In vivo Analysis of Murine Pneumococcal Pneumonia for Mathematical Modelling of Community Aquired Pneumonia
23 February 2017 (online)
Community Acquired Pneumonia (CAP) is one of the leading causes of death worldwide and Streptococcus pneumoniae (S. pn.) is the most prevalent causative pathogen. Through CAPSyS, an interdisciplinary systems medicine approach, we aim to improve the understanding of the underlying mechanisms of the pathogenesis of CAP from pathogen-induced local lesions to systemic inflammation and sepsis by developing a comprehensive dynamic mathematical model of pneumonia.
As contribution to this consortium, we designed a murine pneumonia model that allows for mathematical translation. After transnasal infection with S. pn., we analyzed clinical and inflammatory parameters in infected mice. To resemble the timeframe of patients' antibiotic therapy commencement, we treated experimental groups with antibiotics, starting early (24h) and late (48h) post infection (p.i.).
In our murine S. pn. infection model, we observed bacterial growth at first in alveolar spaces and lungs which progressed to bacteremia in blood. Bacterial loads were reduced by antibiotic treatment as expected. Significant breakdown of the alveolar-epithelial barrier was observed from 48h p.i. and could be inhibited by antibiotic therapy starting 24h p.i.. Inflammatory cytokine and chemokine levels increased with bacterial burden during the disease. Notably, protein levels of vascular endothelial growth factor (VEGF), a signal protein acting on endothelial permeability, were increased and correlated with lung barrier breakdown. Furthermore, recruitment of neutrophils and inflammatory macrophages/monocytes into alveolar spaces correlated well with the levels of cell-type attracting chemokines CXCL1/KC and MCP1/CCL2 measured in bronchoalveolar lavage fluid, respectively.
Collectively, through integration of both murine and human patient data into a combined mathematical model of pneumonia, we aim to discover early targets that can help predict the patients' course of disease and to adjust the individual therapy.
*Geraldine Nouailles and Martin Witzenrath are contributed equally to this work