Molecular Human Reproduction Vol.12, No.11 pp.

687–694, 2006
Advance Access publication September 20, 2006 doi:10.1093/molehr/gal073

An in vitro model of human placental trophoblast

deportation/shedding

M.H.Abumaree1, P.R.Stone and L.W.Chamley

Department of Obstetrics and Gynaecology, University of Auckland School of Medicine, Auckland, New Zealand
1
To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, University of Auckland School
of Medicine, Auckland, New Zealand. E-mail: m.abumaree@auckland.ac.nz

Deportation of trophoblast shed from the placenta into the maternal circulation was first described over 100 years ago. Despite
this, little is known about the quantity or nature of the shed and deported trophoblasts. Neither do we have a clear understanding

Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010

of the fate of deported trophoblasts nor do we have a clear understanding of their effects on the maternal physiology. This defi-
ciency is largely due to the inaccessibility of deported trophoblasts in vivo. This study aimed to produce a model that would allow
us to study deported trophoblasts. We devised a system for culturing placental explants of 12-week gestation in cell culture inserts
with a stainless steel mesh bottom that allowed the ready harvesting of shed/deported trophoblasts. Immunohistochemical and
morphologic investigations demonstrated that these in vitro shed/deported trophoblasts are similar to those found in vivo and that
apoptotic, necrotic and viable trophoblasts were shed from the explants. Inhibiting caspases induced a change from predomi-
nantly apoptotic to predominantly necrotic trophoblast shedding. We have devised an in vitro model that allows the collection of
shed/deported trophoblasts which will significantly enhance our ability to study these cells. Our preliminary investigations con-
firm that apoptosis plays an important role in trophoblast shedding/deportation.

Key words: apoptosis/deportation/explant/model/trophoblast

Introduction
Trophoblast deportation is the process of transporting trophoblasts
that are shed from the placenta to distal sites via the maternal blood.
This process was first reported over 100 years ago by Schmorl (1893)
who described syncytiotrophoblast-like fragments trapped in the lungs
of women who had died of eclampsia. Subsequently, it has been
shown that trophoblast deportation is a physiological feature of nor-
mal pregnancy that may be exacerbated in pre-eclampsia, eclampsia
and other diseases of pregnancy (Fox, 1965; Alvarez et al., 1967;
Boyd and Hamilton, 1970; Chua et al., 1991; Johansen et al., 1999).
Several reports have suggested that programmed cell death (apopto-
sis) may initiate degeneration in the syncytiotrophoblast this layer nat-
urally ages, leading to shedding of multinucleated fragments of the
syncytiotrophoblast called syncytial knots into the maternal circulation
(Nelson, 1996; Huppertz et al., 1998; Marzioni et al., 1998; Chan et al.,
1999; Mayhew et al., 1999). Syncytiotrophoblast has also been noted
to have necrotic features in vivo and in vitro (MacLennan et al., 1972;
Kaufmann, 1985; Palmer et al., 1997), and it has been suggested that
necrotic trophoblasts may also be shed from the placenta, and this
might be especially the case in pathological pregnancies (Huppertz
and Kingdom, 2004).
Although several causes for trophoblast shedding have been pro-
posed, currently it is thought that most trophoblast shedding is simply
the result of the loss of aged or damaged trophoblasts in a process
which is analogous to the shedding of other epithelia, such as occur-
ring in the gut (Zbar et al., 2004). There is relatively little direct
experimental evidence to support this, and estimates of the quantity of
trophoblast shed in normal and diseased pregnancies vary widely.
Thus, there is a large deficit in our knowledge about the nature and
quantity of trophoblasts that are shed and deported during pregnancy,
and we understand little about the mechanisms that lead to this shedding.
This deficit is largely due to the inaccessibility of deported trophob-
lasts that are by and large thought to be trapped in the maternal lungs
and also due to the inaccessibility of the placental site during preg-
nancy. Here, we report the development and characterization of a
model, based on placental villous explant cultures, that allows us to
harvest sufficient quantities of shed trophoblasts to investigate their
nature, origins and potentially their effects on maternal physiology.
Materials and methods
Ethics of experimentation
This study was approved by the regional Ethics Committee, and all placental
tissues were obtained with informed consent.
Placentae
Placentae of 12-week gestations were obtained following elective surgical ter-
mination of pregnancy (TOP). The gestational age and fetal viability of all
pregnancies before TOP were confirmed by ultrasound measurement of
crown–rump length and fetal cardiac activity.
Harvesting trophoblasts shed from placental explants in vitro
Placental tissue from 16 placentae was washed with ice-cold phosphate-
buffered saline (PBS), dissected and rewashed with ice-cold PBS, and triplicate
explants (individual explant weight was ∼40 mg) were transferred into
Net-well™ inserts (400-μm mesh) in 12-well culture plates with 3 ml of
Dulbecco’s modified Eagle’s medium (DMEM)/F12 medium containing 10%
fetal bovine serum (FBS), 5 ng/ml of epidermal growth factor, 5 μg/ml of
insulin, 10 μg/ml of transferin, 100 μg/ml of L-glutamate, 20 nM sodium selenite,

