| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Articles: Hormones and Reproductive Health |
Departments of Obstetrics, Gynecology, and Reproductive Sciences (D.F.B., C.S., K.P.C.) and Cell Biology and Physiology (K.P.C.), University of Pittsburgh School of Medicine, and Magee-Womens Research Institute, Pittsburgh, Pennsylvania 15213; and Nuffield Department of Obstetrics and Gynecology, University of Oxford, John Radcliffe Hospital (A.S., C.W.G.R.), Oxford, United Kingdom OX3 9DU
Address all correspondence and requests for reprints to: Dr. Kirk P. Conrad, Magee-Womens Research Institute, 204 Craft Avenue, Pittsburgh, Pennsylvania 15213. E-mail: rsikpc{at}mail.magee.edu
Abstract
It is postulated that inadequate remodeling of the uterine spiral
arteries in preeclampsia leads to focal ischemia and generation of
inflammatory cytokines, such as tumor necrosis factor (TNF
) and
interleukins (ILs), by the placenta. Our objective was to compare
TNF, IL-1, IL-1ß, and IL-6 levels in placentas from patients
with preeclampsia and normal term pregnancies. Because the placenta is
a large heterogeneous organ, we analyzed multiple sites per placenta.
On the average, there was a 3-fold variation in cytokine protein levels
across the eight sites analyzed for each placenta. However, there were
no significant overall differences among the normal term, preeclamptic,
and preterm placentas from women without preeclampsia. There were also
no significant differences in TNF messenger ribonucleic acid between
the normal term and preeclamptic placentas, although TNF messenger
ribonucleic acid levels were lower in placentas from preterm patients
without diagnosis of preeclampsia than in the normal term placentas.
In vitro, hypoxia stimulated the production of TNF,
IL-1 and IL-1ß, but not that of IL-6, by placental villous
explants from both groups of patients, and this was not exaggerated in
preeclampsia. Finally, although peripheral and uterine venous levels of
TNF were elevated in preeclamptic women compared with normal term
patients, the ratio of uterine to peripheral venous TNF levels was
not significantly different from 1.0 for either patient group. Taken
together, these results suggest that sources other than the placenta
contribute to the elevated concentrations of TNF and IL-6 found in
the circulation of preeclamptic women.
PREECLAMPSIA IS A disease of human
pregnancy diagnosed by the onset of hypertension and proteinuria in the
third trimester (1). Several investigators have suggested
that endothelial dysfunction underlies the disease manifestations
(reviewed in Ref. 2), and the etiology of preeclampsia
relates to deficient physiological remodeling of uterine spiral
arteries (3). Without sufficient changes in the uterine
vasculature in early pregnancy, the placenta is susceptible to the
development of focal regions of ischemia/hypoxia. In turn, reduced
oxygen tension in the placental tissue may lead to the production and
secretion of cytotoxic factors that have been postulated to affect the
maternal vasculature (4). Inflammatory cytokines such as
tumor necrosis factor- (TNF) and interleukin-1ß (IL-1ß) can
produce endothelial dysfunction (5), and synthesis of
these cytokines as well as IL-6 has been documented in the human
placenta (6, 7, 8, 9). Further, we have shown that incubation of
human placental explants under reduced oxygen conditions results in the
elevated production of TNF, IL-1, and IL-1ß
(9).
We and others (reviewed in Refs. 10 and 11)
reported that circulating levels of TNF and IL-6 are increased in
women with preeclampsia. Given that the placenta produces inflammatory
cytokines, some of which can be stimulated by hypoxia in
vitro, our objective was to determine whether placental protein
and messenger ribonucleic acid (mRNA) levels for TNF, IL-1,
IL-1ß, and IL-6 are elevated in women with preeclampsia. Further,
placental tissues from normal pregnant and preeclamptic women were
incubated in 21% or 2% oxygen to determine whether there are
differences in the magnitude of hypoxia-stimulated cytokine production.
A final objective was to compare peripheral circulating levels of
TNF with those in the uterine vein to directly ascertain whether the
uteroplacental unit contributes to the systemic elevation of
inflammatory cytokines in women with preeclampsia.
Subjects and Methods
Human subjects
All placentas were obtained from women undergoing cesarian
section without labor under approval of the institutional review board
at Magee-Womens Hospital. Placentas from nulliparous patients with
normal pregnancies were collected at term (38–41 gestational weeks).
