AF is often treated by ablation to isolate the pulmonary veins where ectopic pacemaker activity or reentry may initiate AF. The relationship of gene expression and histology in the pulmonary vein is unknown. To identify gene expression in pulmonary vein sections and identify regions where gene expression may be indicative of AF pathogenesis. Human pulmonary vein (PV) sleeve tissues form the ostia of the left atrium were obtained from 8 unused lung or heart transplant donors in a protocol approved by the Cleveland Clinic IRB. After tissue sectioning and staining, spatial RNAseq was performed using the 10x Genomics Visium platform. mRNAs from∼1000 to 2500 genes were sequenced in each spatial area. Seurat clustering yielded 15 different clusters. These split into two segregating populations, with one connecting population. Multiple cells and cell types may reside in each 55 μm diameter spatial area. A clear asymmetry of clusters was observed in the PV sections. Cell-type marker genes derived from published human heart single nuclei RNAseq were used to determine the dominant cell types in the 15 different clusters. One group of segregating clusters was enriched in cardiomyocytes, while the other group of segregating clusters was enriched for other cell types including fibroblasts, vascular smooth muscle cells, endothelial cells, and adipocytes. Spatial transcriptomics clearly resolved the venous, cardiomyocyte, and epicardial regions of the PV (Figure, showing histology and spatial expression UMAP clusters for two PV specimens), as well as fibrotic regions. Spatial expression of the AF-associated genes PITX2, SHOX2, and HCN4 was also mapped, confirming higher expression of the cardiac master transcription factor SHOX2 in the PV vs. left atrial appendage (LAA) tissues. Expression of the hyperpolarization-activated ion channel HCN4 was sporadic in both PV and LAA specimens. Spatial transcriptomics identified gene expression profiles in the PV easily identifying the venous vascular smooth muscle, cardiomyocyte, and epicardial layers. The predominant cell type in each spatial region was identified by leveraging heart single nuclei RNAseq data. Spatial transcriptomics may help to discover gene expression and cell types associated with ectopic pacemaker activity in the PV.
Read full abstract