Solution structure of the POU-specific DNA-binding domain of Oct-1

THE transcription factor Oct-1 belongs to a family containing a POU DNA-binding domain. This bipartite domain is composed of a POU-specific domain (POU(s)) and a POU-homeodomain (POU(hd)) connected by a flexible linker. The left half of the optimal POU binding site, the octamer ATGCAAAT, is recognized by POU(s) and the right half by POU(hd). We have determined the solution structure of POU(s) by nuclear magnetic resonance. It consists of four alpha-helices connected by short loops. Helices I and IV are in a parallel coiled-coil arrangement. The folding topology appears to be similar to that of the bacteriophage lambda-repressor and 434 repressor. For the well defined parts of the protein (residues 1-71), the average root-mean square deviation for the backbone atoms is 0.9 angstrom. Based on the observed selective exchange broadening in the (N-15, H-1)-HMQC (heteronuclear multiple quantum coherence) spectrum of the POU(s)-DNA complex we conclude that DNA-binding is mediated by helix III. We propose a model for the POU-DNA complex in which both recognition helices from the two subdomains have adjacent positions in the major groove.