Abstract
The meeting was opened with a fascinating state of the art lecture from AM Carter on the evolution of placentation. Using a comparative physiology approach, the differences in maternal–fetal interface encountered across the animal kingdom served to illustrate the contribution of each cell type to placental function and enlighten our understanding of human placental anatomy. Several other state of the art lectures were delivered over the next few days.
C Weiner gave a lecture entitled ‘From genomics to proteomics’, in which he addressed the use of novel molecular techniques to tackle the old problem of preeclampsia. Genomics and proteomics allow a large number of variables to be studied at once, usually in a relatively small number of samples. DNA microarrays are invaluable discovery tools with acceptable reliability for genome-wide gene expression. The arrays are used for screening candidate molecules that may be involved in the etiology of preeclampsia and positive results are confirmed with quantitative polymerase chain reaction (PCR). Proteomic approaches are more problematic. The set of 38,000 human genes code for over a million proteins, as a result of alternative splicing and post-translational modifications. Unfortunately, protein abundance does not always correlate with gene expression. Sophisticated bioinformatics analysis programs can attempt to identify the function and localization of proteins by analyzing expression patterns of subsets of proteins. The effects of neurokinin B, which is secreted from the placenta in abnormally high amounts in pre-eclampsia, have been investigated using proteomic approaches. The expression of proteins involved in antioxidant defences and anticoagulation is decreased in cultured cytotrophoblast cells exposed to neurokinin B. Surface-enhanced laser desorption/ionization (SELDI) techniques have been used to identify biomarker profiles in the cere-brospinal fluid of women with preeclampsia. Hemoglobin α- and β-chains were identified as important discriminators of preeclampsia. More recent studies have identified α1 antitrypsin in urine as a biomarker for preeclampsia. This protein is highly susceptible to oxidation, which induces fragment generation, and hence its detection may indicate levels of oxidative stress.
CBM Oudejans gave a state of the art lecture on the epigenetic inheritance of preeclampsia. He suggested that the clinical presentations of early- and late-onset preeclampsia imply different genetic origins for these diseases. Early onset (<33 weeks gestation, with associated intrauterine growth restriction [IUGR] and no evidence of preexisting disease) is familial and the pathology seems to be primarily placental, i.e., there is abnormal placentation in the presence of a normal maternal constitution. The primary genetic lesion may be a defect in extravillous trophoblast invasion. Late-onset preeclampsia (>33 weeks gestation, no IUGR, associated with overt diabetes, age and high body mass index [BMI]) is not familial and the underlying pathology is likely to be of maternal origin i.e., there is normal placentation but a predisposing maternal constitution. In this latter scenario it is likely that there are many hidden cis- or transacting genomic variants. Evidence from monozygotic twins shows that when one twin develops hemolysis, elevated liver enzymes and low platelets (HELLP) in pregnancy the other one does not. This suggests that there is some paternal input into the genetic etiology of HELLP. Two recent studies have attempted to understand the genetics of preeclampsia. The Dutch study used sibling-pair analysis to identify susceptibility loci on chromosomes. Using this approach a minimal critical region of 444 kb was identified. STOX-1 was identified as a gene that cosegregated with preeclampsia and it followed a matrilineal inheritance. There are four STOX-1 transcripts expressed in early placenta, including the extravillous trophoblast. Three isoforms of the protein exist, all containing a winged helix domain, and they are DNA-binding transcription factors. STOX-1 expression is restricted to giant cells and it is hypothesized that its expression leads to premature differentiation of trophoblasts into giant cells and loss of invasion. The Genetics Of Pre-EClampsia (GOPEC) study is investigating maternal and fetal genetics in relation to preeclampsia. No positive candidates have yet been identified.
Related to the previous topic, elsewhere in the conference there was a lively debate, ably chaired by M Brown, on whether early- and late-onset pre-eclampsia are the same disease. A pre-debate show of hands indicated that most of the audience believed they were different diseases. JJ Duvekot kicked off the debate by arguing that early- and late-onset preeclampsia are just elements of the same continuum of disease. He argued that there is a gradual increase in the cumulative incidence of preeclampsia with gestational age with no kink in the curve and therefore there are not two distinct groups. Furthermore there is a gradual increase, as gestation increases, in the birthweight of babies born to preeclamptic mothers. He proposed that preeclampsia can be explained by a universal hypothesis of an inadequate fall in vascular tone during pregnancy with or without an inadequate restoration of blood volume. This can lead to failure of trophoblast invasion and IUGR (especially when there is pathology associated with the conceptus or embryo). Alternatively, it can lead to a hyperdynamic circulation, which can result in maternal decompensation and oxidative stress, ending in preeclampsia. Smoking protects against preeclampsia by lowering blood pressure and providing an initial decrease in vascular tone. The counter argument was delivered by P von Dadelszen, often using the same data as JJ Duvekot and making some of the same points as CBM Oudejans above. He proposed that preeclampsia is an important adaptive response to an abnormal pregnancy resulting in lower birthweight for gestational-age babies. There is a point of inflection at 31 weeks in the cumulative incidence of preeclampsia with gestational-age curve, suggesting there are two disease groups. Although P von Dadelszen won the vote on the day, Duvekot had convinced some members of the audience as the margin was decreased from the initial count.
