BIOLOGY OF REPRODUCTION 69, 1–7 (2003)
Published online before print 5 March 2003.
DOI 10.1095/biolreprod.102.014977
M i n i r ev i ew
Endovascular Trophoblast Invasion: Implications for the Pathogenesis of Intrauterine
Growth Retardation and Preeclampsia
Peter Kaufmann,1 Simon Black, and Berthold Huppertz
Department of Anatomy II, University of Technology Aachen, D-52057 Aachen, Germany
ABSTRACT
Maternal uteroplacental blood flow increases during preg-
nancy. Altered uteroplacental blood flow is a core predictor of
abnormal pregnancy. Normally, the uteroplacental arteries are
invaded by endovascular trophoblast and remodeled into dilat-
ed, inelastic tubes without maternal vasomotor control. Dis-
turbed remodeling is associated with maintenance of high utero-
placental vascular resistance and intrauterine growth restriction
(IUGR) and preeclampsia. Herein, we review routes, mecha-
nisms, and control of endovascular trophoblast invasion. The re-
viewed data suggest that endovascular trophoblast invasion in-
volves a side route of interstitial invasion. Failure of vascular
invasion is preceded by impaired interstitial trophoblast inva-
sion. Extravillous trophoblast synthesis of nitric oxide is dis-
cussed in relation to arterial dilation that paves the way for en-
dovascular trophoblast. Moreover, molecular mimicry of invad-
ing trophoblast-expressing endothelial adhesion molecules is
discussed in relation to replacement of endothelium by tropho-
blast. Also, maternal uterine endothelial cells actively prepare
endovascular invasion by expression of selectins that enable tro-
phoblast to adhere to maternal endothelium. Finally, the mother
can prevent endovascular invasion by activated macrophage-in-
duced apoptosis of trophoblast. These data are partially contro-
versial because of methodological restrictions associated with
limitations of human tissue investigations and animal studies.
Animal models require special care when extrapolating data to
the human due to extreme species variations regarding tropho-
blast invasion. Basal plates of delivered placentas or curettage
specimens have been used to describe failure of trophoblast in-
vasion associated with IUGR and preeclampsia; however, they
are unsuitable for these kinds of studies, since they do not in-
clude the area of pathogenic events, i.e., the placental bed.
apoptosis, early development, placenta, syncytiotrophoblast, tro-
phoblast
INTRODUCTION
During the first half of human pregnancy, uteroplacental
arteries undergo a series of pregnancy-specific changes that
1
Correspondence: Peter Kaufmann, Department of Anatomy II, University
Hospital Aachen, University of Technology Aachen, Wendlingweg 2, D-
52057 Aachen, Germany. FAX: 49 241 80 82 472;
e-mail: pkaufmann@ukaachen.de
Received: 23 December 2002.
First decision: 22 December 2003.
Accepted: 26 February 2003.
Q 2003 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
1
Downloaded from https://academic.oup.com/biolreprod/article/69/1/1/2712724 by guest on 01 October 2021
include 1) apparent replacement of endothelium and media
smooth muscle cells by invasive trophoblast, 2) loss of elas-
ticity, 3) dilation to wide, incontractile tubes, and 4) loss
of vasomotor control [1] (Fig. 1A and B). There is general
agreement that spiral artery remodeling reduces maternal
blood flow resistance and increases uteroplacental perfusion
to meet the requirements of the fetus. Moreover, the loss
of contractility and maternal vasomotor control guarantees
maternal blood supply to the placenta, irrespective of ma-
ternal attempts to regulate the blood distribution within the
body [2, 3].
In 1972, Brosens and coworkers [4] described reduced
trophoblast invasion and absence of pregnancy-specific
changes of uteroplacental arteries in placental bed speci-
mens from pregnancies associated with intrauterine growth
retardation (IUGR) often combined with preeclampsia (Fig.
1C) [4, 5]. Since that time, endovascular trophoblast inva-
sion has been one of the major foci of placental research.
The accepted core hypothesis is that reduced endovascular
trophoblast invasion and uteroplacental artery remodeling
are key pathologic features of IUGR and preeclampsia.
However, hypotheses about the molecular mechanisms that
regulate trophoblast invasion and uteroplacental artery re-
modeling are still controversial. The aims of this review are
to reevaluate these controversial hypotheses that explain the
regulation of endovascular trophoblast invasion and utero-
placental artery remodeling, evaluate the various supporting
data, and provide a basis for new research aimed at under-
standing the leading cause of maternal death: preeclampsia.
DEFINITION OF TERMS
During implantation and subsequent trophoblast inva-
sion, fetal trophoblast cells and maternal uterine tissues (en-
dometrium and myometrium) come into intimate contact
with each other. The resulting zone of mixed origin is called
the maternofetal junctional zone (Fig. 1B). Those parts that
adhere to the placenta after delivery and form the bottom
of the intervillous space are called the basal plate (Fig. 1B).
The remaining parts of the junctional zone that adhere to
the uterine wall after delivery make up the placental bed
(Fig. 1B).
