Histones, Like Platelet Factor 4 (PF4), Affect Generation of Activated Protein C: Implications for the Pathogenesis of Severe Sepsis

Abstract 530

Platelet factor 4 (PF4) increases aPC generation by the thrombin (IIa)/thrombomodulin (TM) complex and may impact outcome in sepsis. PF4's effect on aPC generation follows a biphasic curve when tested in solution, on human TM expressing HEK293, and on primary endothelial cells (ECs) with a peak concentration at around 25 μg/ml. Formation of complexes at a specific molar ratio between positively-charged tetramers of PF4 and negatively-charged chondroitin sulfate (CS) on the TM glycosaminoglycan (GAG) is crucial for the increase in aPC generation. Other positively-charged molecules like protamine sulfate (PRT) affect aPC generation in a similar manner, and heparin, which is known to bind PF4 and PRT more avidly than CS, lowers effective PF4 or PRT concentrations. Here we examined whether histones, that are also small positively-charged molecules, affect aPC generation. Histones released from cells in sepsis are cytotoxic toward ECs and lethal when injected into mice, and aPC reverses this lethality. May histones affect aPC generation by the same mechanism as other positively-charged molecules, and how does the presence of PF4 or heparin influence this effect? We have addressed these questions both in solution and with TM-expressing cells, in the absence or presence of endothelial protein C receptor. We found that individual, or mixed histones affect aPC formation following a similar biphasic curve seen with PF4 with a peak effect at around 10 μg/ml but to lesser extent (2-fold maximal increase compared to 6-fold for PF4). Histones and PF4 are additive at low concentrations; however, more importantly, histones only decreased aPC generation when tested in the presence of optimal or higher PF4 concentration (>25 μg/ml). Just as with PF4, added heparin decreased effective histone concentration and shifted the curve for aPC generation to the right, both in the absence or presence of PF4. We hypothesize that normally PF4 released from platelets augments aPC generation and low concentration of histones have similar effect. But when histones are released in sepsis in high concentrations, their interaction with CS on TM blocks formation of complexes between PF4 and TM's CS that are optimal for maximal increase of aPC generation. Further we tested the effect of histones on aPC generation in vivo. Injection of histones in mice increased IIa-induced (2U/kg) aPC generation in plasma. This increase was concentration dependent (at 1 to 20 mg/kg increasing aPC generation up to 10-fold), but injection of higher amount of histones (40 mg/kg) became lethal. Mice that were overexpressing human PF4 had an increased lethality when histones at 40 mg/kg were co-injected with thrombin (2U/kg) over the littermate mice deficient in murine PF4 (60% vs. 0% mortality, respectively, n=5 for each group) suggesting that in vivo histones may also act additively with PF4 on aPC generation. We propose that in severe septic patients, especially those with high levels of released PF4, concurrently available histones suppress aPC generation. By binding to the excess of PF4 and/or histones, heparin may be beneficial in severe sepsis by allowing improved aPC generation.