© The Author 2006. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For
Permissions, please email: journals.permissions@oxfordjournals.org 687
M.H.Abumaree, P.R.Stone and L.W.Chamley
400 U/l hCG, 100 μg/ml streptomycin and 100 U/l penicillin (Keelan et al.,
1994). The explants were cultured at 37°C in a humidified atmosphere contain-
ing 5% CO2 and 95% air for 72 h. Every 24 h, the culture plates were gently
shaken, and the Net-well™ inserts containing the explants were transferred to
fresh culture well containing fresh culture medium. Detached trophoblasts
which had passed from the Net-well™ into the lower chamber were allowed to
settle under gravity and passed from the Net-well™ into the lower chamber of
the culture plate from where they were aspirated and harvested by centrifuga-
tion at 480 × g for 8 min at 20°C. Shed trophoblasts were resuspended in 210 μl
of PBS; then 30-μl aliquots were air-dried onto microscope slides and fixed
with cold acetone (–20°C) for 10 min and then left to air-dry for 1 h. Slides
with acetone-fixed air-dried trophoblasts were stored at –20°C until used in
immunocytochemical staining. Cells shed from the explants were character-
ized by immunocytochemical staining.
The effects of inhibiting caspases on trophoblast shedding
Triplicate explants from 12 placentae were cultured in the presence and absence
of 5 μM caspase inhibitor (DEVD-CHO CPP32/Apopain inhibitor; #235423;
Calbiochem) and harvested as described above. Trophoblasts shed from the
explants were characterized and quantified by immunocytochemistry staining.
Immunohistochemistry staining
Slides with the air-dried cells were thawed for 5 min at room temperature and
were encircled with a Dako pen (DAKO, MedBio, Christchurch, New Zealand).
Non-specific binding was blocked by incubation with 10% normal goat serum
in PBS–Tween for 10 min at room temperature. After three washes with PBS–
Tween, the slides were incubated with primary antibodies and with irrelevant,
species or class-matched, control antibodies diluted in 10% normal goat serum
in PBS–Tween as summarized in Table I for 1 h at room temperature. The
slides were then washed three times with PBS–Tween, and the activity of
endogenous peroxidase was quenched by the addition of 3% H2O2 in methanol
for 5 min at room temperature. The slides were then washed three times with
PBS–Tween. A Zymed Histostain-Plus kit (Zymed, San Francisco, CA, USA)
containing biotinylated secondary antibody and enzyme conjugate was used
according to the manufacturer’s instructions. Aminoethyl carbolyl stain (Med-
Bio) was used according to the manufacturer’s instructions. Slides were then
washed with de-ionized water, immersed in haematoxylin nuclear stain (Surgi-
path; Australian Laboratory Services, Auckland, New Zealand) for 1 min, then
washed with tap water, and coverslips were mounted with Aquamount. The
numbers of cells shed from the explants were quantified by counting the
number of stained cells in 10 random high-power fields (40×) on the immuno-
cytochemistry slides. Then, the number was expressed as the mean number of
shed cells per milligram placental tissue in the explants following the equation:
Mean number of shed trophoblasts per slide × 7 (number of slides)
n= .
120 mg (total weight of three explanted villous tissues)
Depletion of CD45-positive cells from cell suspensions
Shed cells that were harvested from the medium of explant cultures were resus-
pended in 0.1% FBS/PBS (0.1% FBS in sterile PBS, pH 7.4) and then incubated
with a CD45-reactive antibody at a final dilution of 1:200 for 30 min at 4°C.
The unbound antibody was removed by washing twice with 0.1% FBS–PBS
followed by centrifugation at 300 × g for 8 min at 4°C. The cells were then
Table I. Antibodies used in this study
Antibody Host species Dilution Specificity
Cytokeratin-7 Mouse 1:200 Epithelial cells including trophoblast
CD45 Mouse 1:200 Leukocytes
Vimentin Mouse 1:400 Non-epithelial cells
M30 cytodeath Mouse 1:25 Apoptotic trophoblast
Activated caspase-3 Rabbit 1:200 Apoptotic trophoblast
192 Rabbit 1:200 Negative control
G11 Mouse 1:100 Syncytiotrophoblast
BO1D11 Mouse 1:100 Extravillous trophoblast
CD71 Mouse 1:200 Proliferating cells
688
resuspended in 0.1% FBS/PBS and incubated with pre-washed Dynabeads®
Pan mouse IgG (Dynal Biotech Pty Ltd, Australia and New Zealand) (washing
performed according to manufacturers’ instructions) for 30 min at 4°C with
gentle rotation using a blood tube rotator. CD45-positive cells were harvested
by applying the tube containing the cells to a Dynal MPC®-2 Magnetic Particle
Concentrator for 3 min at room temperature. The supernatant, depleted of most
contaminating CD45-positive cells, was aspirated into a fresh tube, and then the
magnetic harvesting step was repeated twice to ensure all contaminating CD45-
positive cells were removed. The final supernatant, containing shed trophoblasts
depleted of CD45-positive cells, was washed twice with 0.1% FBS/PBS, and
the trophoblasts were harvested by centrifugation at 300 × g for 8 min at 4°C
and resuspended in 1 ml of DMEM/F12 medium supplemented with 10% FBS,
100 μg/ml streptomycin, 100 U/l penicillin and 100 μg/ml L-glutamate, pH 7.4.
Determination of the viability of shed trophoblasts with
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
Shed trophoblasts that had been depleted of CD45-positive cells were incu-
bated with 5 mg/ml of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT) in DMEM/F12 medium supplemented with 10% FBS, 100 μg/ml
Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010
streptomycin, 100 U/l penicillin and 100 μg/ml of L-glutamate for 1 h at 37°C.
The cells were washed, fixed with 4% (w/v) paraformaldehyde, air-dried and
visualized immediately by light microscopy using a Nikon Eclipse E400
microscope. The numbers of MTT-positive cells were counted in the same
manner as above. The shed trophoblasts were then further analysed by immu-
nocytochemical staining with antibodies reactive with the M30 cytokeratin
neoepitope (Roche Diagnostics, Auckland, New Zealand) and activated cas-
pase-3 (Sigma, Australia). Shed trophoblasts were also stained with irrelevant,
species or class-matched, control antibodies. The numbers of trophoblasts shed
from the explants were quantified as above.
Statistical analysis
Data were analysed using the Student’s t-test or analysis of variance (ANOVA)
as appropriate. These analyses were performed using Microsoft Excel (Office
97 software). Results were considered to be statistically significant if P < 0.05.
Results
An in vitro model of trophoblast deportation/shedding
Villous explants of ∼40-mg wet weight were cultured in Net-well
inserts, and the cells shed from the explants every 24 h were analysed
by light microscopy and immunocytochemistry, revealing that leuko-
cytes, red blood cells (RBCs) and two distinct types of trophoblasts,
mononuclear trophoblasts and multinucleated syncytiotrophoblast
fragments that we will refer to as syncytial knots, were shed (Figures 1
and 2A–D). The mononuclear trophoblasts varied in size from 7 to 34 μm
in width (mean 15.16 ± 0.69) and from 9 to 54 μm in length (mean
23.39 ± 1.02). The syncytial knots varied greatly in size ranging from
16 to 171 μm in width (mean 43.11 ± 2.75) and from 23 to 366 μm in
length (76.25 ± 6.66) and contained up to several hundred nuclei per
syncytiotrophoblast. The nuclei were usually small and contained
densely packed chromatin. Further characterization of the shed cells
using an antibody (G11), which is reactive with syncytiotrophoblast
Manufacturer
DAKO, MedBio, Christchurch, New Zealand
DAKO, MedBio, Christchurch, New Zealand
DAKO, MedBio, Christchurch, New Zealand
Roche Diagnostics, Auckland, New Zealand
Sigma, Australia
This laboratory
This laboratory
This laboratory
DAKO, MedBio, Christchurch, New Zealand