All trophoblast cells residing outside the placental villi
are summarized under the term extravillous trophoblast
(Fig. 1A). In the basal plate, the extravillous trophoblast
forms proliferating clusters of stem cells, so-called cell col-
umns. The latter connect so-called anchoring villi to the
basal plate (Fig. 1A). The nonproliferative, invasive daugh-
ter cells of the cell columns that invade the uterine inter-
2 KAUFMAN ET AL.

Downloaded from https://academic.oup.com/biolreprod/article/69/1/1/2712724 by guest on 01 October 2021

FIG. 2. Hypothetical routes of endovascular trophoblast invasion. Blue:

fetal tissues, including interstitial trophoblast and its intravasating deriv-
atives. Green: extravasation route of endovascular trophoblast. Red: ma-
ternal tissues. Note that in the hypothetical case of extravasation (A), the
endovascular trophoblast cells invade via the arterial lumen and are de-
rived from an unknown origin. In the rhesus monkey, this source is said
to be the trophoblastic shell, which, however, in the human vanishes
FIG. 1. Schematic representation of interstitial and endovascular tropho- before trophoblast invasion. By contrast, in the more likely case of in-
blast invasion in human pregnancy before Week 6 of gestation (A), after travasation (B), endovascular trophoblast is derived from cell columns via
Week 20 of normal gestation (B), and after Week 20 of gestation in pre- the interstitial invasion route.
eclampsia and/or IUGR (C). Blue: fetal tissues. Red: maternal tissues. ps:
zone of placental separation, where the basal plate (above, attached to
the placenta) separates from the placental bed (remaining in the uterus
after delivery). Note that failure of endovascular trophoblast invasion in
IUGR and preeclampsia (C) is restricted to the placental bed and does
not affect segments of the uteroplacental arteries in the later basal plate
of the placenta.
 ENDOVASCULAR TROPHOBLAST INVASION 3

stitium comprise the interstitial trophoblast (Fig. 1A).

Those invasive extravillous trophoblast cells that infiltrate
arterial walls and lumens make up the endovascular tro-
phoblast (Fig. 1B). Both the route of trophoblast invasion
from the proliferating stem cells of the cell columns into
the placental bed and the route from cell columns into the
uteroplacental arteries are summarized as the invasive path-
way (Fig. 2B).

PHYSIOLOGICAL CHANGES OF SPIRAL ARTERIES

Remodeling of maternal uterine spiral arteries is crucial
for normal growth and development of the fetus. Remod-
eling of uteroplacental arteries, the so-called physiological
changes of uteroplacental arteries, can be divided according

Downloaded from https://academic.oup.com/biolreprod/article/69/1/1/2712724 by guest on 01 October 2021

to structural criteria into three stages: 1) trophoblast inva-
sion-independent vascular changes, 2) vascular remodeling
induced by perivascularly located interstitial trophoblast,
and 3) trophoblast infiltration of vessel walls. IUGR follows
trophoblast malinvasion of the uteroplacental arteries. In
these pregnancies, the failure of arterial remodeling results
in malperfusion of the placenta. The role of each of the
three stages in failure of uteroplacental artery remodeling
remains open. It is very likely that only the full sequence
of events guarantees a degree of arterial dilation adequate
to sufficiently perfuse the placenta throughout all stages of
pregnancy.
Trophoblast Invasion-Independent Vascular Changes
The initial changes to uteroplacental arteries involve a
generalized perturbation of these arteries, endothelial ba-
sophilia and vacuolation, disorganized vascular smooth
muscle, and lumen dilation [6]. The pregnancy-induced
changes in uteroplacental arteries are independent of direct
trophoblast invasion and are considered to involve maternal
activation of local decidual artery renin-angiotensin sys-
tems [6]. Moreover, Craven and coworkers [6] demonstrat-
ed that during intrauterine pregnancies spiral arteries from
both implantation and nonimplantation regions display
these physiological changes. Furthermore, endometrial spi-
ral arteries undergo the same physiological vascular mod-
ifications in ectopic pregnancies.
Vascular Remodeling Induced by Perivascularly Located
Interstitial Trophoblast
Following trophoblast invasion-independent changes, the
uteroplacental arteries within the implantation region are
invaded by extravillous trophoblast cells. In a first step,
extravillous trophoblast cells in juxtaposition against utero-
placental artery structures are associated with further vas-
cular remodeling. The latter findings, described in the guin-
ea pig [2, 7, 8], comprise reduction of media smooth muscle
cells and deposition of fibrinoid material before infiltration
of the media by trophoblast. Respective observations in the
human have not yet been reported.
Trophoblast Infiltration of Vessel Walls
The third stage of uteroplacental vascular remodeling is
characterized by infiltration of the arterial wall by endo-
vascular trophoblast. The uteroplacental arteries undergo
further dilation up to several times the original diameter of
the lumen [1, 3, 9]. Trophoblast infiltration of the media
smooth muscle coincides with loss of elastic fibers [10, 11].
A debate exists regarding whether smooth muscle cells un-
dergo cell death and become replaced by endovascular tro-
FIG. 3. Defense of endovascular trophoblast invasion (via intravasation)
by activated maternal macrophages of the arterial walls in preeclampsia
and/or IUGR. These macrophages force approaching trophoblast cells into
apoptosis by secretion of TNFa and IDO, the latter causing local trypto-
phan depletion. Blue: fetal tissues, including trophoblast. Red: maternal
tissues. Lilac: activated maternal macrophages. Note that in the case of
preeclampsia activated maternal macrophages are accumulated in the
proximal parts of the uteroplacental arteries located in the deeper zones
of the placental bed but are missing in the superficial layers of the junc-
tional zone (basal plate).
phoblast [1] or temporary molecular and structural dedif-
ferentiation [8] during trophoblast invasion. The same ques-
tion was postulated for the replacement of some of the en-
dothelial cells. Large pleomorphic cells that line
uteroplacental arteries within the proximal decidual seg-
ments express factor VIII-related antigen but not human
chorionic gonadotropin, thus suggesting that not all altered
cells without an obvious endothelial phenotype are, in fact,
trophoblast cells [12].
