Patterning the vertebrate ear requires the coordinated expression of genes that are involved in morphogenesis, neurogenesis, and hair cell formation. The zinc finger gene GATA-3 is expressed both in the inner ear and in afferent and efferent auditory neurons. Specifically, GATA-3 is expressed in a population of neurons in rhombomere 4 that extend their axons across the floor plate of rhombomere 4 (r4) at embryonic day 10 (E10) and reach the sensory epithelia of the ear by E13.5. The distribution of their cell bodies corresponds to that of the cell bodies of the cochlear and vestibular efferent neurons as revealed by labeling with tracers. Both GATA-3 heterozygous and GATA-3 null mutant mice show unusual axonal projections, such as misrouted crossing fibers and fibers in the facial nerve, that are absent in wild-type littermates. This suggests that GATA-3 is involved in the pathfinding of efferent neuron axons that navigate to the ear. In the ear, GATA-3 is expressed inside the otocyst and the surrounding periotic mesenchyme. The latter expression is in areas of branching of the developing ear leading to the formation of semicircular canals. Ears of GATA-3 null mutants remain cystic, with a single extension of the endolymphatic duct and no formation of semicircular canals or saccular and utricular recesses. Thus, both the distribution of GATA-3 and the effects of null mutations on the ear suggest involvement of GATA-3 in morphogenesis of the ear. This study shows for the first time that a zinc finger factor is involved in axonal navigation of the inner ear efferent neurons and, simultaneously, in the morphogenesis of the inner ear.

Most cystic fibrosis (CF) patients produce a mutant form (delta F508) of the cystic fibrosis transmembrane conductance regulator (CFTR), which is not properly processed in normal cells but is active as a chloride channel in several experimental systems. We used a double homologous recombination (‘Hit and Run’) procedure to generate a mouse model for the delta F508 mutation. Targeted embryonic stem (ES) cells (Hit clones) were found; of these either 80 or 20% of the clones had lost the delta F508 mutation, depending on the distance between the linearization site in the targeting construct and the delta F508 mutation. Correctly targeted clones underwent a second selection step resulting in ES cell clones (Run clones) heterozygous for the delta F508 mutation with an efficiency of 2–7%. Chimeric mice were generated and offspring homozygous for the delta F508 mutation showed electrophysiological abnormalities in nasal epithelium, gallbladder and in the intestine, and histological abnormalities in the intestine, typical of CF. Our data suggest that the delta F508 mice have residual delta F508 CFTR activity which would explain the mild pathology of the delta F508 mice. The delta F508 mouse may provide a useful model for the study of the processing defect of delta F508 CFTR and for the development of novel therapeutic approaches based on circumvention of the processing block.

The transcriptional control of the differentiation of central serotonergic (5-HT) neurons in vertebrates has recently come under scrutiny and has been shown to involve the homeobox genes Nkx2-2 and Lmx1b, the Ets-domain gene Pet1 (also known as Fev) and the zinc-finger gene Gata3. The basic helix-loop-helix (bHLH) gene Ascl1 (also known as Mash1) is coexpressed with Nkx2-2 in the neuroepithelial domain of the hindbrain, which gives rise to 5-HT neurons. Here we show in the mouse that Ascl1 is essential for the birth of 5-HT neurons, both as a proneural gene for the production of postmitotic neuronal precursors and as a determinant of the serotonergic phenotype for the parallel activation of Gata3, Lmx1b and Pet1. Thus Ascl1, which is essential for noradrenergic differentiation, is also a determinant of the serotonergic phenotype.