Molecular Human Reproduction Vol.7, No.11 pp. 1093–1098, 2001
Abnormal fetal growth is not associated with altered
chorionic villous expression of vascular endothelial growth
factor mRNA
Gendie Lash1,3, Anne MacPherson2, David Liu1, Donna Smith1, Steven Charnock-Jones2
and Philip Baker1
1School
of Human Development, University of Nottingham, City Hospital, Nottingham, UK and 2Department of Obstetrics and
Gynaecology, University of Cambridge, Rosie Maternity Hospital, Cambridge, UK
3To
whom correspondence should be addressed at: Department of Anatomy and Cell Biology, 9th Floor Botterell Hall, Queen’s
University, Kingston, K7L 3N6, Ontario, Canada. E-mail: [email protected]
Altered placental and circulating levels of vascular endothelial growth factor (VEGF) and its receptor (flt-1) may
be associated with pre-eclampsia and intrauterine growth restriction (IUGR). The aim of this study was to determine
whether chorionic villous VEGF or flt-1 mRNA are altered at early gestation in pregnancies subsequently found to
be complicated by abnormal fetal growth. Quantitative reverse transcription–polymerase chain reaction was
performed on chorionic villous samples for VEGF and flt-1 using an internal RNA standard. Using the individualized
birthweight ratio (IBR), the subjects (n ⍧ 51) were divided into three groups; IUGR (IBR <10th centile, n ⍧ 6),
normal (IBR 10th–90th centiles, n ⍧ 41) and macrosomic (IBR >90th centile, n ⍧ 4). There was no correlation
between the mRNA expression of VEGF121 or VEGF165 and gestational age of the normal controls. There was also
no difference in the expression of either of the VEGF isoforms between the IUGR or macrosomic groups and the
normal controls. Expression of flt-1 was below the detection limit of the assay. In conclusion, we have found that
altered chorionic villous expression of VEGF is not associated with the initial stages of development of IUGR or
macrosomia.
Key words: chorionic villous sampling/flt-1/intrauterine growth retardation/pregnancy/VEGF
Introduction
The aetiology of abnormal fetal growth, i.e. intrauterine growth
restriction (IUGR) and macrosomia, is not well understood.
However, it is becoming increasingly apparent that altered
placental development underlies these multifactorial disorders.
From day 21 of development until the end of the first
trimester, the villous vasculature undergoes an increase in the
number of vessels rather than a change in type of vessels. The
number of fetal red blood cell-containing capillaries increases
and sprouting and branching angiogenesis from the existing
capillaries give rise to a primitive capillary network surrounded
by an incomplete layer of pericytes (Benirschke and Kaufmann,
1995). Basal lamina form around the capillaries from the sixth
week, resulting in a web-like arrangement of capillaries within
the stroma of smaller (mesenchymal) villi, while in larger villi
(immature intermediate) most of the capillaries are superficial.
In these larger villi, a few early villous arteries and veins are
located centrally, and from the 15th week, the adventitia of
these central arteries fuse as the stem villous forms (Demir
et al., 1997). From 26 weeks of gestation until term, the villous
© European Society of Human Reproduction and Embryology
vascular growth undergoes a change from branching to nonbranching angiogenesis owing to the formation of the mature
intermediate villi that specialize in gas exchange. These form
at the tips of the villous trees and are long and slender,
containing one or two long, poorly branched capillary loops
(Dunk and Ahmed, 2000).
The type of angiogenesis occurring is dependent in part on
the concentration ratio of the angiogenic growth factors,
vascular endothelial growth factor (VEGF) to placental growth
factor (PlGF). VEGF exerts its effects by binding with high
affinity to two tyrosine kinase receptors VEGFR-1/flt-1 and
VEGFR-2/KDR (de Vries et al., 1992; Terman et al., 1992).
