1999) 1646.
Wht3a — /-like phenotype and limb deficiency in Lefl(—/ — )Tefl(—/ Li Y & Behringer RR (198) Esxl is an X-chromosome-imprinted regulator
—) mice. Genes Dew, 13, 709-717. of placental development and fetal growth. Nar Gener, 20, 309-311.
130 Placenta (2003), Vol. 24
Librach CL, Werb Z, Fitzgerald ML, Chiu K, Corwin NM, Esteves RA, Sapin V, Bouillet P, Oulad-Abdelghani M, Dastugue B, Chambon P &
Grobelny D, Galardy R, Damsky CH & Fisher S (1991) 92-kD type IV Dolle P (2000) Differential expression of retinoic acid-inducible (Stra)
collagenase mediates invasion of human cytotrophoblasts. 7 Cel/ Biol, 113, genes during mouse placentation. Mech Dev, 92, 295-299.
437449, Schmidt C, Bladt F, Gocdecke S, Brinkmann V, Zschiesche W, Sharpe
Lopez MF, Dikkes P, Zurakowski D & Villa-KomaroffL (1996) Insulin- M, Gherardi E & Birchmeicr C (1995) Scatter factor/hepatocyte growth
like growth factor T affects the appearance and glycogen content of glycogen factor is essential for liver development. Nature, 373, 699-702.
cells in the murine placenta. Endocrinology, 137, 2100-2108. Scott IC, Anson-Cartwright L, Riley P, Reda D & Cross JC (2000) The
Luo J, Sladek R, Bader J-A, Matthyssen A, Rossant ] & Giguere V (1997) Hand1 basic helix-loop-helix transcription factor regulates trophoblast
Placental_abnormalitics in mouse embryos lacking the orphan nuclear cell differentitation via multiple mechanisms. Mol Cell Biol, 20, 530-541.
receptor ERR-p.. Nature, 388, 778-782. Sutherland AE, Calarco PG & Damsky CH (1993) Developmental
Ma GT, Soloveva V, Tzeng SJ, Lowe LA, Pfendler KC, Tannaccone PM, regulation of integrin expression at the time of implantation in the mouse
Kuchn MR & Linzer DI (2001) Nodal regulates trophoblast differentiation embryo. Dezelopment, 119, 1175-1186.
and placental development. De Biol, 236, 124-135.
MacAuley A, Cross JC & Werb Z (1998) Reprogramming the cell cycle for Takahashi K, Kobayashi T & Kanayama N (2000) p57(Kip2) regulates
endoreduplication in rodent trophoblast cells. Mol Biol Cell, 9, 793-807. the proper development of labyrinthine and spongiotrophoblasts. Mol Hum
Monkley SJ, Delancy SJ, Pennisi DJ, Christiansen JH & Wainwright B Reprod, 6, 1019-1025.
(1996) Targeted distuption of the Wnt2 gene results in placentation defects. Tanaka M, Gertsenstein M, Rossant ] & Nagy A (1997) Mash2 acts cell
Development, 122, 3343-3353. autonomously in mouse spongiotrophoblast development. Dev Biol, 190,
Morasso MI, Grinberg A, Robinson G, Sargent TD & Mahon KA (1999) 55-65.
Placental failure in mice lacking the homeobox gene DIx3. Proc Natl Acad Tanaka S, Kunath T, Hadjantonakis AK, Nagy A & Rossant J (1998)
Sci USA, 96, 162-167. Promotion of trophoblast stem cell proliferation by FGF4. Science, 282,
Nait-Oumesmar B, Copperman AB & Lazzarini RA (2000) Placental 2072-2075.
expression and chromosomal localization of the human GCMI gene. J Tremblay GB, Kunath T, Bergeron D, Lapointe L, Champigny C,
Histochem Cytochem, 48, 915-922. Bader JA, Rossant ] & Giguere V (2001) Dicthylstilbestrol regulates
Nakayama H, Scott IC & Cross JC (1996) Potential role of Notch/HES trophoblast stem cell differentiation as a ligand of orphan nuclear receptor
signaling in regulation of murine trophoblast differentiation and Mash-2 ERR beta. Genes Dez, 13, 833-838.
