Endothelial FOS expression and pre-eclampsia

Authors


Dr C Delles, Institute of Cardiovascular and Medical Sciences, BHF Glasgow Cardiovascular Research Centre, University of Glasgow, 126 University Place, Glasgow, G12 8TA, UK. Email christian.delles@glasgow.ac.uk

Abstract

Please cite this paper as: Mackenzie RM, Sandrim VC, Carty DM, McClure JD, Freeman DJ, Dominiczak AF, McBride MW, Delles C. Endothelial FOS expression and pre-eclampsia. BJOG 2012;119:1564–1571.

Objective  To study gene expression profiles in human endothelial cells incubated with plasma from women who developed pre-eclampsia and women with normotensive pregnancies.

Design  A case–control study.

Setting  A longitudinal nested case–control study within three maternity units.

Population  A mixed obstetric population attending maternity hospitals in Glasgow.

Methods  Plasma was obtained at both 16 and 28 weeks of gestation from 12 women: six women subsequently developed pre-eclampsia (cases) and six women, matched for age, body mass index (BMI) and parity, remained normotensive (controls). Human umbilical vein endothelial cells (HUVECs) were incubated with plasma for 24 hour before RNA isolation.

Main outcome measures  Gene expression profiles were compared between the two gestational time points using Illumina® HumanHT-12 v4 Expression BeadChips. Differential mRNA expression observed in microarray experiments were validated using quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR), and gene networks were analysed using Ingenuity® pathway analysis.

Results  There was a significant difference in the expression of 25 genes following incubation with plasma from controls, and an increase in the expression of 11 genes following incubation with plasma from cases, with no overlap between the two groups (false discovery rate, FDR < 0.05). There was a 3.74-fold (FDR < 0.001) increase in the expression of the c-Fos gene (FOS) when HUVECs were incubated with control plasma from 16 and 28 weeks of gestation, with no significant difference between the two time points with plasma from cases. Similar findings for FOS were obtained by qRT-PCR.

Conclusions  Plasma from women who subsequently develop pre-eclampsia appears to contain factors that lead to the dysregulation of FOS in endothelial cells during pregnancy. Reduced expression of c-Fos may lead to impaired vasculogenesis, and thereby contribute to the development of pre-eclampsia.

Introduction

The pathogenesis of pre-eclampsia is thought to involve a two-stage process. The first stage is characterised by defective placental angiogenesis. Abnormal placentation and failure in the remodelling of spiral arteries leads to the secretion of placental factors into the maternal circulation in the second stage, and to widespread systemic effects in the mother.1 Among the factors released by the placenta are cytokines, bioactive cellular debris from the trophoblast and a variety of anti-angiogenic factors, including soluble endoglin (sEng) and the soluble form of the vascular endothelial growth factor (VEGF) receptor (soluble FMS-like tyrosine kinase-1, Flt-1).1 VEGF regulates gene transcription via transcription factors such as FOS. The FOS gene family consists of four members that encode leucine zipper proteins: FOS, FOSB, FOSL1 and FOSL2. They dimerise with proteins of the JUN family to form the transcription factor complex AP-1. The c-Fos proto-oncogene is a transcription factor encoded by the FOS gene, and has been implicated as a regulator of cell proliferation, differentiation and transformation.2,3

One of the hallmarks of pre-eclampsia is systemic maternal endothelial dysfunction.4Ex vivo myography studies in subcutaneous resistance arteries demonstrated that normal pregnancy is associated with a marked improvement in endothelium-dependent vasodilation, whereas vessels from women with pre-eclampsia are characterised by an impaired response to vasodilatory stimuli.5,6 Endothelial dysfunction in pre-eclampsia has been confirmed in clinical studies using ultrasound-derived flow-mediated vasodilation.7,8 In women in whom pre-eclampsia develops, impaired endothelium-dependent vasodilation, as assessed by skin microvascular function using iontophoresis of acetylcholine, has also been shown to precede the development of the clinical syndrome.9

