Integrated system biology approaches to fetal medicine problems

Integrated system biology approaches to fetal medicine problems


Jezid Miranda, Fátima Crispi, and Eduard Gratacós


Introduction


Pregnancy is a physiological state that demands intricate and timely hormonal, immunological, metabolic, and physiological adaptations of multiple interconnected molecular and cellular systems. In the past decades, research advances in the molecular pathways implicated in physiological and pathological gestations were achieved by means of the analysis of selected biomarkers using a wide range of available laboratory techniques. However, the introduction of the “omics” technologies has resulted in a revolution in the possibilities for research. The scope of research has expanded, and these approaches offer the means to understand from a much wider perspective the molecular pathways that regulate cells, tissues, and organs. The promise of the omics techniques is to improve radically the understanding of the physiology of pregnancy, the maladaptive processes underlying most of the main obstetrical syndromes, and eventually to enable targeted interventions even in the preclinical phases, long before the clinical onset of the main complications of pregnancy.


However, “omics-based research” is accompanied by new challenges, including the need for new statistical techniques for the interpretation of the massive amount of data available, and the realization that a more complex insight often leads to a more complex reality than we originally imagined. For these reasons, the hypothesis-driven approach that has characterized many studies using omics techniques in the past two decades has not brought us closer to the understanding of the pathogenesis or biomolecular mechanisms implicated in pregnancy complications. A new generation of statistical methods, including bioinformatics and machine/deep learning approaches, has emerged, and with it the evidence that this research is sometimes more in the hands of statisticians and systems biology experts than in those of clinicians. Or, at least, that a strong interaction is required to achieve real advances in the understanding of the mechanisms governing normal and abnormal maternal adaptations and fetal development. In this chapter, we briefly review some examples of the applications of the omics in research of the main complications of pregnancy, and we conclude with some comments and conclusions. The authors of this chapter are clinicians, and consequently, the text here provided cannot and does not intend to address or clarify any technical insight as to the use of these methods, but to illustrate the potential and the caveats to consider on attempting an integrated research approach to pregnancy complications.


Main “omics” methods and their applications in pregnancy complications


Figure 12.1 describes the different questions that omics techniques can answer based on the appraisal employed. In the next paragraphs, we describe successful experiences using techniques for each of the great obstetrical syndromes and the lesson that can be learned from their approaches.



Transcriptomics


The transcriptome is defined as the entire collection of RNA transcripts produced from the genome of a cell in any given state. Transcriptomics techniques are used to describe the global messenger RNA (mRNA) expression of a particular tissue, providing information on how genes are regulated yielding information about which cellular processes are active and which are dormant. Two main techniques lead the field: microarrays, which quantify a set of predetermined sequences, and RNA sequencing (RNA-Seq), which uses high-throughput sequencing to capture all sequences (1). Over the last decade, RNA-seq coupled with computational analysis have led to the recognition that different regulatory noncoding RNAs (microRNAs [miRNAs], long noncoding RNAs [lncRNAs], and circular RNAs [circRNas]) interact with each other and with regulatory proteins to modify gene expression and mRNA to protein translation. Transcriptome analysis has been used to assess placental development and dysfunction in pregnancy complications such as preeclampsia, fetal growth restriction, and preterm labor (24). It is important to highlight that as an alternative to isolated studies with few biological samples, meta-transcriptome analysis of case-control data sets of previous transcriptome studies in preeclampsia (57) have identified consistent alterations in known mediators of clinical symptoms of preeclampsia such as FLT1, endoglin, transcription factor IIH, ATF3, and other genes involved in hypoxia regulation and blood vessel development as drivers of preeclampsia, with the advantages of allowing efficient elimination of false-positive findings pertaining to relatively small study sample sizes, variability in platforms, patient characteristics, and statistical methods used. However, several points should be taken into account in the analysis and interpretation of meta-transcriptomics studies: the tissue origins of the RNA identified in transcriptomics analysis of amniotic fluid cannot be traced; male and female placentas can have different gene expression during development (8); and microarray-based meta-analyses are feasible but have several limitations. Another interesting approach is the use of amniotic fluid supernatant cell-free RNA transcriptome analysis of living fetuses to unravel fetal development (9,10), which can also help to describe molecular pathways altered in fetal conditions such as common autosomal aneuploidies (11,12), monosomy X (13), fetal growth restriction (14), twin-to-twin transfusion syndrome (15), and open myelomeningocele (16). Table 12.1 describes the variety of fetal diseases in which transcriptomics have been used.




















































Table 12.1 Summary of the most consistent biomarkers described by studies applying transcriptomics in fetal disorders


Reference


Pregnancy complication


Sample source


Methodology


Candidate gene identified


Tarui et al. (16)


Open myelomeningocele


Cell-free RNA in amniotic fluid


Microarray


PRICKLE2, GLI3, RAB23, HES1, FOLR1, GAP43 and ZEB1, ACAP1. Pathway analysis demonstrated a significant contribution of inflammation to disease and a broad influence of Wnt signaling pathways (Wnt1, Wnt5A, ITPR1).


Slonim et al. (11)


Down syndrome


Cell-free RNA in amniotic fluid


Microarray


Two sets of genes that differed between the trisomy 21 and euploid samples were identified. Pathway analysis revealed that oxidative stress, ion transport, G-protein signaling, immune and stress response, circulatory system functions, cell structure, sensory perception, and several developmental processes appear to be disrupted in Down syndrome.


Massingham et al. (13)


Turner syndrome


Cell-free RNA in amniotic fluid


Microarray


There were 470 statistically significantly differentially expressed genes identified, most highly represented by the hematologic/immune system, and other genes associated with cardiac and skeletal systems. XIST was significantly downregulated. Identified genes of possible pathologic significance included NFATC3, IGFBP5, and LDLR.