SITES AND ROUTES OF ENDOVASCULAR
TROPHOBLAST INVASION
Endovascular trophoblast invasion is not a homogeneous
process. The density of extravillous trophoblast and the
depth of invasion of uteroplacental arteries are most pro-
nounced in the central region of the placental bed. The
density and depth of invasion of extravillous trophoblast
and the degree of invasion of spiral arteries diminish toward
the placental margin [13]. Central-region placental bed
specimens obtained from normal pregnancies reveal endo-
vascular trophoblast that invaded and dilated uteroplacental
arteries up to the first third of the myometrium, a depth
similar to interstitial trophoblast invasion in this region. The
anatomical pathways taken by endovascular trophoblast
have been a matter of controversy between the extravasa-
tion and intravasation model.
Extravasation
Based on detailed studies on the rhesus monkey, Blan-
kenship and coworkers [14] concluded that endovascular
trophoblast derived from an unknown source gained access
4 KAUFMAN ET AL.
to the arterial lumens via or close to their point of conflu-
ence with the intervillous space. Thereafter, the cells mi-
grate along the arterial lumens retrograde to blood flow by
adhering to and replacing endothelium, locally forming in-
traluminal trophoblastic plugs (Fig. 2A). Finally, a certain
number of these cells were thought to leave the lumen and
centrifugally invade media and adventitia.
Intravasation
Based on studies in the human, most researchers favor
the contrasting concept of intravasation. Structural criteria
[15] and immunohistochemical data [16, 17] revealed that
endovascular trophoblast represents an end stage of differ-
entiation of interstitial trophoblast derived from the cell col-
umns. As a side step, a subpopulation of extravillous tro-
phoblast cells invades the arterial walls from the surround-
ing junctional zone and finally enters the arterial lumens
(Fig. 2B). Whether or not the intravasated cells then mi-
grate inside the arterial lumens and even locally may ex-
travasate remains open.
A combination of both hypotheses was suggested by
Kam et al. [18]. These authors described infiltration and
replacement of arterial media and adventitia by interstitial
trophoblast, followed by the replacement of endothelium
by a separate population of endovascular trophoblast, the
derivation of which was not described.
All of these data on intravasation or extravasation were
collected on uteroplacental arteries. By contrast, Craven et
al. [19] presented convincing evidence that peripheral villi
were directed by the uteroplacental blood flow into mar-
ginal veins. These villi adhered to the endothelial surfaces
and gave rise to cell columns, the cells of which extrava-
sated the venous walls. This villus-to-vein invasion route
requires further investigation. To our best knowledge, it has
never been described in uteroplacental arteries.
The answer to the question of whether extravasation or
intravasation takes place in uteroplacental arteries is crucial
for the understanding of the pros and cons of the various
hypotheses described herein. It is important to recognize
that extravasation was described in the rhesus monkey, a
species with a nearly complete trophoblastic shell that sep-
arates the intervillous space and maternofetal junctional
zone. This trophoblastic shell may be a source for intralu-
minal trophoblast migration. In contrast, the trophoblastic
shell of the very early gestation human placenta becomes
rarified to widely spread cell columns not in contact with
the terminal structures of the maternal arteries [3].
The ability of the early trophoblastic shell from the first
weeks of pregnancy to be the source of endovascular tro-
phoblast extravasating until the end of pregnancy requires
endovascular trophoblast to remain in the cell cycle and to
represent a self-replicating population in those stages of
pregnancy in which the trophoblastic shell is no longer
available. In agreement with the extravasation theory, the
endovascular trophoblast in the rhesus monkey has been
described by King and Blankenship [20] to maintain pro-
liferation based on proliferating cell nuclear antigen
(PCNA) immunohistochemical analysis (PCNA antibody
PC10). The PCNA displays a long half-life (20 h or more)
[21]; therefore, cells may remain immunopositive for days
after leaving the cell cycle. Human endovascular tropho-
blast has not been observed to proliferate according to Ki67
immunohistochemical analysis, 3H-thymidine incorporation
studies, or assessment of mitotic figures. Proliferation of
human extravillous trophoblast has been observed exclu-
sively in the trophoblast that rests on the basal lamina of
cell columns [22].
We conclude from these data that the extravillous tro-
phoblast that emanates from the cell columns provides cells
for the interstitial route of trophoblast invasion. Cells from
the latter route invade (intravasate) uteroplacental arteries
and contribute to the remodeling process by replacing ar-
terial media and endothelium (Fig. 2B).
DOES MISSING TROPHOBLASTIC EXPRESSION
OF A VASCULAR PHENOTYPE CONTRIBUTE
TO MALINVASION OF UTEROPLACENTAL ARTERIES?
The expression of cell adhesion molecules is necessary
for trophoblast invasion because these molecules enable the
Downloaded from https://academic.oup.com/biolreprod/article/69/1/1/2712724 by guest on 01 October 2021
trophoblast to adhere to the extracellular matrix, form col-
onies, and target cells in the vessel wall. Numerous studies
reported that 1) proliferative and early postproliferative
cells within the proximal invasive pathway (immunoreac-
tive with proliferation marker Mib 1) express epithelial ad-
hesion molecules and 2) the nonproliferative trophoblast of
the distal invasive pathway (immunonegative with Mib 1)
acquires an invasive phenotype by switching its repertoire
of adhesion molecules to one resembling that of mesenchy-
mal derivatives [22–24]. Some of these studies [25, 26]
suggested that trophoblast that approaches the uteroplacen-
tal arteries and replaces the endothelium mimicked the ad-
hesion molecule expression patterns found on endothelial
cells.
