Book of Abstracts: Albany 2011

category image Albany 2011
Conversation 17
June 14-18 2011
©Adenine Press (2010)

Intrinsic DNA Shape of Fis-DNA Binding Sites Determines Binding Affinity

The bacterial nucleoid-associated protein Fis (factor for inversion stimulation) binds to DNA of various DNA nucleotide sequences at binding affinities varying in three orders of magnitude. Johnson and coworkers recently reported 11 crystal structures of the Fis protein bound to high- and low-affinity sites (1). A common feature of these Fis-DNA complexes is the narrow minor groove in the central region of the DNA targets. Without the input of any structural data derived from experimental studies, we predicted the shape of the naked Fis-DNA binding sites based on our Monte Carlo algorithm (2). For this purpose, we generated an ideal B-DNA double helix of a given sequence without any structural identity of dinucleotide steps and sampled the three-dimensional structure of high- and low-affinity Fis-DNA binding sites. In agreement with the hypothesis presented by Johnson and coworkers (1), we find that the shape of the naked binding site with the highest Fis-DNA binding affinity already assumes a shape with a minor groove narrowing in its central region similar to the one observed in the crystal structure of the complex. In contrast, the Fis-DNA site with the lowest binding affinity exhibits in its unbound state a shape that is essentially the one of average B-DNA. Although the Fis protein binds DNA largely non-specifically, the finding that DNA shape determines binding affinity indicates a similar mechanism as the one that we previously described for the sequence-specific DNA recognition by the papillomavirus E2 protein. The E2 protein binds to DNA that is intrinsically bent as observed in the protein-DNA complex with high affinity and to DNA that is essentially straight with low affinity (2). The data presented for the larger set of Fis-DNA binding sites further validates our structure prediction method and emphasizes the important role of DNA shape in protein-DNA recognition (3, 4). Moreover, protein-DNA binding affinity was shown to be correlated with gene regulatory control (5), thus suggesting that structural data that explains binding affinity leads to a better understanding of protein-DNA recognition, both for largely non-specific DNA interactions with architectural proteins and highly specific DNA readout by transcription factors.


  1. S. Stella, D. Cascio and R.C. Johnson, Genes Dev 24, 814-26
  2. R. Rohs, H. Sklenar and Z. Shakked, Structure 13, 1499-509 (2005).
  3. R. Rohs, S.M. West, A. Sosinsky, P. Liu, R.S. Mann and B. Honig, Nature 461, 1248-53 (2009).
  4. R. Rohs, X. Jin, S.M. West, R. Joshi, B. Honig and R.S. Mann, Annu Rev Biochem 79, 233-69 (2010).
  5. S.V. Nuzhdin, A. Rychkova and M.W. Hahn, Trends Genet 26, 51-3 (2010).

Tianyin Zhou
Ana Carolina Dantas Machado
Remo Rohs

Molecular and Computational Biology Program
Department of Biological Sciences
University of Southern California
1050 Childs Way, RRI 404C
Los Angeles, CA 90089, U.S.A.

Ph: +1-213-740-0552