Abstract
A clinically effective screening program for the major placenta-related complications of pregnancy (preeclampsia, fetal growth restriction, and abruption) is an area of unmet clinical need. A major hurdle to the implementation of screening is the lack of highly discriminative screening tests. The transcriptome of the placenta demonstrates characteristic changes in preeclampsia and analysis of these changes led to the identification of proteins in the mother’s blood which were associated with the risk of preeclampsia. However, the placenta also releases mRNA into the maternal blood, chiefly packed in microvesicles, and analysis of circulating mRNAs identifies transcripts of placental origin. It is plausible that measurement of specific transcripts (e.g., using PCR-based approaches) or unbiased analysis of transcripts (using plasma RNA-seq) might shed light on the mechanisms of disease and that disease-specific changes in the plasma RNA profile could act as a predictive or diagnostic test for placenta-related complications.
Keywords
Preeclampsia, Fetal growth restriction, Placental abruption, RNA-seq, Plasma, Microvesicle, Placenta
Introduction
Adverse pregnancy outcomes are a major cause of the global burden of disease. Many of these outcomes, such as fetal growth restriction (FGR), preeclampsia, and placental abruption, are related to placental dysfunction. These complications can have profound effects on the mother and child. Preeclampsia is one of the leading causes of maternal death globally . Placental abruption is also associated with hemorrhage , which is another major cause of death. The mother may experience morbidity related to interventions performed in the fetal interest, such as induction of labor or cesarean section. Placental dysfunction can have multiple adverse effects on the fetus. It can lead to acute and chronic fetal hypoxia resulting in severe short-term complications, such as perinatal death or brain damage through asphyxia . It is also recognized that antepartum and intrapartum events are major determinants of long-term adverse neurodevelopmental outcome for the baby. Moreover, placental dysfunction and the management of certain conditions can lead directly to preterm birth , which is a major determinant of mortality and long-term morbidity in the offspring. There is a large body of evidence that a suboptimal intrauterine environment may predispose to a range of diseases in adult life, such as ischemic heart disease and stroke . Finally, placenta-related complications of pregnancy are a marker for later risk of cardiovascular disease in the mother , indicating possible targeted interventions to reduce cardiovascular risk to women experiencing these complications of pregnancy.
Despite these issues, screening and intervention to prevent placenta-related complications of pregnancy remain relatively unsophisticated. The current primary approach to screening for FGR in the United Kingdom and United States is serial measurement of the external size of the uterus (the “symphyseal-fundal height” [SFH]). This reflects the fact that more advanced approaches, such as universal ultrasonic fetal biometry, have not been shown to improve outcomes in meta-analyses of randomized controlled trials . This failure may, in turn, be explained by the fact that purely imaging-based methods do not perform well as a screening test. An ultrasonic estimated fetal weight (EFW) is a crude estimator of fetal size, commonly with errors of > 15% . Moreover, relative fetal smallness for gestational age (SGA) is a proxy for the condition of interest, namely, FGR.
We and others have proposed that measurement of circulating markers of placental function may result in improved clinical risk assessment and, consequently, their incorporation into future trials of screening and intervention may result in better clinical outcomes. The current model for approaching this ( Fig. 1 ) uses analysis of placental RNA to identify differentially expressed transcripts. Assays of the encoded proteins are then evaluated as a screening test. However, with the advent of the ability to analyze DNA and RNA derived from the placenta in the maternal circulation, there is now the potential for identification and assessment of placental biomarkers directly in the maternal blood without the need to study the placenta first. In the present review, we outline the major conditions which could be screened for, the existing evidence for the role of the placenta in their pathogenesis, and the existing evidence for the effectiveness of this approach in the diagnostic and screening pathways.
Adverse Pregnancy Outcome due to Placental Dysfunction
Preeclampsia
Preeclampsia is the manifestation of high blood pressure and multisystem disease (typically renal dysfunction resulting in proteinuria) in pregnancy. The disease can either occur de novo or be “superimposed” on preexisting hypertension and/or renal disease. Maternal risk factors for preeclampsia include nulliparity and obesity. Interestingly, the risk of the disease is lower in smokers. Preeclampsia is a major determinant of maternal and perinatal morbidity and mortality. The disease is commonly associated with FGR, but the two conditions can also exist in isolation. Management of the disease involves monitoring and treatment of its symptoms (e.g., antihypertensives) but effective treatment is, currently, only achieved by delivery.
