NETosis is required for S100A8/A9-induced granulopoiesis after myocardial infarction

Graphical Abstract

Myocardial infarction (MI) provokes a massive systemic inflammatory response characterized by enhanced infiltration of neutrophils to the ischemic heart via granulopoiesis in the bone marrow (BM)¹. In patients with acute coronary syndrome, elevated blood neutrophils and S100A8/A9 is associated with higher incidence of major adverse cardiac events¹. We recently discovered that S100A8/A9 is released by infiltrating neutrophils in the infarct to induce granulopoiesis by interacting with TLR4 on naïve neutrophils, priming the NLPR3-inflammasome for release of IL-1β¹. However, the earliest events that trigger S100A8/A9 release from the infiltrating-neutrophils remain unclear.

Given the abundance of stimuli that can potentially induce cell death in the infarct (i.e. ROS, ATP, Ca²⁺, etc.), we hypothesized that S100A8/A9 is released by NETosis, a form of cell-death that involves extrusion of decondensed chromatin along with its entire granular content including S100A8/A9. To test this hypothesis, first, we interrogated RNA-seq data obtained from adult cardiac CD45⁺ leukocytes 72 hours-post MI (details in Quaife-Ryan et al²). We found robust upregulation of genes encoding NET-associated proteins along with kinases and transcriptional factors (common elements) crucial for NETosis (Figure, A). Because NETosis is initiated by abnormal elevations of either intracellular ROS via NADPH oxidase 2 (i.e. Nox-dependent) or Ca²⁺ (Nox-independent)³, we examined the expression of the kinases and transcription factors that represent these 2 pathways by sorting neutrophils from the infarct (24 hrs. post-MI). The data revealed a genetic signature for Nox-independent NETosis, dominated by Padi4³ (Figure, B), an enzyme that is critical for citrullination of histones and Nox-independent NETosis. Furthermore, we found a robust increase in citrullination of histone moieties (H3Cit⁺ neutrophils), co-localization of S100A8/A9 with H3Cit, and increased S100A8/A9 levels in the heart as early as 6 hours after MI (Figure, C).

Figure. NETosis is essential for S100A8/A9-induced granulopoiesis.

MI was induced in 8–10-week-old male C57BL/6J mice by permanent ligation of the left anterior descending (LAD) coronary artery. A) Cardiac leukocyte heat maps (RNA-seq) displaying log 2-fold change (FDR=0.025) in genes representing NET-associated proteins and elements common to both Nox-dependent and Nox-independent pathways. B) Nox-dependent and Nox-independent NETosis gene expression in cardiac neutrophils (24hr Post-MI). *P<0.05; unpaired student’s t-test (n=4). C) Representative flow cytometry plots (left panel, gated on CD45⁺ cells) and quantification (right panel) of NETosis in cardiac neutrophils by H3Cit staining. Confocal images (X60) of heart sections showing markers of DNA (Dapi, blue), NETosis (H3Cit, red) and S100A8/A9 (green) expression, red arrows: S100A8/A9 and DNA within the neutrophil, white arrows: externalized NETs with extensive H3Cit staining on decondensed nuclei decorated with S100A8/A9 (colocalized = orange/yellow) scale bars=10μm. Heart S100A8/A9 levels (top right) were quantified by ELISA. P<0.05 c.f. sham control (0); Kruskal-Wallis test and Dunn’s multiple comparison test. D) Experimental overview: BM from male C57BL/6J WT (blue) and Padi4^−/− mice (orange) was transplanted to WT recipients and allowed to reconstitute for 6 weeks. MI was induced by LAD ligation and 18hrs later; NETosis, BM hematopoietic stem cells (HSC), common myeloid progenitor (CMP) and granulocyte macrophage progenitor (GMP) cell proliferation (EdU⁺), blood and heart neutrophils (with representative flow cytometry plots), along with circulating and heart neutrophil pro-IL-1β levels were measured by flow cytometry. Quantification of S100A8/A9 levels in the serum and heart (by ELISA). P<0.05 c.f. WT> WT+ MI; unpaired t-test. E) Quantification of dsDNA (blue line) and myeloperoxidase (MPO, red line), and S100A8/A9 (blue line) and MPO-Elastase (red line) in the serum from control (non-STEMI) and STEMI patients on admission (0) and 3, 6, 12 and 24hrs post-revascularization by PCI. n=15–21/group, *P<0.05 c.f. healthy control group; Mann-Whitney test.

To confirm if Nox-independent NETosis is involved in MI-induced granulopoiesis, we transplanted WT or Padi4^−/− BM into WT recipients followed by a MI (Scheme, Figure, D). Padi4, required for Nox-independent NETosis, is expressed almost exclusively in neutrophils. As expected, deletion of Padi4 significantly reduced Nox-independent NETosis. This was associated with suppressed granulopoiesis, fewer neutrophils in the blood and heart and, reduced cardiac and serum S100A8/A9 (Figure, D). Reduced pro-IL-1β expression in cardiac and blood neutrophils from Padi4^−/− mice revealed suppressed granulopoiesis^{1, 4} (Figure D).

To validate these findings in humans, we measured various markers of NETosis and found a marked increase in serum double-stranded DNA (dsDNA), myeloperoxidase (MPO), elastase, which paralleled S100A8/A9 levels in STEMI patients from the time of admission through post-revascularization by percutaneous coronary intervention (PCI) (Figure E). Importantly, neutrophil counts post-MI (<24hrs) in these patients was associated with greater S100A8/A9 levels and incidence of MACE¹.

Currently, no approved anti-inflammatory therapy exists to mitigate MI-induced leukocytosis despite overwhelming evidence demonstrating that neutrophils instigate inflammation and promote heart failure. Furthermore, inhibition of NETs in experimental-MI have been shown to be cardioprotective⁵. We propose that interventions to reduce NETosis and/or S100A8/A9/IL-1β release^{1, 4}, particularly during the acute inflammatory phase could represent a departure from the status quo in managing post-MI inflammation and the subsequent heart failure.