expression. Biol Reprod, $4(suppl 1), 176. Uchara Y, Minowa O, Mori C, Shiota K, Kuno J, Noda T & Kitamura
Norton JD (2000) ID helix-loop-helix proteins in cell growth, differentiation N (1995) Placental defect and embryonic lethality in mice lacking hepato-
and tumorigenesis. 7 Cell Sci, 113, 3897-3905. cyte growth factor/satter factor. Nature, 373, 702-705.
Parr BA, Cornish VA, Cybulsky MI & McMahon AP (2001) Wat7b Wegner M & Riethmacher D (2001) Chronicles ofa switch hunt: gem genes
regulates placental development in mice. Dev Biol, 237, 324-332. in development. Trends Genet, 17, 286-290.
Pijnenborg R, Robertson WB, Brosens I & Dixon G (1981) Review article:
trophoblast invasion and the establishment of hacmochorial placentation in Wendling O, Chambon P & Mark M (1999) Retinoid X receptors are
man and laboratory animals. Placenta, 2, 71-91. essential for early mouse development and placentogenesis. Proc Natl Acad
Plum A, Winterhager E, Pesch J, Lautermann J, Hallas G, Rosentreter Sci USA, 96, 347-551.
B, Traub O, Herberhold C & Willecke K (2001) Connexin3I-deficiency Werling U & Schorle H (2002) Transcription factor gene AP-2 gamma
in mice causes transient placental dysmorphogenesis but docs not impair essential for early murine development. Mol Cell Biol, 22, 3149-3156.
hearing and skin differentiation. Dev Biol, 231, 334-347. Wooding FBP & Flint APF (1994) Placentation. In Marshall’s Physiology of
Riley P, Anson-Cartwright L & Cross JC (1998) The Handl bHLH Reproduction, 4th edition (EA.) Lamming GE, pp. 233-460. New York:
transcription factor is essential for placentation and cardiac morphogenesis. Chapman and Hall
Nat Genet, 18, 271-275. Xu X, Weinstein M, Li C, Naski M, Cohen RI, Ornitz DM, Leder P &
Rinkenberger J, Cross JC & Werb Z (1997) Molecular genetics of Deng C (1998) Fibroblast growth factor receptor 2 (FGFR2)-mediated
implantation in’the mouse. Dev Genet, 21, 6-20. reciprocal regulation loop between FGFS and FGFI0 is essential for limb
Rossant ] (1986) Development of extracmbryonic cell lineages in the mouse induction. Development, 125, 753-765.
embryo. In Experimental Approaches to Mammalian Embryonic Development Yamada K, Ogawa H, Honda S, Harada N & Okazaki T (1999) A GCM
(Ed) Rossant J & Pedersen RA, pp. 97-120. Cambridge: Cambridge motif protein is involved in placenta-specific expression of human aromatase
University Press. gene. 7 Biol Chem, 274, 32279-32286.
Rossant ] & Cross JC (2001) Placental development: lessons from mouse
mutants. Nat Rev Genet, 2, 538-348. Yan J, Tanaka S, Oda M, Makino T, Ohgane J & Shiota K (2001)
Russ AP, Wattler S, Colledge WH, Aparicio SA, Carlton MB, Pearce JJ, Retinoic acid promotes differentiation of trophoblast stem cells to a giant
Barton SC, Surani MA, Ryan K, Nehls MC, Wilson V & Evans MJ cell fate. Dev Biol, 235, 422-432.
(2000) Eomesodermin is required for mouse trophoblast development and Zybina EV & Zybina TG (19%6) Polytene chromosomes in mammalian cells.
mesoderm formation. Natare, 404, 95-99. Int Rev Gytol, 165, 53-119.