A plenary session on obesity and metabolic syndrome and subsequent maternal health joined the International Society for the Study of Hypertension in Pregnancy (ISSHP) meeting with the International Society of Obstetric Medicine. The speakers in this session discussed the impact of preeclampsia on future maternal disease and included an overview of the epidemic of obesity and its potential impact on adverse pregnancy outcome. D Williams reported data from a systematic review of the impact of preeclampsia on future maternal health. There is an approximately twofold increase in ischemic heart disease, a 1.8-fold increase in stroke, a 2.4-times increased risk of hypertension and a 1.79-times increased risk of venous thromboembolism (VTE) associated with a history of preeclampsia. The Cardiovascular Health After Maternal Placental Syndromes (CHAMPS) was a population-based, reterospective study of 1.03 million Canadian women free from cardiovascular disease before their first documented pregnancy. The study concluded that risk of premature cardiovascular disease is higher after a maternal placental syndrome and suggested that affected women should have their blood pressure and weight assessed 6 months' postpartum and a healthy lifestyle should be advised. Of interest, another study showed that an adverse pregnancy outcome is associated with an increased risk of having a renal biopsy later in life, indicating that women with preeclampsia may be at increased risk for future renal disease. N Sattar highlighted the alarming rise of obesity in the western world and discussed the impact of this epidemic on pregnancy outcome. In particular, a higher body mass index (BMI) is associated with the development of preeclampsia. Drawing parallels with research into metabolic syndrome and Type 2 diabetes, N Sattar pointed out that in overweight individuals, lipid does not only accumulate in adipose tissue but also collects in muscle and liver, leading to insulin resistance and exacerbation of the metabolic defects. This may have parallels in the raised liver enzyme levels observed in preeclampsia. There is also evidence that fat can accumulate around blood vessels, leading to increased local cytokine secretion and endothelial and vascular dysfunction. Smoking may protect against preeclampsia by redistributing fat to other tissues. S Kjos discussed which are the best predictors for future development of diabetes after gestational diabetes. These include area under the curve of an oral glucose tolerance test, gestational age at development of gestational diabetes and fasting glucose. A postpartum assessment of oral glucose tolerance was suggested with follow-up of an oral glucose tolerance test and a fasting plasma glucose every 1–3 years. In addition, fasting lipid profile should be checked and a healthy lifestyle encouraged, including weight loss, daily exercise and planned pregnancies.
In a workshop on oxidative stress and inflammatory response L Poston discussed the implications for the oxidative stress hypothesis of preeclampsia of the recent disappointing result of the Vitamins in Pregnancy (VIP) multicenter trial. The trial demonstrated that supplementation with vitamins C and E does not improve pregnancy outcome and may have resulted in an increase in low birthweight babies. The outcome in terms of the oxidative stress hypothesis may be interpreted in a number of ways and L Poston addressed each of these in turn. First, there may be no oxidative stress in preeclampsia. However, the women in the placebo group who did get preeclampsia had lower vitamin C than those who did not. Plasma malondialde-hyde levels were higher in those women who developed preeclampsia and neutrophils in women who got preeclampsia had an increased rate of superoxide generation. Second, vitamins did reduce oxidative stress but this had no effect on pregnancy outcome. Vitamins C and E treatment did reduce plasma levels of uric acid (a by-product of the action of xan-thine oxidase), which implies that they did reduce oxidative stress. Therefore, one could conclude that in the presence of agents that clearly reduce oxidative stress there is no effect on outcome and hence, oxidative stress does not cause preeclampsia. Third, it is possible that the wrong sort of antioxidants were used. There are parallels in the cardiovascular field where the use of antioxidant vitamins to prevent cardiovascular disease has repeatedly failed in large preventions trials to show any effect of vitamin C and E on cardiovascular outcome. However, in this field this has not been interpreted as oxidative stress having no role in the etiology of the disease as there is a large body of evidence in the literature that suggests it does have a role. By contrast, it has been interpreted to mean that the right antioxidants have not been used and, in fact, intracellular antioxidants rather than systemic antioxidants will be more effective. There is also evidence from the cardiovascular field that large doses of vitamin E upset the balance between α- and γ- tocopherol, leading to dramatic and detrimental reductions in γ tocopherol. Furthermore, there is evidence that reduced glutathione levels inhibit the ability of the hexose mono-phosphate shunt pathway to generate nicotinamide adenine dinucleotide phosphate-oxidase (NADPH), thereby impeding the ability to recycle ascorbic acid. In support of this explanation it was found in the VIP trial that antioxidant vitamin treatment had no impact on neutrophil superoxide generation. Finally, it is possible that the timing of vitamin treatment was inappropriate since oxidative damage occurs early in pregnancy and is irreversible. However, the data from the VIP trial suggest that there was less preeclampsia the later the treatment was started. In summary, it is likely that the type of antioxidant treatment was not optimal in the VIP trial. In the same session L Linton discussed the role of caveolin-1 in the trophoblast response to oxidative stress. Caveolin is a target for oxidative stress and in some cells it sensitizes the cell to apoptosis but inhibits apoptosis in other cells. Caveolae are present in villous trophoblasts but not syncytiotrophoblasts at term, and caveolin expression has been shown to be decreased during in vitro syncytialisation. BeWo cells can be induced to syncytialise in response to stimulation by some agents. In subconfluent BeWo cells, caveolin localizes to the mitochondrion but relocalizes to the plasma membrane on confluence. This model should prove useful in elucidating the role of caveolin in the trophoblast and its response to oxidative stress.
