Sepsis is characterized by life-threatening organ dysfunction caused by a dysregulated response to infection. Neutrophils play a crucial role in sepsis during which they release neutrophil extracellular traps (NETs), webs of negatively charged cell-free DNA (cfDNA) complexed with positively-charged histones that kill pathogens but also damage host tissue. While it has been proposed that NET digestion may be beneficial in the treatment of sepsis, it is also possible that this strategy leads to the release of harmful NET degradation products (NDPs), such as cfDNA, histones, and myeloperoxidase (MPO) that cause endothelial damage. An alternative approach would be treatment strategies that prevent NET lysis and the release of NDPs. We posited that NET-bound platelet factor 4 (PF4, CXCL4) may have this effect. PF4 is a highly-positively charged, platelet-specific chemokine released in high concentrations following platelet activation and aggregates polyanionic molecules like heparin, polyphosphates, and DNA. We have found that PF4 causes physical compaction of NETs without inducing histone release. We studied the effect of this compaction using an in vitro microfluidic assay in which NETs released from neutrophils stimulated by tissue-necrosis factor (TNF) α were infused over a layer of cultured human umbilical vein endothelial cells (HUVECs) ± PF4 (25 μg/ml) and incubated for 12 hours. Channels exposed to PF4 were "heathier" with significantly more residual attached endothelial cells (44±24 vs. 150±29 cells/hpf, n=6, p=0.0002). We have previously shown that mice that overexpress PF4 are protected from mortality in lipopolysaccharide (LPS) endotoxemia. Compared to wildtype animals, PF4 null mice have increased circulating levels of NET markers including cfDNA, citrullinated histones, and MPO. Treatment with exogenous PF4 leads to a decrease in plasma levels of NET components and a reduction in mortality. In a microfluidic assay in which channel-adherent NETs were treated with increasing concentrations of PF4 (0-25 µg/ml) and then infused with DNase I (100 U/ml), we found that compacted NET-PF4 complexes become resistant to DNase I digestion. Prior studies have revealed that when PF4 binds to polyanionic molecules, it exposes HIT-like antigenic sites. We confirmed that the NET-PF4 complexes similarly bind HIT IgG isolated from clinical samples and the HIT-like monoclonal antibody KKO. Interestingly, KKO binding further enhanced DNase I resistance, a phenomenon not seen with a polyclonal anti-PF4 antibody (abcam). We hypothesized that KKO which causes PF4 oligomerization unlike the polyclonal antibody, further stabilized the crosslinked NET-PF4 complexes, providing additional protection from DNase I digestion and preventing the release of NDPs. We then asked whether KKO may serve as a targeted therapy in the treatment of sepsis, enhancing NET-PF4 complex resistance to nuclease digestion and leading to the sequestration of NDPs. As KKO is an IgG2bk antibody, we were unable to use pepsin digestion to remove the Fc region that causes platelet activation. We therefore used an IgG-specific endoglycosidase to develop a deglycosylated version of KKO (DG-KKO) that retains the ability to bind to NET-PF4 complexes (data not shown), but has a reduced capacity to interact with hematopoietic cell Fc receptors. DG-KKO has a markedly decreased ability to activate human platelets compared to KKO in the presence of added PF4 as measured by P-selectin level (3710±140 control vs. 1580±300 treated MFI, n = 2). Compared to control, DG-KKO infusion in the murine endotoxemia model prevented thrombocytopenia (mean platelet count 286±16 control vs. 537±69 treated X103/µl, n=4-5, p=0.016), the release of cfDNA (8.5±0.3 vs. 3.7±0.7 µg/ml, n=4-5, p=0.016), and the emission of MPO-DNA complexes (234±14 vs. 99±5 % increase from non-LPS injected controls, n=4-5, p=0.016) As depicted in Figure 1, we propose that in sepsis, NETosis occurs (Step 1) causing the release of harmful NDPs (Step 2). PF4 expelled from activated platelets stabilizes PF4-NET complexes (Step 3). Infused DG-KKO enhances NET stability and decreases release of NDPs following DNase I digestion (Step 4). These studies provide mechanistic insights into the release of NDPs during endotoxemia and offer a targeted, novel therapeutic to prevent their contribution to inflammatory states such as sepsis.


Poncz:Incyte Corporation: Consultancy, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution