Cardiovascular Research 58 (2003) 162–169
www.elsevier.com / locate / cardiores
Microvascular permeability is related to circulating levels of tumour
necrosis factor-a in pre-eclampsia
Nick Anim-Nyame a , *, John Gamble b , Suren R. Sooranna a , Mark R. Johnson a ,
Philip J. Steer a
a
Department of Maternal and Fetal Medicine, Faculty of Medicine, Imperial College of Science, Technology and Medicine,
Chelsea and Westminster Hospital, 369 Fulham Road, London SW10 9 NH, UK
b
School of Sport and Exercise Sciences, University of Birmingham, Birmingham B15 2 TT, UK
Received 9 August 2002; accepted 5 December 2002
Abstract
Introduction: The mechanism for the increased microvascular permeability which, underline many of the complications of
pre-eclampsia, remain unexplained. It has been suggested that a factor present in the maternal circulation in pregnancies complicated by
the disease may be responsible for increased microvascular permeability. In this study, we have investigated the relationship between
filtration capacity (Kf ), an index of microvascular permeability, and maternal levels of VEGF, leptin and TNF-a, all of which are known
permeability factors whose plasma levels are increased in pre-eclampsia. Methods: We used a small cumulative pressure step venous
congestion plethysmography protocol to compare Kf , an index of microvascular permeability, during the third trimester of 20 women with
pre-eclampsia, 18 normal pregnant women and 18 non-pregnant female matched controls. Blood samples were obtained to measure
plasma levels of VEGF, leptin, TNF-a plasma protein concentrations and full blood count. Results: Microvascular filtration capacity (Kf )
was significantly increased in pre-eclampsia compared to the other groups (P<0.0001, ANOVA). Kf was also increased in the normal
pregnant group when compared to the non-pregnant controls (P50.02). Plasma levels of VEGF, leptin and TNF-a were significantly
greater in pre-eclampsia compared to normal pregnancy and non-pregnant controls (P,0.0001, ANOVA, for all three analyses). Total
plasma protein and albumin concentrations were significantly lower in the normal pregnant and pre-eclamptic groups, compared to the
non-pregnant controls (P,0.0001, ANOVA). Kf was significantly related to TNF-a in pre-eclampsia (r50.53, P50.018), and with VEGF
in the non-pregnant controls (r50.6, P50.02). No significant relationship was observed between Kf and VEGF, leptin and TNF-a during
normal pregnancy. There was a significant inverse correlation between plasma albumin concentration and filtration capacity in the normal
pregnant (r520.94, P,0.0001) and non-pregnant (r520.87, P,0.0001) groups but not in the women with pre-eclampsia (r520.18,
P50.8). Conclusions: These data show that that microvascular filtration capacity is significantly increased in pre-eclampsia, and
correlates with circulating levels of TNF-a but not leptin or VEGF.
 2003 European Society of Cardiology. Published by Elsevier Science B.V. All rights reserved.
Keywords: Growth factors; Microcirculation
1. Introduction
Pre-eclampsia is a multisystem disorder of the second
half of pregnancy, characterized by generalized endothelial
cell dysfunction [1]. It appears the release of pro-inflammatory cytokines such as TNF-a [2] and reactive
oxygen species [3] from the ischaemic placenta, which
*Corresponding author. Tel.: 144-208-846-7895; fax: 144-208-8467796.
E-mail address: [email protected] (N. Anim-Nyame).
results from abnormal placentation [4] in pre-eclampsia,
contributes to the endothelial dysfunction. Cytokines may
also contribute directly to oxidative stress induced endothelial dysfunction [5] or indirectly through neutrophil
activation [6]. Reduced maternal plasma volume is a
feature of pre-eclampsia [7] and appears to be secondary to
increased microvascular permeability [8]. The mechanism(s) for the increased microvascular permeability,
however, remains unexplained. Of the Starling forces,
Time for primary review 26 days.
0008-6363 / 03 / $ – see front matter  2003 European Society of Cardiology. Published by Elsevier Science B.V. All rights reserved.
doi:10.1016 / S0008-6363(02)00844-1
N. Anim-Nyame et al. / Cardiovascular Research 58 (2003) 162–169
which influence microvascular fluid exchange, changes in
oncotic [9] and hydrostatic pressure gradients [10] do not
appear to adequately to explain the increased fluid flux
observed in pregnancies complicated by the disease.
