INTRODUCTION Myocardial infarction (MI) inflicts a sterile wound on the heart which recruits hundreds of thousands of monocytes from the blood. The bone marrow (BM) responds to cardiac ischemia with an increase in monocyte production, hence meeting the high demand of these cells. Apart from this quantitative response, little is known whether the BM supplies qualitatively different monocytes during the course of MI.

METHODS&RESULTS We isolated BM Ly6C^high monocytes from non-infarcted mice and from mice which were infarcted 72h prior (surgically induced by permanent coronary ligation) and contrasted monocytic transcriptomes using RNA sequencing. We found 672 differentially expressed genes (DEG) of which 289 were up- and 383 down-regulated (log₂FoldChange³±1.0, adjusted p<0.05). Since most DEG were enriched in “cell adhesion and recruitment” pathways, we tested the functional relevance of our findings in the context of monocyte recruitment to the heart using competitive adoptive transfer experiments. Our data revealed that the BM provides monocytes that are primed for better recruitment (relative recruitment: 56.1% 12h after MI vs. 43.9% steady state, p<0.01) at the early inflammatory phase following coronary ligation (12h after MI). In contrast, monocytes displayed worse recruitment capacities (relative recruitment: 30.4% 72h after MI vs. 65.8% steady state, p<0.01) at the end of the inflammatory phase (72h after MI). We next sought to identify key players in this BM-controlled cardiac recruitment process. Our RNA sequencing data revealed that – in monocytes with poor recruitment capacities – the cell adhesion molecule CD209 was among the most significantly downregulated genes at 72h after MI over steady state. To investigate the molecular role of CD209 in the process of leukocyte recruitment to the myocardium during MI, we cloned human CD209 from a cDNA library into an overexpression plasmid system. After ectopic overexpression of CD209, THP-1 cells (a cell line derived from immortalized human blood monocytes) showed enhanced adhesion to endothelial cells (HUVECs) compared to THP-1 cells under control conditions. Finally, we infarcted wild type (CD209^+/+) and CD209 knockout (CD209^-/-) mice and found decreased numbers of Ly6C^high monocytes in infarcted hearts of CD209^-/- mice in comparison to CD209^+/+mice (491+52 vs. 915+180 Ly6C^high monocytes/mg infarct tissue, mean+SEM, p<0.05) while numbers of blood Ly6C^high monocytes were unchanged.

CONCLUSIONS The BM responds to MI and provides an altered monocyte repertoire in terms of numbers and function and hence remotely controls monocyte/macrophage accumulation in the ischemic heart. Consequently, the heart–bone marrow axis may represent a novel therapeutic target to beneficially modify post-MI cardiac remodeling, a major contributor of MI-related morbidity and mortality.