Fetal Growth Restriction
FGR is a theoretical concept of failure of the fetus to achieve its genetically determined growth potential. The condition has no gold standard. Frequently the babies are small for gestational age, that is, < 10th percentile for sex and week of gestation. However, some cases of FGR may not be SGA and many cases of SGA are not FGR. Corroborative indicators of FGR include the presence of ultrasonic markers (such as abnormal uteroplacental blood flow indices or slowing down of fetal growth), acquired complications of pregnancy (including preeclampsia and preterm birth), and morbidity due to chronic fetal hypoxia. FGR is one of the major causes of stillbirth.
Placental Abruption
Placental abruption is premature separation of the placenta. Placental detachment from the uterine wall normally occurs following delivery of the baby. Abruption is partial or complete detachment when the baby is still in utero. As the placenta is the site of gaseous exchange, abruption leads to fetal asphyxia with the degree of asphyxia related to the proportion of the placenta which has detached. Typically, the separation extends to the edge of the placenta resulting in blood loss vaginally. The abruption can also be “concealed,” that is, the area of detachment is entirely within the placental perimeter and the condition is not associated with vaginal bleeding. The process of abruption can lead to a maternal coagulopathy and disseminated intravascular coagulation. Abruption is a recognized cause of maternal death and accounts for about 10% of stillbirths.
The Placental Transcriptome in Adverse Pregnancy Outcome
The placenta ensures proper fetal growth and development via its endocrine functions, which sustain maternal adaptation to pregnancy and mobilize resources for the fetus, and by mediating feto-maternal exchange of nutrients, oxygen, and waste products. As the interface between the mother and the fetus, the placenta has also a protective role, representing a barrier for pathogens and toxic substances. Due to its role in orchestrating so many fundamental processes during pregnancy, it is not surprising that alterations in placental development, structure, and function are closely associated with poor pregnancy outcome. Both preeclampsia and FGR have a complex etiology, involving various maternal and fetal responses, and placental alterations. This complexity leads to a wide heterogeneity in the case population that, in turn, makes biomarker discovery particularly challenging. Analysis of the placental transcriptome offers a valuable approach to investigate multifactorial placenta-related complications such as preeclampsia and FGR, as it allows to study several potential components at the same time.
There have been many more studies of the placental transcriptome in preeclampsia than in FGR. Meta-analyses of expression microarray studies using placental biopsies from women with preeclampsia, compared with controls, have demonstrated altered placental levels of messenger RNAs (mRNAs) coding for pro- and antiangiogenic factors, such as fms-like tyrosine kinase ( FLT1 ), endoglin ( ENG ), vascular endothelial growth factor A ( VEGFA ), placenta growth factor ( PlGF ) and mir-126, and upregulation of transcripts indicating tissue hypoxia, including hypoxia-inducible factor 2α ( HIF2α ) and mir-210 . Other transcripts frequently identified are involved in cellular growth, metabolism, and endocrine function: pregnancy-associated plasma protein A ( PAPP-A ), insulin-like growth factor binding protein 1 ( IGFBP1 ), HtrA serine peptidase 1 ( HTRA1 ), prolactin ( PRL ), leptin ( LEP ), corticotropin releasing hormone ( CRH ), inhibin subunits ( INHA and INHBA ), and follistatin-like 3 ( FSTL3 ). Several of these genes are also implicated in defective cytotrophoblast invasion of the maternal decidua and impaired spiral artery remodeling, hallmarks of the preeclamptic placenta and frequently observed in other placenta-related complications ( Table 1 ).