There is evidence that sera from women with preeclampsia rather that normal pregnancy increases the
permeability of human umbilical vein endothelial cell
(HUVEC) monolayers [11]. This suggests that there may
be a circulating factor(s) present in pre-eclampsia that may
be responsible for the increased microvascular permeability in pregnancies complicated by the disease. Plasma
levels of vascular endothelial growth factor (VEGF) [12],
leptin [13] and tumour necrosis factor-a (TNF-a) [14] are
elevated in pre-eclampsia. These three substances have
been shown to cause a marked increase in microvascular
permeability in a variety of animal model studies. The role
of VEGF in microvascular permeability in the non-pregnant state is well established [15]. Recent evidence suggests that leptin receptors are expressed by vascular
endothelial cells [16]. Leptin induces angiogenesis and
may increase microvascular permeability [17]. Structurally,
leptin resembles the class I cytokines [18] and may
regulate placental cytokine production during pregnancy
[19]. The increased circulating levels of leptin during
pregnancy appear to be of placental origin [20], although
its functions in the placenta are not known, in addition to
its angiogenic effect, it may increase the exchange of small
molecules between maternal circulation and the fetus by
the induction and maintenance of vascular permeability
[21]. TNF-a is associated with increased capillary leak (for
example in sepsis) and has been shown to increase venular
fluid filtration coefficient (Lp ) [22] in the non-pregnant
state. In this study, we have investigated whether Kf
correlates with plasma levels of VEGF, leptin and TNF-a,
in pre-eclampsia.
2. Methods
2.1. Subjects
We used a small cumulative pressure step venous
congestion plethysmography protocol to compare filtration
capacity (Kf ) during the third trimester of normal and
pre-eclamptic pregnancies. Filtration capacity was measured in the calves of 20 women with pre-eclampsia, 18
normal pregnant women, and 18 non-pregnant controls. All
the women were of similar in age and body mass index
(BMI), and the pregnant women matched for gestational
age. The pregnant women were all nulliparous white
Europeans with spontaneous pregnancies and were recruited from the antenatal clinics or antenatal ward at the
Chelsea and Westminster Hospital, London. The non-pregnant controls were health workers. The normal pregnant
controls were women with no history of medical illnesses,
attending the routine antenatal clinics and who were
163
invited to take part in the study. They were chosen to be
similar to the pre-eclamptic group with regard to the
latter’s booking body mass index (BMI) and, in the
pregnant controls, gestational age. All the women were
non-smokers and were not on any medication. None of the
non-pregnant controls were taking oral contraceptives.
None of the subjects received any intravenous infusion
before or during the study. Women with previous or
present history of peripheral vascular disease, peripheral
neuropathy, chronic hypertension, infective or inflammatory disorders or any other underlying medical disorders
were excluded from the study.
Pre-eclampsia was defined according to the criteria of
hypertension and proteinuria occurring for the first time
after 20 weeks gestation, and reversal of both after
delivery. Hypertension was defined as an absolute blood
pressure greater than 140 mmHg systolic or 90 mmHg
diastolic taken twice, 6 h apart. The first and fifth
Korotkoff sounds were used to determine the systolic and
diastolic components, respectively. Proteinuria was defined
as more than 0.5 g urinary protein excretion in a 24-h urine
sample [23]. The urine specimens were collected into
plastic jugs containing phenyl mercuric acetate as a
preservative. Protein was measured by colorimetric reactions using an autoanalyser, as described by Watanabe et
al. [24]. The obstetric records of all pregnant groups were
reviewed after delivery to confirm reversal of hypertension
and proteinuria. The investigation conformed to the principles outlined in the Declaration of Helsinki. The Local
Ethics Committee approved the study and informed consent was obtained from each patient.
2.2. Study protocol
Studies were performed in a quiet room at a steady
temperature (23–24 8C). Subjects rested for at least 15 min
before the study. Observations were made in the left lateral
position, to prevent aorto-caval compression and with the
right mid-calf supported at the level of the heart. In
addition to the blood pressure assessment at diagnosis,
arterial blood pressure was measured, non-invasively, on
the ipsi-lateral calf and arm, using a Dinamap Vital Sign
Monitor (Type 1800, Critikon, Tampa, FL, USA). Averaged values of systolic, diastolic and mean arterial blood
pressures were calculated from triplicate measurements.
Filtration capacity (Kf ), an index of vascular permeability,
was estimated using the Filtrass strain gauge plethysmograph (Filtrass, DOMED, Munich, Germany) [25]. The
device, a modification of a standard strain gauge
plethysmograph, has been fully described previously [26].
Briefly, the congestion pressure cuff, which is attached to a
compressor pump built into the apparatus, was placed
around the right thigh and enclosed in a rigid corset to
reduce filling volume and thus filling time. Changes in calf
circumference in response to a rapid increase in cuff
pressure ( pcuff ), were measured using a passive inductive
164
N. Anim-Nyame et al. / Cardiovascular Research 58 (2003) 162–169
transducer with an accuracy of 65 mm. The files were
recorded and saved for subsequent ‘off-line’ analysis.
