Markert UR (ed): Immunology of Pregnancy.

Chem Immunol Allergy. Basel, Karger, 2005, vol 89, pp 4961

Immunology of Preeclampsia
Leif Matthiesen
a
, Gran Berg
a
, Jan Ernerudh
b
, Christina Ekerfelt
b
,
Yvonne Jonsson
b
, Surendra Sharma
c
a
Department of Molecular and Clinical Medicine, Division of Obstetrics and
Gynaecology and
b
Department of Molecular and Clinical Medicine, Division of
Clinical Immunology, University Hospital, Linkoping, Sweden;
c
Department of
Pediatrics and Pathology, Women and Infants Hospital of Rhode Island, Brown
University, Providence, R.I., USA
Abstract
Preeclampsia is a placenta-dependent disorder with both local and systemic anomalies
with neonatal and maternal morbidity. It is manifested late in pregnancy, but the onset is during
early stages of gestation. The current hypothesis regarding the aetiology of preeclampsia is
focused on maladaptation of immune responses and defective trophoblast invasion. Thus, an
excessive maternal inflammatory response, perhaps directed against foreign fetal antigens,
results in a chain of events including shallow trophoblast invasion, defective spiral artery
remodelling, placental infarction and release of pro-inflammatory cytokines and placental frag-
ments in the systemic circulation. During normal pregnancy, trophoblasts interact in the
decidua with the unique uterine NK cells, modifying their cytokine repertoire, regulating adhe-
sion molecules and matrix metalloproteinases. The inability of trophoblasts to accomplish these
changes might be a critical factor for the onset of preeclampsia. Several cytokines, produced at
the maternal-fetal interface, have an impact on trophoblast invasion. It is suggested that defi-
ciency of interleukin-10 may contribute to enhanced inflammatory responses towards the tro-
phoblasts elicited by e.g. tumour necrosis factor- and interferon-. Consequently, trophoblasts
subjected to a high rate of apoptosis are hampered in their invasive capacity resulting in defec-
tive transformation of spiral arteries, hypoxia, thrombosis and infarction of the placenta. The
ensuing infarction of placenta leads to leakage of increasing amounts of placental fragments
and cytokines in the maternal circulation and an exaggerated systemic endothelial activation as
identified in preeclampsia. So far, treatment of preeclampsia is focused on signs like hyperten-
sion, whereas attempts of modifying immune responses may be a possibility in the future.
Copyright 2005 S. Karger AG, Basel
Introduction
Preeclampsia is a complication that is detected in the second half of preg-
nancy, but most probably has its onset during the early stages of gestation. This
Matthiesen/Berg/Ernerudh/Ekerfelt/Jonsson/Sharma 50
pregnancy-associated disorder is histologically characterized by restrained
trophoblast invasion, vasculitis, thrombosis and ischaemia of the placenta.
These features may also be apparent in other obstetric complications like recur-
rent spontaneous abortion, intrauterine growth retardation, fetal death, and
abruptio placentae. The seemingly disparate clinical entities might have their
common aetiology in the immune responses including local subclinical inflam-
mation at the placental bed and systemically (in preeclampsia) in the maternal
circulation. Preeclampsia is hard to detect in its early form and predictors that
can be used to identify the women at risk of preeclampsia would be of value for
the clinician. This paper deals with preeclampsia in humans and associated
immunological changes and is an overview of recent important findings of this
important but still poorly understood condition.
Clinical Preeclampsia
Preeclampsia occurs after the 20th week of gestation and is a heteroge-
neous disease. Since termination of pregnancy cures the disease, preeclampsia
is a placenta-dependent disorder with both local intrauterine and systemic signs
and symptoms. The hallmark signs are hypertension and proteinuria (table 1).
The incidence of preeclampsia is 35% of all pregnancies depending on the
population studied [1].
A number of risk factors are thought to increase the risk of developing
preeclampsia: maternal vascular disease, autoimmune disorders, maternal and
paternal genetic causes, diabetes mellitus, primiparity and twin pregnancy.
