Fig. 6. Spatiotemporal EVT distributions suggest that intravasation is the predominant route of EVT invasion in superficial decidua.
a, MIBI overlay of anchoring villous and associated cell column EVT populations. Inset, cell column EVTs. Representative image of n = 60 FOVs. b, MIBI overlay of spiral arteries and associated perivascular EVT populations. Inset, perivascular EVT breaching artery wall. Representative image of n = 54 FOVs. c, MIBI overlay of remodelled spiral arteries and associated intravascular EVT populations. Inset, intravascular EVTs in a clump. Representative image of n = 23 FOVs. d, Percentage of arteries with scores less than or equal to a given SAR (δ) threshold by perivascular or intravascular EVTs present. Arteries were considered to have perivascular or intravascular EVT if ≥ 5 EVTs were present. e, Lineage and functional marker trends of EVT populations by anatomical location using MIBI data. Lineage marker (CD57, HLA-G and CD56) trends are mean expression values. Functional marker (Ki67 and PD-L1) trends are mean positive cell frequencies. Columns Z-scored and hierarchically clustered. f, Expression (Z-score) of top 35 DEGs by log(fold change) (adjusted P value < 0.05) between interstitial and intravascular EVT populations using NanoString whole transcriptome data. Genes also differentially expressed in preeclamptic decidua samples1,43 are indicated in bold. g, Application of NicheNet algorithm to artery and intravascular EVT whole transcriptome data to predict EVT–artery interactions and downstream signalling targets. h, Outcome of ligand activity prediction according to NicheNet on DEGs on intravascular EVTs. Results are shown for the ten EVT ligands that best predict receivers expressed in arteries, ranked by Pearson’s correlation coefficient or the EVT ligand activity ranking metric. Ligands, receivers and targets also differentially expressed in preeclamptic decidua samples1 are indicated in bold. Prolif. and diff., proliferation and differentiation.