Zhou et al. [25, 26] introduced the hypothesis that im-
paired invasion of uteroplacental arteries is due to tropho-
blastic failure to acquire the vascular repertoire of adhesion
molecules. In normal pregnancies, these authors reported a
generally reduced expression of E-cadherin in extravillous
trophoblast, whereas up-regulating expression of VE-(en-
dothelial) cadherin, platelet-endothelial adhesion molecule-
1 (PECAM-1), vascular endothelial adhesion molecule 1
(VECAM-1), and a4-integrins. Endovascular trophoblast
continues to express these receptors and, like activated en-
dothelial cells, acquires avb3 [25]. The same group re-
ported that extravillous trophoblast in preeclampsia failed
to express most of these endothelial markers and hypothe-
sized that expression of vascular phenotyped trophoblast is
required for successful endovascular invasion [26].
However, the trophoblast-endothelial mimicry model has
not been supported by other investigators. For example, af-
ter studying placental bed biopsy specimens, Lyall and co-
workers [27] reported PECAM-1 expression was not de-
tected in extravillous trophoblast but observed only in en-
dothelial cells. Moreover, no differences in cell-type pat-
terns of PECAM-1 expression were observed between
normal pregnancy, preeclampsia, and IUGR. Also regard-
ing integrins, the patterns found in normal pregnancies were
comparable to those in preeclampsia [28]. Finally, the
down-regulation of E-cadherin during trophoblast invasion,
described by Zhou et al. [25, 26], was not supported in a
later study [29].
A contrasting vascular adhesion hypothesis has been pre-
sented by other authors [30–32]. King and Loke [30] de-
scribed a model of endotheliotrophoblastic interaction,
where the maternal endothelium at the implantation site un-
dergoes pregnancy-induced changes that allow their re-
placement by trophoblast. These authors reported that en-
dovascular trophoblast expresses the cell-surface carbohy-
drate sialyl-Lewisx, normally located on leukocytes. This
carbohydrate is the cognate ligand of E- and P-selectins.
 ENDOVASCULAR TROPHOBLAST INVASION
Both these lectins are expressed by endothelium during in-
flammatory reactions. Leukocytes attach to endothelium via
interactions between sialyl-Lewisx and selectins and sub-
sequently migrate through vessel walls. During pregnancy,
maternal endothelial E- and P-selectin expression occurs
exclusively at the implantation site [31] and may provide a
mechanism for maternal and fetal cell interaction to enable
trophoblast to home within the uteroplacental vessel lu-
mens. The sialyl-Lewisx- E-selectin interaction is also in-
volved in adhesion of cancer cell lines to human umbilical
vein endothelial cells in vitro [32].
DOES MISSING TROPHOBLASTIC SECRETION
OF NITRIC OXIDE CONTRIBUTE TO MALINVASION
OF UTEROPLACENTAL ARTERIES ONLY IN RODENTS?
Pregnancy-induced dilation of uteroplacental arteries
was considered to result from the destruction of musculo-
elastic structures of vessels by invading trophoblast [1, 33].
This widely accepted hypothesis is contradicted by the ob-
servations that arterial dilation in human pregnancy starts
at 8 wk of gestation before invasion of vessels by tropho-
blast [2, 6, 7, 15]. Guinea pig uteroplacental arteries com-
mence dilating when approached by interstitial trophoblast
expressing nitric oxide synthase (endothelial nitric oxide
synthase—eNOS—and possibly also inducible nitric oxide
synthase—iNOS) [8]. Only the already dilated arteries were
subsequently invaded by trophoblast. Moreover, in this spe-
cies the trophoblast invasion of uteroplacental arteries is not
associated with removal of media smooth muscle cells, but
rather muscle cells dedifferentiate to myoblasts [8]. These
myoblasts redifferentiated to an intact media within a few
days after delivery [34].
Lyall et al. [35] could not detect endovascular tropho-
blast expression of eNOS or iNOS in placental bed biopsy
specimens from Weeks 8 to 19 of human pregnancy and
have questioned whether the guinea pig data on nitric oxide
secretion can be transferred to the human. However, Martin
and Conrad [36] reported eNOS expression in human in-
terstitial trophoblast using immunohistochemical and in situ
hybridization analysis. Moreover, using a human extravil-
lous trophoblast cell line that expresses both the constitu-
tive (eNOS) and the inducible isoforms (iNOS), Cartwright
and coworkers [37] have shown that trophoblast cell mo-
tility and invasion strongly depend on trophoblast-derived
NOS in vitro.
The question of whether pregnancy-induced dilation by
trophoblast-derived nitric oxide is specific only for rodents
and human trophoblast in vitro or also occurs in the human
in vivo has implications for understanding the pathogenesis
of preeclampsia. Chwalisz and Garfield and coworkers [38,
39] reported preeclampsia like biological responses, includ-
ing hypertension, proteinuria, and fetal growth retardation
in rats and guinea pigs following long-term inhibition of
NOS with L-NAME.