Hui et al. (15)


Twin-twin transfusion syndrome


Cell-free RNA in amniotic fluid


Microarray


Differential expression of 801 genes (472 genes were downregulated, and 329 genes were upregulated), which were significantly enriched for neurological disease and cardiovascular system pathways.


Cho et al. (14)


Fetal growth restriction


Cell-free RNA in amniotic fluid


Microarray


Differential expression of 411 genes (95 genes were downregulated and 316 genes were upregulated). Upregulated genes were highly related to metabolic process and protein synthesis. LRP10 and IGF-2 were significantly increased (6- and 17-fold, respectively) in IUGR.


Vora et al. (43)


Preterm birth


Cord blood


RNA-seq


148 genes were differentially expressed in the cord blood of neonates that were delivered preterm. Two pathways, cell cycle/metabolism and immune/inflammatory signaling, represent the predominant pathways with significant differential expression in preterm neonates.


Proteomics


The proteome is the whole set of proteins that is or can be expressed by a given genome in cells, tissues, or organisms. Proteomics is the study of proteomes and their functions. The interest in proteomics increased substantially with the recognition that there are many more proteins than gene products (more than 100,000 human proteins have been recognized so far) (17). Specifically, during pregnancy, proteomics analysis can be applied to test maternal blood, placental tissue, trophoblast cells, amniotic fluid, and cervicovaginal fluid (Table 12.2). The most important challenge in the study of the proteome is that it is a highly complex and dynamic system, since protein expression varies within the cell depending on its physiological or pathological state. In addition, posttranslational modifications as well as chemical damages such as glycation, oxidation, and nitration took place, increasing the diversity and heterogeneity of the proteome.




































































































Table 12.2 Summary of the most consistent biomarkers described by studies applying proteomics in preterm birth and fetal growth restriction


Reference


Pregnancy complication


Sample source


Proteomic method


Candidate protein/altered protein expression


Pereira et al. (44)


Preterm birth


Cervicovaginal fluid


2D-PAGE
shotgun proteomics


There were 34 biomarkers described including calgranulins, annexins, S100 calcium-binding protein A7, and epidermal fatty acid binding protein were abundant in CVF and differentially present in PTL and SPTB samples, as were the serum proteins α-1-antitrypsin, α1-acid glycoprotein, haptoglobin, serotransferrin, and vitamin D binding protein.


Butt et al. (45)


Preterm birth


Placental tissue


2D-PAGE


Eleven proteins were identified by tandem mass spectrometry, falling into three distinct functional classes: structural/cytoskeletal components, ER lumenal proteins with enzymatic or chaperone functions, and proteins with anticoagulant properties.


Cobo et al. (46)


Preterm birth


Amniotic fluid


Targeted SDS-ELFO, Western blot


Interleukin-6 and proteomic candidate biomarkers were predictors of intraamniotic infection in patients with preterm labor and intact membranes.


Stella et al. (47)


Preterm birth


Maternal serum


2D-PAGE, MS profiling


Peaks at 7781 and 3164 m/z, present on CM10, IMAC30, and H50 surfaces, were significantly different when comparing PLPTD versus PLTD.


Bujold et al. (48)


Preterm birth


Amniotic fluid


2D-LC, protein level


In patients with intra-amniotic infection/inflammation, fibrinopeptide B, transferrin, MHC class 1 chain-related A antigen fragment, transcription elongation factor A, sex-determining region Y (SRY) box 5 protein, Down syndrome critical region 2 protein (DSCR2), IGFBP-1, and human peptide 8 (HP8) were overexpressed.


Gravett et al. (21)


Preterm birth and chorioamnionitis


Amniotic fluid


2D SELDI-TOF MS


IGFBP-1 Calgranulin B


Buhimschi et al. (25)


Preterm birth


Amniotic fluid


SELDI-TOF ELISA


HNP 1 and 2, Cal A and C


Buhimschi et al. (24)


Preterm birth


Amniotic fluid


SELDI-TOF


Discriminatory profile 10–14 kDa and 2D-DIGE-derived identities


Rüetschi et al. (49)


Preterm birth


AF, CVF


SELDI-TOF, MS


HNP 1–3, Cal A and B


Romero R et al. (50)


Preterm birth


Amniotic fluid


 SELDI-TOF


Analysis was performed with a testing and training set of samples. Then 39 potential biomarkers were able to distinguish patients with preterm labor and intra-amniotic inflammation.


Wölter et al. (51)


Fetal growth restriction


Umbilical cord blood


MALDI-TOF mass spectrometric profile


The best differentiating protein between IUGR cases and controls was apolipoprotein C-III0.


Wölter et al. (52)


Fetal growth restriction


Maternal blood


MALDI-TOF mass spectrometric profile


The two best differentiating proteins in mothers of IUGR fetuses were apolipoprotein C-II and apolipoprotein C-III0. Together with three robustly expressed protein proteoforms proapolipoprotein C-II, apolipoprotein C-III1, and apolipoprotein C-III2.


Lynch et al. (53)


Preterm birth


Maternal blood


SOMA scan proteomic assay


Coagulation factors IX ab and IX; factor B; PECAM-1; complement factor H, VEGF SR2; Cathepsin Z; GHR; Ficolin-3 and Cadherin.


Ezrin et al. (54)


Preterm birth


Microparticles in maternal blood


LC-MS


α1-antitrypsin, antithrombin III, α2-macroglobulin, α1B-glycoprotein, albumin, apolipoproteins, serotransferrin, IGHM2, complement factors.

Stay updated, free articles. Join our Telegram channel

May 10, 2020 | Posted by in GYNECOLOGY | Comments Off on Integrated system biology approaches to fetal medicine problems

Full access? Get Clinical Tree

Get Clinical Tree app for offline access