In primary isolates of first trimester trophoblast cells and in a
first trimester extravillous trophoblast cell line, the VEGFR-1/
flt-1 receptor is expressed, as detected by reverse transcription–
polymerase chain reaction (RT–PCR) (Lash et al., 1999). In
human placenta, VEGF expression has been analysed by in-situ
hybridization and immunohistochemical techniques and has
been demonstrated to be localized to the villous trophoblast
and macrophages of both fetal and maternal origin (Ahmed
1093
G.Lash et al.
et al., 1995; Shore et al., 1997). A recent study on the
expression of VEGF, VEGFR-1, VEGFR-2/KDR and VEGFR3/flt-4 in the placentae of normotensive women demonstrated
by immunohistochemistry that VEGF and VEGFR-2 were
localized to the endothelial cells of capillaries in the villi and
of larger vessels (Helske et al., 2001). VEGFR-1 and VEGFR-3
were localized to the syncytial layer of the chorionic villi.
VEGFR-1 was also localized to the placental endothelial cells.
PlGF shares 53% homology with VEGF and is expressed both
in the villous syncytiotrophoblast and in the media of larger
placental stem vessels (Shore et al., 1997; Dunk and Ahmed,
2000). As VEGF is known to act in a paracrine manner on
endothelial cells, it seems likely to be involved in the initiation
of placental vasculogenesis. Correlation of the effects of the
growth factors (Dunk and Ahmed, 2000) and their expression
patterns throughout gestation (Kaufmann et al., 1985; Leiser
et al., 1985) have suggested that VEGF and VEGFR-2/KDR
are involved in the first two trimesters of pregnancy in the
establishment of the richly branched capillary beds of the
mesenchymal and intermediate villi, whereas PlGF and flt-1
are more likely to be involved in the formation of the long
poorly branched terminal capillary loops in the last trimester.
It appears that there may be two phenotypes of IUGR. In
one phenotype, the placenta appears to be in a state of relative
hypoxia and in the other, severe early-onset IUGR, it appears
to be in a state of hyperoxia (Kingdom and Kaufmann, 1999).
In the majority of placentae from women with IUGR there is
morphological evidence of continued branching angiogenesis.
However, in a small subset of women with severe early-onset
IUGR, there is reduced fetal–placental blood flow resulting in
or from elongated maldeveloped villous capillaries (Kohnen
and Kingdom, 2000).
Macrosomia (or large for gestational age infants) may be
associated with diabetes and the majority of investigations
have concentrated on insulin, the insulin-like growth factors I
and II and their binding proteins. However, very little consideration has been given to the idea of altered placentation as a
possible underlying factor of macrosomia in pregnancies not
complicated by diabetes. It is possible that inappropriate
placental development is associated with non-diabetes-associated macrosomia.
In this study, we have attempted to measure the expression
of VEGF and flt-1 mRNA in first and second trimester
placental tissue from normal pregnancies and from pregnancies
complicated by IUGR or macrosomia. The only ethically
acceptable method of obtaining tissue was to use that surplus
to diagnostic purposes, after chorionic villous sampling.
Materials and methods
Patient details
Written consent was obtained from women attending Nottingham
City Hospital for diagnostic first trimester chorionic villous sampling.
In the majority of cases, sampling was on the basis of increased
maternal age; occasionally patients were sampled due to a previous
child having had chromosomal abnormalities. Approximately 10 mg
of tissue was removed for cytogenetic studies, and tissue surplus to
these requirements was utilized in the study. Using the individualized
1094
birth ratio (IBR) (Wilcox et al., 1993), the patients were divided into
three groups; IBR ⬍10th centile, IBR 10th–90th centile, IBR ⬎90th
centile. Patient demographics for the three groups are shown in
Table I. The IBR is a ratio relative to a predicted birthweight,
calculated using independent coefficients for gestation at delivery,
fetal sex, parity, ethnic origin, maternal height and booking weight.
This ratio enables a more accurate prediction of pregnancy outcome
than birthweight for gestational age alone. All women in the study
were Caucasian and there was a normal distribution of fetal sex
within each group. None of the patients had diabetes mellitus. None
of the IUGR pregnancies were complicated by early onset disease
(⬍32 weeks) or Doppler umbilical artery absent end-diastolic flow.
Sample collection
Tissue samples that were surplus to clinical diagnostic requirements
were collected into liquid nitrogen following the sampling procedure
and stored until required for analysis. Samples were collected over a
3 year period; however, there was no difference in the average storage
time for the three groups.