DO ACTIVATED MATERNAL MACROPHAGES
PREVENT PREGNANCY-INDUCED ADAPTATION
OF UTEROPLACENTAL ARTERIES BY INHIBITING
INVASION VIA INDUCTION OF TROPHOBLAST
APOPTOSIS?
Our discussion on adaptation of uteroplacental arteries to
pregnancy conditions focused largely on intrinsic tropho-
blastic features associated with invasion. The trophoblastic
orientation of the respective hypotheses implied trophoblast
failure that caused impaired endovascular trophoblast inva-
5
sion associated with IUGR or preeclampsia. However, well-
known maternal risk factors for preeclampsia and IUGR (re-
nal disease, diabetes, obesity, and psychosocial stress) render
it unlikely that trophoblastic failure is the sole pathogenic
mechanism. A dominant maternal-factor model used to ex-
plain preeclampsia is dysfunction of normal trophoblast im-
mune privilege. A promising maternal pathogenic link might
be presented by macrophages.
Maternal macrophages are normal constituents of the
placental implantation site and can be demonstrated by an-
tibodies to CD14 or CD68 [40, 41]. Greater numbers of
macrophages are found in the decidua basalis compared
with the decidua parietalis, where trophoblast invasion is
limited. The observations of differential macrophage dis-
tributions hint at interactions between trophoblast and mac-
Downloaded from https://academic.oup.com/biolreprod/article/69/1/1/2712724 by guest on 01 October 2021
rophages [42]. Macrophages produce and respond to a wide
range of cytokines and may be involved in decidual para-
crine networks that regulate trophoblast invasion [43, 44].
Activated macrophages produce high levels of tumor ne-
crosis factor a (TNFa) [44]. One of the cognate receptors
of TNFa, TNF receptor 1 (TNF-R1), is expressed by tro-
phoblast cells [45], and interactions between TNFa and
TNF-R1 were described to induce trophoblast apoptosis in
vitro [46].
Using an immortalized extravillous trophoblast cell line,
we have further substantiated and extended these data [47].
The experiments revealed that activated macrophages in-
duce trophoblast apoptosis by the concerted action of two
mechanisms: 1) by secretion of TNFa that binds to the
trophoblastic TNF-R1 and 2) by secretion of indolamine
2,3-dioxygenase (IDO) that catabolizes and depletes local
levels of tryptophan (Fig. 3). These data explain the im-
munohistochemically evident inverse relation between the
amount of endovascular trophoblast and macrophages in
the wall of uteroplacental arteries [41] (Fig. 3). The fact
that macrophages induce trophoblast apoptosis in vitro ren-
ders it unlikely that the increased macrophage population
in the arterial walls in preeclampsia is simply due to apo-
ptotic attraction of macrophages [47]. However, combining
both possible interactions between macrophages and the
trophoblast, it could be hypothesized that macrophage-in-
duced trophoblast apoptosis attracts and activates more
macrophages, leading to a vicious cycle. In normal preg-
nancy, the walls of uteroplacental arteries are largely de-
void of macrophages and become invaded by the tropho-
blast. In contrast, preeclampsia is associated with reduced
trophoblast invasion of uteroplacental vessels, and accu-
mulation of apoptotic interstitial trophoblast juxtaposing the
arteries correlate with maternal macrophages in the arterial
media [47]. In addition, murine macrophage function is in-
hibited by high-dose progesterone. In particular, expression
of iNOS and TNFa was reduced [48].
IUGR AND PREECLAMPSIA: INCOMPETENT INVASION
OF EXTRAVILLOUS TROPHOBLAST OR EXAGGERATED
MATERNAL DEFENSE AGAINST INVASION?
It is generally believed that preeclampsia is associated
with a generalized impairment of trophoblast invasion, that
is, both interstitial and endovascular trophoblast invasion
are reduced. Statements such as ‘‘preeclampsia is associated
with abnormally shallow placentation’’ [49], ‘‘hypoinvasive
placental phenotype characteristic of preeclampsia;’’ [50],
and ‘‘shallow trophoblast invasion . . . predisposing the
pregnancy to preeclampsia’’ [51] suggest impaired/shallow
interstitial trophoblastic invasion results in impaired endo-
6 KAUFMAN ET AL.
vascular trophoblast invasion and leads to maladaptation of
uteroplacental arteries. Recent quantitative studies on inter-
stitial trophoblast invasion in hysterectomized uteri from
patients with preeclampsia have revealed that both invasive
depth and numerical density of interstitial extravillous tro-
phoblast are significantly reduced compared with normal
[52]. However, in contrast to previous studies on basal
plates from delivered placentas ([53] These authors obvi-
ously have studied the basal plate, but erroneously have
called it ‘‘placental bed.’’) interstitial trophoblast apoptosis
within the placental bed was not increased but rather re-
duced in preeclampsia [52], whereas endovascular apopto-
sis was increased [41].
These contrasting apoptosis features let us doubt that
shallow trophoblast invasion per se is the cause of impaired
endovascular invasion. As already discussed, mere intrinsic
trophoblastic phenomena (such as missing expression of a
vascular phenotype, reduced nitric oxide secretion by the
trophoblast, and altered trophoblast behavior caused by de-
ficient oxygenation [49, 51], the latter issue not discussed
in this review) are unlikely to be the exclusive causes of
malinvasion of uteroplacental arteries with subsequent
IUGR and/or preeclampsia. Rather, clinical and basic re-
search data discussed herein suggest that maladaptation and
malinvasion of uteroplacental arteries characteristic of
IUGR and preeclampsia result from 1) intrinsic factors,
namely abnormal biology of the extravillous trophoblast,
acting in concert with 2) extrinsic, maternal uterine factors
operating around the uteroplacental arteries, such as im-
paired decidual remodeling [54, 55], macrophage-based de-
fense mechanisms [41, 47], impaired function of uterine
NK cells [56], and maternal endothelial failure to express
selectins [30, 31]. Furthermore, it is feasible these factors
may interact, resulting in a cascade of events.
