D14-04 Efficacy of Mesenchymal Stromal Cells for Bacterial Pneumonia in Mice Mediated by Mitochondrial Transfer to Both Macrophages and Neutrophils

Abstract Introduction The therapeutic potential of mesenchymal stromal cells (MSCs) for ARDS may depend on identifying treatment-responsive biologic profiles. 1 In our mouse model of pneumococcal pneumonia, dexamethasone and IL-6 receptor antagonist reduced lung injury only in hyper-inflammatory mice identified by latent class analysis (LCA) 2. Because the majority of patients with ARDS are hypo-inflammatory, therapies for this phenotype are needed. We evaluated the efficacy of MSCs in this model and tested whether their effects differed by inflammatory phenotype. We also tested the contribution of mitochondrial transfer from MSCs to host immune cells, because this process has been shown to enhance cellular bioenergetics and to promote bacterial clearance. 3 methods C57BL/6 female mice (n = 75) were intranasally inoculated with 2x10⁸ colony forming units (cfu) of Streptococcus pneumoniae. Human clinical grade MSCs from the STAT trial were used1. At 18 hours post-infection, submandibular blood was collected, and 500, 000 MSCs cells or vehicle were injected retro-orbitally. Ceftriaxone 150mg/kg was given intraperitoneally at 18 and 30 hours. We monitored oxygenation with SpO2. At 36 hours, the mice were sacrificed and bronchoalveolar (BAL) fluid were collected. LCA was done following established methodologies from prior studies. To increase the statistical power of the LCA, we combined these mice (n = 75) with mice from a previous study that followed the same experimental protocol (n = 222). The variables included in the LCA were body weight, presence of bacteremia, and the 18 hours biomarker levels (IL-6, KC, TNFR-1, and MMP-8) 2. In a separate cohort of 30 mice, MSCs stained with Mitotracker Deep-Red, a mitochondria-labeling dye, were injected. Lungs were collected, digested into single-cell suspensions and analyzed by flow cytometry. Results LCA classified mice into two phenotypes. Class 1 reflected the hypo-inflammatory phenotype, while Class 2 the hyper-inflammatory phenotype, similar to those reported in human ARDS. MSCs reduced BAL protein, hypoxemia, and bacterial load at 36 hours for Class 1 but not for Class 2 (Figure 1A). Flow cytometric analysis of the single-cell suspensions demonstrated uptake of Mitotracker dye by macrophages, monocytes, and also by neutrophils (Figure 1B-C), indicating that MSCs transfer their mitochondria to these cells in vivo. Conclusions MSCs significantly reduced alveolar protein permeability, bacterial load and the degree of hypoxemia in a mouse model of ARDS from pneumococcal pneumonia in the hypo-inflammatory phenotype, but not in the hyper-inflammatory phenotype. The therapeutic effect of MSCs is mediated in part through mitochondrial transfer from MSCs to host macrophages and neutrophils. This abstract is funded by: NIH