| Gene Name | Expression in Preeclampsia | Number of Publications | |
|---|---|---|---|
| Gene symbol | |||
| LEP | Leptin | Upregulated | 12 |
| FLT1 | fms-like tyrosine kinase 1 | Upregulated | 11 |
| INHBA | Inhibin β A subunit | Upregulated | 9 |
| ENG | Endoglin | Upregulated | 8 |
| INHA | Inhibin α subunit | Upregulated | 6 |
| SIGLEC6 | Sialic acid binding Ig-like lectin 6 | Upregulated | 6 |
| BCL6 | B-cell CLL/lymphoma 6 | Upregulated | 5 |
| CGB | Chorionic gonadotropin β subunit | Upregulated | 5 |
| HTRA1 | HtrA serine peptidase 1 | Upregulated | 5 |
| FSTL3 | Follistatin-like 3 | Upregulated | 4 |
| miRNA name | |||
| hsa-miR-210 | Upregulated | 3 | |
| hsa-miR-1 | Downregulated | 2 | |
| hsa-miR-139-5p | Downregulated | 2 | |
| hsa-miR-150 | Downregulated | 2 | |
| hsa-miR-181a | Upregulated | 2 | |
| hsa-miR-542-3p | Downregulated | 2 | |
| hsa-miR-625 | Downregulated | 2 | |
| hsa-miR-638 | Upregulated | 2 | |
Understanding the placental pathological processes associated with pregnancy diseases might offer a chance for intervention and improved outcome. But the placental undergoes profound changes during gestation and it is only available for analysis during pregnancy via invasive procedures, such as chorionic villus sampling. Nevertheless, these tissues offer the chance to study placental gene expression early in gestation. Similarly, to what observed in the term preeclamptic placenta, altered levels of mRNAs encoding regulators of angiogenesis, trophoblast invasion, oxidative stress, and inflammation were measured in first trimester biopsies collected from women who subsequently developed preeclampsia . Therefore studies conducted on first trimester biopsies support the idea that placental dysfunction leading to complications in the second half of pregnancy often has its origins early in gestation . As these types of samples are not easily accessible, attention has focused on circulating maternal biochemical markers of placental insufficiency to address early detection of pregnancy diseases, but several reports describe the low predictive value of the majority of the candidate proteins analyzed . Hence, in more recent years circulating nucleic acids have being investigated, with the hope that the more sensitive techniques available for the detection of these molecules (targeted PCR-based methodologies and, lately, unbiased massively parallel sequencing (MPS)) would lead to screening tests with improved sensitivity and accuracy. Initial studies focused on transcripts highly expressed in the placenta or involved in the etiology of preeclampsia and FGR, such as those mentioned previously.
Placental Origin of Maternal Circulating Nucleic Acids
Placental nucleic acids are released into the maternal bloodstream bound to specific proteins or packaged in extracellular vesicles (EVs). EVs are primarily categorized based on their size, but they also differ by their cargo and how they are released into the blood stream ( Fig. 2 ). EVs of placental origin include exosomes (with a size of 30–100 nm), microvesicles (0.1–1 μm), and apoptotic bodies (1–5 μm) . In addition, syncytial nuclear aggregates of syncytiotrophoblast cells (20–500 μm) also shed from the placental surface. Routine methods for vesicle isolation are based on their size or density and include differential centrifugation, density gradient ultracentrifugation, microfiltration, and size exclusion chromatography. In order to purify specific vesicles subtypes, immunocapture with antibodies directed against specific surface antigens such as alkaline phosphatase have been used to enrich for EVs originating from the syncytiotrophoblast layer. The choice of what type of EVs to isolate depends on the biological question, as it is now becoming evident that these vesicles not only facilitate removal of cellular waste material, but they also play a role in intercellular signaling and mediate feto-maternal communication. This function is obviously influenced by the EVs content, which includes immunomodulatory proteins (e.g., antiinflammatory cytokines and syncytin-1, regulating maternal immune reaction), vasoactive proteins (e.g., FLT1, ENG and tissue factor, influencing the maternal vascular system), lipids (e.g., cholesterol and sphingomyelins, modulating coagulation, and hemostasis), and nucleic acids (reviewed by Tong and Chamley ). Due to their role in intercellular and feto-maternal communication, the release of EVs is the result of physiological processes and the normal turnover of placental syncytiotrophoblast cells, and their concentration increases with gestational age . In case of placenta-related complications, which are associated with abnormal cellular death and shedding of cellular debris, an increased level of these vesicles is observed. For example, higher maternal plasma concentrations of placenta-derived exosomes were measured throughout gestation in pregnancies with gestational diabetes mellitus (GDM) and in women who subsequently developed preeclampsia as compared to control pregnancies . Therefore the overall abundance of circulating EVs offers a first layer of evidence that altered levels of circulating nucleic acids could be associated with pregnancy complications, and examples of specific feto-placental DNA, RNA, and microRNA (miRNA) molecules are provided later.