TISSUES AND MODELS FOR THE STUDY
OF ENDOVASCULAR TROPHOBLAST INVASION
The phenomenon of endovascular trophoblast invasion
is an area of reproductive biology research in which knowl-
edge remains partly superficial and controversial. The main
reason for this is the limitations of research tools available
to study endovascular trophoblast in combination with
missing awareness of these limitations.
Descriptive studies on human material are handicapped
by several facts:
● Easily available, delivered placentas [53] do not cover
the zone of interest, the placental bed.
● Placental bed biopsy specimens [25–27, 35, 41, 47] are
available to only a few groups, have been obtained large-
ly at the time of cesarean section, represent only a small
segment of the placental bed, and require special exper-
tise to evaluate.
● Hysterectomized uteri [6, 15, 52] are rare and often col-
lected under heterogeneous and pathological conditions.
Moreover, it is extremely difficult to collect control cas-
es.
● In general, all of the listed samples are available in those
early stages of human pregnancy in which the pathogen-
esis of IUGR and preeclampsia have their origin. How-
ever, in these stages, adequate separation between path-
ological and control groups is impossible since the po-
tential outcome of an interrupted pregnancy is untestable.
Alternatively, samples from later stages of pregnancy
with obvious signs of disease when compared with nor-
mal material will only provide data during overt disease
and cannot be used to describe causation.
In vitro studies with human cells, cell lines, and tissue
explants [25, 45–47, 50, 57] provide well-defined experi-
mental models; however, they are usually limited to one or
two cellular players (e.g., cytotrophoblast, cytotrophoblast
plus endothelial cells, cytotrophoblast plus macrophages)
and therefore cannot mimic the complex interplay of tro-
phoblast, mesenchyme, various maternal immune cells, and
the diverse cellular components of and within the vessels.
Even the use of tissue explants where complex fetal and
maternal tissue structures are maintained cannot solve this
problem, since not all of the cellular players remain in the
respective state of differentiation during in vitro culture.
Animal experiments [2, 7, 8, 14, 20, 34, 38, 39, 48] solve
the problems regarding availability of samples, pregnancy
stages, and experimental conditions. However, assuming
Downloaded from https://academic.oup.com/biolreprod/article/69/1/1/2712724 by guest on 01 October 2021
that trophoblast, endothelium, and macrophages in other
species are biologically similar, there are doubts that their
interplay in pregnancy is really comparable to that in the
human. The seemingly basic phenomenon of endovascular
trophoblast invasion is highly variable among ruminants
(nonexistent), myomorph rodents (trophoblast invasion is
blocked by formation of hyperploid giant cells), human
(reaching the superficial myometrium), and caviomorph ro-
dents (with trophoblast invasion extending into extrauter-
ine, intraperitoneal arteries). These facts prevent simple ex-
trapolation of data from any other species to the human.
At present, we cannot solve the general methodological
problems inherent in the area of endovascular trophoblast
biology. However, we should be aware of the limitations of
the various tissues and experimental models when ap-
proaching the next generation of research emerging from
high-throughput genomic and proteomic strategies.
ACKNOWLEDGMENT
The authors thank Professor Joan Hunt, University of Kansas Medical
Center, Kansas City, KS, for initiating this review.
REFERENCES
1. Brosens IA, Robertson WB, Dixon HG. The physiological response
of the vessels of the placental bed to normal pregnancy. J Pathol
Bacteriol 1967; 93:569–579.
2. Moll W, Nienartowicz A, Hees H, Wrobel K-H, Lenz A. Blood flow
regulation in the uteroplacental arteries. Trophoblast Res 1988; 3:83–
96.
3. Benirschke K, Kaufmann P. Pathology of the Human Placenta, 4th ed.
New York: Springer; 2000.
4. Brosens IA, Robertson WB, Dixon HG. The role of the spiral arteries
in the pathogenesis of preeclampsia. Obstet Gynecol Annu 1972; 1:
177–191.
5. Brosens IA, Dixon HG, Robertson WB. Fetal growth retardation and
the arteries of the placental bed. Br J Obstet Gynaecol 1977; 84:656–
663.
6. Craven CM, Morgan T, Ward K. Decidual spiral artery remodelling
begins before cellular interaction with cytotrophoblasts. Placenta
1998; 19:241–252.
7. Hees H, Moll W, Wrobel KH, Hees I. Pregnancy-induced structural
changes and trophoblastic invasion in the segmental mesometrial ar-
teries of the guinea pig (Cavia porcellus L). Placenta 1987; 8:609–
626.
8. Nanaev AK, Chwalisz K, Frank HG, Kohnen G, Hegele-Hartung C,
Kaufmann P. Physiological dilation of uteroplacental arteries in the
guinea pig depends on nitric oxide synthase activity of extravillous
trophoblast. Cell Tissue Res 1995; 282:407–421.
9. Hirano H, Imai Y, Ito H. Spiral artery of placenta: development and
pathology: immunohistochemical, microscopical, and electron-micro-
scopic study. Kobe J Med Sci 2002; 48:13–23.