Although the exact aetiology remains to be delineated, all of the associated
causes converge into a common pathophysiological denominator: endothelial
dysfunction. Thus, it has been suggested that an excessive maternal inflamma-
tory response, perhaps directed against foreign fetal antigens, results in an
impaired trophoblast invasion with a defective spiral artery remodelling ensued
by high-resistance vessels and a reduced placental perfusion. The consequences
are placental hypoxia and infarction with release of pro-inflammatory cytokines
and placental fragments into the maternal circulation with ultimately general-
ized maternal, and possibly fetal, endothelial activation [1].
Trophoblast Invasion
An adequate trophoblast invasion is possible only after a proper endome-
trial decidualization of the uterine wall has occurred. The decidualization is ini-
tiated immediately after ovulation in order to receive the embryo. The
production of progesterone from the corpus luteum stimulates the decidua to
Preeclampsia and the Immune System 51
increase the vascularization and secretory activity of the endometrial glands.
The leukocytes in the decidua consist mainly of unique uterine natural killer
(uNK) cells (6570%) and monocyte/macrophages (1520%), whose exact
function is unknown. A small number of T cells are also present, whereas B
cells are almost absent. In the endometrial extracellular matrix (consisting of
different types of collagens, proteoglycans, and glycoproteins), changes occur
facilitating the invasive properties of trophoblasts creating a safe anchor of the
placenta in the decidua and the vascular remodelling of the spiral arteries [2, 3].
The invading cytotrophoblasts are a subpopulation of villous cytotro-
phoblasts, which in turn differentiate into an outer layer of multinucleated cells,
the syncytiotrophoblasts. The syncytiotrophoblasts cover the fetal mesenchyme
and blood vessels and are in direct contact with maternal circulating blood.
Across this syncytiotrophoblast cell membrane, nutrients and oxygen are deliv-
ered to the fetus and waste products are returned to the maternal circulation.
Table 1. Diagnosis of preeclampsia
Definition of preeclampsia according to WHO
Preeclampsia is a syndrome defined by hypertension and proteinuria and may be
associated with other signs and symptoms
Preeclampsia occurs after the 20th gestational week
Moderate preeclampsia
Systolic blood pressure 140mm Hg and/or a diastolic pressure 90 mm Hg measured
on separate occasions at least 4h apart
Proteinuria in a 24-hour protein excretion 300mg or 1on two random urine samples
collected 4 h apart
Severe preeclampsia
Systolic blood pressure 160 mm Hg and/or diastolic 110 mm Hg measured on
separate occasions at least 4h apart
Proteinuria in a 24-hour protein excretion 5 g or 3on two random urine samples
collected 4h apart
Cerebral dysfunction (blurred vision, scotoma, headache, cerebrovascular accidents)
Epigastric or right upper quadrant pain
Renal failure or oliguria 500ml in 24h
Pulmonary oedema
Impaired liver function (serum transaminase levels 2 times normal or greater)
Thrombocytopenia (100,000 platelets/mm
3
)
Coagulopathy
Fetal growth restriction
Eclampsia (generalized convulsions)
HELLP
Matthiesen/Berg/Ernerudh/Ekerfelt/Jonsson/Sharma 52
The cytotrophoblasts that differentiate into extravillous cytotrophoblasts
are designed to develop a migratory capacity to invade deep into the decidual
matrix and the maternal spiral arteries. The musculoelastic media of the spiral
arteries are replaced by the invading cytotrophoblasts and fibrinoid material.