Placentas were obtained from nulliparous women with preeclampsia at
28–38 gestational weeks. The diagnosis of preeclampsia was made by
strict criteria: onset of hypertension during the third trimester
(
140/90 mm Hg on two occasions), detectable urinary protein (1+ by
dipstick or 300 mg/24 h), and plasma uric acid levels exceeding 1
SD from the mean for gestational age (in five of six
patients). For placental samples used for cytokine protein and mRNA
analyses, two preeclamptic patients of less than 35 weeks gestation
received ß-methasone, one of whom also had a diagnosis of HELLP
syndrome (hemolysis, elevated liver enzymes, and low platelets), and
two of the preeclamptic patients had growth-restricted infants. For
placental samples used in explant culture, three of the six patients
were less than 30 gestational weeks and received ß-methasone, and two
had growth-restricted infants.
As an additional control group, six placentas were obtained from nulliparous women with preterm delivery and without preeclampsia (28–32 gestational weeks). These patients with preterm deliveries were also delivered by cesarian section (one with labor), and all neonates were of appropriate size for gestational age. Only one patient had chorioamnionitis, which was reported as mild by the perinatal pathologist.
Patients from whom uterine and peripheral venous blood samples were taken at cesarean delivery were selected in Oxford, using different criteria. All gave their written and informed consent to take part in the study, which was approved by the Central Oxford research ethics committee. Women with preeclampsia (n = 8) had clear documentation of no proteinuria earlier in pregnancy after which at least 2+ protein, as a new development, was detected using dipstick testing. Their diastolic pressures had increased from less than 90 mm Hg before 20 weeks gestational age to more than 90 mm Hg before delivery, recorded on at least two occasions. There was no restriction with regard to parity, and plasma uric acid measurements were not used as selection criteria. Normal pregnant women (n = 8) required cesarean delivery for nonurgent reasons, such as breech presentation or elective repeat cesarean delivery. They never had proteinuria, and at no time during their uncomplicated pregnancies were diastolic pressures of 90 mm Hg or more recorded. None was taking regular medications for any purpose. The cases and controls were not matched in any respect, except for delivery by cesarean section.
Collection of placental samples
Immediately after delivery of the placenta by cesarian section,
samples were collected, and the total length of processing time was
less than 15 min. As the placenta is a large heterogeneous organ, and
sampling bias is a potential hazard (12), a square grid of
16 holes in a 4 x 4 design within a circular template was
constructed to overlay the placenta. Templates of various sizes were
prepared to best fit placentas of different diameters. After blotting
the maternal surface of blood, a sterile nickel cork borer was inserted
into each of the 16 holes (
1 cm) to obtain a full thickness biopsy
(0.5 g tissue, which excluded the chorionic plate). Each sample was
briefly rinsed 3 times in cold saline, blotted dry, and flash-frozen in
individual bags in liquid nitrogen.
Preparation of placental homogenates
The frozen placental sample was pulverized on dry ice, and approximately 100 mg were transferred to 5-mL polypropylene tubes. Each tube had a cocktail of protease inhibitors added (phenylmethylsulfonylfluoride, leupeptin, antipain, pepstatin A, and soybean trypsin inhibitor; Sigma, St. Louis, MO). Placental samples were then homogenized with a Tissumizer (Tekmar, Cincinnati, OH) for 30 s on ice in a 2-fold volume of homogenizing buffer [50 mmol/L Tris (pH 7.4) and 1 mmol/L ethylenediamine tetraacetate; Fisher Scientific, Pittsburgh, PA]. The homogenate was centrifuged at 3,000 x g for 10 min at 4 C, and then the collected supernatant was centrifuged at 10,000 x g for 15 min at 4 C. Two 10-µL aliquots were analyzed for protein concentration by the Lowry method, and the remaining sample was aliquoted and stored at -80 C until enzyme-linked immunosorbent assay (ELISA) analysis.
For ELISA analysis, samples were used from odd-numbered sites, and thus
eight samples per placenta were analyzed. Samples were run in duplicate
on two separate ELISA plates, and the normal and preeclamptic samples
were pipetted into alternating wells. To control for any gestational
age effect, homogenates were prepared from a single placental biopsy
site from preterm deliveries without preeclampsia (only one sample was
available from these patients). In this final study normal term and
preeclamptic placental homogenates were prepared as a pool of four
sites per placenta and assayed for TNF and IL-6 levels in duplicate.
Placental homogenates were diluted in the manufacturer’s serum diluent
to fall within the linear portion of the respective standard curves as
determined in preliminary studies. Results were then normalized per mg
protein.
ELISA of placental homogenates
All ELISAs were performed using kits obtained from R\|[amp ]\|D Systems, Inc. (Minneapolis, MN). For TNF, IL-1ß, and IL-6,
Quantikine High-Sensitivity kits were used, with sensitivities of 0.5,
0.125, and 0.156 pg/mL, respectively, whereas the sensitivity for the
IL-1 assay was 3.9 pg/mL. A limited number of samples were also
evaluated for IL-12 by an ELISA that had a sensitivity of 0.781 pg/mL.
