Klebsiella pneumoniae induces dose-dependent shock, organ dysfunction, and coagulopathy in a nonhuman primate critical care model

Abstract

Nonhuman primate models that closely emulate the disease course, pathogenesis, and supportive care provided to human patients in the modern intensive care unit with bacterial sepsis are urgently needed to study pathogenesis and assess novel therapies. We therefore developed a non-human primate model of septic shock that includes supportive care akin to a modern intensive care unit. In this study, we characterized pathogen kinetics and evaluated the physiologic, immunologic, and pathologic responses in this model of septic shock induced by the clinically relevant pathogen Klebsiella pneumoniae across a three-log dose range. We observed dose-dependent bacteremia and circulating levels of Klebsiella pneumoniae DNA and endotoxin. Tachycardia and hypotension occurred in all animals and the study endpoint occurred in 8 of 12 animals that were euthanized. The infused bacterial dose was significantly associated with the severity of renal insufficiency and coagulopathy. Neutrophil activation evidenced by increased CD11b expression, decreased CD62L expression, and increased circulating levels of myeloperoxidase, lactoferrin, and neutrophil extracellular traps; monocyte activation evidenced by increased circulating levels of interleukin-6, tumor necrosis factor-alpha, granulocyte-macrophage colony-stimulating factor, and monocyte chemotactic protein-1; and endothelial activation evidenced by increased circulating levels of syndecan-1 and angiopoietin-II were all consistent with human sepsis. Our model provides an opportunity to study pathogenesis and investigate novel therapeutics for the treatment of bacterial sepsis in the setting of modern supportive care.IMPORTANCEThere is currently a disconnect between the efficacy of sepsis therapies in pre-clinical animal models and human clinical trials. Therefore, developing nonhuman primate models that closely mimic human sepsis pathogenesis to study novel host-targeted therapeutics is a priority. In this study, we developed a model of septic shock with a clinically relevant bacteria (Klebsiella pneumoniae) that provides standard supportive care including mechanical ventilation, invasive hemodynamic monitoring, volume resuscitation, vasopressors, antibiotics, and steroids. In a dose-dependent manner, we observed that this model closely emulates the hemodynamic, end-organ dysfunction, and cellular and soluble responses associated with human sepsis. This validated model provides a unique opportunity to study the pathogenesis of acute septic shock and evaluate host-directed therapeutics in a large animal model that closely emulates the modern-day intensive care unit and supportive critical care.

Keywords: nonhuman primate; pathogenesis; sepsis; therapeutics.

Fig 1

Klebsiella pneumoniae dose correlates with organ dysfunction, coagulopathy, and neutropenia. Correlation of change in baseline of different clinical laboratory parameters in the first 12 hours with bacterial dose. Abbreviations: AUC, area under the curve; PT, prothrombin time; WBC, white blood cell count; ANC, absolute neutrophil count.

Fig 2

Markers of neutrophils activation, migration, and maturity vary by dose. Quantification of cellular markers of neutrophil maturity and activation. All values represent geometric mean fluorescent intensity (MFI) quantified in neutrophils by FlowJo version 10.8.1.

Fig 3

Soluble markers of neutrophil activation vary by dose. Quantification of NETs, soluble MPO (myeloperoxidase) and lactoferrin. Soluble NETs, MPO, and lactoferrin were quantified in plasma. Lines represent moving averages of two technical replicates. NETs data are presented as fold change from baseline NETs for each animal. Error bars display standard deviation. Intracellular MPO and lactoferrin represent the geometric mean of intensity quantified in neutrophils by FlowJo version 10.8.1.

Fig 4

Soluble markers of monocyte, macrophage, and NF-κB-associated activation vary by dose. Legend: Quantification of cytokines longitudinally across study period. Data represent average of two technical replicates.