The spiral arteries are thereby modulated into low-resistance flow channels
allowing increased blood volume to the intervillous space [3]. The invasion of
cytotrophoblasts relies on their expression of cell adhesion molecules and
secretion of proteolytic enzymes, matrix metalloproteinases (MMP). Integrins
are cell membrane adhesion receptors that adhere to different matrix glycopro-
teins depending on their expression of tissue-specific subunits. When tro-
phoblasts migrate across the basement membrane and into the decidua towards
the spiral arteries their expression of integrins is modulated according to the
structure of the surrounding tissue. The surrounding matrix is digested by pro-
teolytic enzymes secreted by the trophoblasts. Thus, the integrins and proteases
together give trophoblasts a migratory capacity, which is a significant physio-
logical adaptation for a successful pregnancy outcome. A shallow trophoblast
invasion results in a poor placenta vascularization and deficient anchor in the
matrix tissue. This is associated with a high risk of preeclampsia, intrauterine
growth retardation and abruptio placentae [35 ].
Balancing Act between Inflammatory and Anti-Inflammatory
Immune Responses
The trophoblast invasion is under the influence of several cytokines pro-
duced at the maternal-fetal interface by several cells of immune and non-immune
origin, such as leucocytes including NK cells, trophoblasts, stromal cells and
glandular endothelium [6]. Thus, the current hypothesis regarding the aetiology
of preeclampsia should focus on maladaptation of immune responses and defec-
tive trophoblast invasion (fig. 1). The activation of the adaptive immune response
is characterized according to the phenomenon of polarized cytokine secretion by
T helper (Th) cells. These are primarily divided into two subsets: Th1 and Th2. In
humans, Th1 cells secrete inflammatory cytokines such as interferon- (IFNI-)
and tumour necrosis factor- (TNF-), whereas Th2 cells secrete anti-inflamma-
tory cytokines such as IL-4, IL-5, and IL-9. Both Th1 and Th2 cells as well as
non-lymphoid cells, including macrophages, secrete IL-10. Although the
Th1/Th2 model is too simple to encompass all the complex differentiation pro-
files of cytokine-producing cells, it still provides a useful framework to explain
the immune responses imparted either by immune cells or non-immune cells [7].
An important decisive factor for the induction of either the Th1 or Th2
pathway is the presence of certain cytokines during the initial process when
Preeclampsia and the Immune System 53
antigens are recognized. IL-4 dictates the immune response to Th2 and the
effects of IL-4 have been shown to dominate over those of IFN- [8]. Thus, it is
possible that the presence of the trophoblasts in a uterine cavity with a poor res-
ident anti-inflammatory milieu initiates an incompatible activation of the decid-
ual immune cells that direct the local immune activity towards inflammation.
Mechanisms of placentation
Spontaneous
abortion
Fetal
growth
retardation
Fetal
death
Abruptio
placentae Preeclampsia
Placenta ischaemia
& inflammation
I

t
r
i
m
e
s
t
e
r
I
I

t
r
i
m
e
s
t
e
r
I
I
I

t
r
i
m
e
s
t
e
r
'Abnormal'
Inadequate
trophoblast invasion
into decidua and
defect
remodelling of spiral
arteries
'Normal'
Pregnancy-
physiologic
trophoblast invasion
into decidua and
remodelling of spiral
arteries
Balance of
cytokine-,
hormonal-, and
neuronal networks
Interactions
between
uterine NK cells and
trophoblast cells
Cellular
homeostasis
Angiogenesis
Other
mechanisms
?
Endothelial dysfunction
Free
radicals
Release of
toxic substances:
Pro-inflammatory cytkines?
Syncytial knots?
Systemic
Inflammation:
TNF- > IL-10, TGF-
IFN-> IL-4 Apoptosis
Uncomplicated
pregnancy
Fig. 1. Flow chart showing mechanisms of placental development in uncomplicated
pregnancies (normal) and of pathological placentation (abnormal), as in preeclampsia.
Other pregnancy complications, spontaneous abortion, fetal death and growth retardation,
may also be clinical signs of placental ischaemia and inflammation as shown.
Matthiesen/Berg/Ernerudh/Ekerfelt/Jonsson/Sharma 54
Subsequently, the systemic cytokine production and immune responses are
likely to be predominant in their inflammatory functions which might initiate
the pathology associated with preeclampsia.
Cytokines and Preeclampsia
A set of cytokines have so far been of particular interest in the pathological
pregnancy outcome, including preeclampsia (fig. 1).
Transforming Growth Factor-b
Transforming growth factor- (TGF-) is secreted by decidual stroma
cells, macrophages and T cells and is present locally at the maternal-fetal inter-
face. This cytokine exerts a regulatory role by a potent negative effect on tro-
phoblast invasiveness by induction of tissue inhibitors of matrix proteases and
increased adhesiveness to matrix proteins [5, 6]. However, the impact of an
overexpression of TGF- on a shallow cytotrophoblast invasion at the
fetal-placental unit has been disputed since no difference was found either in
the placental bed or in the placenta in preeclamptic patients compared with nor-
mal pregnancies [9].
Tumour Necrosis Factor-a
TNF- is a proinflammatory cytokine produced e.g. by NK cells, mono-
cytes/macrophages and trophoblasts. TNF- promotes apoptosis and leakage of
the endothelial vessels, leading to systemic endothelial activation and thereby
signs associated with preelampsia [10]. In conjunction with an overexpression
and secretion of TNF- in the placenta and in plasma as observed in
preeclampsia an enhanced plasma and placental expression of IL-1 has been
reported. IL-1 and TNF- both promote structural and functional changes in
endothelial cells including oxidative stress, activation of the complement cas-
cade, secretion of vasoconstrictors, microthrombosis and infarction, and ele-
vated thromboxane levels. All these changes are seen in preeclampsia and the
effects of increased expression of TNF- seem to be involved in the pathophys-
iological mechanisms leading to the clinical signs [1, 11]. Thus, TNF- is a
major contributor to many of the local and systemic changes that characterize
preeclampsia. TNF- has also been shown to elevate leptin protein levels, a
phenomenon associated with preeclampsia. Interestingly, microarray analysis
of differentially expressed genes in placental tissue of preeclampsia revealed
that one of the most upregulated transcripts in preeclampsia tissue was the
obese leptin gene [12].
Preeclampsia and the Immune System 55
Interferon-g
IFN- released by activated T cells activates the specialized uNK cells
which possess regulatory properties for physiological trophoblast invasion in
the decidua. However, excessive amounts of IFN- in conjunction with TNF-
and IL-1 can lead to apoptosis of trophoblasts [2, 13]. This may indeed also be
the case in unexplained spontaneous abortions [14]. In an inflammatory envi-
ronment, macrophages secrete high levels of IL-12 that stimulate IFN- secre-
tion by NK cells, thereby inhibiting angiogenesis [6].
IL-10
IL-10 is an important anti-inflammatory cytokine in pregnancy that inhibits
upregulation of MMP-2 and MMP-9 and promotes the termination of Th1
inflammatory rejection reactions against the fetal-placental unit. In a small num-
ber of preeclampsia cases, high levels of IL-10 are seen both in the placenta and
in peripheral blood, which might be a compensatory response to elevated levels
of IFN-, TNF-, IL-2 and IL-12 [5, 8, 15]. On the other hand, IL-10 deficiency
and an increase of TNF- expression in the placenta and decidua are observed in
preeclampsia compared to those with a normal pregnancy. This was interpreted
as a modified immune balance consistent with inflammatory responses in
preeclampsia [16]. This suggests that coupling of IL-10 deficiency and inflam-
matory signals at different stages of pregnancy may contribute to disparate clin-
ical conditions, including preeclampsia [17, Sharma, unpubl. observations].