All assays were conducted according to manufacturer’s protocols.
The cytokine ELISAs of placental homogenates were extensively validated
for dilutional parallelism and monitored for recovery of recombinant
human cytokine standards in homogenates to assure no interference by
substances in the crude samples. Linearity was assessed for TNF,
IL-1ß, and IL-6 in placental samples (prepared as a pool of
homogenates from three each normal and preeclamptic samples), which
were assayed in three dilutions. These experiments were performed in
duplicate, and the data are expressed as a percentage of the measured
values (observed) divided by the calculated value from assay of the
initial pooled sample (expected), as shown in Table 1
. Dilutional analysis was not performed
for IL-1 and IL-12, because the samples were undetectable at
dilutions greater than 1:2. Recovery of samples spiked with the
respective recombinant human cytokine standard was performed with three
concentrations of the standard added to diluted, pooled samples (n
= 3 each for normal and preeclamptic placental homogenates).
|
For determination of TNF and ß-actin mRNA expression in the
preeclamptic and normal term placentas, total RNA was pooled from four
biopsy sites per placenta, and six placentas in each group were
analyzed. Only one biopsy site was available for the six placentas from
preterm deliveries without preeclampsia. Total RNA was isolated
from placental tissues using RNAwiz (Ambion, Inc.,
Austin, TX). RNA from the four biopsy sites per placenta were then
combined, and contaminating genomic DNA was removed using DNA-free
(Ambion, Inc.). The concentration of pooled RNA was
measured by spectrophotometry. Before performing RT-PCR, we verified
that each pool of RNA was of high quality by ethidium bromide staining
of total RNA (2.5 µg) separated on a formaldehyde-containing 1.5%
agarose gel, and well defined bands were observed for both 18S and 28S
RNA with no visible degradation (data not shown).
Human TNF and ß-actin primers were purchased from R\|[amp ]\|D Systems, Inc. (Minneapolis, MN), or manufactured by the DNA
Synthesis Facility at the University of Pittsburgh. For TNF, the
forward and reverse sequences were GTGACAAGCCTGTAGCCCA and
ACTCGGCAAAGTCGAGATAG, respectively. Using these primers, the RT-PCR
fragment was 414 bp, whereas amplification of genomic DNA would have
yielded a 714-bp fragment. For ß-actin, the forward and reverse
sequences were CTACAATGAGCTGCGTGTGG and AAGGAAGGCTGGAAGAGTGC,
respectively. Using these primers, the RT-PCR fragment was 528 bp,
whereas amplification of genomic DNA would have yielded a 969-bp
fragment.
The semiquantitative RT-PCR technique was modified based on that previously described (13), and all reactions were prepared in an Airclean 600 Workstation with a UV light to prevent contamination (Airclean Systems, Raleigh, NC). Total RNA (5 µg) was annealed with 30 ng of each of the reverse primers in a total volume of 10.5 µL. Using a Genius thermocycler (Techne, Princeton, NJ), the mixture was incubated for 10 min at 75 C, cooled to 42 C at 0.2 C/min, maintained at 42 C for 15 min, and then held at 4 C until further processing. Next, an RT master mix was prepared, and 9.5 µL were added to each annealing reaction consisting of 4 µL RT buffer (5x stock), 2 µL dithiothreitol (100 mmol/L), 2 µL deoxy (d)-NTP (10 mmol/L), 1 µL AMVRT (10 U/µL), and 0.5 µL RNasin (40 U/µL). The mixture was incubated for 45 min at 42 C, heated to 95 C for 5 min, and then held at 4 C until further processing.
Separate PCR master mixes were then prepared containing either the
TNF or ß-actin primer pairs (kept on ice). One µL of the RT
reaction was added to 19 µL of the ß-actin PCR master mix, and 2
µL of the RT reaction was added to 18 µL of the TNF master mix.