2.3. Assessment of filtration capacity
Microvascular filtration capacity (Kf ) was measured
using an established small cumulative pressure step strain
gauge plethysmography protocol [25,26]. Briefly, a series
of five to seven small (8 mmHg) cumulative pressure steps
were applied to the venous congestion cuff and the
resulting changes in tissue volume, derived from alterations in calf circumference (Fig. 1a), were recorded. Using
this protocol, no change in the recorded signal is observed
until the ambient venous pressure in the limb is exceeded.
At congestion cuff pressures greater than this value, each
additional pressure increment causes a change in limb
volume that is attributed to vascular filling (Va ) (Fig. 1a).
The volume change representing Va can be fitted by
exponential analysis and it takes 30–45 s to reach a
steady-state [26]. When congestion cuff pressure exceeds
the isovolumetric venous pressure ( pvi ); the equilibrium
pressure at the microvascular interface reflecting the
transmicrovascular balance of Starling forces, a steadystate change in volume is observed, reflecting fluid filtration (Jv ) (Fig. 1a).
Interpretation of these data is, perhaps, made easier by
referring to the Starling equation itself. Movement of fluid
and plasma proteins between the vascular and interstitial
compartments is governed by the Starling forces, which
are described in Eq. (1):
Jv 5 Kf ([ pc 2 pt ] 2 a [pc 2 pt ])
(1)
where Kf is the fluid filtration capacity; pc and pt are the
hydrostatic pressures in the capillaries and the tissue,
respectively; a is the osmotic reflection coefficient, an
index of the vascular permeability to plasma proteins; and
tc and tt are plasma and tissue oncotic pressures, respectively. The filtration capacity Kf reflects the product of the
area available for fluid filtration and the permeability per
unit surface area. The isovolumetric venous pressure ( pvi )
is an index of the equilibrium pressure at the microvascular
interface, reflecting the local plasma oncotic pressure ap
(Fig. 1b).
Computer-based analysis enables differentiation between
volume and filtration responses [27]. The value of Kf is
determined by linear regression, based on the pcuff and Jv ,
co-ordinates obtained at pressures above those causing
measurable increase in Jv (see Fig. 1b). The slope of this
relationship is Kf (Fig. 1b) and the units are expressed as
KfU (ml min 21 100 ml 21 mmHg 21 310 23 ) [26]. The
intercept of the line on the abscissa, that is where Jv ,
represents pvi (Fig. 1a), the pressure that has to be
exceeded to induce net filtration at the level of the strain
gauge.
Fig. 1. (a) Calf volume response (top panel) to a step increase in
congestion cuff pressure ( pcuff ). The upper trace of the top panel reflects
the whole volume response and the superimposed dotted line, the
regression slope (Jv ). The lower trace of the top panel shows the volume
response after the regression slope (Jv ) has been subtracted. The volume
(Va ) reflects the compliance volume change in response to this pressure
step. (b). The relationship between Jv and pcuff (top panel) and the
vascular compliance (lower panel) of the subject in (a).
2.4. Sample collection and analysis
The venous blood samples used to assay for circulating
levels of total VEGF, leptin, TNF-a, plasma albumin, total
protein, uric acid and creatinine concentrations and also
full blood count, were obtained from the antecubital vein.
N. Anim-Nyame et al. / Cardiovascular Research 58 (2003) 162–169
Care was taken in sample handling to avoid platelet
activation, since this is known to increase VEGF levels in
vitro. The heparinized samples were centrifuged for 10 min
at 3000 rpm and 4 8C, the plasma separated within an hour
of venepuncture and stored at 270 8C until assayed. Total
immunoreactive VEGF was assayed by quantitative sandwich enzyme immunoassay technique, using the Quantikine human VEGF commercial kit (R&D Systems Europe
limited, Abingdon, Oxon, UK). The assays were done in
duplicate, using an enzyme-linked polyclonal antibody
specific to VEGF. Total VEGF (pg / ml) was measured
because of the effects of pregnancy on its free form.
Factors produced by the placenta results in underestimation
of the free forms of VEGF [28]. Total circulating leptin
concentrations (ng / ml) were measured in duplicate by
radioimmunoassay as described previously [29]. The assay
employed a polyclonal (rabbit) antibody raised against
recombinant human leptin. Standards and 125 I-tracer were
also made from recombinant human leptin [29]. The
average intra- and inter-assay coefficients of variation were
3 and 8%, respectively. TNF-a was assayed using a solid
phase sandwich commercial ELISA kit (Diaclone Research, France) with inter- and intra-assay coefficients of
variation of 1.7 and 6.0%, respectively. Full blood count,
plasma albumin, total protein, uric acid and creatinine
concentrations were also measured, using an autoanalyser.
