guanylyl cyclase br Materials and methods br Results br Disc

Materials and methods
Results
Discussion The main goal of current investigation was to study the role of oxygen in the maintenance of LESC phenotype under conditions of biochemically undefined and a serum-free highly defined growth environment. The oxygen concentrations used in the present study were chosen to mimic the natural oxygen gradient from the superficial epithelial layer composed of PMCs and TDCs to the niche harboring LESCs. In vivo measurements with oxygen sensing electrodes, penetrating the central cornea in rabbits breathing normal air (Kwan et al., 1972), showed below the tear film an oxygen concentration of 16–17%, decreasing further at the epithelio-stromal junction to 8–9%. As the limbal epithelium has more layers than the central corneal epithelium (Feng and Simpson, 2008), and as human epithelium is thicker than rabbit epithelium (Reiser et al., 2005), the theoretical pericellular oxygen concentration at the human limbal basement membrane could easily reach around 2–5%. Finally, the closure of eye lid during sleep could shift the gradient to more hypoxic levels (Shimmura et al., 1998). The limbal capillary system should not change the hypoxic nature of the niche as the oxygen concentration in most peripheral guanylyl cyclase is in the range of 0.5–2.6% (Brahimi-Horn and Pouysségur, 2007). There are obviously great biological implications of such gradient, as ample evidence has been accumulated to suggest a key role for hypoxia in maintaining different types of stem cells (reviewed in Mohyeldin et al., 2010; Toussaint et al., 2011; Zachar et al., 2011). Although attempts have previously been done to evaluate the role of hypoxia in the control of LESCs (Miyashita et al., 2007; O\'Callaghan et al., 2011), the current study represents a progress in that it uses a hypoxic workstation that provides for a stable gaseous containment. The potentially damaging effect of re-oxygenation (Shimmura et al., 1998) has thus been eliminated. The panel of markers in the current study was selected to identify both the stem cell and the differentiated phenotypes. The stem cell marker ABCG2 is a membrane transporter expressed by a “side” population of cells excluding Hoechst 33342. ABCG2 (M. Kim et al., 2002) has been found in limbus (de Paiva et al., 2005; Watanabe et al., 2004) and it is postulated to confer a cell survival advantage to cells growing in hypoxic conditions by a hypoxia-inducible factor-1 (HIF-1) dependent pathway (Krishnamurthy and Schuetz, 2005; Krishnamurthy et al., 2004). The p63α has also been proposed to be a marker of LESCs, as it has been linked to holoclones in the 3T3 system (Pellegrini, 2001). It has also been shown that a high rate of bright p63α+ cells in holoclones increases success-rates of transplantation (Rama et al., 2010). The cytokeratin CK3 stains the corneal cells except for the basal cell layer in limbus, and is thus accepted as a differentiation marker (Pitz and Moll, 2002; Schermer, 1986). Our data seem to raise a question about the significance of p63α as a marker to identify LESCs. Previous research suggested a role for this marker in feeder-based cultures (Di Iorio, 2005), and indeed our own data demonstrate high expression of p63α, along with ABCG2, in hypoxia. In the EpiLife system, on the other hand, the major regulatory effect on p63α was probably exerted by the proliferative status of cultures, as this marker seemed downregulated with increasing confluence, irrespective of the level of hypoxia. Furthermore, the proportion of P63α expressing cells in EpiLife was too high to be considered stem-cell specific, ranging from 46% in 20% oxygen to an almost complete expression in 15% oxygen. Also, cell types could be found in both culture systems to co-express p63α and CK3. Taking into account previous data corroborating our observation about the effect of cell density on the expression of p63α in EpiLife (Salehi-Had et al., 2005), our findings indicate that ABCG2 might be a more relevant marker to identify the LESC phenotype. More work will be needed to obtain full understanding of the subpopulations of cultured corneal epithelial cells that are positive for p63α, and whether there are differences between the p63α+ phenotypes across the culture systems.