RNA extraction
Total RNA was isolated from the chorionic villous tissue samples
using the Purescript RNA isolation kit (Flowgen, Staffordshire, UK)
following the manufacturer’s instructions for tissue samples.
RT–PCR
First strand cDNA was synthesized using the Reverse-iT kit (Advanced
Biotechnology Ltd, UK) which utilizes Murine Molony Leukemia
Virus (MMLV) reverse transcriptase and an anchored oligo dT primer.
Approximately 1 µg of total RNA was used in the RT reaction.
RT–PCR was performed for β-actin and HLA-G (Kirszenbaum
et al., 1994) using the primer pairs described in Table II. A standard
PCR reaction mix was used to amplify 2 µl of each cDNA sample
using a PCR master mix (Advanced Biotechnology Ltd, UK) with a
final concentration of 2.0 mmol/l MgCl2 and 0.2 µl of each primer
(50 pmol/l starting concentration). The PCR for HLA-G was a seminested reaction, first using the 257 and 1225 primer pair and secondly
amplifying 2 µl of the first round product with the 526 and 1225
primer pair. In both cases, 35 cycles of denaturation, annealing and
extension were performed.
Quantitative RT–PCR
Constructs for standard artificial mRNA (construct mRNA, cRNA)
for VEGF and flt-1 were made in pBlueScript (Stratagene, Amsterdam,
NL). A schematic diagram for the design of both constructs is shown
in Figure 1. A three-step cloning protocol was used for the insertion
of each component of the construct. Oligonucleotides were made for
the complementary and identical sequences for the primers to be used
with appropriate restriction site sequences at their ends for ease in
cloning. The reverse primer oligonucleotide also contained a polyT
stretch to produce a polyA tail in the resultant RNA for use in the
reverse transcriptase reaction. Stuffer DNA from lambda DNA was
used to give the construct its expected length, i.e. suitably different
from the wild type PCR product length to be distinguished from it
during electrophoresis, but close enough that both the construct and
unknown product were amplified under the same kinetics. At each
step of the cloning procedure, clones were verified by sequence
analysis (ABI automated sequencer unit, Department of Biochemistry,
University of Nottingham, UK).
RNA was produced from the plasmids containing each construct
using a T3 RNA polymerase assay (Promega, Southampton, UK) and
measured spectrophotometrically to determine the concentration of
the cRNA. Different concentrations of the cRNA were reverse-
Placental VEGF expression and fetal growth
Table I. Patient demographics for the three subject groups in this study as divided by the individualized birth ratio (IBR)
Maternal age
Maternal weight at booking (kg)
Maternal height at booking (m)
Parity
Gestation at time of sampling (weeks ⫹ days)
Gestation at delivery (weeks ⫹ days)
Fetal weight at birth (kg)*
Centile*
IBR ⬍10th centile (n ⫽ 6)
IBR 10th–90th centile (n ⫽ 41)
IBR ⬎90th centile (n ⫽ 4)
39 (36.8–39.8)
66 (62.6–68.3)
1.62 (1.57–1.67)
1 (1–1)
11 ⫹ 5 (11 ⫹ 4 to 12)
39 (38 ⫹ 6 to 39 ⫹ 3)
2.73 (2.56–2.92)
5 (2.5–7.5)
38 (36–39)
65 (59.2–73.7)
1.63 (1.6–1.68)
1 (0–2)
11 ⫹ 3 (11 ⫹ 1 to 12)
40 ⫹ 2 (39 to 41)
3.44 (3.22–3.74)
42 (25.8–63.8)
37.5 (37–38.8)
59.4 (47.2–74.3)
1.60 (1.58–1.64)
1.5 (0.8–2)
13⫹2 (12 ⫹ 6 to 13 ⫹ 3)
39 ⫹ 5 (39 ⫹ 2 to 40)
4.17 (4.12–4.27)
98 (97.8–98.3)
Data are shown as median (interquartile range). There was no statistical difference between the ⬍10th centile or ⬎90th centile groups and the 10th–90th
centile group for any of the factors apart from fetal weight at birth and IBR centile; *P ⬎ 0.001, Mann–Whitney U-test.