Plasma of women in which pre-eclampsia develops has been found to cause impaired endothelium-dependent vasodilatation in in vitro studies.10 Other studies highlight a role of microparticles in the maternal plasma of women with pre-eclampsia, causing endothelial dysfunction.11 Some of the effects of plasma from women with pre-eclampsia on endothelial cells are the result of the activation of monocytes and subsequent inflammatory responses.12,13 A previous study by Donker et al.14, using microarray technology, investigated the effects of the plasma factors present in severe early-onset pre-eclampsia on gene expression profiles in human umbilical vein endothelial cells (HUVECs). This study failed to detect substantially altered endothelial gene expression on exposure to pre-eclamptic plasma factors.

We aimed to improve upon previous investigations by using HUVECs from a single donor and plasma obtained at both early (16 weeks of gestation) and mid pregnancy (28 weeks of gestation). We investigated whether plasma contains soluble factors that have effects on gene expression in endothelial cells. Our study was not, however, designed to identify factors that lead to impaired placentation.

Methods

Subjects

The Proteomics in Pre-eclampsia (PIP) study was a longitudinal study designed to identify clinical and biochemical markers involved in the pathogenesis of pre-eclampsia. Recruitment protocols are described in detail elsewhere.15 A subset of 180 pregnant women with two or more risk factors for pre-eclampsia were invited to attend for study visits at 16 and 28 weeks of gestation.16 Specific risk factors assessed at the booking visit were: age > 35 years; body mass index > 30 kg/m²; nulliparity; history of pre-eclampsia in a previous pregnancy; family history of pre-eclampsia in mother or sister; detectable proteinuria at booking. Women with a history of chronic hypertension, diabetes or renal disease were not included in the study. Of these 180 women, 17 (9.4%) developed pre-eclampsia, seven (3.9%) developed pregnancy-induced hypertension without significant proteinuria and 156 (87%) remained normotensive throughout their pregnancy. For the present study, we selected a convenience sample of six women who developed pre-eclampsia (cases), and an age, parity and body mass index-matched control cohort of six women with normotensive pregnancies (controls). Pregnancy-induced hypertension was defined in accordance with the International Society for the Study of Hypertension in Pregnancy (ISSHP) criteria: systolic blood pressure ≥ 140 mmHg and diastolic blood pressure ≥ 90 mmHg on two occasions more than 6 hours apart, occurring after 20 weeks of gestation, but before the onset of labour.17 Pre-eclampsia was defined as pregnancy-induced hypertension in association with proteinuria (≥300 mg/24 hours, protein:creatinine ratio ≥ 30 mg/mmol or, if neither were available, ++ or more on dipstick measurement). Delivery information was obtained from labour ward databases; casenotes for all women who developed pre-eclampsia were reviewed by the same author (D.M.C.). The clinical characteristics of these women are available in Table S1. All participants gave written informed consent. The study was approved by the West of Scotland Ethics Committee, and complies with the principles of the Declaration of Helsinki.

Collection and preparation of plasma samples

At the study visit, blood was collected via venesection of the antecubital vein using the Vacutainer® system (BD, Franklin Lakes, NJ, USA), with EDTA as the anticoagulant. Whole blood was immediately placed on ice and centrifuged at 4°C. Plasma was stored at −80°C in 1-ml aliquots.

HUVEC cell culture and plasma incubation conditions

Single-donor HUVECs (PromoCell GmbH, Heidelberg, Germany) were cultured at 37°C in 5% CO2 in complete Large Vessel Endothelial Cell Basal Medium (TCS Cellworks Ltd, Buckinghamshire, UK), supplemented with 20% (v/v) fetal calf serum (FCS), 100 iu/ml penicillin, 100 μg/ml streptomycin and 2 mmol/l l-glutamine (endothelial cell medium). Cells were grown to confluency in 75-cm2 tissue culture flasks (Corning, Costar, Netherlands), before being detached from the flask surface by trypsin-EDTA (0.5/0.2 mg/ml in phosphate-buffered saline, PBS), resuspended in complete medium and split at a 1:3 ratio into additional 75-cm2 flasks. At passage 4, HUVECs were resuspended in endothelial cell medium and re-plated in 10 μg/ml fibronectin-coated (Sigma-Aldrich, Poole, UK) six-well tissue culture plates (Corning), where they were grown to confluency for incubation with patient plasma.