Another workshop addressed the role of immunology in preeclampsia. S Saito discussed the Th1/Th2 switch in pregnancy and the role of T-helper cells. Memory T-cells have a lifetime of 10 years and this may explain why the length of time to the next pregnancy increases the risk of preeclampsia. A delicate balance between all T cells, T regulatory cells as well as T helper cells, is required for healthy pregnancy. The most extreme immunological response is when T regulatory cell activity is reduced and toll-like receptor activation provokes Th1 cells. Normal pregnancy is associated with increased Tregulatory and Th2 cell function, whereas preeclampsia is associated with decreased Th2 and T regulatory cell function and increased Th1 cell function. J Morris discussed the importance of nuclear factor κB (NFκB) in the Th1 response and allograft rejection. However, his group found evidence that the entire NFκB pathway in peripheral blood mononuclear cells was suppressed in pregnancy. In preeclampsia, NFκB activity was even more downregulated. In monocytes NFκB was upregulated and even more so in preeclampsia. Thus, the involvement of the immune system in pregnancy via NFκB activation is complex. CWG Redman proposed that it is not just T helper cells but whole Type 1 or Type 2 immune responses that are important in pregnancy. Natural killer cells are part of the innate (inflammatory) system and have an important role. The immune interfaces that exist between the mother and baby include:
Placenta/decidua (tissue/trophoblast);
Villous trophoblast/maternal blood;
An extended interface 2 resulting from fragmented syncytiotrophoblast membrane fragments shed into the maternal circulation.
Interface 1 is important in early pregnancy and poor placentation can lead to early pregnancy loss. Cytotro-phoblasts express human leukocyte antigens (HLAs) and these are probably important in establishment of pregnancy. Interfaces 2 and 3 are important for preeclampsia. The syncytiotrophoblast does not express HLAs and hence these antigens are not important in late pregnancy and preeclampsia. In a workshop on early placentation, S Hahn gave an interesting talk on the role of neutrophil extracellular traps (NETs) in harbouring shed syncytiotrophoblast membrane material in preeclampsia. On activation by inflammatory signals, neutrophils generate extracellular DNA containing fibrous lattices. It was proposed that these NETs may contribute to the increased maternal circulatory cell-free DNA seen in preeclampsia.
Summary
The discussions regarding classification and diagnosis of the complex disease preeclampsia continue. New genetic, genomic and proteomic approaches are beginning to identify novel candidate genes and proteins that may be involved in the etiology of preeclampsia, such as neurokinin B and STOX-1. Preeclampsia is associated with increased risk of future maternal vascular and renal disease. The approaching obesity epidemic is likely to have a huge impact on pregnancy complications, including the incidence of preeclampsia. The oxidative hypothesis of the etiology of preeclampsia has suffered a setback with the failure of the antioxidant vitamin trials in preeclampsia. However, parallels with cardiovascular disease research would suggest that intracellular antioxidants might prove more effective. Understanding the role of inflammation at the maternal–fetal interface remains an important goal in pregnancy research.
Future perspective
Collaboration between researchers to generate large biobanks of DNA and tissue will help to resolve issues of classification and different disease pathways in preeclampsia. It is likely that, again with cooperation, “omics” approaches will start to bear fruit regarding understanding preeclampsia. Problems surrounding the technology for proteomics and metabonomics and the processing of the large amounts of information generated will be addressed. Research into the oxidative stress theory of preeclampsia will increase rather than decline as new antioxidant approaches are considered. Sadly, the increasing rates of obesity in adolescents and children would indicate that we are too late to prevent the adverse impact of obesity on pregnancy outcome in the coming child-bearing generation.
Highlights
Recent research has focused on genomic and proteomic techniques that allow the screening of large numbers of candidate molecules that are potentially involved in preeclampsia.
A history of preeclampsia increases a woman's subsequent risk of ischemic heart disease, stroke, hypertension and venous thromboembolism. Blood pressure and weight should be assessed 6 months postpartum and advice given regarding following a healthy lifestyle.
The oral glucose tolerance, fasting plasma glucose and fasting lipid profile of women with a history of gestational diabetes should be assessed postpartum and every 1–3 years thereafter.
Supplementation with the antioxidant vitamins C and E does not improve pregnancy outcome and may result in an increase in low birthweight babies.
The approaching obesity epidemic will have a large impact on the clinical management of pregnancy and its complications.