Other Cytokines
Recently, several other cytokines have been identified in the
immunopathological cascade of preeclampsia. Since these cytokines do not
adjust to the original concept of Th2 as beneficial and Th1 as deleterious to
pregnancy, it has been proposed that caution should be observed with the
immunotrophism theory stated by Wegmann et al. [18]. Nevertheless, the
Th1/Th2 paradigm in its simplistic form may still be part of complex
immune-endocrine interactions locally or systemically. In this context,
Chaouat et al. [19] suggest that the preclinical cytokine network has come
closer to the patient bedside, showing a correlation between the evaluation of
uterine blood flow, ultrasonographic morphology of uterine-placental vessels
and immunohistochemical localization and levels of IL-12, IL-18 and counts
of uNK cells. They showed, in a group of patients enrolled in an in vitro
fertilization programme, that a correlation exists between cytotoxic cytokine
profiles and vascular anomalies in implantation failures. This scenario is in
contrast with the proper activation and localization of uNK cells and vas-
culature seen in implantation success. Pro-inflammatory cytokines trigger
Matthiesen/Berg/Ernerudh/Ekerfelt/Jonsson/Sharma 56
activation of the coagulation cascade leading to vasculitis and infarction and
may further deteriorate the early placental development and hamper the tro-
phoblast invasion [1, 5, 10].
Recently, an elegant way of measuring cytotoxic responses, by means of
granulysin levels in serum, was reported to be associated with the occurrence
and clinical manifestations of preeclampsia [20]. The real challenge is to find
early markers of subsequent preeclampsia. In this context, soluble IL-2 receptor
in plasma was elevated in the 1st trimester of women that later developed
preeclampsia compared with controls [21].
Maternal-Fetal Interactions
In the uterine cavity, the extravillous cytotrophoblast cells reveal themselves
by the expression of the unusual HLA class I molecules: HLA-E, and HLA-G
together with HLA-C. At present, the only receptors that have been found to
these HLA class I molecules are located on the unique uNK cells (fig. 1). uNK
cells are CD56bright CD16 compatible with a low cytotoxic potential com-
pared with the classical killer NK cells in peripheral blood that express
CD56dim CD16. The syncytiotrophoblast, covering the placental villi and
thereby exposed to maternal blood, expresses no HLA molecules [2].
The uNK cells show a variation over the menstrual period. During the
luteal phase and until midgestation uNK cells increase in number and they
accumulate around the invading cytotrophoblasts. After initial development of
the placenta, levels of uNK cells decline and cease to be present at term [2].
The interaction between extravillous cytotrophoblasts and uNK cells, pos-
sibly after stimulation by IFN-, has recently been suggested to have an influ-
ence on the remodelling of spiral arteries [22]. A high expression of receptors
signalling inhibition of cytotoxic activity of uNK cells interacts with HLA-E,
HLA-C, and HLA-G [2].
The inability of cytotrophoblasts to modify the cytokine repertoire of uNK
cells and their regulation of adhesion molecules, MMPs and sufficient neovas-
cularization may be critical factors for the onset of pregnancy complications
including preeclampsia [2, 3, 6, 17, 18].
Apoptosis and Syncytial Knots
Programmed cell death or apoptosis plays an important role in cell home-
ostasis and tissue remodelling, particularly placental development. Importantly,
placental degeneration observed in preeclampsia may be due to unscheduled
Preeclampsia and the Immune System 57
apoptosis of trophoblasts. The pregnancy-associated remodelling of the spiral
arteries is mediated by invasive cytotrophoblasts. However, if these trophoblasts
are subjected to a high rate of apoptosis, this defective transformation of spiral
arteries may result in local ischaemia, thrombosis and infarction (fig. 1). The
exact causes of enhanced apoptosis in preeclampsia are currently unknown.
Likewise, increased apoptosis of syncytiotrophoblasts may increase the amount
of syncytiotrophoblast debris, syncytial knots, that leak into the maternal circu-
lation and generate an exaggerated systemic endothelial activation [23]. Sargent
et al. [24] have proposed that when syncytial knots break off in increasing
amounts from the placenta and are shed into the maternal circulation they may
be the cause of the systemic endothelial activation that is seen in preeclampsia
(fig. 1). The deported trophoblast debris can, in vitro, activate maternal sources
of TNF- and IL-12 from monocytes, which further pushes the systemic
immune response towards extensive inflammation instead of the normal innate
immune reactivity that syncytial knots usually accomplish during pregnancy.