Figure 1. Phase-contrast photomicrograph showing an example of the
medium harvested from the chamber below a villous explant cultured for 24 h
in a Net-well insert demonstrating shed syncytial knots (arrowheads), mononu-
clear cells (arrow) and red blood cells (dashed arrow). Bar = 20 μm.
Figure 2. Photomicrographs demonstrating the characterization of cells shed
from placental explants in vitro. (A) Mononuclear trophoblasts and syncytial
knots stained with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
mide (MTT), (B) G11 antibody stains syncytial knots (red) but not mononu-
clear trophoblasts (arrow), (C) mononuclear trophoblast stained with a
cytokeratin-7-reactive antibody (red) and (D) an isotype-matched CD45-
reactive antibody stains leukocytes (red) but not a syncytial knot (arrow). The
shed cells were counterstained with haematoxylin. Bar = 30 μm.
but not with cytotrophoblast, as well as the lack of expression of
CD71 by these mononuclear trophoblasts suggested that the shed
mononuclear trophoblasts were not derived from the syncytium
(Figure 2B). The shed RBCs were not nucleated.
Quantification of trophoblast shedding
Semi-quantitative analysis revealed that, after 24 h, there were a mean
of 3.6 mononuclear trophoblasts and 1.9 syncytial knots shed per
milligram of explant. The numbers of both mononuclear trophoblasts
and syncytial knots then decreased at 48 h (P < 0.0001) but increased
In vitro trophoblast shedding
again at 72 h (P < 0.0001) (Figure 3). Conversely, the number of shed
CD45-positive cells and vimentin-positive cells decreased signifi-
cantly (P < 0.0001) with increasing time in culture. However, there
was no significant difference between the numbers of vimentin-positive
and CD45-positive mononuclear cells shed from the explants at any
time point (P > 0.1), suggesting that all of the non-trophoblast mono-
nuclear cells were leukocytes (data not shown).
Viable, apoptotic and necrotic trophoblasts were shed from
the placental explants
Semi-quantitative analysis of shed cells stained with an antibody reac-
tive with the M30 cytokeratin neoepitope revealed that 38% (SE ±
1.42) of the mononuclear trophoblasts and 25% (SE ± 1.63) of the
syncytial knots shed were apoptotic after 24 h in culture. The percent-
age of both types of shed trophoblast that were apoptotic increased
significantly (P < 0.0001) with time in culture (Figure 4). The number
of syncytial knots stained with an activated caspase-3-reactive anti-
body was not significantly different from those staining with the M30
Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010
antibody (data not shown). Counting MTT-stained shed cells (Figures 1
and 4) revealed that after 24 h, ∼59% (SE ± 1.31) of the shed mononu-
clear trophoblasts and 71% (SE ± 1.89) of the syncytial knots were
strongly MTT positive, suggesting that they were viable, but this per-
centage decreased significantly (P < 0.0001) with time in culture
(Figure 4). The number of apoptotic trophoblasts and viable trophob-
lasts did not add up to the total number of trophoblasts shed from the
explants, and we estimated the numbers of non-viable, non-apoptotic
(necrotic) trophoblasts shed using the equation below:
Necrotic trophoblasts = total shed trophoblasts – (viable trophoblasts
+ apoptotic trophoblasts).
Approximately 5% of both the mononuclear trophoblast and the
syncytial knots shed from the cultured explants were necrotic, and this
percentage remained relatively unchanged throughout the 3-day study
period (Figure 4).
The effects on trophoblast shedding of inhibiting caspases
To further investigate the role of apoptosis in trophoblast shedding,
we quantified and compared trophoblasts shed from villous explants
that had been incubated in the presence or absence of a broad-spectrum
caspase inhibitor. This analysis demonstrated that the caspase inhibitor
significantly (P < 0.0001) reduced the numbers of both mononuclear
trophoblasts and syncytial knots shed at all time points (Figure 5) but
that the shedding of mononuclear trophoblasts was more dramatically
effected than the shedding of syncytial knots. The caspase inhibitor
significantly reduced the numbers of shed viable (MTT positive)
mononuclear trophoblasts, by ∼4-, 5- and 6-fold at 24, 48 and 72 h,
respectively (Figure 6). Whereas the inhibitor reduced the numbers of
viable syncytial knots by only approximately 2-fold at all time points;
but this reduction was significant (P < 0.002). Conversely, the caspase
inhibitor caused a striking increase in the percentage of necrotic tro-
phoblasts shed especially at the 72-h time point when 66% (SE ±
6.15) of the mononuclear trophoblasts and 52% (SE ± 5.99) of the
syncytial knots shed were necrotic (Figure 6).
Discussion
Despite trophoblast deportation having been first described by
Schmorl over 100 years ago, little is known about the processes that
lead to trophoblast deportation or the consequences of deported tro-
phoblasts on maternal physiology, and this is largely because deported
trophoblasts are inaccessible in vivo. Trophoblast deportation involves
689
M.H.Abumaree, P.R.Stone and L.W.Chamley

Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010

Figure 3. The mean numbers of cytokeratin-positive mononuclear trophoblasts or syncytial knots shed per milligram of villous explant. The number of shed mono-
nuclear trophoblasts and syncytial knots changed significantly with time [analysis of variance (ANOVA), P < 0.0001]. Each experiment was conducted using
triplicate villous explants from 16 separate placentae. Bars represent SE.

Figure 4. The percentage of viable (white), apoptotic (grey) and necrotic (striped) trophoblasts that were shed per milligram of placental tissue from explant cul-
tures. Experiments were conducted using triplicate explants from 16 separate placentae.

at least two separate physical processes. One of these processes is the study, we have developed an in vitro model of trophoblast shedding,
transportation of trophoblasts from the uteroplacental site to distal which allowed us to study the biological factors involved in the shedding
sites in the maternal body via the maternal blood. However, preceding of trophoblasts and the nature of the shed trophoblasts. In establishing
this step, trophoblasts are first shed from the placental surface. In this this model, we carefully considered several issues.

690
 In vitro trophoblast shedding

Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010

Figure 5. The mean numbers of cytokeratin-positive trophoblasts shed per milligram of villous explant from explants cultured in the presence (white) or in the
absence (grey) of 5 μM caspase inhibitor. The mean number of trophoblasts shed per milligram of placental tissue decreased significantly in the presence of the cas-
pase inhibitor (t-test, *P < 0.00001). Data were obtained from triplicate explants at each time point from 12 separate placentae. Bars represent SE.

100%
Viable, apoptotic or necrotic trophoblasts as a percentage of

80%
total shed trophoblasts

60%

40%

20%

0%
T24 T48 T72 T24 T48 T72 T24 T48 T72 T24 T48 T72

Absence of caspase inhibitor Presence of caspase inhibitor Absence of caspase inhibitor Presence of caspase inhibitor

Mononuclear Trophoblasts Syncytial knots

Characteristics of shed trophoblasts

Figure 6. Percentage of viable (white), apoptotic (grey) and necrotic (striped) trophoblasts that were shed per milligram of placental tissue.

