Longitudinal Changes to Tight Junction Expression and Endothelial Cell Integrity in a Mouse Model of Sterile Corneal Inflammation

Stromal edema commonly accompanies corneal inflammation in a range of disorders including infectious keratitis. While edema would be expected in any focal tissue inflammation, in cornea it is possible that edema is not a consequence of stromal inflammation alone but also focal subjacent malfunction of corneal endothelium, the cell monolayer that functions to prevent hydration and swelling of the stroma. Is it the case that changes in the endothelium result from inflammation in the overlying stroma even though separated by a relatively thick Descemet's membrane? 

The article by Downie and colleagues¹ describes a rodent model that addresses this question. Using correlation of optical coherence tomography (OCT) measurements of stromal thickness with intrastromal CD45⁺ cell density, endothelial monolayer cell density, and morphological changes, they demonstrate a response in the endothelium to topical application of a TLR (Troll-like Receptor)-9 agonist following corneal abrasion that is independent of tight junction protein ZO (Zonula Occludens)-1 expression. Accordingly we now have information that disturbance of endothelium does contribute to corneal thickening in stromal inflammation, and this model provides some information about mechanisms on which there has been little earlier research. This work builds on earlier reports of functional LPS (lipopolysaccharide) receptor proteins CD14 and TLR-4 in human corneal cells² and demonstration that CD14⁺ monocyte-derived macrophages can directly induce cytokine-mediated corneal endothelial cell death ex vivo.³ 

The importance of this research question in human corneal disease lies in the fact that the endothelium in man is essentially nonreplicative. One consequence is that severe or chronic stromal inflammation, whether or not sterile and in a wide range of disorders, causes visually significant decline in endothelial cell function and irreversible loss of stromal transparency. These findings have implications for our understanding and treatment of innate immunologic responses to disorders such as corneal infections.