Published Research - Department of Chemistry


Date of this Version



Published in Biochemistry 48 (2009), pp 1801–1809; doi: 10.1021/bi8018724 Copyright © 2009 American Chemical Society.
Note: The American Chemical Society, publishers of this article, permit the archiving of only the abstract, tables, and figures from their publications. The full text of this article has been deposited with NIH PubMed Central and will appear there soon. [3/22/2010]


Studies of the binding and bending of the AdMLP TATA sequence (TATAAAAG) by the core domain of yeast TBP allow quantitation of the roles of the N-terminal domains of yeast and human TBP. All three proteins bind DNA via a three-step mechanism with no evidence for an initially bound but unbent DNA. The large enthalpy and entropy of activation for the first step in yTBP binding can now be assigned to movement of the NTD from the DNA binding pocket and not to energetics of DNA bending. The energetic patterns for hTBP and cTBP suggest that the 158-amino acid NTD in hTBP does not initially occupy the DNA binding pocket. Despite the appearance of similar energetics for hTBP and cTBP, order of magnitude differences in rate constants lead to differing populations of intermediates during DNA binding. We find that the NTDs destabilize the three bound forms of DNA for both yTBP and hTBP. For all three proteins, the DNA bend angle (θ) depends on the TATA sequence, with θ for cTBP and hTBP being greater than that for yTBP. For all three proteins, θ for the G6 variant (TATAAGAG) varies with temperature and increases in the presence of osmolyte to be similar to that of AdMLP. Crystallographic studies of cTBP binding to a number of variants had shown no dependence of DNA bending on sequence. The results reported here reveal a clear structural difference for the bound DNA in solution versus the crystal; we attribute the difference to the presence of osmolytes in the crystals.

Included in

Chemistry Commons