Preliminary experiments were conducted to determine the optimal amount
of dNTP and MgCl2 in these master mixes. For
ß-actin, the PCR master mix consisted of 1 µL buffer (20x stock),
1.5 µL dNTP (2 mmol/L), 1.2 µL MgCl2 (25
mmol/L), 6 µL enhancer (10x), 5.1 µL nuclease-free water, 0.2 µL
Thermus flavus DNA polymerase (1 U/µL), and 4 µL
of the ß-actin primer pair (2 µmol/L). For TNF, the PCR master
mix consisted of 1 µL buffer (20x stock), 1 µL dNTP (2 µmol/L),
0.8 µL MgCl2 (25 mmol/L), 6 µL enhancer
(10x), 5 µL nuclease-free water, 0.2 µL Tf1 (1 U/µL), and 4 µL
of the TNF primer pair (2 µmol/L). The mixtures were titurated and
placed in the thermal cycler at 4 C. Then the lid was heated at 94 C
for 2 min, followed by heating of the reactions at 94 C for 5 min. Each
cycle consisted of 1 min each at 94, 55, and 72 C. After completion of
the PCR, the reactions were maintained at 4 C until further processing.
To determine the optimal number of PCR cycles for ß-actin and TNF,
we tested 15–33 cycles in intervals of 3 for ß-actin and 24–42
cycles for TNF and used 21 and 31 cycles, respectively, for
subsequent analyses (Fig. 1). Further, we
determined a linear relationship between the input of complementary DNA
(cDNA) and the amount of PCR products generated with cDNA equivalent to
2.5, 1.25, and 0.625 µg total RNA (data not shown).
|
(ATCC 39894, American Type Culture Collection, Manassas, VA; PstI fragment
of 1.1 kb) or ß-actin (ATCC 65128, EcoRI
fragment of 1.1 kb) insert, 50 µCi [32P]dCTP
(3000 Ci/mol; NEN Life Science Products-DuPont, Boston,
MA), and 2 U Klenow polymerase using the Multiprime DNA labeling kit
(Amersham Pharmacia Biotech, Piscataway, NJ).
Unincorporated 32P was removed using a spin
column (Biospin P30, Bio-Rad Laboratories, Inc., Hercules,
CA). After decanting the prehybridization solution, labeled probe and 6
mL High Efficiency Hybridization System with 50% formamide (MRC,
Cincinnati, OH) were added, and hybridization was performed overnight
at 42 C. For posthybridization, the Washing/Prehyb solution was again
used at room temperature and at 50 C as needed. The membranes were
exposed to Kodak BioMax MR film (Eastman Kodak Co., Rochester, NY) for 10–20 min. The authenticity of the
RT-PCR fragment for TNF was verified by restriction enzyme digestion
for total RNA from the placental pool and from human monocytic HL-60
cells stimulated with lipopolysaccharide. Villous explant culture
Villous explant cultures were prepared as previously described (9). Briefly, blunt dissection of placental cotyledons (3–5/placenta) was performed to remove decidual tissue and large blood vessels. This was followed by fine dissection of 5- to 10-mg pieces of villous placenta bathed in 0.9% saline, which were placed into 24-well plates (35–50 mg total tissue) containing 1 mL medium 199 (Mediatech, Herndon, VA) with 10% FCS (Summit Technology, Ft. Collins, CO) and penicillin-streptomycin-gentamicin. Explants were incubated at 37 C with either no preincubation period or a 24-h preincubation. Incubations were carried out on an orbital shaker (60 rpm) under either standard tissue culture conditions of 5% CO2-95% room air or hypoxia (2.1% O2-5% CO2-balance N2). At the end of the experiment, tissue weight was recorded so that cytokine values could be corrected per wet weight, and the conditioned medium was stored at -80 C. The viability of villous explants was monitored at experiment termination by assessment of lactate dehydrogenase release into spent medium as previously described (9).
Peripheral and uterine venous blood samples
TNF was measured in plasma or serum using an ELISA kit
obtained from R\|[amp ]\|D Systems, Inc., as previously described
(11). Blood was taken from a superficial vein on the lower
lateral uterine segment. If the placenta was lateralized (as predicted
by ultrasonography), the placental side was chosen. The vein was
cannulated by a 21-gauge butterfly needle, and up to 20 mL were drawn
into standard blood collection vials (Becton Dickinson and Co., Mountain View, CA). Less than 10 min earlier an equivalent
blood sample was drawn from a peripheral arm or foot vein and processed
in an identical manner. As soon as the samples without anticoagulant
had clotted, they were centrifuged at 2500 x g, and
the serum or plasma was separated and stored at -80 C. The samples
were transported to Pittsburgh, unthawed, on dry ice for assay.
Data analysis
Cytokine protein levels in the multiple samples of normal and
preeclamptic placental homogenates were analyzed by two-factor repeated
measures ANOVA. For comparison of placental TNF/ß-actin mRNA
expression in preeclamptic, preterm (without preeclampsia), and term
samples, we employed a one-factor randomized block design ANOVA.
