Hair cells of the inner ear are champion mechanosensors, opening their ion channels in response to the slightest displacement of the tufts of stereocilia perched on their apical ends. With them we can hear a whisper, get a thrill on a roller coaster, and find our balance when rising in the dead of night. But age can take its toll, such that the whisper fades, the coaster nauseates and the dark menaces. The hair cells are dying, sometimes in large numbers. Birds are able to replace the dead cells, but for mammals like us, this had seemed an impossible feat. To solve this problem in the old, one can look to the young. A flurry of recent papers reveals that embryos can make extra hair cells quite readily, often whole rows of them. Manipulations have ranged from treatment with retinoic acid to knocking out genes, including a cell cycle inhibitor or a ligand for the transmembrane signaling protein, Notch. Now two new papers1xNotch signaling regulates the pattern of auditory hair cell differentiation in mammals. Zine, A. et al. Development. 2000; 127: 3373–3383PubMedSee all References, 2xOverexpression of Math1 induces robust production of extra hair cells in postnatal rat inner ears. Zheng, J.L. and Gao, W-Q. Nature Neuroscience. 2000; 3: 580–586Crossref | PubMed | Scopus (363)See all References, both using cultured hearing organs of normal perinatal rats, can be added to this growing list.Zine et al.1xNotch signaling regulates the pattern of auditory hair cell differentiation in mammals. Zine, A. et al. Development. 2000; 127: 3373–3383PubMedSee all References1 are continuing an emerging story about Notch and hair cell fate specification. In a loss-of-function approach, cultured organs of Corti were transfected with antisense oligonucleotides directed against messages for the Notch1 receptor or one of its ligands, Jagged1, using lipofection. Both oligos induced many rows of extra hair cells. The extra hair cells arose right next to the normal rows, and in some cases these might have originated from a conversion of the supporting cells that normally separate hair cells from one another. The response was best in the youngest specimens, declining to low levels by postnatal day three.Last year, Bermingham and colleagues3xMath1: an essential gene for the generation of inner ear hair cells. Bermingham, N.A. et al. Science. 1999; 284: 1837–1841Crossref | PubMed | Scopus (615)See all References3 used a knockout approach to prove that Math1, a basic helix-loop-helix transcription factor related to the fly atonal gene, is absolutely required to make hair cells in mice. Zheng and Gao2xOverexpression of Math1 induces robust production of extra hair cells in postnatal rat inner ears. Zheng, J.L. and Gao, W-Q. Nature Neuroscience. 2000; 3: 580–586Crossref | PubMed | Scopus (363)See all References2 have taken the reverse tack, using a gain-of-function approach to express Math1 in places where ordinarily it is absent. Math1 DNA was placed in an expression plasmid that also encoded green fluorescent protein (GFP). This was driven into the organ cultures using electroporation. Glowing ‘non-sensory’ cells producing GFP alone did nothing unusual. But with Math1, within 12 days, many became hair cells, complete with ultrastructural and molecular hallmarks. Curiously, the only transfected cells able to accomplish this were located about 30–100 microns away from the nearest normal hair cells, in a region where many cells normally die within a few weeks after birth. Determining whether the converted hair cells persist will require an intact animal model.There is a negative regulatory link between the Notch pathway and other nuclear transcription factors related to Math1. This suggests that the two studies are probably tweaking a common pathway directing hair cell fate, despite obvious differences in their transfection methods and target populations. While the therapeutic value of such studies is still a few years away, the future is looking brighter (or sounding louder, as the case may be).
Read full abstract