Table II. Primer pairs used for reverse transcription–polymerase chain reaction (RT–PCR)
β-actin
HLA-G
Sense
Antisense
Annealing
Expected product
temperature (°C) length (bp)
5⬘-AAGGATTCCTATGTGGGC-3⬘
257 5⬘-GGAAGAGGAGACACGGAACA-3⬘
526 5⬘-CCAATGTGGCTGAACAAAGG-3⬘
5⬘-CATCTCTTGCTCGAAGTC-3⬘
54
600
59
64
5⬘-GGCTCTAGAGCGCAGAGTCTCTTC-3⬘
60
5⬘-GGCTCTAGATCACCGCCTCGGCTTGTCAC-3⬘ 64
5⬘-AGCCACAGTCCGGCACGTAGGTGATT-3⬘
62
400
VEGF I⫹J
5⬘-GGCTCTAGATCGGGCCTCCGAAACCAT-3⬘
VEGF C⫹H 5⬘-GAGTGTGTGCCCACTGAGGA-3⬘
flt-1
5⬘-GTCACAGAAGAGGATGAAGGTGTCTA-3⬘
1225 5⬘-TGAGACAGAGACGGAGACAT-3⬘
400, 300 and 200
200
Cycle conditions used for PCR were: 96°C for 2 min then 96°C for 30 s, annealing temperature for 30 s, 72°C for 1 min for 30–35 cycles then 72°C min and
held at 4°C.
30 cycles were performed and the forward primer was tagged with a
Cy5 molecule for use in an ALF sequencer. All products were run
on a polyacrylamide gel and visualized with an ALF sequencer. The
intensity of fluorescence for each product was assessed as the area
under a curve as determined by the ALF software. The area under
the curve for each product in each sample was plotted against the
dilution factor on a log–log plot to obtain the ratio of sample to
construct. This ratio was then applied to the known amount of cRNA
added to the RT reaction and expressed as the pg mRNA/µg total
RNA in the sample.
Results
Figure 1. Schematic representation of the constructs used for
quantitative reverse transcription–polymerase chain reaction (RT–
PCR).
transcribed as previously described and used in a PCR reaction to
determine the optimum concentration of cRNA to use in each of the
VEGF and flt-1 experiments.
The optimum concentration of cRNA for VEGF and ftl-1 was
added to 1 µg of RNA for each sample and reverse-transcribed as
described earlier. VEGF mRNA was amplified using two primer pairs
(Table II) in a nested PCR reaction. Two µl of cDNA for each sample
was added to a PCR master mix as previously described and one in
three serial dilutions were performed such that there were six reactions
for each sample. In the PCR for VEGF, 30 cycles were performed in
the first round with the I ⫹ J primer pair and a subsequent 19 cycles
were performed in the second round with the C ⫹ H primer pair
using 2 µl of the first round product. The VEGF-C primer was tagged
with a Cy5 molecule for use in an ALF sequencer. In the flt-1 PCR,
β-actin and HLAG RT–PCR
Initially 127 samples were collected for analysis. All of these
were screened by RT–PCR for the presence of β-actin. Of the
initial 127 samples, 75 had a product for β-actin when run on
an ethidium bromide agarose gel. Unfortunately there was
insufficient patient data to determine the pregnancy outcome
or IBR, or the pregnancy in 24 of these cases had been
terminated due to detected chromosomal abnormalities. The
remaining 51 samples were studied for the presence of HLAG by RT–PCR to determine the presence of trophoblast tissue
within the sample. All samples tested positive for HLA-G, a
specific marker of trophoblast cells.
VEGF quantitative RT–PCR
A typical graph of PCR band intensity obtained from the ALF
sequencer is shown in Figure 2. There was no correlation
between the mRNA expression of VEGF121 (Figure 3) or
VEGF165 (Figure 4) and the gestational age of the normal
controls [gestation range 9 to 18 ⫹ 6 weeks, VEGF121, P ⫽
1095
G.Lash et al.
r2 ⫽ 0.04; VEGF165, P ⫽ 0.7, r2 ⫽ 0.09 (Spearman correlation)]
pregnancies.