On reaching confluency in six-well plates, endothelial cell medium was removed and HUVECs were washed twice in PBS to remove traces of FCS. Cells were then incubated at 37°C, 5% CO2, for 24 hours in endothelial cell medium devoid of FCS, but supplemented with 20% (v/v) plasma from pre-eclamptic patients at 16 and 28 weeks of gestation, or matched control subjects at 16 and 28 weeks of gestation. During the course of the 24-hour incubation period, cells were viewed microscopically, and were consistently found to be adherent and viable.

HUVEC total RNA extraction, quantification and validation

Following the 24-hour incubation of HUVECs with patient and control plasma, total RNA was isolated from the cells using the RNeasy® Mini Kit (QIAGEN, Leusden, the Netherlands), and contaminating DNA was removed using the TURBO DNA-free™ kit (Ambion, Huntingdon, UK). Total RNA was then quantified using the NanoDrop® ND-100 Spectrophotometer and nd-1000 3.1.0 (Labtech International Ltd, Lewes, UK), and sample integrity was verified by electrophoresis of 500 ng on an Agilent Bioanalyzer 2100 (Figure S1).

Biotinylated cRNA synthesis

The Illumina® TotalPrep RNA Amplification Kit (Illumina, Saffron Walden, UK) was used to generate biotinylated, amplified complementary RNA (cRNA) for hybridization with Illumina® Sentrix arrays. Purified cRNA was quantified using the NanoDrop® ND-100 Spectrophotometer, and quality was assessed via electrophoresis of 500 ng on an Agilent Bioanalyzer 2100 (Figure S2).

Microarray hybridization and data collection

cRNA was prepared for hybridization with Illumina® HumanHT-12 v4 Expression BeadChips, using the Illumina® Whole-Genome Expression Direct Hybridization Assay. Following completion of the hybridization procedure, BeadChips were scanned and images extracted using specific protocols on an Illumina® BeadArray Reader.

Validation with quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR)

TaqMan® qRT-PCR (Applied Biosystems, Warrington, UK) was used to validate the differential mRNA expression observed in microarray experiments. The same total RNAs isolated for microarray experiments were used as templates for the synthesis of complementary DNA (cDNA). First-strand cDNA was synthesised from 1 μg of DNase-treated total RNA (TaqMan® Reverse Transcription Reagents). Applied Biosystems Custom TaqMan® Gene Expression Assay for FOS (Hs00170630_m1) was used with TaqMan® Endogenous Control, β-actin (ACTB, Hs99999903_m1). β-actin expression was stable in the HUVECs (cycle threshold, Ct, values of 23 ± 1). The relative quantitation of FOS was calculated using the comparative (ΔΔCt) method.18

Pathway analysis

Ingenuity® Pathway Analysis (Ingenuity Systems, Redwood City, CA, USA) was used to identify a network of genes that were connected to FOS and significantly regulated in the healthy controls at 28 weeks of gestation, versus controls at 16 weeks of gestation.

Microarray data analysis

Two Illumina® HT-12 v4 Expression BeadChip microarrays were used to assess the gene expression of the 24 samples. Two samples were obtained from each of 12 subjects, one at 16 weeks of gestation and a second at 28 weeks of gestation. To minimise unwanted sources of error, the design was blocked and nested to ensure that each chip had both the samples from half of the pre-eclamptic and healthy women. After quality control checks, one sample from each group was removed.

The data were quantile-normalised and background subtracted using Beadstudio® (Illumina, San Diego, CA), and rank products were used to assess the statistical significance of pairwise intergroup differences.19 Paired analyses were carried out to investigate changes over time, and unpaired analyses were used to compare experiments with plasma from women with normotensive pregnancies and with plasma from women who developed pre-eclampsia at each gestational time point.