Differences in cytokine production by villous explants from normal and
preeclamptic placentas cultured under standard tissue culture
conditions or hypoxia were compared by two-factor randomized block
design ANOVA. Fisher’s least significant difference test was used for
post-hoc comparisons of individual means. Finally, an
unpaired t test and one-sample sign test, respectively, were
used to compare TNF in blood samples from preeeclamptic and normal
term women and to determine whether the ratio of TNF in the uterine
and peripheral venous blood was different from unity. P
< 0.05 was taken as significant.
Results
Placental cytokine protein and mRNA
To test whether the use of crude placental homogenates in cytokine
ELISAs would introduce cross-reacting or interfering substances, all
assays were validated by dilutional parallelism (Table 1
) and recovery
of spiked recombinant human cytokine standards. All cytokine levels
were detectable within a linear portion of the standard curves with
appropriate sample dilution. The overall recovery of cytokine standards
spiked into pooled placental homogenates ranged from 90–107% for the
TNF, IL-1, IL-1ß, IL-6, and IL-12 assays.
Table 2 depicts cytokine protein levels
in placental homogenates from normal term and preeclamptic placentas
after multiple site biopsy. Each of eight biopsies per placenta was
analyzed by specific assay, and then the values obtained for the eight
sites were averaged for each placenta. There were no significant
differences in levels of TNF, IL-1ß, IL-1, or IL-6 in the
placentas from normal term compared with preeclamptic pregnancies. In
addition, in a limited number of samples (three sites per placenta,
four patients), IL-12 levels did not differ (term, 4.3 ± 1.0
pg/g; preeclamptic, 4.1 ± 1.0 pg/g; mean ±
SEM). Within each placenta, there was a 3-fold range in
cytokine values obtained over the eight sites assayed (Fig. 2). This variation within a placenta was
similar between patient groups and was not related to the anatomical
location of the biopsy.
|
|
39 weeks) and preeclamptic patients (33.5 weeks), single site
biopsies were obtained from a group of preterm patients without
preeclampsia, and placental homogenates were assayed for TNF and
IL-6 levels. Again, there were no significant differences in placental
TNF and IL-6 concentrations (Table 3).
|
shows the confirmation of TNF
mRNA expression in the human placenta by restriction enzyme digestion
using HincII. Two fragments of 169 and 245 bp were
predicted. This pattern was observed for RT-PCR products in the total
RNA pooled from three normal term placentas and in
lipopolysaccharide-stimulated HL-60 cells that served as a positive
control for TNF production. Figure 4
portrays the results for TNF and ß-actin mRNA expression for the
normal term and preeclamptic placentas. There were no significant
differences between the two groups of placentas. In contrast,
normalized TNF expression was significantly lower in the preterm
placentas without preeclampsia than in the normal term placenta
(P < 0.05; Fig. 5).
|
|
|
production between term and preeclamptic placental
villous explants (91 ± 12 vs. 97 ± 32 pg/g
wt) or in IL-1ß (173 ± 78 vs. 107 ± 27 pg/g
wt). After this 24-h incubation, villous explants were placed into
fresh medium and incubated for an additional 24 h under standard
tissue culture conditions (21% oxygen) or under 2.1%
O2 (hypoxia). As presented in Fig. 6, exposure to reduced oxygen
significantly stimulated the production of TNF, IL-1ß, and IL-1
in villous explants obtained from both normal term and preeclamptic
placentas. Hypoxia did not affect IL-6 production by villous explants.
IL-6 and TNF levels were significantly reduced, however, in
villous explants prepared from preeclamptic vs. normal term
placenta. Finally, in two placentas from each patient group, decidual
basal plate was incubated in a similar fashion; however, no
differences in TNF or IL-6 levels were observed (data not
shown).
|
levels in the preeclamptic women, and
this difference was maintained across uterine circulation (Table 4). To determine whether this elevation
in peripheral levels was of uteroplacental origin, uterine venous
levels were taken as a ratio of peripheral venous levels. This ratio
did not differ significantly from 1.0 in either patient group (term,
1.04 ± 0.05; preeclamptic, 0.98 ± 0.03).
|
The failure of the uterine vasculature to undergo adequate
physiological remodeling in women with preeclampsia has led a number of
investigators to postulate that a consequence of reduced placental
perfusion is the generation of cytotoxic factors that circulate and
injure the maternal endothelium (1, 2, 3, 4). Indeed, we
previously reported that the normal human placenta can produce more
inflammatory cytokines, including TNF, when incubated under low
oxygen tension (9). Inflammatory cytokines are notorious
for affecting the endothelium in a fashion similar to events reported
in preeclamptic women (2, 5), and TNF levels are higher
in the circulation of women with preeclampsia (11).