2.5. Statistical analysis
All the data were normally distributed data and are
presented as mean6S.D. Statistical differences between the
groups were compared using analysis of variance with
Bonferroni correction for multiple comparisons. The relationships between Kf and the clinical parameters, VEGF,
leptin and TNF-a concentrations were determined using
Pearson correlation coefficients. Multiple regression analysis was performed to determine which of the parameters
were independently related to Kf and to assess associations
between Kf and measured clinical and biochemical variables. Statistical significance was assumed at a P value less
165
than 0.05. The Statistical Package for Social Sciences
(SPSS, version 10) was used for these analyses.
3. Results
The clinical and demographic characteristics for the
three groups are shown in Table 1. There were no
significant differences in age or BMI among the three
groups, or in gestational age between the pregnant groups.
Babies born to women with pre-eclampsia were smaller
compared to the normal pregnant controls (P50.02). As
expected from the recruitment criteria, the women with
pre-eclampsia had higher systolic and diastolic blood
pressures (P50.001), higher serum uric acid levels (P5
0.001) and lower platelet counts (P50.002) than the
pregnant control group.
Microvascular filtration capacity was significantly increased in pre-eclampsia (6.3260.3 KfU), compared to the
normal pregnant and non-pregnant controls (4.3560.18
and 3.1360.16 KfU, respectively, P<0.0001, ANOVA
with Bonferroni correction). Kf was also increased in the
normal pregnant group when compared to the non-pregnant controls (P50.02) (Fig. 2). Plasma levels of VEGF,
leptin and TNF-a were significantly greater in pre-eclampsia (149.7967.9 pg / ml, 29.5963.04 mg / l and
191.02612.3763.89 mg / l, respectively) compared to normal pregnancy (53.0364.9 pg / ml, 12.6260.97 mg / l and
59.7561.59 mg / l, respectively) and non-pregnant controls
(38.565.65 pg / ml, 10.05 mg / l and 34.8964.0 mg / l,
respectively) (P,0.0001, ANOVA, for all three factors)
(Fig. 3a–c). Total plasma protein and albumin concentrations were significantly lower in the normal pregnant
and pre-eclamptic groups, compared to the non-pregnant
controls (P,0.0001, ANOVA) (Table 1). Although plasma
protein levels were lower in pre-eclampsia than normal
pregnancy, the differences were not statistically significant
(P50.05) (Fig. 3d).
Kf was significantly related to TNF-a in pre-eclampsia
(Fig. 4b, r50.53, P50.018), and with VEGF in the
non-pregnant controls (Fig. 3a, r50.6, P50.02). No
Table 1
Clinical and demographic characteristics of the three groups
Variables
Non-pregnant
(n518)
Normal pregnant
(n518)
Pre-eclampsia
(n520)
P value
Age (years)
Gestation (weeks)
BMI (kg / m 2 )
SBP (mmHg)
DBP (mmHg)
Birth weight (kg)
Plasma uric acid
(mmol / l)
Plasma albumin (g / l)
Total protein (g / l)
Platelets (310 6 )
29.964.2
N /A
22.762.6
11366.0
6766.2
N /A
29.065.1
35.661.6
23.463.3
119.4610.9
72.366.9
3.4460.15
29.664.4
35.862.0
24.664.0
141.3613.3
93.064.7
2.9560.12
NS
NS
NS
0.001
0.001
0.02
0.2460.04
38.861.1
74.360.9
257.9621.5
0.2260.05
26.262.6
69.864.1
230.9642.5
0.3960.03
24.061.6
64.1963.4
138.8632.7
0.001
NS
NS
0.002
166
N. Anim-Nyame et al. / Cardiovascular Research 58 (2003) 162–169
Fig. 2. Dot plots comparing microvascular filtration capacity in women
with pre-eclampsia (n520), normal pregnant controls (n518) and nonpregnant control women (n518). The dots represent Kf values for women
in each group and the bars mean values.
Fig. 4. This graph demonstrates the relationship between birth weight
(kg) and microvascular filtration capacity (KfU) in pre-eclampsia (d),
normal pregnancy (s).
Fig. 3. Graphs demonstrating the relationship between microvascular filtration capacity and circulating levels of (a) TNF-a (mg / l), (b) VEGF (pg / l), (c)
leptin (pg / l), and (d) plasma albumin (g / l) in pre-eclampsia (d), normal pregnancy (s), and non-pregnant controls (m).
N. Anim-Nyame et al. / Cardiovascular Research 58 (2003) 162–169
significant relationship was observed between Kf and
VEGF, leptin and TNF-a during normal pregnancy. There
were highly significant inverse correlations between plasma albumin concentration and filtration capacity in the
normal pregnant (r5 20.94, P,0.0001) and non-pregnant
(r5 20.87, P,0.0001) groups, but not in the women with
pre-eclampsia (r5 20.18, P50.8) (Fig. 3d). A significant
inverse relationship was also observed between centile
birth weight and filtration capacity in pre-eclampsia (r5 2
0.67, P50.001) but not in normal pregnancy (r50.07,
P50.77) (Fig. 4). Plasma uric acid concentrations correlated significantly with Kf , and plasma levels of VEGF,
leptin and TNF-a in pre-eclampsia in pre-eclampsia (r5
0.64, 0.68 and 0.71; respectively, P50.001).