10. Robertson WB. Uteroplacental vasculature. J Clin Pathol 1976;
29(suppl):9–17.
11. Robertson WB, Manning PJ. Elastic tissue in uterine blood vessels. J
Pathol 1974; 112:237–243.
 ENDOVASCULAR TROPHOBLAST INVASION
12. Tuttle SE, O’Toole RV, O’Shaughnessy RW, Zuspan FP. Immunohis-
tochemical evaluation of human placental implantation: an initial
study. Am J Obstet Gynecol 1985; 153:239–244. 36.
13. Brosens IA. The uteroplacental vessels at term: the distribution and
extent of physiological changes. Trophoblast Res 1988; 3:61–68.
14. Blankenship TN, Enders AC, King BF. Trophoblastic invasion and the 37.
development of uteroplacental arteries in the macaque: immunohis-
tochemical localization of cytokeratins, desmin, type IV collagen, lam-
inin, and fibronectin. Cell Tissue Res 1993; 272:227–236. 38.
15. Pijnenborg R, Bland JM, Robertson WB, Brosens IA. Uteroplacental
arterial changes related to interstitial trophoblast migration in early
human pregnancy. Placenta 1983; 4:397–413.
16. Damsky CH, Fitzgerald ML, Fisher SJ. Distribution patterns of extra- 39.
cellular matrix components and adhesion receptors are intricately
modulated during first trimester cytotrophoblast differentiation along
the invasive pathway, in vivo. J Clin Invest 1992; 89:210–222.
17. Fisher SJ, Damsky CH. Human cytotrophoblast invasion. Semin Cell
Biol 1993; 4:183–188. 40.
18. Kam EPY, Gardner L, Loke YW, King A. The role of trophoblast in
the physiological change in decidual spiral arteries. Hum Reprod 41.
1999; 14:2131–2138.
19. Craven CM, Zhao L, Ward K. Lateral placental growth occurs by
trophoblast cell invasion of decidual veins. Placenta 2000; 21:160–
169. 42.
20. King BF, Blankenship TN. Expression of proliferating cell nuclear
antigen (PCNA) in developing macaque placentas. Placenta 1993; 14: 43.
A36.
21. Bravo R, Macdonald-Bravo H. Existence of two populations of cyclin/ 44.
proliferating cell nuclear antigen during the cell cycle: associated with
DANN replication sites. J Cel Biol 1987; 105:1549–1554. 45.
22. Kaufmann P, Castellucci M. Extravillous trophoblast in the human
placenta. Trophoblast Res 1997; 10:21–65.
23. Vicovac L, Aplin JD. Epithelial-mesenchymal transition during tro-
phoblast differentiation. Acta Anat 1996; 156:202–216. 46.
24. Merviel P, Challier JC, Carbillon L, Foidart JM, Uzan S. The role of
integrins in human embryo implantation. Fetal Diagn Ther 2001; 16:
364–371. 47.
25. Zhou Y, Fisher SJ, Janatpour MJ, Genbacev O, Dejana E, Wheelock
M, Damsky CH. Human cytotrophoblasts adopt a vascular phenotype
as they differentiate: a strategy for successful endovascular invasion?
J Clin Invest 1997; 99:2139–2151. 48.
26. Zhou Y, Damsky CH, Fisher SJ. Preeclampsia is associated with fail-
ure of human cytotrophoblasts to mimic a vascular adhesion pheno-
type: one cause of defective endovascular invasion in this syndrome?
J Clin Invest 1997; 99:2152–2164. 49.
27. Lyall F, Bulmer JN, Duffie E, Cousins F, Theriault A, Robson SC.
Human trophoblast invasion and spiral artery transformation: the role
of PECAM-1 in normal pregnancy, preeclampsia, and fetal growth
restriction. Am J Pathol 2001; 158:1713–1721. 50.
28. Divers MJ, Bulmer JN, Miller D, Lilford RF. Beta 1 integrins in third
trimester human placentae: no differential expression in pathological
pregnancy. Placenta 1995; 16:245–260.
29. Floridon C, Nielsen O, Holund B, Sunde L, Westergaard JG, Thomsen 51.
SG, Teisner B. Localization of E-cadherin in villous, extravillous and
vascular trophoblasts during intrauterine, ectopic and molar pregnan-
cy. Mol Hum Reprod 2000; 6:943–950. 52.
30. King A, Loke YW. Differential expression of blood-group-related car-
bohydrate antigens by trophoblast subpopulations. Placenta 1988; 9:
513–521.
31. Burrows TD, King A, Loke YW. Expression of adhesion molecules 53.
by endovascular trophoblast and decidual endothelial cells: implica-
tions for vascular invasion during implantation. Placenta 1994; 15:21–
33. 54.
32. Takada A, Ohmori K, Yoneda T, Tsuyuoka K, Hasegawa A, Kiso M,
Kannagi R. Contribution of carbohydrate antigens sialyl Lewis A and
sialyl Lewis X to adhesion of human cancer cells to vascular endo- 55.
thelium. Cancer Res 1993; 53:354–361.
33. Pijnenborg R, Dixon GH, Robertson WB, Brosens IA. Trophoblastic
invasion of human decidua from 8 to 18 weeks of pregnancy. Placenta 56.
1980; 1:3–19.
34. Nanaev AK, Kosanke G, Reister F, Kemp B, Frank HG, Kaufmann P.
Pregnancy-induced de-differentiation of media smooth muscle cells in
uteroplacental arteries of the guinea pig is reversible after delivery. 57.