Circulating Nucleic Acids and Adverse Pregnancy Outcome
Cell-Free DNA
After the initial discovery of cell-free “fetal”—although more accurately “placental” DNA (cfpDNA) in maternal plasma and serum by Prof. Dennis Lo’s team in 1997, it has been established that cfpDNA fragments represent approximately 3%–10% of the total cell-free DNA (cfDNA) in maternal circulation in pregnancy . cfpDNA molecules can reliably be measured from 5 to 6 weeks of gestation . Their levels increase with gestational age, paralleling the increase in placental size, and there is rapid clearance from the maternal circulation after delivery . Initially cfpDNA measurements were performed by quantitative PCR, aimed at the detection of Y-chromosome genes for fetal sex determination, or Rhesus D ( RHD ) specific sequences for fetal RHD genotyping in RhD − mothers. However, the real breakthrough in cfpDNA-based noninvasive prenatal testing (NIPT) has been the development of massively parallel sequencing (MPS) techniques. These allow accurate counting of fetal DNA fragments in maternal blood and hence the detection of fetal aneuploidies. This application has been a strong incentive to improve cfpDNA-based prenatal testing methods based on placental circulating nucleic acids. Sequencing maternal plasma cfDNA has been particularly successful in Down syndrome and trisomy 18 screening . cfpDNA NIPT is now an attractive alternative to chorionic villus sampling and amniocentesis—procedures disliked by pregnant women and associated with increased risk pregnancy loss .
It is possible that there is an association between cfpDNA levels and pregnancies complicated by preeclampsia and FGR. The rationale is that the placental insufficiency underlying these pathological conditions might lead to an increased release of necrotic and apoptotic fragments containing cfpDNA from the syncytiotrophoblast layer of the placenta into the maternal circulation . In several studies elevated levels of cfpDNA were detected prior to the clinical symptoms of preeclampsia and FGR, fitting the hypothesis that this phenomenon might have a causal role in the onset of the disease Reduced clearance from the maternal circulation might also contribute to higher concentrations of circulating cfpDNA in women with preeclampsia . Studies in mice addressing this issue have been inconclusive .
Although several studies demonstrated higher maternal cfpDNA levels in pregnancies complicated by preeclampsia and FGR compared to normal pregnancies , more recent data are not always consistent with these initial findings . A recent systematic review and meta-analysis reported that the effectiveness of cfpDNA as a biomarker for preeclampsia is promising only when measurements are performed in the second trimester and for cases with early onset of the disease . These issues led the authors to conclude that quantifying total cfpDNA concentrations does not seem to be as useful for preeclampsia screening purposes as other biochemical markers, such as the sFLT1/PlGF ratio . Nevertheless, the authors recognized the heterogeneity and the fragmentary nature of the data presented in the studies included in their review. Contradictory results might also derive from the different DNA markers used, and this reflects once more the need for standardized protocols in studies aiming at identifying potential disease biomarkers.
miRNAs in Maternal Plasma
A comprehensive placental miRNA profile was first reported by Liang and colleagues . The villous trophoblast layer is the main source of placental miRNAs, which are expressed in these cells in a spatiotemporal manner throughout pregnancy . Placental miRNAs are released in the maternal bloodstream bound to proteins and high-density lipoproteins or packaged in EVs (i.e., exosomes, microvesicles, and apoptotic bodies), leading to increased stability of the circulating miRNAs. An increase of these vesicles is detected in maternal blood from complicated pregnancies. Moreover, exosomes isolated from the maternal circulation from GDM pregnancies promote migration and release of proinflammatory cytokines from endothelial cells . This suggests that EVs and their cargo (miRNAs, proteins, growth factors, etc.) not only regulate the function of adjacent cells, but also contribute to feto-maternal communication by targeting distant tissues. Consistent with this, trophoblast-derived exosomes confer viral resistance to nonplacental recipient cells, such as human primary endothelial cells which would represent an important line of defense against viral infections . More recently, Cronqvist and colleagues demonstrated that placental EVs are internalized by primary human endothelial cells. Treatment of cells with placental EVs derived from pregnancies complicated by preeclampsia led to significant downregulation of several transcripts, including FLT1 . This is of particular interest as the upregulation of FLT1 mRNA, which encodes a vascular endothelial growth factor receptor, was identified as one of the key changes in the preeclamptic placenta and this was paralleled by elevated levels of sFLT1 protein in the maternal circulation . Furthermore, inhibition of the FLT1 signaling pathway sensitizes endothelial cells to inflammatory mediators and may contribute to the widespread endothelial dysfunction in PE .
Studies of the relationship between placental or circulating miRNAs and pregnancy complications have focused on candidates with high or specific expression in the placenta and were grouped into three main clusters ( Fig. 3 ) .