Therefore, in the present work we investigated whether the placenta is
the source of increased circulating TNF and IL-6 in preeclampsia.
Placental concentrations of other inflammatory cytokines, such as
IL-1, IL–1ß, and IL-12, were also evaluated.
As the placenta is a large heterogeneous organ, we sampled each
placenta in a systematic and unbiased fashion (12). We
constructed circular grids, each with 16 holes to fit over a placenta
for multiple sampling. This allowed us to obtain punch biopsies without
bias toward the condition of the placenta (although if sections were
taken near areas of gross infarctions, this was noted), and sample
collection was completed within 15 min of delivery. Also, in these
studies only placentas obtained after cesarian section in women without
labor were used, because others have reported elevated levels of TNF
and IL-6 in amniotic fluid with labor (14).
Inflammatory cytokines were readily detected in homogenates prepared
from all biopsies in the normal term and preterm placentas (with and
without preeclampsia). Comparison of cytokine protein levels within a
single placenta demonstrated a 3-fold variation among sites that was
independent of anatomic location or patient group. In retrospect, our
strategy of obtaining multiple biopsies in a systematic and unbiased
fashion (as opposed to single sampling) was warranted in view of this
highly variable expression. In contrast to a previous report
(15), we did not find an elevation in TNF protein
levels in the preeclamptic placenta, nor did we detect increased TNF
protein levels in the conditioned media of villous explants from
preeclamptic placentas; in fact, significantly lower levels were
detected in media from samples incubated under standard tissue culture
conditions. Further, our evaluation of TNF mRNA using
semiquantitative RT-PCR did not demonstrate any significant differences
in TNF mRNA between normal term and preeclamptic placentas, although
differences have been reported by others (15, 16). Clear
explanations for the different results obtained in our study and those
of others are not readily apparent, although technical considerations
such as placental sampling, mode of delivery (15), as well
as differences in ELISAs and approaches to semiquantitative RT-PCR are
possibilities. Of note is that in our study TNF message was
detectable in all normal term placentas, whereas in other studies
TNF mRNA was negligible (15, 16). Furthermore, other
studies have shown the normal placenta to be a source of TNF
(6, 17). Finally, in a study by Opsjon and colleagues of
amniotic fluid cytokine levels, it was mentioned (data not shown) that
there were no differences in protein levels of TNF, IL-1ß, or IL-6
in preeclamptic vs. normal placentas (14).
We have previously shown that incubation of human placental explants
under reduced oxygen conditions results in the elevated production of
TNF, IL-1, and IL-1ß (9), and we have postulated
that focal ischemia/hypoxia due to insufficient changes in the uterine
vasculature may elevate placental inflammatory cytokine production in
preeclampsia. Although results from the current study via direct
measurement of placental cytokines and uterine venous sampling do not
support our original hypothesis, they do not discount the central dogma
that reduced placental perfusion may result in the production of
cytotoxic factors by the placenta. In fact, our prior work
(9) and current studies on placental expression of
hypoxia-inducible transcription factors (18) provide
evidence that oxygen-sensing pathways are present in the human
placenta.
We also assessed TNF mRNA and protein levels in preterm placentas
from women without a diagnosis of preeclampsia to match gestational age
to some of the preeclamptic women. Importantly, TNF protein levels
did not differ among preterm, preeclamptic, or normal term groups,
although TNF mRNA was less in the preterm group compared with the
normal term placenta.
We investigated IL-1 and IL-1ß placental protein levels, because
the production of these cytokines can be stimulated in villous explants
by hypoxia (Ref. 9 and the current study), and placental
IL-1ß mRNA was reported to be increased in women with preeclampsia
(16). We failed to detect any significant differences in
protein levels of these cytokines in placentas obtained from normal
term and preeclamptic women, nor were concentrations in the conditioned
media from villous explants cultured under 21% oxygen any different
between these patient groups. Moreover, circulating IL-1ß is not
significantly increased in preeclampsia (11). Because the
status of placental IL-12 and its role in preeclampsia are uncertain
(19, 20), we also evaluated IL-12 levels using an ELISA
that detects both the p35 and p40 subunits of IL-12 in a limited number
of samples. Again, our results suggested no differences.
Protein levels of IL-6 were comparable in placentas delivered at term
or preterm with or without preeclampsia. At the mRNA level, others have
also reported no differences (16, 21). Amniotic fluid
levels of IL-6 are reportedly lower in nonlabored, preeclamptic
pregnancies (14) and those with small for gestational age
fetuses (22). We and others (the current study and Ref.