Significance was determined using the false discovery rates (FDRs) multiple testing correction method,20 with an FDR cut-off of 5%.

Statistical analyses

For clinical parameters, continuous data are given as means ± standard deviations. For comparisons of normally distributed continuous variables, paired and unpaired Student’s t-tests were applied as appropriate. The Mann–Whitney U-test and the Wilcoxon signed-rank test were used for analysis of non-normally distributed data, including qRT-PCR. Categorical data were analysed by Fisher’s exact test. A P value of <0.05 (two-tailed) was considered to be significant.

Results

Characteristics of study participants

All women were primigravidae. Pre-eclampsia was late onset in all affected women: none required anti-hypertensive medication or delivery before 36 weeks of gestation. Four of the six women had blood pressure in the ‘severe’ category (i.e. over 160 mmHg systolic or 110 mmHg diastolic). In line with the selection criteria, women who developed pre-eclampsia and women with normotensive pregnancies were of similar age, parity and body mass index. As observed previously in the overall study cohort, blood pressure was already higher in the first trimester in women who subsequently developed pre-eclampsia, compared with women who had normotensive pregnancies.15 Gestation at delivery and birthweights were lower in women with pre-eclampsia compared with controls (Table S1).

Endothelial cell gene expression profile induced by plasma of pregnant woman

Analysis of paired data revealed the differential endothelial expression of 25 genes on incubation of cells with plasma from control subjects (Table 1), and that of 11 genes on incubation with plasma from cases (Table 2). No overlap was observed between the two groups. The most striking result was a 3.74-fold increase in the expression of FOS, encoding the transcription factor, c-Fos, in cells incubated with control plasma obtained at 28 weeks of gestation, compared with cells incubated with control plasma obtained at 16 weeks of gestation (FDR < 0.001). A similar increase was not observed in cells incubated with plasma from cases, where there was a non-significant 1.11-fold increase in the expression of FOS (FDR = 0.10).

Table 1. Differential gene expression in normotensive pregnancy.
Probe IDGene nameAccession numberFDR HP for 16 vs 28 weeks of gestationFold change
  1. Human umbilical vein endothelial cells (HUVECs) were incubated with plasma from women with healthy pregnancies (HP), obtained at 16 and 28 weeks of gestation. Differentially expressed genes, their false discovery rates (FDRs) and fold changes from microarray experiments are displayed. Note the 3.74-fold upregulation of FOS expression.

ILMN_1669523FOS NM_005252.2 <0.0053.7
ILMN_1751607FOSB NM_006732.1 <0.0052.2
ILMN_1747759WSB1 NM_134264.2 <0.0051.9
ILMN_1720829ZFP36 NM_003407.2 <0.0051.7
ILMN_1760556C1ORF63 NM_207035.1 <0.0051.7
ILMN_2415748WSB1 NM_134265.2 <0.0051.7
ILMN_2054297PTGS2 NM_000963.1 <0.0051.7
ILMN_1781285DUSP1 NM_004417.2 0.011.6
ILMN_1670572IDO2 NM_194294.2 0.041.5
ILMN_1737561LOC88523 NM_033111.2 0.031.5
ILMN_1682636CXCL2 NM_002089.3 0.011.5
ILMN_2195821C5ORF41 NM_153607.1 <0.0051.4
ILMN_1664802WSB1 NM_134265.2 0.011.4
ILMN_1712798ZNF608 NM_020747.2 0.031.4
ILMN_1721621NKTR NM_005385.3 0.051.4
ILMN_1794638VIP NM_194435.1 0.031.4
ILMN_1774901GDPD3 NM_024307.2 0.041.4
ILMN_1710284HES1 NM_005524.2 0.021.4
ILMN_1657870ABL2 NM_005158.3 0.031.4
ILMN_2096970MYO3B NM_138995.1 0.041.3
ILMN_1751234C1GALT1C1 NM_152692.3 0.04–1.4
ILMN_1746972LOC653321 XM_932385.1 0.01–1.4
ILMN_2367469CARS NM_001014438.1 0.03–1.5
ILMN_1733559LOC100008589 NR_003287.1 0.03–1.5
ILMN_2287276FAM177A1 NM_173607.3 <0.005–1.6
Table 2. Differential gene expression: pre-eclamptic pregnancy.
Probe IDGene nameAccession numberFDR PE for 16 vs 28 weeks of gestationFold change
  1. Human umbilical vein endothelial cells (HUVECs) were incubated with plasma from women who subsequently developed pre-eclampsia (PE), obtained at 16 and 28 weeks of gestation. Differentially expressed genes, their false discovery rates (FDRs) and fold changes from microarray experiments are displayed.

