Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
???displayArticle.abstract???
We have identified Math5, a mouse basic helix-loop-helix (bHLH) gene that is closely related to Drosophila atonal and Xenopus Xath5 and is largely restricted to the developing eye. Math5 retinal expression precedes differentiation of the first neurons and persists within progenitor cells until after birth. To position Math5 in a hierarchy of retinal development, we compared Math5 and Hes1 expression in wild-type and Pax6-deficient (Sey) embryos. Math5 expression is downregulated in Sey/+ eyes and abolished in Sey/Sey eye rudiments, whereas the bHLH gene Hes1 is upregulated in a similar dose-dependent manner. These results link Pax6 to the process of retinal neurogenesis and provide the first molecular correlate for the dosage-sensitivity of the Pax6 phenotype. During retinogenesis, Math5 is expressed significantly before NeuroD, Ngn2 or Mash1. To test whether these bHLH genes influence the fates of distinct classes of retinal neurons, we ectopically expressed Math5 and Mash1 in Xenopus retinal progenitors. Unexpectedly, lipofection of either mouse gene into the frog retina caused an increase in differentiated bipolar cells. Directed expression of Math5, but not Xath5, in Xenopus blastomeres produced an expanded retinal phenotype. We propose that Math5 acts as a proneural gene, but has properties different from its most closely related vertebrate family member, Xath5.
Fig. 5. Overexpression of Math5 in Xenopus embryos by RNA injection. (A) Injection into one cell of a 2-cell Xenopus embryo with RNA for nβgal alone (i) or in combination with Math5 (ii). Stage 14-15 embryos were stained to detect β-galactosidase activity (magenta) and probed by whole-mount in situ hybridization for N- tubulin expression (purple). Embryos are oriented in a dorsal view with anterior at the top and injected side on the right. (i) Control embryo expressing nβgal alone demonstrating the normal pattern of N-tubulin expression in the neural plate. (ii) Embryo expressing both Math5 and nβgal with ectopic N-tubulin on the injected side.
(B) Injection into blastomere D.1.1 of a 16-cell Xenopus embryo with RNA for GFP alone (i-iii) or with RNA for Math5 plus GFP (iv- vi). The embryos were fixed and cryostat sectioned at stage 41. Sections were immunostained for NCAM and labeled with Hoechst to visualize the retinal cell layers. (i-iii) Retinal section from a control embryo injected with GFP RNA alone. Hoechst staining reveals normal lamination (i). Within the retina, a large cluster of cells derived from the injected blastomere is labeled by GFP (ii) and robust NCAM staining is observed (iii). (iv-vi) A retinal section from an embryo injected with GFP and Math5 RNAs. Hoechst staining highlights a disruption in the normal arrangement of retinal cell layers (iv). This disrupted region corresponds to a cluster of GFP- labeled cells within the retina (v). The GFP-positive cells, including those extruding from the back of the retina, are NCAM-positive (vi). Scale bar 1 μm.
Fig. 6. In vivo lipofection of Math5 promotes retinal bipolar cell fates. Embryos were transfected at stage 18 in the region of the optic vesicle with DNA for GFP alone (A) or GFP in combination with Xath5 (B), Math5 (C) or Mash1 (not shown). At stage 41 these embryos were fixed, cryosectioned and stained with Hoechst dye to visualize the retinal cell layers. GFP-labeled cells were scored in retinal sections by laminar position and morphology. In section of embryos transfected with Xath5, there are more labeled cells in the RGC layer (B), while sections of embryos transfected with Math5 show both a larger number of GFP-labeled cells overall and more labeled bipolar cells (C). Bipolar cells were scored as cells with distinct cell bodies in the INL and with thin processes extending radially but not extending into the ONL or GCL. These can be distinguished from Müller glial cells, which are radially oriented but have a more complex morphology (see Dorsky et al, 1995). (D) Percentage of retinal cell types labeled following transfection with DNA for GFP (green, n=432 cells from 5 eyes), GFP plus Xath5 (red, n=939 cells from 5 eyes), Math5 (navy, n=866 cells from 7 eyes) or Mash1 (light blue, n=440 cells from 6 eyes). Xath5 data are derived from Kanekar et al. (1997) and are included here for comparison. Expression of Xath5 biases progenitors toward RGC fate, while both Math5 and Mash1 promote the formation of bipolar cells. The per cent representation of each cell type was calculated as a weighted average. Error bars indicate SEM; asterisk, P<0.01 by Student t-test. Cell types on the graph are listed from left to right in rough order of birth. Abbreviations: G, retinal ganglion cell; B, bipolar cell; P, photoreceptor; A, amacrine cell; H, horizontal cell. Scale bar 0.5 μm.