21) found that IL-6 production by villous placental tissue
from preeclamptic women maintained in culture was decreased compared
with that by villous explants from normal term placenta. Although IL-6
levels are increased in the circulation from preeclamptic women
(11, 23), the source of this IL-6 may be activated
leukocytes or the maternal endothelium, which can produce IL-6 upon
exposure to stimuli such as TNF (5).
As previously mentioned, several investigators have identified
elevations in circulating levels of TNF in preeclampsia (reviewed in
Refs. 10 and 11) and in women early in
pregnancy who later developed preeclampsia (24). Serum
concentrations of the TNF-p55 soluble receptor are also increased
before clinical manifestations (25). As an alternative
approach to investigate the placenta as a source of circulating TNF,
we obtained uterine venous and peripheral blood samples from normal
term and preeclamptic women at cesarian section. TNF levels were
elevated in peripheral circulation in preeclamptic women as previously
reported (11). However, the ratio of uterine/peripheral
venous TNF was not significantly different from 1.0 for both groups
of patients. Because the half-life of TNF in the circulation is
relatively short (<30 min) (26), these results suggest
that the placenta does not contribute significantly to the elevated
circulating levels in the disease. Other sources for elevated
circulating cytokines might include leukocytes, which are at a
heightened activation state during preeclampsia (27) and
express greater TNF mRNA in women with preeclampsia
(28).
In conclusion, there were no significant alterations in protein or mRNA
expression of TNF in placentas from women with preeclampsia, nor was
there a gradient for TNF across the uteroplacental unit in
vivo during the disease. Placental levels of other inflammatory
cytokines (IL-1, IL-1ß, and IL-6) were not altered in
preeclampsia. Based on these results, we suggest that tissues other
than the placenta, such as activated leukocytes or the endothelium, may
contribute to the elevated concentration of inflammatory cytokines
found in the circulation of preeclamptic women.
Acknowledgments
We gratefully acknowledge Theresa Miles for technical assistance with tissue procurement and cytokine ELISAs, Sue Kauffman for the preparation of graphs, Cynthia Schatzmann for patient recruitment, and Drs. Phillip Heine and Deb Draper for providing preterm placental samples.
Footnotes
1 This work was supported by NIH Grants P01-HD-30367 and
K04-HD-01098. 
Received November 2, 2000.
Revised January 19, 2001.
Accepted February 2, 2001.
References
mRNA and protein are
present in human placental and uterine cells at early and late stages
of gestation. Am J Pathol. 139:327–335.[Abstract]
and interleukin-1ß proteins in human
placentas. J Reprod Immunol. 22:257–268.[CrossRef][Medline]
concentrations and
mRNA expression are increased in preeclamptic placentas. J Reprod
Immunol. 32:157–169.[CrossRef][Medline]
, interleukin 1ß, and interleukin
10 is increased in preeclampsia. Am J Obstet Gynecol. 181:915–920.[CrossRef][Medline]
levels before the clinical
manifestations of preeclampsia. Am J Reprod Immunol. 38:89–93.
-soluble
receptor p55 (sTNFp55) and subsequent risk of preeclampsia. Am J
Epidemiol. 149:323–329.: preclinical studies and results from early clinical trials. Immunol Ser. 56:567–587.[Medline]
(TNF-) gene polymorphism
and expression in pre-eclampsia. Clin Exp Immunol. 104:154–159.[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  D. D. Briana and A. Malamitsi-Puchner Reviews: Adipocytokines in Normal and Complicated Pregnancies Reproductive Sciences, October 1, 2009; 16(10): 921 - 937. [Abstract] [PDF]
|
|
|
|
|
|
I. M. Bernstein, D. Damron, A. L. Schonberg, and R. Shapiro The Relationship of Plasma Volume, Sympathetic Tone, and Proinflammatory Cytokines in Young Healthy Nonpregnant Women Reproductive Sciences, October 1, 2009; 16(10): 980 - 985. [Abstract] [PDF]
|
|
|
|
 |
|