ILMN_1821517HS.508682 AV762101 <0.0051.9
ILMN_2181432SPC24 NM_182513.1 0.011.8
ILMN_1881909HS.579631 BU536065 0.011.7
ILMN_1704537PHGDH NM_006623.2 0.031.7
ILMN_1717793C19ORF33 NM_033520.1 0.041.6
ILMN_1672650PKM2 NM_002654.3 0.041.5
ILMN_1688500PCDH10 NM_032961.1 0.03–1.7
ILMN_1715458PCDH10 NM_020815.1 0.04–1.7
ILMN_1782922PDE4B NM_002600.3 0.03–1.8
ILMN_1672124C4ORF18 NM_016613.4 0.01–1.9
ILMN_2396672ABLIM1 NM_001003407.1 <0.005–2.0

With FOS identified as a potential candidate gene, analyses of unpaired microarray data were performed, and revealed 2.2-fold higher FOS expression in HUVECs incubated with control plasma obtained at 28 weeks of gestation, as compared with HUVECs incubated with plasma isolated from cases at the same time point (FDR < 0.001; Table S2). A 1.6-fold higher expression of FOS was observed in cells incubated with plasma obtained from cases at 16 weeks of gestation, as compared with cells incubated with control plasma from the same time point (FDR = 0.002; Table S3).

Validation of differential expression by qRT-PCR

The upregulation of endothelial FOS mRNA expression observed upon incubation of HUVECs with control plasma obtained at 28 weeks of gestation, as compared with the incubation of HUVECs with control plasma obtained at 16 weeks of gestation, was validated by TaqMan® qRT-PCR (Figure 1A). Also in concordance with findings of the microarray study, qRT-PCR revealed no differential expression of FOS mRNA in HUVECs incubated with plasma obtained at 16 or 28 weeks of gestation from cases (Figure 1B). We confirmed increased FOS mRNA expression in cells incubated with plasma obtained from cases, compared with controls, obtained at 16 weeks of gestation (Figure 1C). In contrast, no significant difference in FOS mRNA expression was observed between cells incubated with plasma obtained at 28 weeks of gestation from cases and controls (Figure 1D).

Figure 1.

FOS gene expression in qRT-PCR experiments. Expression of FOS relative to β-actin (ACTB) was investigated in human umbilical vein endothelial cells stimulated with plasma obtained at 16 and 28 weeks of gestation from women with healthy pregnancies (HPs) or from women who developed pre-eclampsia (PE). The inline image comparative method was used for the relative quantitation (RQ) of gene expression. Scatter plots of individual data are shown. (A) Comparison within the HP group at 16 and 28 weeks of gestation (n = 5, = 0.022). (B) Comparison within the PE group at 16 and 28 weeks of gestation (n = 4 or 5, = 1.000). (C) Comparison between HP and PE groups at 16 weeks of gestation (n = 5, = 0.022). (D) Comparison between HP and PE groups at 28 weeks of gestation (n = 4 or 5, = 0.111).

Pathway analysis

We performed pathway analysis on the data set obtained from control plasma at 16 and 28 weeks of gestation. The analysis was centred on FOS, which was the gene with the highest fold change in this experiment. Differential expression of genes upstream and downstream of FOS in our data set is illustrated in Figure S3, and a speculative and focused pathway is shown in Figure 2.

Figure 2.

