Modeling dry eye with an air–liquid interface in corneal epithelium-on-a-chip

Rodi Kado Abdalkader, Romanas Chaleckis, Takuya Fujita, Ken Ichiro Kamei

Research output: Contribution to journalArticlepeer-review

Abstract

Dry eye syndrome (DES) is a complex ocular condition characterized by an unstable tear film and inadequate tear production, leading to tissue damage. Despite its common occurrence, there is currently no comprehensive in vitro model that accurately reproduce the cellular characteristics of DES. Here we modified a corneal epithelium-on-a-chip (CEpOC) model to recapitulate DES by subjecting HCE-T human corneal epithelial cells to an air–liquid (AL) interface stimulus. We then assessed the effects of AL stimulation both in the presence and absence of diclofenac (DCF), non-steroidal anti-inflammatory drug. Transcriptomic analysis revealed distinct gene expression changes in response to AL and AL_DCF, affecting pathways related to development, epithelial structure, inflammation, and extracellular matrix remodeling. Both treatments upregulated PIEZO2, linked to corneal damage signaling, while downregulating OCLN, involved in cell–cell junctions. They increased the expression of inflammatory genes (e.g., IL-6) and reduced mucin production genes (e.g., MUC16), reflecting dry eye characteristics. Metabolomic analysis showed increased secretion of metabolites associated with cell damage and inflammation (e.g., methyl-2-oxovaleric acid, 3-methyl-2-oxobutanoic acid, lauroyl-carnitine) in response to AL and even more with AL_DCF, indicating a shift in cellular metabolism. This study showcases the potential use of AL stimulus within the CEpOC to induce cellular characteristics relevant to DES.

Original languageEnglish (US)
Article number4185
JournalScientific reports
Volume14
Issue number1
DOIs
StatePublished - Dec 2024

ASJC Scopus subject areas

  • General

Fingerprint

Dive into the research topics of 'Modeling dry eye with an air–liquid interface in corneal epithelium-on-a-chip'. Together they form a unique fingerprint.

Cite this