Gap junctional communication in microinjected human limbal and peripheral corneal epithelial cells cultured on intact amniotic membrane.

The aim of the study was to determine if human limbal epithelial cells (HLEC) do not form gap junctions (GJ) during ex vivo expansion on preserved and intact human amniotic membrane (AM). Thereby, we attempt to evaluate if characteristic features of the limbal epithelial progenitor cells are preserved on AM. Primary human limbal (HLEC) and peripheral corneal (HPCEC) epithelial cells from limbal and peripheral corneal explants were cultured with SHEM either on intact AM or plastic. After 3-4 weeks, cell cultures were terminated and processed for immunofluorescence. In all cell cultures, formations of GJs were analyzed with a mouse monoclonal antibody to connexin 43 (Cx43) and a rabbit affinity purified antibody against connexin 26 (Cx26). Sections of human limbus and cornea served as positive control. Lucifer yellow (LY) known to be a GJ permeant dye was used to analyse functionality of GJ. Microinjection of LY into single cells was performed with a pressure microinjection device under visual control and with the aid of phase contrast optics. Dye spread of LY into adjacent cells indicating intercellular communication was compared between HLEC and HPCEC cultured either on AM or plastic. In vivo, a punctate pattern of Cx43 was typically found in basal and suprabasal corneal epithelial cells and labeling for Cx26 was observed in all cell layers of the human corneal epithelium, however, subpopulations of limbal basal epithelial cells lacked detectable fluorescence signals for both connexins. In HLEC cultured on AM, a scanty immunolabeling for Cx43 (12.6%) was noted, but HPCEC cultured on AM as well as HLEC cultured on plastic showed a higher labeling index (LI) for Cx43 (42.7 and 52.3%, respectively). A significant lower immunostaining for Cx26 was observed in HLEC cultured on AM (LI: 35.16%) in comparison to HLEC cultured on plastic (68.4%), as well as, HPCEC cultured either on AM or plastic (61% and 79.3%, respectively; p<0.001). Gap junctional communication was evidenced more frequently in HLEC cultured on plastic (51%, p<0.05) in contrast to HLEC cultures on AM, which exhibited a limited dye spread in 29.7% of injected cells. A significant difference in dye coupling was also evidenced between HPCEC on AM (52.9%; p<0.05) and HLEC on AM. Subpopulations of HLEC cultured on AM remain Cx43 and Cx26 negative and without functional GPs indicating that characteristic features of limbal epithelial progenitor cells might be preserved during ex-vivo expansion on AM. These data provide support to the use of the ex-vivo expansion of HLEC as an alternative therapeutic strategy for corneal surface reconstruction in distinct ocular surface diseases.