Background: Sepsis is a syndrome in which infection triggers a systemic inflammatory and pro-coagulant response, with a prevalence of up to 1 case per 1000 and a mortality rate of up to 40%. Cell-free DNA (cfDNA) is elevated in sepsis, and correlates with mortality. This DNA may come from nuclear, mitochondrial, or bacterial sources. Nuclear and mitochondrial DNA may come from activated neutrophils which release neutrophil extracellular traps (NETs). CpG motifs on bacterial and mitochondrial DNA can stimulate inflammatory responses via TLR9, which is present on neutrophils, monocytes, and recently shown to be expressed on platelets. cfDNA can activate coagulation via the intrinsic pathway. cfDNA may thus play an important pathogenic role in sepsis. This study elucidates the relative effects of nuclear, mitochondrial, and bacterial DNA on inflammatory and pro-coagulant pathways.
Methods: Mitochondrial DNA concentrations were measured by PCR using plasma samples from septic patients. Nuclear and mitochondrial DNA were purified from human embryonic kidney 293 cells, and bacterial DNA was from E. coli. Neutrophils from healthy donors were cultured with purified bacterial, mitochondrial, or nuclear DNA at 15µg/mL for 20h. IL-6 levels in the supernatants were measured by ELISA, and neutrophil viability was measured by flow cytometry for annexin-V binding and propidium iodide exclusion. The three types of DNA were added to either citrated normal human platelet-poor plasma or platelet-rich plasma, and continuous thrombin generation was measured (Technothrombin, Vienna, Austria). Light transmission aggregometry was performed in citrated platelet-rich plasma with co-treatment of a sub-threshold concentration of ADP and varying concentrations of DNA. Markers of platelet activation were measured by flow cytometry for P-selectin and activated integrin αIIbβ3. All reagents contained less than 0.06EU/mL of LPS by limulus amoebocyte lysate assay.
Results: Cell-free mitochondrial DNA was elevated in plasma from septic patients compared to healthy controls. Bacterial, but not mitochondrial or nuclear, DNA increased neutrophil IL-6 secretion. Both mitochondrial and bacterial DNA increased neutrophil viability at 20h. At concentrations found in the plasma of critically-ill patients, mitochondrial, nuclear, and bacterial DNA increased thrombin generation in both platelet-poor plasma and platelet-rich plasma to a similar degree, and this effect was abolished by corn-trypsin inhibitor and reduced in FXII-depleted plasma, indicating dependence on the intrinsic pathway of coagulation. Independently of coagulation, nuclear DNA at high concentrations, such as may be seen in the NET micro-environment, was capable of causing aggregation of ADP pre-stimulated platelets in citrated plasma, which was accompanied by activation of integrin αIIbβ3and surface expression of P-selectin. This effect also occurred with synthetic phosphodiester oligonucleotides, and was abolished by DNase pre-digestion.
Conclusions: cfDNA of bacterial origin can stimulate neutrophil IL-6 release, while both mitochondrial and bacterial DNA prolonged neutrophil viability. All types of DNA can activate coagulation via the contact pathway. DNA at high concentrations may be able to directly stimulate platelets. Total plasma cfDNA and cell-free mitochondrial DNA specifically are elevated in sepsis. Thus, nuclear, mitochondrial, and bacterial DNA may play distinct roles in the pathogenesis of sepsis.
Bhagirath:Pfizer: Research Funding.
Asterisk with author names denotes non-ASH members.