DNA-SPECIFIC BINDING BY HIN AND FIS. Thang Chiu, Reid Johnson, Richard E. Dickerson, MBI, UCLA, CA 90095

We are interested in learning how dna-binding proteins recognize their target sequences. One system of particular interest is the Hin/Fis system of recombination. Both proteins bind their target sequences via a Helix-Turn-Helix. Hin belongs to a family of proteins that catalyses site-specific DNA inversion in enteric bacteria. Its binding site consists of a highly conserved inverted repeat of 12-bp separated by a central 2-bp 'core'. (consensus sequence: A/T G G T T T A/T G G A G/T A A) The availability of comprehensive mutagenesis data of the binding sites makes it a highly attractive system for studying protein-dna interactions. The crystal structure of a 52 aa peptide consisting the dna-binding-domain of E.coli Hin bound to the hixL half-site tGTTTTTGATAAGA/aTCTTATCAAAAAC has been solved (Feng et al, Science '94). We are interested in solving the crystal structures of Hin bound to various mutant binding sites in order to understand the mechanism by which its dna-binding specificity is determined.

Fis (Factor for Inversion Stimulation) is a recombinational activator of the Hin family of dna invertases and is also involved in phage lambda site-specific recombination, in transcriptional activation of rRNA and tRNA operons, in repression of its own synthesis, and in oriC-directed DNA replication. Although it is a site-specific dna binding protein, a comparison of over 35 Fis binding sites reveals a 25-bp binding site which has, at most, a degenerate 15-bp core. Gel electrophoresis of Fis-dna complexes indicate that Fis induces a bend (in the dna) of 40-90 degree upon dna binding. This bending is required for its activity and the degree of bending (and binding) is dependent on both the sequence of the 15-bp core and the sequence of the flanking regions. Given the degenerate nature of the binding sites, the first question one might ask is "how does Fis differentiate a true binding site from a false binding site". A first step towards addressing this question is to crystallize (at least one) Fis binding site, alone and complexed to Fis protein, and solve their crystal structures. Efforts toward achieving this goal are underway.