Mapping the maternal-fetal interface at a single-cell resolution to interrogate the aetiology of severe pre-eclampsia and identify potential disease

During pregnancy there is a highly coordinated dialogue between mother and fetus and this communication, once established, ensures the correct development of the baby and allows the mother to both tolerate her unborn child, and remain well.

A common and dangerous pregnancy complication is pre-eclampsia (PE), where a pregnant woman develops high blood pressure that can lead to organ failure. Unfortunately, there is no treatment available and the underlying cause of this condition remains unknown. Recent research by others and us suggests dysregulation of the immune response at the mother-fetal interface, which affects how fetal tissues connect to the maternal blood system.

We therefore aim to study cell dialogue at the interface of the mother and her unborn child 'the fetus' known as the maternal-fetal interface (MFI). The MFI consists of different tissues belonging to both mother and fetus and includes; the maternal uterine 'womb' wall, the placenta, which supplies the fetus, along with the membrane that covers the wall of the womb and placenta. We will use technologies that allow us to interrogate the cells in these tissues at a single cell resolution, identifying both the specific cell type that are present, as well as what it is making or signalling. These approaches allow the quantification of the expression of thousands of genes in their native 'original' context.

The comparison between patients with severe PE (which poses the greatest clinical burden including maternal and fetal death and disability) versus healthy controls will help us detect specific genes expressed differently in the pathological condition. The groups will enable us to understand the differences seen and how gestational age affects these changes. With this information, we will look for these markers in maternal blood samples, aiming to develop a new way to diagnose PE, that reveals what is the cause of the disease. This approach lends itself to a more personalised form of treatment.

Our research may contribute to finding a treatment for this severe pregnancy complication as well as investigating whether its presence can be identified through a more simple blood test.

This project aims to achieve a better understanding of maternal-fetal interface interactions between cell types in pregnancies with severe PE.
Cases and controls are matched by gestational age, delivery mode, ethnicity, age, BMI and smoking status.
Sampling. From the mid-point of the largest distance between the cord insertion site and the edge of the placenta two samples are taken. One is placed in Hypothermosol Frs for single cell sequencing (SC) and the other is snap frozen for spatial transcriptomics (ST). From the mirrored area of the placental bed is two samples of are taken via a punch biopsy, for Sc and ST as above. Rolled chorio-amniotic membranes (CAM) are collected and divided for SC and ST as above. Maternal peripheral blood is sampled just before delivery.
Tissues are dissociated in Accutase or Accumax, on an automated dissociator. PBMCs are isolated using density gradient media. Cell suspensions are processed using the Chromium Next GEM Single Cell 5' Reagent Kits (v2 Chemistry Dual Index). Snap frozen tissues are processed using 10X Genomics Visium Spatial Gene Expression Slides. Resulting libraries from multiple patients are pooled and sequenced on Novaseq S4 v1.5 (200cycle) kits.
Maternal and fetal cells are separated using genetic variants. Unsupervised methods are used to obtain clusters, cell types will be assigned based on differentially expressed markers. Cell type composition is compared using scCODA. Gene expression for each cell type is compared using DESeq2. GSEA analysis is performed. Identification of genetic programs ("topics") that significantly change in severe PE is performed using ConutClust. Trajectories as well as their speed in pseudotime of different cell lineages, are inferred using Slingshot. Spatial mapping of cell types is inferred, using Seurat. Local distribution of cell types are quantified using spatial networks. We infer cell-to-cell signalling mediated by ligand-receptors pairs using CellPhoneDBv3 and Giotto.