Identification of a biologically significant DNA-binding peptide motif by use of a random phage display library

A peptide library approach was used to identify peptides that could bind to different DNA structures. A 23-mer random peptide library was displayed in the context of the pIII protein of M13 filamentous phage. Double-stranded (ds) oligodeoxyribonucleotides (oligos) were immobilized in 96-well plates using either chemical conjugation or a biotin-avidin linking method. Individual phage clones capable of binding to immobilized oligos were selected from the phage library. Using a plaque dilution assay for rapid screening of binding preferences, four groups of oligo-binding (OB) phage were tentatively identified as showing preference for: (1) single-stranded (ss) oligos irrespective of sequence; (2) ds oligos irrespective of sequence; (3) sequence-specific binding to ss oligos; and (4) weak non-specific binding to all types of oligos tested. A quantitative solution-phase competition assay was used to confirm the ability of certain phage to discriminate ss from ds oligos. A consensus motif, FGRA, was found in those phage clones that preferentially bound ss oligos; this motif has previously been noted in the binding domains of several ribonucleoproteins and ss DNA-binding proteins. Peptides based on the FGRA motif, but not scrambled controls, were able to inhibit the binding of appropriate phage clones or of Escherichia coli ss DNA-binding protein to oligos. This suggests that amino acid sequences that are capable of affecting biologically significant protein-DNA interactions can be identified from random peptide libraries using phage display techniques.