Corneal macrophages: an investigation of the cell-cell and neuro-immune interactions of local macrophages in homeostatic and injured murine corneas

Abstract

The cornea is the most anterior portion of the corneoscleral envelope and, together with the conjunctiva, provides a physical and immunological barrier to foreign particles and pathogens. Like other epithelial tissues, sensory nerves densely innervate the corneal epithelium. The innervation serves a reflex function to protect the eyes and to regulate tear film secretion. The cornea also plays a major role in the refraction of light. Accordingly, its transparency is crucial in maintaining vision. An inflammatory response to infection or trauma can, however, cause the oedema of the collagenous stroma, altering the particular arrangement of keratocytes embedded within the collagenous lamellae, thus compromising vision. Despite the existence of resident macrophages and dendritic cells (DCs), distributed throughout the corneal epithelium and stroma, very little is known about their biological role, movement, method of communication over long distances, and interactions with other cell types in the cornea following injury to the epithelium. This project, therefore, looked to investigate cellular communication between resident immune cells within the corneal stroma. The first contribution of this study to the field of ocular immunology was the in vivo characterisation of membrane nanotubes (MNTs) in the murine cornea using live cell imaging. In vitro studies have provided strong evidence that MNTs enable communication and the exchange of proteins and organelles between connected immune cells. Analysis of corneal flatmounts from Cx3cr1GFP and CD11ceYFP transgenic mice revealed that both DCs and macrophages were capable of forming MNTs following abrasion of the corneal epithelium and exposure to pathogenic stimuli (lipopolysaccharide – LPS), in response to a clinically relevant HSV-1 infection, as well as other stress stimuli. Myeloid derived cells in the iris and dura mater were also found to be capable of forming MNTs, suggesting MNTs in vivo are not a phenomenon limited to the cornea. Analysis of live-cell imaging of fresh explanted corneas revealed MNTs to form de novo, extending from the cell soma at a greater rate than those previously reported in cell culture studies. Whilst the Cx3cr1GFP expression was apparent in all MNTs captured, indicating cytoplasmic continuity within the MNT and between connected cells, the narrow diameter of MNTs in the mouse cornea likely precludes the transfer of large cellular organelles. A second contribution of this study to the field of ocular immunology was the identification of a novel sub-population of resident macrophages that intimately associated with peripheral corneal nerves. This sub-population of cells was therefore nominated as “nerve-associated macrophages” (NAMs). Investigations of other ocular and non-ocular connective tissues revealed some interaction between resident immune cells and large βIII-tubulin+ peripheral nerve fibres in the skin, dura, iris and connective tissue of the cremaster muscle. This association was not as distinctive as the peripheral nerve trunks of the cornea, indicating that the lack of a myelin sheath around nerves in the cornea might facilitate a more distinct interaction between immune cells and nerves than in other myelinated nerves. This study also revealed NAMs disassociate from peripheral corneal nerve trunks within two hours of injury to the central corneal epithelium; this disassociation was found to be Cx3cr1-dependent. Application of anaesthetic to a naive cornea also revealed a decrease in NAM density, suggesting that alterations in the corneal sensory nerves affect NAMs. NAM density returned to baseline 72 hours after corneal injury. These findings may implicate NAMs to possibly be the early responders to corneal damage, and as such, a potential alternative ocular surface defence mechanism. In summary, the findings presented here support a role for NAMs in the peripheral cornea and MNTs following injury to the central corneal epithelium. Furthermore, it highlights that MNTs and NAMs both respond to stress stimuli and likely play an important role in the early stages of the innate immune response in the mammalian cornea. In conclusion, these data provide evidence to refute the notion that resident mononuclear phagocytes in the cornea are immunologically inert, as well as providing direct evidence to suggest that such cells are interconnected with one another and with sensory nerves of the cornea, and may contribute to shaping an adaptive immune response. A collective understanding of the cell biology of resident corneal immune cells, such as the early response of NAM and role of MNTs in corneal infectious disease progression, can ultimately lead to the development of potential therapeutic interventions aimed at regulating and reducing the harmful effects of inflammatory responses in the cornea.