Previous experiments with mouse chimaeras

Previous experiments with mouse chimaeras and X-inactivation mosaics (Collinson et al., 2002, 2004; Mort et al., 2009) suggested that the numbers of active LESC clones decline with age in wild-type mice and are reduced in Pax6+/− heterozygotes. Although this would imply that LESC function is deficient in Pax6+/− and older mice the analysis of stripes in X-inactivation mosaics and chimaeras is related to the number of active coherent clones of LESCs and does not relate directly to the actual number of LESCs (Collinson et al., 2002; Mort et al., 2009). Keratin 5, K5 (also known as cytokeratin 5, CK5) is an intermediate filament protein, which is produced by the Krt5 gene in the basal epithelial layer of the skin and most stratified epithelial tissues including the conjunctival and limbal epithelia (Byrne and Fuchs, 1993) but its expression in the cornea has been more controversial. It was originally suggested that the mouse Krt5 gene was not expressed in the corneal epithelium and K5 immunostaining in this tissue detected long-lived keratin protein that was actually synthesised earlier in the limbal stem GDC0941 (Byrne and Fuchs, 1993). However, more recent RT-PCR experiments have shown that the Krt5 gene is expressed in the mouse cornea (Lu et al., 2006). We have produced several transgenic mouse lines in which the human KRT5 (keratin 5) promoter drives expression of β-galactosidase (β-gal) from the LacZ gene (Whitaker, 2004). Adult hemizygous KRT5 mice showed variegated β-gal expression in the pinna, back skin, tail skin and tongue and more variable staining in the trachea and tracheal submucosal glands. During the characterisation of one of these KRT5 lines we noted that β-gal was also expressed in a mosaic pattern in the conjunctiva and limbus of the ocular surface and that its expression sometimes extended to the corneal epithelium where it formed rare, individual radial stripes of β-gal-positive cells. This raised the possibilities that (1) analysis of the distribution of KRT5 stripes may help evaluate the new corneal epithelial stem cell hypothesis (Majo et al., 2008) versus the conventional limbal epithelial stem cell hypothesis in unmanipulated tissues and (2) the frequency of rare β-gal-positive stripes may provide a useful approach for comparing stem cell function in different groups of mice.
Results
Discussion
Materials and methods
Acknowledgments We thank Brendan Doe for performing pronuclear injections, staff at BRR, University of Edinburgh and the Transgenic Unit, MRC Human Genetics Unit for expert animal husbandry and specialised technical services, Dirk A. Kleinjan for providing Pax6 mice, Ted Pinner for help with the illustrations and Richard Mort for critical comments on the manuscript. This work was supported, in part, by a PhD studentship for PD from Fight for Sight (UK) and the RS MacDonald Charitable Trust, awarded to JW.
Introduction In the process of neural induction, a portion of ectodermal cells is influenced by signals from surrounding tissues [e.g., somites (S) and notochord (N)] which enable them to create neural ectoderm (Jones & Woodland, 1989; Hemmati-Brivanlou et al., 1990; Vincent Tropepe et al., 2001). Later, this neuroepithelia develops into the neural plate, fold and tube which contribute to the formation of the central nervous system (CNS) of which they need to be correctly patterned in order to generate distinct classes of neural progenitors (NPs) that will compose the forebrain, midbrain, hindbrain, and spinal cord. The initial neuroepithelial cells have a default forebrain or rostral identity that are differentiated, patterned, and specified into caudal neural cells by signals secreted from surrounding tissues such as the somites, notochord and dorsal ectoderm after a period of time (Pankratz et al., 2007; Stern, 2001). Early neuronal patterning has been retrieved largely from studies using Xenopus, chicks, and mice (Londin & ER, 2005; Weinstein, 1997; Wittler, 2004). However, neuronal patterning is still not completely understood in human. It remains to be determined as to how NPs are specified, distributed and become restricted to certain mature neurons over time along two major axes: rostrocaudal (RC) and dorsoventral (DV) (Temple, 2001).