Back

Bacterial histone-like proteins released during antibiotic treatment mediate vascular injury in meningococcal sepsis.

McGuinness, D.; Johnson, C.; Lemgruber Soares, L.; Bidmos, F.; Yates, E. A.; Turnbull, J. E.; Evans, T.; McHugh, R. E.; Roe, A. J.; Douce, G.; Smith, A.; Ure, R.; Toh, C.-H.; Abrams, S.; Herberg, J.; Wright, V. J.; Nijman, R.; HEYDERMAN, R. S.; Faust, S. N.; Levin, M.; Moxon, C.

2026-07-16 microbiology
10.64898/2026.07.15.738198 bioRxiv
Show abstract

Vascular injury and coagulopathy are key drivers of mortality in bacterial sepsis. In Neisseria meningitidis infection, endothelial adhesion and thrombosis cause the characteristic petechial rash and, in the most severe cases, purpura fulminans. Although antibiotics rapidly kill bacteria, inflammation and vascular injury often persist or worsen after bacterial clearance, suggesting ongoing toxicity from released bacterial components. Here we identify bacterial histone-like proteins (HLPs), small positively charged DNA-binding proteins conserved across bacterial species, as previously unrecognized mediators of vascular damage. In vitro HLPs are released following antibiotic exposure, disrupting endothelial integrity. In patients with severe sepsis, they are detectable in plasma and tissue, colocalising with areas of vascular leak and coagulopathy. Non-anticoagulant heparins and anti-HLP antibodies neutralize HLP-induced endothelial disruption and toxicity in vitro and in vivo. These findings reveal HLPs as antibiotic-released bacterial toxins and suggest new therapeutic strategies to prevent vascular injury in sepsis. One sentence summaryAntibiotic treatment of meningococcal and other sepsis causative bacteria triggers the release of histone-like proteins (HLPs) that damage the endothelium and drive coagulopathy, but these effects can be neutralized by non-anticoagulant heparins or anti-HLP antibodies.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.