Bipartite DNA recognition by the human Oct-2 POU domain: POUs-specific phosphate contacts are analogous to those of bacteriophage lambda repressor.

The POU motif defines a family of eukaryotic transcription factors broadly involved in tissue-specific gene expression and developmental regulation. The motif contains two DNA-binding domains: an N-terminal POU-specific domain (POUs) and C-terminal homeodomain (POUHD). Surprisingly, POUs has recently been found to be similar in structure to helix-turn-helix (HTH) domains of phage repressor and Cro proteins [Assa-Munt, N., Mortishire-Smith, R., Aurora, R., Herr, W., & Wright, P.E. (1993) Cell 73, 193-205; Dekker, N., Cox, M., Boelens, R., Verrijzer, C.P., van der Vliet, P.C., & Kaptein, R. (1993) Nature 362, 852-855]. Because POUHD and POUs are expected to bind DNA differently, we have used "methylphosphonate interference" to investigate the alignment of their HTH elements in a specific DNA complex. This neutral phosphate analogue, originally developed for applications in antisense drug design [Miller, P. S., & Ts'o, P. O. P. (1987) Anti-Cancer Drug Des. 2, 117-128], is shown to provide a sensitive probe for sites of backbone-specific protein-DNA interaction. Inferred POUs-phosphate contacts are in striking accord with cocrystal structures of bacteriophage repressor and Cro proteins. Alignment of POUHD and POUs in successive major grooves in each case predicts unique HTH-adenine contacts. This prediction is verified using DNA base analogues to effect interchange of AT functional groups by the method of 2'-deoxyinosine/5-methyl-2'-deoxycytosine substitution [McLaughlin, L. W., Benseler, F., Graeser, E., Piel, N., & Scholtissek, S. (1987) Biochemistry 26, 7238-7245]. Our results strongly support the hypothesis that the DNA-binding properties of POUs are analogous to those of bacteriophage lambda repressor.