Redirecting full-length FLT1 expression towards its soluble isoform by targeting alternative polyadenylation promotes postischaemic angiogenesis

Abstract Aims Vascular endothelial growth factors (VEGFs) and their endothelial tyrosine kinase receptors are crucial mediators of vasculogenesis and angiogenesis, with FLT1 (VEGFR1) acting as a decoy receptor for VEGFs. Alternative polyadenylation (APA) of Flt1 transcripts produces a truncated mRNA transcript encoding soluble (s) FLT1, which contains only the extracellular ligand-binding domain. The relative roles of full-length (fl) FLT1 and sFLT1 in angiogenesis are incompletely understood. Alternative polyadenylation is an important post-transcriptional mechanism of gene regulation that is increasingly recognized to affect the expression of many genes related to cardiovascular diseases. However, there are no in vivo models available to target APA and to examine the physiological roles of gene isoforms. Methods and results We here generated mice by introducing silent mutations into the Flt1 exon 13/intron 13 5′ splice site through homology-directed repair with CRISPR/Cas9 gene editing. Embryonic development, postnatal retinal vascular development, and reparative angiogenesis were evaluated in models of hindlimb ischaemia and myocardial infarction, complemented by assessments of cardiac functional recovery and cell phenotypes through snRNAseq, as well as in vitro assays examining endothelial cell phenotype and function. The introduction of specific splice site mutations within Flt1 to activate APA led to a complete loss of flFLT1 and an increase in sFLT1 expression. Using this mouse model, we demonstrate that sFLT1 alone is sufficient for developmental vasculogenesis, whereas the specific loss of flFLT1 enhances reparative revascularization and angiogenesis and improves cardiac recovery after myocardial infarction. Conclusion Our findings show that APA can be effectively targeted to modulate FLT1 isoform expression in vivo. Our mouse model may help to understand diseases driven by an imbalance in FLT1 isoform expression. Given the widespread presence of APA signals, our approach could furthermore serve as a model for studying the function of other truncated gene isoforms.