First, we choose to use placentae of 12-week gestation. This was
because there is evidence suggesting that the oxygen concentration
around the human placenta increases near the end of the first trimester
of pregnancy when the trophoblast plugs in the spiral arteries dissi-
pate, allowing maternal blood to enter the intervillous space (Rodesch
et al., 1992; Merce et al., 1996; Jaffe, 1998; Watson et al., 1998;
Jauniaux et al., 2000). Furthermore, we have recently shown, by stud-
ying >3500 explants from placentae ranging from 8 to 12 weeks of
gestation, that although oxygen concentration affects the behaviour of
trophoblasts from placentae of 10-week gestation or less, there is little
or no effect of oxygen concentration on explants from placentae of
≥11-week gestation (James et al., 2006). In addition, although placentae
of 12-week gestation are less susceptible to oxidative damage than
placentae of earlier gestation, they retain a full/continuous mononu-
clear trophoblast layer beneath the syncytiotrophoblast and thus should
have significant regenerative capacity (Rodesch et al., 1992; Merce
691
M.H.Abumaree, P.R.Stone and L.W.Chamley
et al., 1996; Jaffe, 1998; Watson et al., 1998; Jauniaux et al., 2000).
Therefore, it seemed likely to us that placentae of 12-week gestation
would be well suited to use in this model because they do not require
culture in reduced oxygen levels, as would be required for earlier
gestation placentae; yet, they have substantial regenerative capacity
that might have been reduced in later gestation placentae.
Second, the tissues were carefully washed to remove contaminating
maternal cells. However, this washing was not entirely effective as
demonstrated by the continued shedding of CD45-positive cells from
the explants. That these CD45-positive cells were likely to be of
maternal origin is suggested by their decreased numbers with time in
culture, but we cannot rule out the possibility that these CD45-positive
cells were of fetal origin. Regardless of whether these were fetal or
maternal leukocytes, their shedding from the explants means that cau-
tion must be taken to avoid confusing these leukocytes with shed
mononuclear trophoblasts. In this study, we either used trophoblast-
specific markers to avoid this confusion or depleted the leukocytes
using magnetic beads before the final counting of shed cells.
Third, the Net-well inserts we used allowed the suspension of the
explants in the medium in a manner modelling the suspension of float-
ing villi in the maternal blood lakes and allowed us to move the
explants into fresh wells without the need to touch the delicate
explants, thereby reducing experimental artefacts. This ability to
move the explants allowed us ready access to the shed trophoblasts
that fell through the 400-μm mesh at the bottom of the inserts.
Fourth, by gently shaking the explants before moving them to fresh
culture wells each day, we hoped to imitate the effects of the move-
ment of maternal blood across the villous surface, which is likely to be
one of the forces involved in trophoblast shedding and deportation.
Although this may introduce variation between individual experi-
ments, the shaking was gentle and as uniform as possible. In addition,
the large number of experimental replicates used should have dimin-
ished the effects of inter-experiment variation.
We and others have previously shown that the syncytiotrophoblast
undergoes significant artefactual degradation during the first 24 h of
villous explant culture and that it is subsequently regenerated (Palmer
et al., 1997; James et al., 2005). Confirming the artefactual degenera-
tion of the syncytiotrophoblast during the first 24 h of explant culture,
we demonstrated that there was significantly more trophoblast shed at
the 24-h time point than at later time points in our model.
However, surprisingly, we found that most of these artefactually
shed trophoblasts were still able to metabolize MTT, suggesting that
they were viable. We had anticipated that these cells would be shed
because of exposure to excess (atmospheric) oxygen levels or the
effects of the physical trauma of the TOP and that, consequently, they
would be necrotic. Although the ability of a cell to metabolize MTT is
often used as a marker of cell viability, it is actually a marker of mito-
chondrial function, and it may be that many of these artefactually shed
trophoblasts were undergoing a death process but that this process was
not far enough advanced to have caused either mitochondrial dysfunc-
tion or caspase activation. Future studies using other markers of cell
viability/death will be required to clarify this issue.
These results suggest that the process by which syncytial knots are
extruded from the placental surface is not absolutely dependent upon
the ‘death’ of either the syncytiotrophoblast or the mononuclear tro-
phoblasts. We believe that it is crucial to bear this artefactual shedding