The reason for this strong apoptosis is unknown, but it has been shown that
pro-inflammatory cytokines are capable of upregulating Fas/FasL genes, while
anti-inflammatory cytokines protect trophoblasts against Fas-induced apoptosis
[24, Sharma, unpubl. observations].
Free Radicals
Other mediators of inflammation are also important in the pathogenesis of
preeclampsia, including reactive oxygen species, in particular superoxide
anions. These agents are increased in preeclampsia, where the equilibrium of
antioxidants (vitamin E, ascorbic acid, glutathione peroxidase, superoxide
catalase/mutase, and caeruloplasmin) is disturbed. Antioxidants are produced
by many cells, also trophoblasts and leucocytes, to protect them from free rad-
icals or as part of cellular homeostasis and ageing. Free radicals and levels of
lipid peroxidation are increased in preeclampsia and capable of evoking
systemic endothelial activation, including platelet consumption, altered throm-
boxane/prostacyclin ratio, increased TNF- production and promotion of the
coagulation cascade [25].
During normal pregnancy, a rise in antioxidants is detected in blood with
increasing gestational age. However, if the inflammation is strong or the produc-
tion of the antioxidants is low, the predominating condition inevitably favours oxi-
dizing species. This is the case in preeclampsia, where free radicals are present at
significantly higher levels than during normal pregnancy [25] (fig. 1). In the
haemolysis, elevated liver enzymes, low platelet (HELLP) syndrome, haemoly-
sis of erythrocytes might occur due to a high degree of oxidation of glutathione,
Matthiesen/Berg/Ernerudh/Ekerfelt/Jonsson/Sharma 58
which causes cell damage. As a consequence, it has been suggested that treatment
with inhibitors of cyclooxygenase to block oxidative stress on erythrocytes as
well as nutritional supplements with antioxidants, vitamin E and C, might reduce
the incidence of preeclampsia in high-risk pregnancies [25].
Lymphocyte Populations in Blood
Preeclampsia is also characterized by systemic changes in the distribution
of lymphocyte populations in peripheral blood. Increased levels of
activated/memory cells (CD4CD45RO and CD4CD29) and decreased
levels of nave/suppressor cells (CD4CD45RA) have been noted. The
interpretation is that antigens have activated the T cells observed in preeclamp-
sia. In contrast, lymphocytes in normal pregnancy are switched towards a pre-
dominance of CD4CD45RA nave/suppressor T cells. The level of
cytotoxic CD8 T cells expressing the S6F1 marker, which represent killer
effector functions, is increased in preeclamptic pregnancies compared with nor-
mal pregnancies, again indicating inflammatory activity [26].
The mechanisms behind leucocyte activation in preeclampsia are
unknown, but the changes are similar to those observed in humans after viral
or bacterial infections. Low doses of bacterial endotoxin injected into preg-
nant rats resulted in a condition resembling preeclampsia including the
appearance of T cell activation markers [15]. This presents an intriguing basis
to probe the role of clinical and subclinical infections in the pathogenesis of
preeclampsia. These observations also indicate that preeclampsia is associ-
ated with both the innate and the adaptive immune activity in the peripheral
blood [6, 18, 24, 26].
Toxic Substances of Preeclampsia
What is the nature of the toxic substances that escape from an obvious
sick placenta, swim out into the maternal circulation and gain access to and
disturb almost every organ in the human body and reveal their presence by the
characteristic signs and symptoms of preeclampsia (fig. 1)? Many candidates
(fig. 1) have been suggested although no complete agreement has been reached
[1, 6, 10, 11, 16, 18, 2325].