Placenta 2000; 21:306–312.
35. Lyall F, Bulmer JN, Kelly H, Duffie E, Robson SC. Human tropho-
7
blast invasion and spiral artery transformation: the role of nitric oxide.
Am J Pathol 1999; 154:1105–1114.
Martin D, Conrad KP. Expression of endothelial nitric oxide synthase
by extravillous trophoblast cells in the human placenta. Placenta 2000;
21:23–31.
Cartwright JE, Holden DP, Whitley GS. Hepatocyte growth factor reg-
ulates human trophoblast motility and invasion: a role for nitric oxide.
Br J Pharmacol 1999; 128:181–189.
Chwalisz K, Ciesla I, Garfield RE. Inhibition of nitric oxide (NO)
synthesis induces preterm parturition and preeclampsia-like conditions
in guinea pigs. In: Program of the 41st Meeting of the Society of
Gynecologic Investigation; 1994; Chicago, IL. Abstract O 36.
Garfield RE, Yallampalli C, Buhimschi IA, Chwalisz K. Reversal of
pre-eclampsia symptoms induced in rats by nitric oxide inhibition with
L-arginine, steroid hormones and an endothelin antagonist. In: Pro-
gram of the 41st Meeting of the Society of Gynecologic Investigation;
1994; Chicago, IL. Abstract P 384.
Downloaded from https://academic.oup.com/biolreprod/article/69/1/1/2712724 by guest on 01 October 2021
Bulmer JN, Johnson PM. Macrophage populations in the human pla-
centa and amniochorion. Clin Exp Immunol 1984; 57:393–403.
Reister F, Frank HG, Heyl W, Kosanke G, Huppertz B, Schroder W,
Kaufmann P, Rath W. The distribution of macrophages in spiral arter-
ies of the placental bed in pre-eclampsia differs from that in healthy
patients. Placenta 1999; 20:229–233.
Loke YW, King A. Human Implantation—Cell Biology and Immu-
nology. Cambridge: Cambridge University Press; 1995.
Hunt JS. The role of macrophages in the uterine response to pregnan-
cy. Placenta 1990; 11:467–475.
Hunt JS. Macrophages in human uteroplacental tissues: a review. Am
J Reprod Immunol 1989; 21:119–122.
Yui J, Hemmings D, Garcia Lloret MI, Guilbert LJ. Expression of the
human p55 and p75 tumor necrosis factor receptors in primary villous
trophoblasts and their role in cytotoxic signal transduction. Biol Re-
prod 1996; 55:400–409.
Yui J, Garcia-Lloret MI, Wegmann TG, Guilbert LJ. Cytotoxicity of
tumour necrosis factor-alpha and gamma-interferon against primary
human placental trophoblasts. Placenta 1994; 15:819–835.
Reister F, Frank HG, Kingdom JCP, Heyl W, Kaufmann P, Rath W,
Huppertz B. Macrophage-induced apoptosis limits endovascular tro-
phoblast invasion in the uterine wall of preeclamptic women. Lab
Invest 2001; 81:1143–1152.
Miller L, Alley EW, Murphy WJ, Russell SW, Hunt JS. Progesterone
inhibits inducible nitric oxide synthase gene expression and nitric ox-
ide production in murine macrophages. J Leukocyte Biol 1996; 59:
442–450.
Zhou Y, Genbacev O, Damsky CH, Fisher SJ. Oxygen regulates hu-
man cytotrophoblast differentiation and invasion: implications for en-
dovascular invasion in normal pregnancy and in pre-eclampsia. J Re-
prod Immunol 1998; 39:197–213.
Caniggia I, Grisaru-Gravnosky S, Kuliszewsky M, Post M, Lye SJ.
Inhibition of TGF-beta (3) restores the invasive capability of extra-
villous trophoblasts in preeclamptic pregnancies. J Clin Invest 1999;
103:1641–1650.
Caniggia I, Winter J, Lye SJ, Post M. Oxygen and placental devel-
opment during the first trimester: implications for the pathophysiology
of pre-eclampsia. Placenta 2000; 21:S25–S30.
Kadyrov M, Schmitz C, Black S, Kaufmann P, Huppertz B. Pre-
eclampsia and maternal anemia display reduced apoptosis and oppo-
site invasive phenotypes of extravillous trophoblast. Placenta; (in
press).
DiFederico E, Genbacev O, Fisher SJ. Preeclampsia is associated with
widespread apoptosis of placental cytotrophoblasts within the uterine
wall. Am J Pathol 1999; 155:293–301.
Aplin JD. Implantation, trophoblast differentiation and haemochorial
placentation: mechanistic evidence in vivo and in vitro. J Cell Sci
1991; 99:681–692.
Brosens JJ, Pijnenborg R, Brosens IA. The myometrial junctional
zone spiral arteries in normal and abnormal pregnancies: a review of
the literature. Am J Obstet Gynecol 2002; 187:1416–1423.
Ashkar AA, Di Santo JP, Croy BA. Interferon gamma contributes to
initiation of uterine vascular modification, decidual integrity, and uter-
ine natural killer cell maturation during normal murine pregnancy. J
Exp Med 2000; 192:259–270.
Cartwright JE, Kenny LC, Dash PR, Crocker IP, Aplin JD, Baker PN,
Whitley GS. Trophoblast invasion of spiral arteries: a novel in vitro
model. Placenta 2002; 23:232–235.