DEFINING BACTERIAL EXTRACELLULAR VESICLE INTERACTIONS WITH TARGET CELLS: IMPLICATIONS FOR WOUND REPAIR AND GUT-BRAIN AXIS COMMUNICATION

Abstract

Bacterial extracellular vesicles (BEVs) are lipid nanoparticles loaded with bioactive cargo that mediate microbe-microbe and microbe-host cell communication. Recent reports suggest that probiotic bacteria can play a beneficial role in wound healing and are essential in the gut-brain axis. Due to their secondary role in wound healing and their long-distance effect on the brain, we hypothesize that probiotic BEVs are the main effector in this phenomenon. Therefore, this study evaluates the biological effects of probiotic BEVs in wound healing and neuropod-like enteroendocrine signaling. Three probiotic strains, Lactiplantibacillus plantarum, Limosilactobacillus reuteri, and Bifidobacterium longum subsp. infantis, were selected as models due to their established beneficial probiotic capabilities. In the context of wound healing, the pro-migratory, pro-angiogenic, and anti-inflammatory activity of the BEVs were assessed using scratch assays, tube formation assays, and macrophage cytokine release analysis, respectively. Here, it was found that the vesicles promoted endothelial migration and angiogenesis in a strain- and dose-dependent manner, with Lactiplantibacillus plantarum showing strong effects on migration and Bifidobacterium longum subsp. infantis enhancing angiogenic responses. At higher dosages, inflammatory signaling was reduced, suggesting immunomodulatory potential.

In neuropod-like models, STC-1 and NCI-H716 cells were used to evaluate vesicle uptake, viability, calcium signaling, and gene expression. Both cell lines efficiently internalized vesicles without inducing cytotoxicity. Despite successful uptake, vesicle treatment did not produce strong calcium flux or significant transcriptional changes in genes associated with neuroendocrine signaling. These findings suggest that host cell-vesicle interactions alone are not sufficient to drive rapid signaling responses in these models.