Concerning cytokines as potential villains and as potential diagnostic tools
in the prediction of preeclampsia, we addressed this question (like many others)
by measuring cytokine levels in serum using the Luminex

assay (Camarillo,
Calif., USA) in preeclamptic patients (n 15) and compared them with normal
Preeclampsia and the Immune System 59
pregnancies (n 15). In preeclampsia, we observed an upregulated systemic
innate immune reactivity with increased levels of TNF-, IL-6, and IL-8. When
we stimulated peripheral blood mononuclear cells with paternal antigens (fetus-
specific) or recall antigens (purified protein derivates of Mycobacterium tuber-
culosis or tetanus toxoid) similar levels of induced secretions of IL-4, IL-10,
IL-12 and IFN- (detected by the highly sensitive ELISPOT assay) were
detected in preeclampsia and normal pregnancies. This does not exclude local
cytokine aberrations at the placental level that are compatible with inflammatory
activity. However, the results agree with the main concept of preeclampsia being
an inflammatory phenomenon [1, 6, 15, 18, 23, 24], but with a much more com-
plex picture than a Th1 deviation only [19].
Conclusion
Preeclampsia is a multisystem disorder based on a cascade of immuno-
pathological events originating from the placenta. No single candidate mecha-
nism exists to explain the complex pathogenesis. As of now, there is no reliable
marker or predictor of preeclampsia. Clearly, however, local as well as systemic
inflammatory activity occurs in preeclamptic patients. To identify these com-
plex immune factors and arrange them in a test where the diverted inflamma-
tory activity will be detected should be the target in future research concerning
preeclampsia.
To further elucidate the mechanisms underlying preeclampsia, it is our
hope that animal models can be developed in the very near future, wherein
depletion or the overwhelming presence of key players in the aetiology of the
disease can be studied developmentally.
Acknowledgment
This study was supported by the US Public Health COBRE grant 1 P20 RR018728
(SS) NIH, a grant of the County Council of stergtland, Sweden and a grant of the Health
Research Council in the South-East of Sweden.
References
1 Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA, McLaughlin MK: Preeclampsia: An
endothelial cell disorder. Am J Obstet Gynecol 1999;54:133142.
2 Moffett-King A: Natural killer cells and pregnancy. Nat Rev Immunol 2002;2:656663.
Matthiesen/Berg/Ernerudh/Ekerfelt/Jonsson/Sharma 60
3 Bischof P, Campana A: A model for implantation of the human blastocyst and early placentation.
Hum Reprod Update 1996; 2/3:262270.
4 Pijnenborg R: The placental bed. Hypertens Pregnancy 1996;15:723.
5 Zhou Y, Fisher SJ, Janatpour M, Genbacev O, Dejana E, Wheelock M, Damsky CH: Human
cytotrophoblast adopt a vascular phenotype as they differentiate. A strategy for successful endovas-
cular invasion? J Clin Invest 1997;99:21392151.
6 Saito S, Sakai M: Th1/Th2 balance in preeclampsia. J Reprod Immunol 2003;59:161173.
7 Mosmann TR, Sad S: The expanding universe of T-cell subsets: Th1, Th2 and more. Immunol
Today 1996;17:138146.
8 Swain SL: IL-4 dictates T-cell differentiation. Res Immunol 1993;144:616620.
9 Lyall F, Simpson H, Bulmer JN, Barber A, Robson SC: Transforming growth factor- expression
in human placenta and placental bed in third trimester normal pregnancy, preeclampsia, and fetal
growth restriction. Am J Pathol 2001;159:18271838.
10 Rinehart BK, Terrone DA, Lagoo-Deenadayalan S, Barber WH, Hale EA, Martin JN, Bennett WA:
Expression of the placental cytokines tumor necrosis factor , interleukin 1 and interleukin 10 is
increased in preeclampsia. Am J Obstet Gynecol 1999;181:915920.
11 Anim-Nyame N, Gamble J, Sooranna SR, Johnson MR, Steer PJ: Microvascular permeability is
related to circulating levels of tumour necrosis factor- in pre-eclampsia. Cardiovasc Res 2003;
58:162169.
12 Reimer T, Koczan D, Gerber B, Richter D, Theisen HJ, Friese K: Microarray analysis of differen-
tially expressed genes in placental tissue of pre-eclampsia: Up-regulation of obesity related genes.