4. Discussion
In this study we compared microvascular permeability in
normal pregnancy, pregnancies complicated by pre-eclampsia, and in the non-pregnant state. Whereas microvascular
permeability increased during normal pregnancy and preeclampsia, the increase was significantly greater in preeclampsia. Furthermore, we observed that changes in
microvascular permeability in pre-eclampsia correlated
significantly with circulating levels of TNF-a but not with
either VEGF or leptin. No significant relationship was
observed between permeability and these permeabilityinducing factors in the normal pregnant controls. This is
the first study to compare changes in microvascular
permeability with those of specific circulating permeability
factors in pregnancy. However, while suggestive, these
data do not prove a causal relationship between microvascular permeability and TNF-a in pre-eclampsia.
Although Haller et al. [11], reported that a circulating
factor(s) present in maternal circulation in pre-eclampsia
increased endothelial cell permeability, no specific permeability factors were identified. The relationship between
microvascular permeability and circulating TNF-a observed in the present study is consistent with the hypothesis that pre-eclampsia may be an exaggeration of inflammatory responses common to all pregnancies [30]. It is
also consistent with the role of TNF-a in inducing
capillary leak by increased fluid filtration coefficient (Lp )
[22] in the non-pregnant state, particularly in sepsis. The
placenta expresses TNF-a receptors, the expression increases with gestational age [31], and is known to be up
regulated in pre-eclampsia [32]. It appears that the increased circulating levels, during pregnancy, may be in
response to trophoblast–decidual cell interactions, and may
thus play a role in trophoblast differentiation and invasion
of decidua and spiral arteries [33,34].
Although, this study did not investigate the mechanisms
by which TNF-a may increase microvascular permeability
in pre-eclampsia several are possible. TNF-a has been
shown to up-regulate the expression of the cell adhesion
167
molecules VCAM-1 and ICAM-1 on the endothelial cells
[35]. Furthermore, circulating levels of TNF-a correlate
with VCAM-1 expression [36]. Thus, the increase in
VCAM-1 and ICAM-1 expression may increase
leucocyte–endothelial cell interactions and, in addition,
increase microvascular permeability. TNF-a may also
increase endothelial permeability by inducing oxidative
stress in pre-eclampsia [37]. This is because TNF-a has
been shown to interfere with mitochondria electron flow
with resultant release of oxidizing free radicals, leading to
lipid peroxidation [38].
VEGF is known to be one of the most potent microvascular permeability [15] inducing agents and circulating levels are elevated in pre-eclampsia [12]. Thus, the
lack of significant correlation between the level of VEGF
and microvascular permeability is an interesting observation. Hypoxia is known to be a very potent stimulus for
upregulation of VEGF production [39] and we have
previously shown that tissue blood flow is reduced in
pre-eclampsia [40] and precedes the clinical onset of the
disease [41]. Although, the increase in VEGF might have
been related to the reduced tissue blood flow, we did not
observe any correlation between circulating levels of
VEGF and Kf . There are other possible explanations for
the apparent lack of correlation between VEGF and Kf in
the pre-eclamptic group. In animal studies, Bates and
co-workers [15,42], investigated differences in vascular
permeability following acute and chronic exposure to
VEGF. They observed that, whereas the greatest permeability increases followed acute exposure to VEGF, the
effects were transient. By contrast, chronic exposure
resulted in more sustained increases in permeability. Preeclampsia presents a chronic exposure state, and may
therefore explain the findings in this study. It is also
possible that the mechanism(s) by which VEGF exerts its
action had been maximally up regulated. If this is a correct
interpretation of these data, it must be assumed that the
further increases in permeability in response to TNF-a,
were operating via another pathway. Beynon et al. [43]
have demonstrated that cytokines have synergistic effects
on permeability of endothelial cell monolayers. Therefore
the diverse array of agonists or growth factors released in a
complex pathophysiological state such as pre-eclampsia,
may give rise to responses via synergistic mechanisms. A
significant correlation was observed between VEGF and Kf
in the non-pregnant controls. Whilst the plasma VEGF
levels were much lower, it is possible that we were
observing the effect of VEGF on permeability, without the
synergistic action of the other factors normally raised in
pre-eclampsia. Haller et al. [11] demonstrated that the
circulating factor(s) present in maternal circulation in preeclampsia which induced increased microvascular permeability was mediated by the protein kinase C (PKC)
pathway. In their study, incubation of pre-eclamptic sera
induced translocation of a and ´ isoforms of PKC within
endothelial cells. This is consistent with recent evidence
168
N. Anim-Nyame et al. / Cardiovascular Research 58 (2003) 162–169
that VEGF increased microvascular permeability in vivo
by phospholipase C (PLC) stimulation but not by the PKC
pathway [44], observations that are in line with our own
results.