in mind when looking at our results (as well as the results of others)
and believe that the results of the trophoblast shedding that occurred at
48 and 72 h in our model are more likely to be a truer representation of
shedding as it occurs in vivo. We have also shown in other work
(James et al., 2005), as well as by examining the explants used in this
study, following MTT staining (data not shown), that the viability of
most cells in villous explants is good between 24 and 72 h of culture
692
but that the viability of most of the cells in villous explants is signifi-
cantly reduced after 96 h in culture, and consequently, for the remain-
der of our discussion, we will consider only those trophoblasts shed at
48 and 72 h. Both mononuclear cytotrophoblasts and multinucleated
trophoblasts (syncytial knots) were shed from the explants, and the
trophoblasts that were shed in our model closely resemble those that
have previously been reported by others to be shed in vivo (Chua
et al., 1991; Hawes et al., 1994; Johansen et al., 1999). That the
mononuclear trophoblasts were not fragments of the syncytiotrophob-
last was confirmed by their failure to stain with the G11 antibody,
which reacts with syncytiotrophoblast but not with cytotrophoblasts,
as well as by the absence of CD71 expression by the shed mononuclear
trophoblasts (data not shown). It is most likely that these mononuclear
trophoblasts were villous cytotrophoblasts because there were likely
to be very few extravillous cytotrophoblasts associated with the
explants we used. It is not clear to us why villous cytotrophoblasts
should be shed from the placenta because they are usually overlain by
the syncytiotrophoblast, but it is possible to speculate that they are
Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010
shed accidentally after being exposed at syncytiotrophoblast disconti-
nuities during the process of syncytiotrophoblast shedding and subse-
quent regeneration (Nelson, 1996). Alternatively, it is possible that
cytotrophoblasts are shed ‘deliberately’ as part of the mechanism for
extruding syncytial knots from the placental surface. Further work
will be required to establish why, both in vitro and in vivo, cytotro-
phoblasts are shed from the placenta.
Based on published mean placental weights (Kaufmann, 2000) and
the mean numbers of trophoblasts shed from the explants in our study,
we estimate that 105 mononuclear trophoblasts and 4.7 × 104 syncytial
knots would be shed daily from a placenta of 12-week gestation. The
numbers of trophoblasts shed daily by an average placenta at 9 months
of gestation would rise to 1.8 × 106 mononuclear trophoblasts and 8.5 ×
105 syncytial knots. Our estimates are greater than the widely quoted
estimate of 150 000 trophoblasts shed daily during normal preg-
nancy, but we have been unable to determine how Chua et al. (1991)
derived the quoted estimates, and they have acknowledged that
150 000 may be a substantial underestimate of daily trophoblast shed-
ding. Our estimates are based on the assumptions that (i) our model
accurately reflects the levels of trophoblast shedding occurring in vivo
(these estimates were based on the mean numbers of trophoblast shed
in our model at 48 and 72 h) and (ii) that the rate of shedding is
directly related to placental weight. This latter assumption is some-
what questionable because the internal structures of the placenta, in
particular the relative amount of mononuclear trophoblast, change
with increasing gestation age. It would be of significant value to
repeat experiments such as those we describe here using mid-trimester,
as well as term placentae to address this issue directly.
The manner in which we have presented our data may lead the casual
observer to believe that shedding of mononuclear trophoblasts is quantita-
tively more important than that of syncytial knots. This is because we
have reported the numbers of each of these cell types. It must be borne in
mind that each syncytial knot contains up to several hundred nuclei and
vastly more cytoplasm than mononuclear trophoblasts, and had we been
able to quantify the numbers of nuclei, it would be quite clear that the total
amount of cellular material shed as syncytial knots greatly exceeds the
amount of material shed as mononuclear trophoblasts. However, because
of the large physical size of the shed syncytial knots, it was impossible to
accurately quantify the numbers of shed nuclei in these structures.
Microscopic examination demonstrated that the nuclei of the shed
trophoblasts were densely packed, suggesting that they may have been
apoptotic. We confirmed that these cells were undergoing an apop-
totic process by staining for both the M30 cytokeratin neoepitope and
the activated caspase-3. This suggested that, as has been hypothesized
by others, apoptosis plays a major role in the shedding of these cells

