Abstract 15720: Single Cell Spatial Transcriptomics of the Ischemic Borderzone

Introduction: Myocardial infarction (MI), leads to development of an infarct borderzone (BZ) separating areas of cell death (ischemic zones, IZ) from surviving myocardium (remote zones, RZ). The BZ is a geographically complex and biologically enigmatic territory thought to play critical roles in infarct expansion, ventricular remodeling, arrhythmias, and provisional regeneration; however, its cellular and molecular processes are not well understood because previous studies have relied on microdissection, which combines zones in uncontrolled and unmeasured proportions. Hypothesis: We hypothesize that the BZ can be redefined using single cell spatial transcriptomics. Methods: We used single cell and single nuclei RNA-seq, capture-based spatial transcriptomics, and multiplexed RNA fluorescence in situ hybridization (mFISH) to redefine the BZ into functional subunits. Results: Transcriptomic data from more than 300,000 cells reveal that the BZ is comprised of multiple spatially defined niches characterized by distinct structural, morphological, transcriptional, and functional subsets of cardiomyocytes, fibroblasts, monocytes/macrophages, and neutrophils. We identified two distinct BZ populations of cardiomyocytes (CMs): BZ1 (Nppa+Xirp2-), which forms a hundreds-of-microns-thick transitional layer adjacent to the RZ, and BZ2 (Nppa+Xirp2+), which forms a tens-of-microns-thick layer that decorates the complex peninsulas and islands adjacent to the dead myocardium of the IZ. Compared to BZ1 CMs, BZ2 CMs have significantly reduced contact area with neighboring CMs; express elevated levels of genes involved in mechanotransduction; and preferentially colocalize with matricellular-protein-expressing myofibroblasts. The transcriptional BZ is detectable within an hour of ischemia, prior to immune infiltration or fibroblast activation. We propose a unifying "loss of neighbor" hypothesis to mechanistically explain the results. Conclusions: Our work demonstrates that harmonizing multi-modality single cell and spatially resolved datasets has the potential to uncover new biology in the ischemic borderzone.