Haematopoietic loss of KDM6A impairs cardiac recovery post myocardial infarction via pro-inflammatory myeloid cells

Abstract

Abstract Background Ischaemic heart failure (IHF) remains a leading cause of death worldwide. Clonal haematopoiesis has emerged as a major risk factor for IHF and patients with CH have worse outcomes after acute myocardial infarction (AMI). CH is characterized by clonal expansion of a haematopoietic stem cell that carries a leukaemogenic mutation. Recently, acquired loss-of-function mutations in the histone demethylase KDM6A were identified as CH-drivers in IHF patients. KDM6A is frequently mutated in human cancer and acts as a tumour suppressor in acute myeloid leukaemia. However, the role of KDM6A in the regulation of cardiac damage after MI is not understood. Purpose Investigate the effect of leukocyte specific loss of KDM6A in the development of IHF after AMI in mice and humans. Methods/Results A mouse model of haematopoietic specific loss of Kdm6a (VaviCre-Kdm6aflox) combined with a model of myocardial injury was employed to investigate the role of Kdm6a in leukocytes in post-MI cardiac recovery. Firstly, following 28 days (D) post-MI, mice lacking Kdm6a in haematopoietic cells (KDM6A KO) showed significantly worse cardiac ejection fraction compared to controls (controls 43.98 ± 2.55 % vs KDM6A KO 31.86 ± 3.23 % recovery from baseline, P=0.009). This was accompanied by an increase in cardiac scar size at D28 post-MI (controls 22.46±5.25% vs KDM6A KO 56.83±11.1 % area, P<0.0001). Moreover, KDM6A KO mice showed peripheral blood monocytosis (controls 33793±5644 vs KDM6A KO 76227±14035 monocytes/ml blood, P =0.008) at D3 post-MI, which was accompanied by an increased influx of CD11b+ myeloid cells in the injured heart (7804±1023 vs 11444±1436 CD11b+ cells/mg tissue, P=0.05), as well as by an increased expression of pro-inflammatory marker TNF-α (controls 1.01±0.14 vs KDM6A KO 2.16±0.45 fold change, P=0.04) in the ischaemic myocardium. These data were corroborated by single-cell RNA sequencing (scRNA-seq) profiling of CD45+ cardiac leukocytes 3D post-MI. Here, transcriptomically distinct subsets of macrophages/monocytes showed upregulation of pro-inflammatory markers such as Il1-β, Nlrp3, Cxcl2, and Nfkb1 (P < 0.05, Bonferroni correction). Finally, scRNA-seq of circulating blood cells from IHF patients carrying KDM6A CH-driver mutations showed elevated levels of pro-inflammatory genes such as IL32, IL7R and CCL5 in CD14+ classical monocytes (P < 0.05, Bonferroni correction) compared to non-carriers. Conclusions In summary, this study shows that loss of KDM6A in myeloid cells amplifies acute cardiac inflammation and impairs recovery of left ventricular function after AMI accompanied by increased pro-inflammatory myeloid cells in the heart. Further interrogation of genes identified in the aforementioned murine and human sequencing data and subsequent in vitro mechanistic studies will be essential to identify the mechanism by which loss of KDM6A results in altered leukocyte function and worse HF prognosis post-MI.