Introduction Heart failure incidence has seen a 92% increase since 1990, with the majority of cases arising as a consequence of ischaemic and hypertensive heart disease. This is in part due to lack of effective treatments targeting the chronic additive remodelling within diseased hearts. Therefore, increased understanding of the underlying mechanisms is required to inform therapeutic development. Within our group we have identified ectopic expression of SRY-Box 9 (SOX9) as a key regulator of liver and kidney fibrosis. Recent studies have also implicated SOX9 in driving cardiac fibrosis. We aim to integrate spatial transcriptomics (ST) with single cell (sc) RNA sequencing and computational approaches to deconvolute intercellular populations and processes. These approaches will provide a mechanistic insight into the role of SOX9 in remodelling myocardial tissue. Materials and Methods: Fibrosis and hypertrophy were induced over 2 weeks using transverse aortic constriction (TAC) on wild type and SOX9-null mice. Echocardiography was used to analyse cardiac function. ST was performed using the 10X Genomics platform on cryosections of WT Sham, WT TAC and SOX9-null TAC hearts. The data was analysed by squidpy on python. Structural changes were evaluated with immunohistochemistry and immunofluorescence.

Results Following TAC, ectopic SOX9 expression was detected in myofibroblasts associated with interstitial fibrosis and in cardiomyocytes immediately adjacent. Significantly, Sox9-null mice improved TAC-induced hypertrophy and fibrosis. To uncover the functional role of SOX9 in heart fibrosis and provide novel mechanistic insight we carried out single cell ST. ST resolved the damaged heart tissue into discrete gene expression clusters that correlated with histological landmarks. We integrated published scRNA-seq data to increase the resolution of the ST data. Through Cell2location computational approaches we defined multi-cell type clusters to reveal disease associated molecular signatures. We identified various subtypes of cardiomyocytes and fibroblasts within and surrounding areas of high fibrosis that have a set of distinguished differentially expressed genes. Interestingly, the gene expression profile of the fibrosis associated cardiomyocyte subpopulations suggested a profibrotic non-canonical cardiomyocyte subtype, which was reduced in Sox9-null fibrotic hearts.

Conclusion In conclusion our study provides evidence to support a key role for SOX9 in the propagation of cardiac remodelling. Using cutting edge ST we have spatially resolved cell populations suggesting a SOX9-dependent role for profibrotic cardiomyocyte-fibroblast crosstalk in mediating disease progression.