Thus it seems that VEGF is unlikely to be the only
circulating factor responsible for the increased microvascular permeability in pre-eclampsia. It is also possible that
other factors present in maternal circulation in pre-eclamptics inhibit the vascular permeability effects of VEGF.
Angiopoietins, for instance, are vascular endothelial cellspecific growth factors that play an important role principally during the last stages of angiogenesis after the
induction of new capillaries by VEGF. There is evidence
that over-expression of angiopoietin-1 (Ang1) results in
non-leaky vessels by inhibiting the effects of VEGF [45].
Whereas combined overexpression of VEGF and Ang1 had
an additive effect on new vessel formation, this combination results in leak resistant-vessels typical of Ang-1
[46,47]. Furthermore, Ang1 suppresses VEGF-induced
expression of the cell adhesion molecules VCAM-1,
ICAM-1 and E-selectin, thus inhibiting endothelial adhesiveness [48]. It is therefore possible that the effect of
increased VEGF on vascular permeability is suppressed by
the presence of other circulating factor(s), such as Ang1, in
pregnancies complicated by pre-eclampsia. Further work is
required to investigate circulating levels of angiopoietins in
normal pregnancy and pre-eclampsia and their relationships with VEGF.
Although leptin has both angiogenic and vascular permeability effects [21], circulating levels did not correlate
with Kf in all three groups of women in this study. Since
the increased level of leptin during pregnancy is predominantly placental in origin, it is possible it acts locally
to increase placental exchange between the mother and
fetus. Such effects would be consistent with previous
evidence that placental expression of leptin mRNA is
upregulated during hypoxic conditions including pre-eclampsia [49] and diabetes [50]. The lack of correlation
between Kf and leptin could also be explained by synergism between the different circulating permeability factors
in all three groups of women. Another observation worth
further attention is the striking correlation between plasma
albumin concentration and permeability in the pregnant
and non-pregnant controls. Since albumin is one of the
major contributors to plasma oncotic pressure, the major
expected effect of changes in concentration would be in
terms of the value of sp, not Kf (Fig. 1b). That the
pre-eclamptics do not also show this, despite their high
permeability values, probably reflects the intervention of
other causative agents, as discussed above. Increased
microvascular permeability is associated with reduced
plasma volume, which is a characteristic feature of the
disease [7]. Thus, the inverse correlation between filtration
capacity and birth weight is consistent with previous
observations of inverse relationship between plasma volume and birth weight [51].
In summary, we have demonstrated that the increased
microvascular permeability in pre-eclampsia significantly
correlates with increased circulating levels of TNF-a but
not VEGF and leptin. However, the data do not prove a
causal relationship between Kf and TNF-a.
Acknowledgements
This work was supported by a grant from The Henry
Smith Charity.
References
[1] Roberts JM, Taylor RN, Musci TJ et al. Pre-eclampsia: an endothelial cell disorder. Am J Obstet Gynecol 1989;16:1200–1204.
[2] Conrad KP, Benyo DF. Placenta cytokine and the pathogenesis of
pre-eclampsia. Am J Reprod Immunol 1997;37:240–249.
[3] Walsh SW, Wang Y. Secretion of lipid peroxides by the human
placenta. Am J Obstet Gynecol 1993;169:1462–1466.
[4] Bosens I, Dixon HG, Robertson WB. The role of the spiral arteries
in the pathogenesis of pre-eclampsia. J Pathol Bacteriol
1967;93:581–592.
[5] Taylor RN. Immunobiology of preeclampsia. Am J Reprod Immunol
1997;37:79–87.
[6] Greer IA, Haddad NG, Dawes J, Johnstone FD, Calder AA.
Neutrophil activation in pregnancy-induced hypertension. Br J
Obstet Gynaecol 1989;96:978–982.
[7] Silver HM, Seebeck M, Carlson R. Comparison of total blood
volume in normal, pre-eclamptic and nonproteinuric gestational
hypertensive pregnancy by simultaneous measurement of red blood
cell and plasma volume. Am J Obstet Gynecol 1998;179(1):87–93.
[8] Brown MA, Zammit VC, Lowe SA. Capillary permeability and
extra cellular fluid volumes in pregnancy-induced hypertension. Clin
Sci 1989;77(6):599–604.
[9] Valenzuela GJ. Is decreased plasma oncotic pressure enough to
explain the oedema of pregnancy? Am J Obstet Gynecol
1989;161(1):1624–1627.
[10] Oian P, Matau JM. Calculated capillary hydrostatic pressure in
normal pregnancy and pre-eclampsia. Am J Obstet Gynecol
1987;157(1):102–106.