from the placenta (Nelson, 1996; Huppertz et al., 1998). To further
examine the role of apoptosis in the process of trophoblast shedding,
we studied the effects of simultaneously inhibiting caspases 3, 6, 7, 8
and 10. This resulted in substantial reductions in the number of tro-
phoblasts shed from explants, confirming that most shed trophoblasts
were apoptotic. The mechanism that physically drives the shedding of
trophoblasts from the villous surface is at present unknown. That
inhibiting caspases caused a significant reduction in the total number
of trophoblasts shed indicates that the apoptosis pathway may be
important in driving the shedding process. However, because shed-
ding was not abolished by the caspase inhibitor, it seems that other
pathways must also be involved in the process of trophoblast shed-
ding, and there may be redundancy between these alternative pathways.
Although several mechanisms have been proposed to explain why
trophoblasts are shed into the maternal blood, our results and those of
others (Ku et al., 1997; Mayhew et al., 1999; Kadyrov et al., 2001)
suggest that the most likely explanation for shedding of trophoblasts is
simply that shedding is the final stage in the normal cellular ageing
and turnover of the trophoblast layers and is analogous to shedding of
other epithelia as has been suggested previously (Huppertz and
Kingdom, 2004; Zbar et al., 2004).
Although several other workers have suggested that shed trophob-
lasts are likely to be apoptotic (Huppertz et al., 2003; Huppertz and
Kingdom, 2004), we believe our study provides the first direct exper-
imental documentation that shed trophoblast is predominantly apop-
totic. Two further points worthy of discussion arise from these
experiments. First, the shedding of mononuclear trophoblasts was
reduced to a much greater extent than that of syncytial knots. This
could be explained if the apoptotic pathway had already been par-
tially initiated in the syncytiotrophoblast such that there was a smaller
range of caspase-dependent steps that remained to be inhibited in the
syncytiotrophoblast than in the cytotrophoblasts and supports the
suggestion that syncytial knots, before their extrusion from the pla-
centa, are proapoptotic (Mayhew et al., 1999). Second, although the
caspase inhibitor caused notable and significant reductions in the
total numbers of both syncytial knots and cytotrophoblasts shed, a
more striking change occurred in the nature of the dead trophoblasts
that changed from being predominately apoptotic to being predomi-
nately necrotic. This suggests to us that the trophoblasts were com-
mitted to die and that, when the apoptotic pathway was unavailable,
the ageing trophoblasts had to die by an alternative mechanism such
as necrosis or by a combination of death mechanisms, which may be
analogous to aponecrotic death as proposed by Huppertz et al.
(2003). Alternatively, caspase-independent programmed cell death
may also contribute to the shedding of trophoblasts when the caspase
pathways are inhibited. Regardless of the exact mechanism of death,
this finding strongly suggests that trophoblasts have a programmed
life span.
Although most of the shed trophoblasts were apoptotic, we were
interested to know more about the condition of the non-apoptotic cells.
Consequently, following the depletion of contaminating leukocytes, we
determined whether any of the shed trophoblasts were viable, by
staining with MTT, and found that 6% of both the mononuclear
trophoblasts and the syncytial knots shed from the cultured explants
were viable after 72 h of culture (Figure 4). Some of these viable syn-
cytial knots may be true syncytial sprouts which have detached from
the placental surface as suggested by Boyd and Hamilton (1970), but
because the placentae we studied were of 12-week gestation, this
seems unlikely. There is growing evidence that fetal cells remain via-
ble in maternal organs and/or the maternal circulation for prolonged
periods of time after pregnancy (Bianchi et al., 1996; O’Donoghue
et al., 2004). This phenomenon is called chronic microchimerism.
While other cell types, such as fetal mesenchymal stem cells, are
In vitro trophoblast shedding
known to be involved in fetomaternal microchimerism, our results
suggest that trophoblasts could also contribute to the chronic micro-
chimerism induced by pregnancy. Given the possibility that some of
these viable trophoblasts could be true syncytial sprouts and the
possibility that they could contribute to chronic microchimerism, it
would be interesting, in future studies, to determine exactly how long
trophoblasts remain viable, in vitro, after they have been shed.
The cumulative numbers of viable and apoptotic trophoblasts shed
from the explants was less than the total number of shed trophoblasts,
suggesting that a small percentage (∼5%) of shed trophoblasts were
neither viable nor apoptotic, and we suggest that these trophoblasts
died by a caspase-independent mechanism and may have been
necrotic. Unfortunately, there are no objective markers that can con-
firm that these trophoblasts were necrotic, but this finding is poten-
tially very significant to the outcome of pregnancy because evidence
from other systems suggests that apoptotic cells are phagocytosed
without stimulating an inflammatory immune response or may lead to
active suppression of immune responses to antigens from the apop-
Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010
totic cells (Fadok and Chimini, 2001). In contrast, phagocytosis of
necrotic cells can lead to development of inflammatory immune
responses. Thus, phagocytosis of shed apoptotic trophoblasts by
maternal immune cells may produce a suppressive type of immune
response, which would be beneficial to the maintenance of the preg-
nancy. Indeed, we have recently shown that trophoblasts shed from
our model produce an anti-inflammatory, immunosuppressive type of
immune response (Abumaree et al., 2006). We have also recently
shown that endothelial cells can phagocytose dead trophoblasts (Chen
et al., 2006). Although phagocytosis of apoptotic trophoblasts had no
effect on the endothelial cells, phagocytosis of necrotic trophoblasts
led to endothelial cell activation (Chen et al., 2006). Given that
increased trophoblast shedding is thought to occur in pre-eclampsia and
that both endothelial activation and aberrant inflammatory responses
are hallmarks of pre-eclampsia, increased shedding of necrotic tro-
phoblasts may be important to the pathogenesis of this disease.
