Characterization of Brn-3.0 and identification of a highly related member (Brn-3.1) of the class IV POUdomain family suggest potential roles of Brn-3.0 in the development of retinal ganglion cells and sensory neurons, as well as potential roles in the pituitary gland and the immune system. Brn-3.0 is expressed in the pituitary gland and in a corticotroph cell line. A functional DNA response element has been identified in the proopiomelanocortin promoter. In contrast to previously described mammalian POU-domain proteins, Brn-3.0 binds relatively ineffectively to known octamer DNA motifs, but instead binds with high affinity to a distinct set ofDNA elements, functioning as a transcriptional activator. Brn-3.0, Brn-3

Investigation of the large POU domain family of developmental regulators has revealed a molecular mechanism by which highly related transcription factors sharing common DNA-binding motifs act to functionally discriminate their cognate DNA sequences. Studies of two classes of neuron-specific POU domain factors {III and IV) indicate that functional specificity on their native response elements is achieved by accommodating different nucleotide spacing between variably oriented bipartite core DNA-binding motifs. The preferred orientation of the POU-specific domain of the neuronal factors on their native response elements appears to be opposite that of Pit-1 and Oct-1. Members of POU-III {Brn-2) class exhibit remarkable flexibility in DNA site recognition {tolerating core motifs spaced by 0, 2, or 3 nucleotides), whereas POU-IV IBm-3) class is highly constrained {tolerating core motifs with a spacing of 3 nucleotides). The molecular determinant of the constraint in DNA site selection appears to be imparted by 3 amino acid residues in the amino-terminal basic region in concert, with helix 2 of the POU homeo domain which together are involved in minor groove and possibly phosphate backbone contacts. Similar mechanisms may underlie differential flexibility in spacing and orientation for diverse families of transcription factors.

Nerve regeneration across a 10-mm gap was delayed in streptozotocin diabetic rats 3 and 4 weeks after transecting the sciatic nerve. Opposite ends of each cut nerve were introduced into a silicone tube, leaving a 10-mm gap. Electron microscopy was used to evaluate the progress of regeneration in sections at 2-mm intervals across the 10-mm gap. After 3 weeks, control axons had bridged the 10-mm gap, and myelin sheaths extended for 6-8 mm. By contrast, axons and their myelin sheaths were seen no further than 2 mm from the proximal stump in diabetic animals. By 4 weeks, axons had bridged the gap in diabetics; however, they appeared immature and showed dystrophic changes. The findings suggest that although regeneration does occur in diabetic nerves, it is significantly delayed and qualitatively impaired.