There was no significant difference in the expression of
either of the VEGF isoforms between the IUGR or macrosomic
groups and the normal controls [VEGF121: IUGR, 0.06 (0.02–
0.09) (P ⫽ 0.2, Mann–Whitney U-test), normal, 0.15 (0.02–
0.27), macrosomic, 0.12 (0.06–0.2) (P ⫽ 0.6, Mann–Whitney
U-test); VEGF165: IUGR, 0.09 (0.03–0.23) (P ⫽ 0.2, Mann–
Whitney U-test), normal, 0.34 (0.04–0.53), macrosomic, 0.27
(0.16–0.39) (P ⫽ 0.6, Mann–Whitney U-test)] (Figures 5
and 6).
flt-1 quantitative RT–PCR
The level of flt-1 in all of the samples (n ⫽ 51) was below
the level of detection of the assay system.
Figure 2. Typical graph of polymerase chain reaction band
intensity results from running products on the ALF sequencer.
Figure 3. Correlation of amount of VEGF121 mRNA with
gestational age in chorionic villous samples from women with
normal-sized infants for gestational age.
0.6, r2 ⫽ 0.08; VEGF165, P ⫽ 0.4, r2 ⫽ 0.07 (Spearman
correlation)]. There was also no correlation between the
expression of VEGF121 (Figure 5) or VEGF165 (Figure 6) and
the gestational age of the IUGR [gestation range 6 ⫹ 4 to
19 ⫹ 4 weeks, VEGF121 P ⫽ 0.7, r2 ⫽ 0.04; VEGF165,
P ⫽ 0.7, r2 ⫽ 0.09 (Spearman correlation)] or macrosomia
[gestation range 11 ⫹ 2 to 13 ⫹ 3 weeks, VEGF121 P ⫽ 0.7,
1096
Discussion
This study investigated the expression of two isoforms of
VEGF-A, 165 and 121, and one of their receptors, VEGFR-1/
flt-1, in first and second trimester placental tissue in pregnancies
complicated by abnormal fetal and placental growth (IUGR
and macrosomia). VEGFR-1/flt-1 mRNA expression in the
chorionic villous samples was below the level of detection of
the assay system. We found no correlation between infant size
at birth and VEGF expression. We also found no evidence of
altered VEGF expression with gestation in the early stages of
pregnancy.
Recently, it has been demonstrated that there is no difference
in the intensity or localization of staining for VEGF between
placental sections from women with pre-eclampsia or IUGR
those from normal pregnancies (Helske et al., 2001). They
did, however, observe an increase in the intensity of staining
for VEGFR-1/flt-1, but not for VEGFR-2 or VEGFR-3, in a
proportion of the same groups of affected placentae compared
with normal placentae. Additionally, the related protein, PlGF,
has been shown to increase in term placentae complicated by
fetal growth restriction (Khaliq et al., 1999). In contrast, serum
PlGF has been shown to be lower in women with pre-eclampsia
or IUGR compared to normal controls at term (Helske et al.,
2001). It would be interesting to repeat this study to investigate
the expression of PlGF in these samples and to determine if
the previously reported increase in expression of this growth
factor at term is an effect of IUGR or a cause of it. The six
cases of IUGR in the study had late onset disease, a condition
characterized by increased branching angiogenesis (Kohnen
and Kingdom, 2000). VEGF has been reported to be important
for branching angiogenesis, whereas PlGF has been shown to
be associated with non-branching angiogenesis. We would thus
have anticipated an increase in VEGF expression. Although
the sample size in this study was low, the results do not
support a role for chorionic villus-derived VEGF in the
pathogenesis of the condition. It would be interesting to
investigate samples from women with early-onset IUGR, given
the likelihood of a different aetiology of this disease.
In previous reports, increased levels of serum VEGF have
been associated with pre-eclampsia (Baker et al., 1995; Sharkey
et al., 1996; Kupferminc et al., 1997; Brockelsby et al., 1999).
Placental VEGF expression and fetal growth
Figure 4. Correlation of amount of VEGF165 mRNA with gestational age in chorionic villous samples from women with normal-sized
infants for gestational age.
