STRUCTURE OF THE B-MYB DNA-BINDING DOMAIN IN SOLUTION AND EVIDENCE FOR MULTIPLE CONFORMATIONS IN THE REGION OF REPEAT-2 INVOLVED IN DNA-BINDING - IMPLICATIONS FOR SEQUENCE-SPECIFIC DNA-BINDING BY MYB PROTEINS

A range of double and triple resonance heteronuclear NMR experiments h as been used to obtain nearly complete sequence-specific N-15, C-13 an d H-1 resonance assignments for a 110-residue protein corresponding to the B-Myb DNA-binding domain (B-MybR2R3) and to determine its seconda ry structure in solution. The protein was found to contain two stable helices in repeat-2 (R2) and three in repeat-3 (R3), involving residue s K12-K24 (R2-1), W30-H36 (R2-2), E64-V76 (R3-1), W81-L87 (R3-2) and D 93-K105 (R3-3). In addition, the chemical shift and nuclear Overhauser effect data suggest that amino acids Q44-W49 near the C-terminus of R 2 form an unstable or nascent helix, which could be stabilised on bind ing to a specific DNA target site. The two N-terminal helices in R2 an d R3 occupy essentially identical positions in the two domains, consis tent with the high level of sequence similarity between these regions. In contrast, the C-terminal region forming the third helix in R3 show s low sequence similarity with R2, accounting for the differences in s econdary structure. In the case of B-MybR2R3, there is clear chemical shift and line-broadening evidence for the existence of multiple confo rmations in the C-terminal region of R2, which is believed to form one half of the DNA-binding site. We propose that conformational instabil ity of part of the DNA-binding motif is a way of increasing the specif icity of Myb proteins for a relatively short (6-bp) DNA target site by reducing their affinity for non-specific DNA sequences compared to sp ecific sites.