[11] Haller H, Hempel A, Homuth V, Mandelkow A et al. Endothelialcell permeability and protein kinase C in pre-eclampsia. Lancet
1998;351:945–949.
[12] Baker PN, Krasnov J, Roberts JM, Yeo KT. Elevated serum levels of
vascular endothelial growth factor in patients with pre-eclampsia.
Obstet Gynecol 1995;86(5):815–821.
[13] Anim-Nyame N, Sooranna SR, Steer PJ, Johnson MR. Longitudinal
changes in maternal plasma levels of leptin during normal pregnancy and pre-eclampsia. Hum Reprod 2000;15(9):2033–2036.
[14] Visser W, Beckman I, Bremer HA, Lim HL, Wallenburg HCS.
Bioactive tumour necrosis factor alpha in pre-eclamptic patients
with and without the HELLP syndrome. Br J Obstet Gynaecol
1994;101:1081–1082.
[15] Bates DO, Lodwick D, Williams B. Vascular endothelial growth
factor
and
microvascular
permeability.
Microcirculation
1999;6(2):83–96.
[16] Sierra-Honigmann MR, Nath AK, Murakami C, Garcia-Cardena G,
Papapetropoulos A, Sessa WC, Madge JL et al. Science
1998;5383(28):1683–1686.
[17] Cohen B, Barkan D, Levy Y, Goldberg I, Fridman E, Kopolovic J,
Rubinstein M. Leptin induces angiopoietin-2 expression in adipose
tissues. J Biol Chem 2001;276(11):697–700.
N. Anim-Nyame et al. / Cardiovascular Research 58 (2003) 162–169
[18] Madej T, Bogguski MS, Bryant SH. Threading analysis suggests
that the obese gene product may be a helical cytokine. FEBS Lett
1995;373:13–18.
[19] Soh EB, Mitchell MD, Keelan JA. Does leptin exhibit cytokine-like
properties in tissues of pregnancy? Am J Reprod Immunol
2000;43:292–298.
[20] Masuzaki H, Ogawa Y, Sagawa N, Hosoda K et al. Non adipose
tissue production of leptin: leptin as a novel placenta-derived
hormone in humans. Nat Med 1997;3:1029–1033.
[21] Cao R, Brakenheim E, Wahlestedt C, Thyberg J, Cao Y. Leptin
induces vascular permeability and synergistically stimulates angiogenesis with FGF-2 and VEGF. Proc Natl Acad Sci USA
2001;98:6390–6395.
[22] Skinner RA, Tucker VL, Curry FR. Acute effects of tumour necrosis
factor on hydraulic conductivity of mammalian postcapillary venules. Trauma 1999;47(3):486–491.
[23] Report on the National High Blood Pressure Educational Programme Working Group. Report on the National High Blood
Pressure Educational Programme Working Group on High Blood
Pressure in Pregnancy. Am J Obstet Gynecol 2000;183(1):S1–S22.
[24] Watanabe N, Kamel S, Ohkubo A, Yamanaka M, Ohsawa S, Makino
K. Urinary protein as measured with pyrogallol red-molybdate
complex manually and in Hitachi 726 automated analyser. Clin
Chem 1986;32:1551–1554.
[25] Christ F, Bauer A, Brugger D et al. A novel automatically
calibrating, liquid metal free measuring device for venous congestion plethysmography. J Appl Physiol 2000;88:1577–1583.
[26] Gamble J, Gartside IB, Christ F. A reassessment of mercury strain
gauge plethysmography for microvascular permeability assessment
in man. J Physiol 1993;464:407–422.
[27] Gamble J, Bethel D, Day NPJ, Loc PP, Phu NH, Gartside IB, Farrar
JF, White NJ. Age-related changes in microvascular permeability: a
significant factor in susceptibility of children to shock. Clin Sci
2000;98:211–216.
[28] Anthony FW, Evans PW, Wheeler T, Woods PJ. Variation in
detection of VEGF in maternal serum by immunoassay and the
possible influence binding proteins. Ann Clin Biochem
1997;34:276–280.
[29] Ma Z, Gingerich RL, Santiago JV, Klein S, Smith GH, Landt M.
Radioimmunoassay of leptin in human plasma. Clin Chem
1996;42:942–946.
[30] Redman CW, Sacks GP, Sargent IL. Preeclampsia: an excessive
maternal inflammatory response to pregnancy. Am J Obstet Gynecol
1999;180:499–506.
[31] Yelavarthi KK, Hunt JS. Analysis of p60 and p80 tumour necrosis
factor-alpha receptor messenger RNA and protein in human placentas. Am J Pathol 1993;143(4):113–1141.
[32] Vince GL, Starkey PM, Ausrgulen R, Kwiatkowski D, Redman
CWG. Interleukin-6, tumour necrosis factor and soluble tumour
necrosis factor receptors in women with. pre-eclampsia. Br J Obstet
Gynaecol 1995;102:20–25.