Figure 5. Graph of the median (interquartile ranges) of VEGF121
expression in samples from women with intrauterine growth
retardation (IUGR), macrosomia or normal-sized infants.
Several other studies have found depressed levels of VEGF in
the sera of pre-eclamptic women (Lyall et al., 1997; Reuvekamp
et al., 1999), although the discrepancies may be explained by
the fact that quantification of VEGF in pregnancy is affected
by interference from binding proteins (Anthony et al., 1997).
A decrease in placental VEGF mRNA in women with preeclampsia has been demonstrated (Cooper et al. 1996); however, another study found no difference in placental VEGF
protein levels between normotensive and pre-eclamptic pregnancies (Helske et al., 2001). These studies have been conducted at term or on clinical presentation of the condition, and
it is likely therefore that these altered levels may be an effect
of the inefficient placentation rather than the cause of it.
Figure 6. Graph of the median (interquartile ranges) of VEGF165
expression in samples from women with IUGR, macrosomia or
normal-sized infants.
The small sample size, resulting from inadequate tissue for
analysis and in some cases inadequate patient details is an
important caveat in the interpretation of our results. It would
have been optimal to collect multiple samples from the same
women over gestation, and so due to the nature of the tissue
collection this was not possible, but we may not have collected
a truly representative sample of chorionic villi from each
woman. A second caveat that should be placed on this study
is how representative the samples obtained are of the population
as a whole. Although advancing maternal age increases the
incidence of pre-eclampsia, it is likely that we did not sample
any women with pre-eclampsia as the increased maternal age
meant that the majority of subjects were multiparous.
This study has indicated that chorionic villi VEGF mRNA
levels do not increase early in gestation, over a period when
1097
G.Lash et al.
oxygen tension is rising in the placenta and one would expect
to see an increase in VEGF mRNA. Nor could we find any
evidence for the role of chorionic villi VEGF expression in
the early placenta of pregnancies later complicated by fetal
growth abnormalities such as IUGR or macrosomia. It is
possible, however, that altered maternal myometrial or decidual
VEGF production may lead to these conditions. One of the
receptors for VEGF, VEGFR-1/flt-1, was not detectable in any
of the samples studied. The study of the other receptors for
VEGF and the expression of PlGF in similar samples and in
samples of maternal myometrium and decidua would be very
interesting. Further research is required on the factors involved
in early placental development.
Acknowledgements
We wish to acknowledge the financial support of Action Research
(G.E.L). They also wish to thank Dr Gareth Cross, Clinical Genetics,
Nottingham City Hospital for use of the ALF sequencer. We would
also like to acknowledge the technical support of Miss Suzanne
Cooper and the assistance of Dr Bryony Strachan in patient detail
collection.
References
Ahmed, A., Li, X.F., Dunk, C. et al. (1995) Colocalisation of vascular
endothelial growth factor and its Flt-1 receptor in human placenta. Growth
Factors, 12, 235–243.
Anthony, F.W., Evans, P.W., Wheeler, T. et al. (1997) Variation in detection
of VEGF in maternal serum by immunoassay and the possible influence of
binding proteins. Ann. Clin. Biochem., 34, 276–280.
Baker, P.N., Krasnow, J., Roberts, J.M. et al. (1995) Elevated serum levels of
vascular endothelial growth factor in patients with preeclampsia. Obstet.
Gynecol., 86, 15–21.
Benirschke, K. and Kaufmann, P. (1995) Pathology of the Human Placenta,
3rd edn. Springer, New York.
Brockelsby, J.C., Hayman, R., Warren, A.Y. et al. (1999) VEGF mimics preeclamptic plasma in inhibiting bradykinin-induced relaxation of isolated
pregnancy myometrial vessels via VEGFR-1. Lab. Invest., 79, 1101–1111.
Cooper, J.C., Sharkey, A.M., Charnock-Jones, D.S. et al. (1996) VEGF mRNA
levels in placentae from pregnancies complicated by pre-eclampsia. Br. J.
Obstet. Gynaecol., 103, 1191–1196.
de Vries, C., Escobedo, J.A., Ueno, H. et al. (1992) The fms-like tyrosine
kinase, a receptor for vascular endothelial growth factor. Science, 255,
989–991.
