Z25 - Lipopolysaccharide and Bacterial Extracellular Vesicles Upregulate Endothelial Intercellular Adhesion Molecule-1 but do not Increase Permeability of a Blood-Brain Barrier Lab-on-Chip

Introduction: : Sepsis occurs when the host immune system becomes hyperactive in response to infection and is responsible for 11 million deaths every year. The blood-brain barrier (BBB) is particularly threatened during sepsis since the immune response can destabilize this barrier and allow unwanted molecules and cells to enter the central nervous system to induce changes in brain chemistry. Furthermore, the BBB remains compromised following sepsis recovery, increasing the chances for sepsis survivors to develop neurological disorders. In order to study the initiation of the cascade that leads to sepsis, we investigated how bacterial products such as lipopolysaccharide (LPS) and bacterial extracellular vesicles (BEVs) could inflame the BBB and prime the endothelial cells for interactions with immune cells. This was accomplished in our lab-on-chip model featuring isogenic stem cell-derived brain microvascular endothelial cells (BMECs) and pericytes.

Materials and

Methods: : Purified LPS was obtained from a commercial source, while BEVs were isolated from E. coli-conditioned LB Broth via ultracentrifugation. Preliminary experiments with BMEC monocultures were conducted in multiwell plates. For co-culture experiments, we used lab-on-chip devices featuring an ultrathin silicon nitride nanoporous membrane positioned between two media-filled compartments. Stem cell-derived brain pericyte-like cells were seeded on the underside of the membrane and BMECs were seeded on top of the membrane. Following one week of culture, the cells were subjected to 16 - 18 h treatments with LPS, BEVs, or a pro-inflammatory cytomix composed of equal parts TNF-α, IL-1β, and IFN-γ. We used immunofluorescence labeling to examine BMEC expression of intercellular adhesion molecule-1 (ICAM-1), an endothelial cell surface marker of inflammation responsible for attracting passing immune cells. In addition, we labeled the basement membrane proteins collagen IV, laminin, and fibronectin to see if they were disrupted by pro-inflammatory treatments. Finally, we inspected barrier permeability by adding lucifer yellow dye to the top compartment of the lab-on-chip devices and allowing it to diffuse through the nanoporous membrane for 1 h before collection and measurement of fluorescence intensity.

Results, Conclusions, and Discussions:: In multiwell plate experiments, treatment with 100 ng/ml LPS caused BMEC expression of ICAM-1 to increase during the first 10 h and remain elevated at least 24 h after treatment onset. However 0.5 ng/ml LPS, which is equivalent to higher-end concentrations measured in the blood of sepsis patients, was insufficient to compromise barrier function in our lab-on-chip model (Fig 1A,B). There was no significant increase in BMEC expression of ICAM-1 until exposure to supraphysiological concentrations of LPS (Fig 1A,B). Furthermore, LPS failed to induce significant increases in permeability to lucifer yellow dye (Fig 1C), nor did it cause a significant loss of basement membrane proteins. Similarly, high concentrations of BEVs failed to cause changes in permeability to lucifer yellow dye within the assay period (Fig 1D). It is probable that, although these bacterial factors can help prime the endothelium for interactions with immune cells via upregulating ICAM-1 expression, these factors do not induce substantial BBB disruption directly. Rather, the pro-inflammatory factors that are primarily responsible for compromising the BBB appear to be host cytokines secreted by immune cells, which have demonstrated an ability to promote increased permeability in our model. Having tested the direct effects of the bacterial products, we plan to engineer a more complete sepsis model that incorporates immune cells to determine whether priming the endothelium with bacterial products can increase immune cell transmigration through the membrane and exacerbate barrier disruption.

Acknowledgements and/or References (Optional)::