[33] Terranova PF, Hunter VJ, Roby KF, Hunt J. Tumour necrosis
factor-a in the female reproductive tract. Proc Soc Exp Biol Med
1995;209:325–342.
[34] Kupfermic MJ, Peaceman AM, Wigton TR, Tamura RK, Rehnberg
KA, Socol ML. Immunoreactive tumour necrosis factor-a is elevated in maternal plasma but undetected in amniotic fluid in the
second trimester. Am J Obstet Gynecol 1994;171:976–979.
[35] Slowik MR, De Luca LG, Fiers W, Pober JS. Tumor necrosis factor
activates human endothelial cells through the p55 tumor necrosis
factor receptor but the p75 receptor contributes to activation at low
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
169
tumor
necrosis
factor
concentration.
Am
J
Pathol
1993;143(6):1724–1730.
Beckmann I, Visser W, Struijk PC, van Dooren M, Glavimans J,
Wallenburg HC. Circulating bioactive tumor necrosis factor-alpha,
tumor necrosis factor-alpha receptors, fibronectin, and tumor necrosis factor-alpha inducible cell adhesion molecule VCAM-1 in
uncomplicated pregnancy. Am J Obstet Gynecol 1997;177(1):1247–
1252.
Stark JM. Pre-eclampsia and cytokine induced oxidative stress. Br J
Obstet Gynaecol 1993;100:105–109.
Fiers W. Tumor necrosis factor. Characterization at the molecular,
cellular and in vivo level. FEBS Lett 1991;285(2):199–212.
Minchenko A, Bauer T, Salceda S, Caro J. Hypoxic stimulation of
vascular endothelial growth factor expression in vitro and in vivo.
Lab Invest 1994;71:374–379.
Anim-Nyame N, Sooranna SR, Johnson MR, Gamble J, Steer P.
Resting peripheral blood flow in normal pregnancy and in preeclampsia. Clin Sci 2000;99:505–510.
Anim-Nyame N, Sooranna PJ, Johnson MR, Gamble J, Steer PJ.
Longitudinal changes in resting blood flow during normal pregnancy
and pregnancies complicated by chronic hypertension and preeclampsia. Cardiovasc Res 2001;50:603–609.
Bates DO, Curry FE. Vascular endothelial growth factor increases
hydraulic conductivity of isolated perfused microvessels. Am J
Physiol 1996;271:H2520–H2528.
Benyon HLC, Haskard DO, Davies KA, Haroutunian MJ, Walport
MJ. Combinations of low concentrations of cytokines and acute
agonists synergise in increasing the permeability of endothelial cell
monolayers. Clin Exp Immunol 1993;91:314–319.
Pocock TM, Bates DO. In vivo mechanisms of vascular endothelial
growth factor-mediated increased hydraulic conductivity of Rana
capillaries. J Physiol 2001;534(2):479–488.
Dvorak HF, Brown LP, Detmar M, Dvorak AM. Vascular permeability factors / vascular endothelial growth factor, microvascular
hyper permeability and angiogenesis. Am J Pathol 1995;146:1029–
1039.
Thurston G, Rudge JS, Suri C, Smith K, MacClain J, Sato TN,
Yancopoulos GD, McDonald DM. Leakage-resistant blood vessels
in mice transgenically over expressing Angiopoietins 1. Science
1999;286:2511–2514.
Thurston G, Rudge JS, Joffe E, Zhou H, Ross L, Croll SD, Glaser L,
Holash J, McDonald DM, Yancopoulos GD. Angiopoietin-1 protects
the adult vasculature against plasma leak. Nat Med 2000;6:460–463.
Kim I, Moon SO, Park SK, Chae SW, Koh GY. Angiopoietin-1
reduces VEGF-stimulated leukocyte adhesion to endothelial cells by
reducing ICAM-1, VCAM-1 and E-selectin expression. Cir Res
2001;89:477–479.
Mise H, Sagawa N, Matsumoro N, Yura S, Nanno H, Itoh H, Mori
T, Masuzaki H, Hosoda K, Ogawa Y, Nakao K. Augmented
placental production of leptin in pre-eclampsia: possible involvement of placental hypoxia. J Clin Endocrinol Metab
1998;83:3225–3229.
Lepercq J, Cauzac M, Lahlou N, Timsit J, Girard J, Auwex J,
Hauguel-de Mouzon S. Overexpression of placental leptin in
diabetic pregnancy: a critical role for insulin. Diabetes
1998;47(5):847–850.
Zamudio S, Palmer SK, Dahms TE, Berman JC, McCullough RG,
McCullough RE, Moore LG. Blood volume expansion, pre-eclampsia, and infant birth weight at high attitude. J Appl Physiol
1993;75:1566–1573.