Demir, R., Kosanke, G., Kohen, G. et al. (1997) Classification of human
1098
placenta stem villi: review of structural and functional aspects. Micros.
Res. Tech., 38, 29–41.
Dunk, C. and Ahmed, A. (2000) Growth factor regulators of placental
angiogenesis. In Kingdom, J. and Baker, P.N. (eds), Intrauterine Growth
Restriction. Springer, London, pp. 149–165.
Helske, S., Vuorela, P., Carpen, O. et al. (2001) Expression of vascular
endothelial growth factor receptors 1, 2 and 3 in placentae from normal
and complicated pregnancies. Mol. Hum. Reprod., 7, 205–210.
Kaufmann, P., Bruns, U., Leiser, R. et al. (1985) The fetal vascularisation of
term human placental villi. II. Intermediate and terminal villi. Anat.
Embryol. (Berl)., 173, 203–214.
Khaliq, A., Dunk, C., Jiang, J. et al. (1999) Hypoxia down-regulates placenta
growth factor, whereas fetal growth restriction up-regulates placenta growth
factor expression: molecular evidence for ‘placental hyperoxia’ in uterine
growth restriction. Lab. Invest., 79, 151–170.
Kingdom, J.C. and Kaufmann, P. (1999) Oxygen and placental vascular
development. Adv. Exp. Med. Biol., 474, 259–275.
Kirszenbaum, M., Moreau, P., Gluckman, E. et al. (1994) An alternatively
spliced form of HLA-G mRNA in human trophoblasts and evidence for
the presence of HLA-G transcript in adult lymphocytes. Proc. Natl Acad.
Sci. USA, 91, 4209–4213.
Kohnen, G. and Kingdom, J. (2000) Villous development and the pathogenesis
of IUGR. In Kingdom, J. and Baker, P.N. (eds), Intrauterine Growth
Restriction. Springer, London, pp. 131–147.
Kupferminc, M.J., Daniel, Y., Englender, T. et al. (1997) Vascular endothelial
growth factor is increased in patients with pre-eclampsia. Am. J. Reprod.
Immunol., 38, 302–306.
Lash, G.E., Cartwright, J.E., Whitley, G. StJ. et al. (1999) The effects of
angiogenic growth factors on extravillous trophoblast invasion and motility.
Placenta, 20, 661–667.
Leiser, R., Luckhardt, M., Kaufmann, P. et al. (1985) The fetal vascularisation
of term human placental villi. I. Peripheral stem villi. Anat. Embryol.
(Berl.), 173, 71–80.
Lyall, F., Greer, I.A., Boswell, F. et al. (1997) Suppression of serum vascular
endothelial growth factor immunoreactivity in normal pregnancy and in
pre-eclampsia. Br. J. Obstet. Gynaecol., 104, 223–228.
Reuvekamp, A., Velsing-Aarts, F.V., Poulina, I.E. et al. (1999) Selective
deficit of angiogenic growth factors characterizes pregnancies complicated
by pre-eclampsia. Br. J. Obstet. Gynaecol., 106, 1019–1022.
Sharkey, A.M., Cooper, J.C., Balmforth, J.R. et al. (1996) Maternal plasma
levels of vascular endothelial growth factor in normotensive pregnancies
and in pregnancies complicated by pre-eclampsia. Eur. J. Clin. Invest., 26,
1182–1185.
Shore, V.H., Wang, T.H., Wang, C.L. et al. (1997) Vascular endothelial growth
factor, placenta growth factor and their receptors in isolated human
trophoblast. Placenta, 18, 657–665.
Terman, B.I., Dougher-Vermazen, M., Carrion, M.E. et al. (1992) Identification
of the KDR tyrosine kinase as a receptor for vascular endothelial cell
growth factor. Biochem. Biophys. Res. Commun., 187, 1579–1586.
Wilcox, M.A., Johnson, I.R., Maynard, P.V. et al. (1993) The individualised
birthweight ratio: a more logical outcome measure of pregnancy than
birthweight alone. Br. J. Obstet. Gynaecol., 100, 342–347.
Received on April 18, 2001; accepted on September 5, 2001