 Speculative FOS gene-centred pathways in normal pregnancy. Genes in which expression was upregulated when human umbilical vein endothelial cells were incubated with plasma from women with normal pregnancies obtained at 28 weeks of gestation, compared with plasma from 16 weeks of gestation, are depicted in the shaded symbols. Fold changes are given in Table 1, and more detailed networks are provided in Figure S3. The rectangle illustrates intranuclear processes, namely regulation of transcription by c-Fos and FOSB. The upstream activation of vascular endothelial growth factor receptor-1 (VEGFR-1, Flt-1) by vascular endothelial growth factor (VEGF) and placental growth factor (PLGF) are also shown. Pre-eclamptic pregnancy is characterised by reduced PLGF expression and increased levels of soluble Flt-1, which further contributes to the reduced activation of Flt-1. Note that the figure is based on microarray data showing subtle but statistically significant differences in gene expression affecting this regulatory pathway.

Discussion and conclusion

We found that HUVEC FOS expression was markedly increased upon incubation of cells with plasma from women with normotensive pregnancy at 28 weeks of gestation, compared with plasma from the same women at 16 weeks of gestation. c-Fos is a member of a family of transcription factors that are involved in the regulation of angiogenesis.21 Moreover, members of the Fos family can dimerise with c-Jun and form the activating protein-1 (AP-1) family of transcription factors, known to regulate the transcription of genes involved in proliferation and differentiation.22 It appears plausible that these processes contribute to placentation, which itself involves extensive vasculogenesis.23 Our data suggest that plasma from women who are in the process of developing pre-eclampsia contains factors that lead to the dysregulation of the c-Fos pathway, and that this dysregulation may occur before 16 weeks of gestation.

Impaired remodelling of uterine spiral arteries is one of the hallmarks in the pathogenesis of pre-eclampsia and leads to the secretion of placenta-derived factors into the maternal circulation that cause systemic responses, including endothelial dysfunction and renal failure.1 Expression of one of the most important angiogenic factors in normal pregnancy, placenta growth factor (PLGF), is decreased in women who develop pre-eclampsia.24 Lower levels of PLGF in maternal plasma, and thereby reduced activation of the VEGFR-1 (Flt-1), could be responsible for the reduced stimulation of FOS expression in our study. Another mechanism leading to reduced VEGFR-1 activation is competition of VEGFR-1 with soluble Flt-1 (sFlt-1). Increased levels of sFlt-1 are consistently observed in pre-eclampsia, even at stages in pregnancy where the full clinical syndrome has not yet developed (Figure 2).24,25 In plasma samples of PIP study participants, we have previously demonstrated increased sFlt-1 levels at 28 weeks of gestation in women who subsequently develop pre-eclampsia.26 These mechanisms could explain why we found differential FOS regulation in our study of women with late-onset pre-eclampsia, although the links between impaired vasculogenesis and development of the syndrome may be even tighter in early-onset pre-eclampsia.

Previous human placenta gene and protein studies have indicated a role for c-Fos in the regulation of placentation. Marzioni et al. 27 recently described a correlation between c-Fos protein expression and trophoblast invasiveness in normal pregnancy, and a dissociation between FOS gene and c-Fos protein expression, especially in pre-eclampsia with fetal growth restriction. In an earlier study, Faxén et al.28 also observed FOS mRNA upregulation in placental tissue of women with pre-eclampsia with fetal growth restriction. We used a longitudinal prospective design to perform pairwise comparisons in order to reduce the ‘noise’ in unpaired comparison of human gene expression data. The differences between stimulation with plasma from gestational week 28 versus gestational week 16 within each group (cases and controls) are therefore statistically more robust than the comparisons at the same time points between groups. The apparent higher expression of FOS at 16 weeks of gestation in cases compared with controls could, however, indicate that FOS is indeed upregulated in early pregnancy in women who subsequently developed pre-eclampsia, with no further increase during pregnancy being possible. Rather than demonstrating a failure of upregulation of FOS in women with pre-eclampsia, our data therefore should be interpreted as dysregulation of the c-Fos signalling pathway in pre-eclampsia compared with normotensive pregnancy. We appreciate that the process leading to pre-eclampsia may be well under way at 16 weeks of gestation, and that we may not find all pathogenetic factors with the design of our present study. Differential regulation of a whole pathway may explain the subtle changes in expression of individual genes that act together to cause downstream effects.