 C. T. Ndao, A. Dumont, N. Fievet, S. Doucoure, A. Gaye, and J. Y. Lehesran Placental Malarial Infection as a Risk Factor for Hypertensive Disorders During Pregnancy in Africa: A Case-Control Study in an Urban Area of Senegal, West Africa Am. J. Epidemiol., October 1, 2009; 170(7): 847 - 853. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kotani, A. Iwase, K. Ino, S. Sumigama, E. Yamamoto, H. Hayakawa, T. Nagasaka, A. Itakura, S. Nomura, and F. Kikkawa Activator Protein-2 Impairs the Invasion of a Human Extravillous Trophoblast Cell Line Endocrinology, September 1, 2009; 150(9): 4376 - 4385. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Lockwood, C.-F. Yen, M. Basar, U. A. Kayisli, M. Martel, I. Buhimschi, C. Buhimschi, S. J. Huang, G. Krikun, and F. Schatz Preeclampsia-Related Inflammatory Cytokines Regulate Interleukin-6 Expression in Human Decidual Cells Am. J. Pathol., June 1, 2008; 172(6): 1571 - 1579. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Rajakumar, A. Jeyabalan, N. Markovic, R. Ness, C. Gilmour, and K. P. Conrad Placental HIF-1{alpha}, HIF-2{alpha}, membrane and soluble VEGF receptor-1 proteins are not increased in normotensive pregnancies complicated by late-onset intrauterine growth restriction Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2007; 293(2): R766 - R774. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Tsukimori, H. Nakano, and N. Wake Difference in Neutrophil Superoxide Generation During Pregnancy Between Preeclampsia and Essential Hypertension Hypertension, June 1, 2007; 49(6): 1436 - 1441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Paternoster, S. Fantinato, A. Stella, K. N. Nanhorngue, M. Milani, M. Plebani, U. Nicolini, and A. Girolami C-Reactive Protein in Hypertensive Disorders in Pregnancy Clinical and Applied Thrombosis/Hemostasis, July 1, 2006; 12(3): 330 - 337. [Abstract] [PDF]
|
|
|
|
|
|
J. M. Roberts and H. S. Gammill Preeclampsia: Recent Insights Hypertension, December 1, 2005; 46(6): 1243 - 1249. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Lam, K.-H. Lim, and S. A. Karumanchi Circulating Angiogenic Factors in the Pathogenesis and Prediction of Preeclampsia Hypertension, November 1, 2005; 46(5): 1077 - 1085. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Soleymanlou, I. Jurisica, O. Nevo, F. Ietta, X. Zhang, S. Zamudio, M. Post, and I. Caniggia Molecular Evidence of Placental Hypoxia in Preeclampsia J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 4299 - 4308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Freeman, F. McManus, E. A. Brown, L. Cherry, J. Norrie, J. E. Ramsay, P. Clark, I. D. Walker, N. Sattar, and I. A. Greer Short- and Long-Term Changes in Plasma Inflammatory Markers Associated With Preeclampsia Hypertension, November 1, 2004; 44(5): 708 - 714. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Davison, V. Homuth, A. Jeyabalan, K. P. Conrad, S. A. Karumanchi, S. Quaggin, R. Dechend, and F. C. Luft New Aspects in the Pathophysiology of Preeclampsia J. Am. Soc. Nephrol., September 1, 2004; 15(9): 2440 - 2448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Novak, A. Rajakumar, T. M. Miles, and K. P. Conrad Nitric Oxide Synthase Isoforms in the Rat Kidney During Pregnancy Reproductive Sciences, July 1, 2004; 11(5): 280 - 288. [Abstract] [PDF]
|
|
|
|
|
|
T.-H. Hung, D. S. Charnock-Jones, J. N. Skepper, and G. J. Burton Secretion of Tumor Necrosis Factor-{alpha} from Human Placental Tissues Induced by Hypoxia-Reoxygenation Causes Endothelial Cell Activation in Vitro: A Potential Mediator of the Inflammatory Response in Preeclampsia Am. J. Pathol., March 1, 2004; 164(3): 1049 - 1061. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Thadhani, W. P. Mutter, M. Wolf, R. J. Levine, R. N. Taylor, V. P. Sukhatme, J. Ecker, and S. A. Karumanchi First Trimester Placental Growth Factor and Soluble Fms-Like Tyrosine Kinase 1 and Risk for Preeclampsia J. Clin. Endocrinol. Metab., February 1, 2004; 89(2): 770 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Orshal and R. A. Khalil Reduced Endothelial NO-cGMP-Mediated Vascular Relaxation and Hypertension in IL-6-Infused Pregnant Rats Hypertension, February 1, 2004; 43(2): 434 - 444. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Khalil and J. P. Granger Vascular mechanisms of increased arterial pressure in preeclampsia: lessons from animal models Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2002; 283(1): R29 - R45. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ito, A. Itakura, Y. Ohno, M. Nomura, T. Senga, T. Nagasaka, and S. Mizutani Possible Activation of the Renin-Angiotensin System in the Feto-Placental Unit in Preeclampsia J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1871 - 1878. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
,
TNF
; n = 5) or hypoxia (2% oxygen; 
; n =
5). The cytokines evaluated were TNF
, significant difference between incubations of
normal term and preeclamptic villous explants (P <
0.05).