Mol Hum Reprod 2002;8:674680.
13 Ashkar AA, DiSanto JP, Croy BA: Interferon-gamma contributes to initiation of uterine vascular
modification, decidual integrity, and uterine natural killer cell maturation during normal murine
pregnancy. J Exp Med 2000;192:259270.
14 Plevyak M, Hanna N, Mayer S, Murphy S, Pinar H, Fast L, Ekerfelt C, Ernerudh J, Berg G,
Matthiesen L, Sharma S: Deficiency of decidual IL-10 in first trimester missed abortion: A
lack of correlation with the decidual immune cell profile. Am J Reprod Immunol 2002;47:
242250.
15 Murphy S, Fast L, Sharma S: IL-10, uterine NK cells, inflammation, and pregnancy. Am J Reprod
Immunol 2004;51:434.
16 Henessy A, Pilmore HL, Simmons LA, Painter DM: A deficiency of placental IL-10 in preeclamp-
sia. J Immunol 1999;163:34913495.
17 Roth I, Fisher SJ: IL-10 is an autocrine inhibitor of human placental cytotrophoblast MMP-9 pro-
duction and invasion. Dev Biol 1999;205:194204.
18 Wegmann TG, Lin H, Guilbert L, Mossman TH: Bidirectional cytokine interactions in the
materno-fetal relationship: Is successful pregnancy a Th2 phenomenon? Immunol Today 1993;
14:353355.
19 Chaouat G, Ledee-Bataille N, Zourbas S, Dubanchat S, Sandra O, Martal J, Ostojojic S, Frydman R:
Implantation: Can immunological parameters of implantation failure be of interest for preeclampsia?
J Reprod Immunol 2003;59:205217.
20 Sakai M, Ogawa K, Shiozaki A, Yoneda S, Sasaki Y, Nagata K, Saito S: Serum granulysin is a
marker for Th1 type immunity in pre-eclampsia. Clin Exp Immunol 2004;1:114119.
21 Eneroth E, Remberger M, Vahlne A, Ringden O: Increased serum concentrations of interleukin-2
receptor in the first trimester in women who later developed severe preeclampsia. Acta Obstet
Gynecol Scand 1998;77:591593.
22 Croy BA, Ashkar AA, Minhas K, Greenwood JD: Can murine uterine natural killer cells give
insights into the pathogenesis of pre eclampsia? J Soc Gynecol Invest 2000;7:1220.
23 Neale D, Demasio K, Illuzi J, Chaiworapongsa T, Romero R, Mor G: Maternal serum of women
with pre-eclampsia reduces trophoblast cell viability: Evidence for an increased sensitivity to Fas-
mediated apoptosis. J Maternal Fetal Neonatal Med 2003;13:3944.
24 Sargent IL, Germain SJ, Sacks GP, Kumar S, Redman CWG: Trophoblast deportation and the
maternal inflammatory response in pre-eclampsia. J Reprod Immunol 2003;59:153160.
Preeclampsia and the Immune System 61
25 Walker JJ: Antioxidants and inflammatory cell responses in preeclampsia. Semin Reprod Endocrinol
1998;16/1:4755.
26 Matthiesen L, Berg G, Ernerudh J, Skogh T: Lymphocyte subsets and autoantibodies in pregnan-
cies complicated by placental disorders. Am J Reprod Immunol 1995;33:3139.
27 Jonsson Y, Rubr M, Matthiesen L, Berg G. Nieminen K, Sharma S, Ernerudh J, Ekerfelt C:
Cytokine mapping of serum from women with preeclampsia and from women with normal preg-
nancy. Am J Reprod Immunol, submitted.
Leif Matthiesen
Department of Molecular and Clinical Medicine
Division of Obstetrics and Gynaecology, University Hospital
SE581 85 Linkoping (Sweden), Tel. 46 13 22 31 37
Fax 46 13 14 81 56, E-Mail [email protected]