In summary, we report a novel placental villous explant model that
allows the study of trophoblast shedding from the human placenta.
This model is likely to be useful for studying the causes and conse-
quences of trophoblast deportation.
Acknowledgements
We thank the Staff of Epsom Day Unit for their assistance in collecting placen-
tae. This research was funded by grants from The University of Auckland Staff
Research Fund, Auckland Uniservices and the New Zealand Lotteries Board
(Health).
References
Abumaree MH, Stone PR and Chamley LW (2006) The effects of apoptotic,
deported human placental trophoblast on macrophages: possible conse-
quences for pregnancy. J Reprod Immunol, in press.
Alvarez H, Benedetti WL and De Leonis VK (1967) Syncytial proliferation in
normal and toxemic pregnancies. Obstet Gynecol 29,637–643.
Bianchi DW, Zickwolf GK, Weil GJ, Sylvester S and DeMaria MA (1996)
Male fetal progenitor cells persist in maternal blood for as long as 27 years
postpartum. Proc Natl Acad Sci USA 93,705–708.
Boyd JD and Hamilton WJ (1970) The Human Placenta. W. Helfer & Sons,
Cambridge.
Chan CC, Lao TT and Cheung AN (1999) Apoptotic and proliferative activities
in first trimester placentae. Placenta 20,223–227.
Chen Q, Stone PR, McCowan LM and Chamley LW (2006) Phagocytosis of
necrotic but not apoptotic trophoblasts induces endothelial cell activation.
Hypertension 47,116–121.
Chua S, Wilkins T, Sargent I and Redman C (1991) Trophoblast deportation in
pre-eclamptic pregnancy. Br J Obstet Gynaecol 98,973–979.
Fadok VA and Chimini G (2001) The phagocytosis of apoptotic cells. Semin
Immunol 13,365–372.
693
M.H.Abumaree, P.R.Stone and L.W.Chamley
Fox H (1965) The significance of villous syncytial knots in the human pla-
centa. J Obstet Gynaecol Br Commonw 72,347–355.
Hawes CS, Suskin HA, Petropoulos A, Latham SE and Mueller UW (1994) A
morphologic study of trophoblast isolated from peripheral blood of pregnant
women. Am J Obstet Gynecol 170,1297–1300.
Huppertz B and Kingdom JC (2004) Apoptosis in the trophoblast—role of
apoptosis in placental morphogenesis. J Soc Gynecol Invest 11,353–362.
Huppertz B, Frank HG, Kingdom JC, Reister F and Kaufmann P (1998)
Villous cytotrophoblast regulation of the syncytial apoptotic cascade in the
human placenta. Histochem Cell Biol 110,495–508.
Huppertz B, Kingdom J, Caniggia I, Desoye G, Black S, Korr H and Kaufmann P
(2003) Hypoxia favours necrotic versus apoptotic shedding of placental
syncytiotrophoblast into the maternal circulation. Placenta 24,181–190.
Jaffe R (1998) First trimester utero-placental circulation: maternal–fetal inter-
action. J Perinat Med 26,168–174.
James JL, Stone PR and Chamley LW (2005) Cytotrophoblast differentia-
tion in the first trimester of pregnancy: evidence for separate progenitors
of extravillous trophoblasts and syncytiotrophoblast. Reproduction
130,95–103.
James JL, Stone PR and Chamley LW (2006) The effects of oxygen concentra-
tion and gestational age on extravillous trophoblast outgrowth in a human
first trimester villous explant model. Hum Reprod, in press.
Jauniaux E, Watson AL, Hempstock J, Bao YP, Skepper JN and Burton GJ (2000)
Onset of maternal arterial blood flow and placental oxidative stress. A possible
factor in human early pregnancy failure. Am J Pathol 157,2111–2122.
Johansen M, Redman CW, Wilkins T and Sargent IL (1999) Trophoblast deporta-
tion in human pregnancy—its relevance for pre-eclampsia. Placenta 20,531–539.
Kadyrov M, Kaufmann P and Huppertz B (2001) Expression of a cytokeratin
18 neo-epitope is a specific marker for trophoblast apoptosis in human pla-
centa. Placenta 22,44–48.
Kaufmann P (1985) Influence of ischemia and artificial perfusion on placental
ultrastructure and morphometry. Contrib Gynecol Obstet 13,18–26.
Kaufmann P and Benirschke K (2000) Pathology of the Human Placenta.
Springer-Verlag, New York.
Keelan J, Song Y and France JT (1994) Comparative regulation of inhibin,
activin and human chorionic gonadotropin production by placental trophob-
last cells in culture. Placenta 15,803–818.
Ku NO, Liao J and Omary MB (1997) Apoptosis generates stable fragments of
human type I keratins. J Biol Chem 272,33197–33203.
694
MacLennan AH, Sharp F and Shaw-Dunn J (1972) The ultrastructure of
human trophoblast in spontaneous and induced hypoxia using a system of
organ culture. A comparison with ultrastructural changes in pre-eclampsia
and placental insufficiency. J Obstet Gynaecol Br Commonw 79,113–121.
Marzioni D, Muhlhauser J, Crescimanno C, Banita M, Pierleoni C and
Castellucci M (1998) BCL-2 expression in the human placenta and its corre-
lation with fibrin deposits. Hum Reprod 13,1717–1722.
Mayhew TM, Leach L, McGee R, Ismail WW, Myklebust R and Lammiman
MJ (1999) Proliferation, differentiation and apoptosis in villous trophoblast
at 13–41 weeks of gestation (including observations on annulate lamellae
and nuclear pore complexes). Placenta 20,407–422.
Merce LT, Barco MJ and Bau S (1996) Color Doppler sonographic assessment
of placental circulation in the first trimester of normal pregnancy. J Ultra-
sound Med 15,135–142.
Nelson DM (1996) Apoptotic changes occur in syncytiotrophoblast of human
placental villi where fibrin type fibrinoid is deposited at discontinuities in
the villous trophoblast. Placenta 17,387–391.
O’Donoghue K, Chan J, de la Fuente J, Kennea N, Sandison A, Anderson JR,
Roberts IA and Fisk NM (2004) Microchimerism in female bone marrow
and bone decades after fetal mesenchymal stem-cell trafficking in preg-
nancy. Lancet 364,179–182.
Downloaded from molehr.oxfordjournals.org at Science Library on September 7, 2010
Palmer ME, Watson AL and Burton GJ (1997) Morphological analysis of
degeneration and regeneration of syncytiotrophoblast in first trimester
placental villi during organ culture. Hum Reprod 12,379–382.
Rodesch F, Simon P, Donner C and Jauniaux E (1992) Oxygen measurements
in endometrial and trophoblastic tissues during early pregnancy. Obstet
Gynecol 80,283–285.
Schmorl G (1893) Pathologisch-anatomische Untersuchungen über Paerperal-
Eklampsie, Verlag FCW Vogel, Leipzig.
Watson AL, Skepper JN, Jauniaux E and Burton GJ (1998) Susceptibility
of human placental syncytiotrophoblastic mitochondria to oxygen-
mediated damage in relation to gestational age. J Clin Endocrinol Metab
83,1697–1705.
Zbar AP, Simopoulos C and Karayiannakis AJ (2004) Cadherins: an integral
role in inflammatory bowel disease and mucosal restitution. J Gastroenterol
39,413–421.
Submitted on June 19, 2006; accepted on July 21, 2006