Our study in endothelial cells differs from both these and other studies that have examined placental gene expression.29,30 We have specifically investigated the effect of maternal plasma on endothelial cell gene expression, and from our data one cannot immediately draw conclusions on unstimulated or baseline gene expression. It is also important to note that HUVECs are fetal cells, and although they are widely used as models for other endothelial cells, our data may not be directly transferable to the maternal endothelium. It is a strength of our study that cases and controls were matched for body mass index, as obesity in pregnancy, independent of pre-eclampsia, impairs endothelial function.31 We are aware that the small sample size is a limitation of our study. Furthermore, it should be highlighted that most of our interpretation is based on the microarray data, which we could not fully confirm in qRT-PCR experiments. Although global gene expression studies are useful tools to develop hypotheses, we acknowledge that we have not rigorously validated and explained our findings.

Findings of our study are not in line with the previous data of Donker et al.14, who failed to find significant differences in HUVEC gene expression upon incubation of cells with plasma from women with either normal pregnancy or overt pre-eclampsia. This is perhaps a result of limitations of the earlier study, wherein cells were from pooled donors, plasma samples were obtained at only one time point (around 28 weeks of gestation) and a more stringent cut-off value of 2.7-fold differences in expression were used in their analysis. They did, however, identify (by qRT-PCR) a modest induction of IL8 gene expression in cells incubated with plasma from women with pre-eclampisa, which is in agreement with the results from the present study (Table S2). Advances in technology and data analysis, in addition to reduced costs, have made our more complex experiment possible. Our two time points, well before the clinical diagnosis of pre-eclampsia, may have further helped to discover pathways that are involved in earlier stages of the disease.

We have demonstrated that a relatively simple in vitro model can provide insight into the regulation of endothelial cell gene expression during pregnancy. Our study identified targets within c-Fos-related pathways that are altered in pre-eclampsia, as compared with normal pregnancy. These targets require further molecular and functional studies to dissect their role in the pathogenesis of pre-eclampsia. Although our data point towards a critical role of the c-Fos pathway in the development of pre-eclampsia, it is too early to propose therapies based on this principle. Irrespective of the functional relevance, a test that comprehensively studies factors in maternal plasma, and their effect on endothelial cell gene expression, could aid in the early diagnosis of pre-eclampsia. More comprehensive strategies using transcriptomic, proteomic or metabolomic methods could lead to better predictive and diagnostic tests in the future.15,32

Disclosure of interests

None to declare.

Contribution to authorship

RMM and VCS contributed equally to this article. RMM designed and performed gene expression experiments; VCS had the idea to perform this study and designed the experiments; DMC recruited patients into the PIP study; JDM designed the experiments and analysed the microarray data; DJF critically reviewed the study design and interpreted the data; AFD contributed to the design of the PIP study and provided funding for the present work; MWM designed and performed gene expression studies, and analysed the data; CD designed the PIP study and the present work, provided funding and had overall responsibility for the project. The article was drafted by RMM and CD, and all authors critically revised the article.

Details of ethics approval

The study was approved by the West of Scotland Research Ethics Committee 2 (REC reference 07/S0709/79; R&D reference WN07CA022), and complies with the principles of the Declaration of Helsinki.

Funding

This work was supported by a Strategic Research Development grant from the Scottish Funding Council to AFD, an NHS Greater Glasgow and Clyde Endowment Award 07REF003W to CD and the European Commission’s 7th Framework Programme grant ‘EURATRANS’ to AFD and CD. VCS was supported by Fundação de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG; Brazil), the British Council Exchange Programme and the L’Oréal-UNESCO-ABC ‘For Women in Science’ grant.

Acknowledgements

We would like to thank Miss Wendy Crawford and Dr Wai Kwong Lee for their help with the microarray experiments.

Ancillary