Abstract
Purpose :
The cornea is an essential transparent barrier that protects the eye from exposure to microbes on a daily basis. Wound healing is critical for maintaining this protective barrier. We recently demonstrated that certain bacteria prevent corneal wound healing in vitro and ex vivo. With the contact lens-associated keratitis pathogen Serratia marcescens, LPS was responsible for this inhibition. This study sought to determine the impact of LPS-containing bacterial secretomes on the host response of corneal epithelial cells in vitro.
Methods :
Growth medium (mock) and sterile bacterial culture filtrates (secretomes) were added to human corneal limbal epithelial (HCLE) cells. HCLE supernatants from mock and secretome treated cells were collected 24 hours post challenge and analyzed by cytokine array, ELISA, microarray, and metabolomic analysis. Contents of challenged HCLE cells were also metabolomically analyzed. To identify the role of TLR4 in S. marcescens inhibition of wound healing, HCLEs were independently treated with OxPAPC and CLI-095 inhibitors and tested in cell migration assays.
Results :
HCLEs treated with S. marcescens secretomes resulted in increased IL-1α, IL-1β, GM-CSF, and IL-6 as determined by cytokine array, ELISA for IL-1 β, and microarray. Metabolomics from secretome treated HCLEs had elevated levels of phosphoethanolamine, a phosphorylated amino alcohol associated with autophagy, lipid signaling, and apoptosis. TLR4 inhibition of HCLE cells did not prevent S. marcescens secretome inhibition of corneal wound healing, suggesting this inhibition is TLR4 independent.
Conclusions :
Using a combination of transcriptomics and metabolomics this study provides insight into the impact of bacteria on human corneal epithelial cells. Together these data support when HCLEs are treated with S. marcescens secretomes, there is an increase in pro-inflammatory markers and small molecules consistent with activation of the autophagy pathway correlating with altering epithelial cell wound healing.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.