IDENTIFICATION OF PEPTIDE LIGANDS SPECIFIC FOR THE SUGAR-BINDING SITEOF CONCANAVALIN-A BY SCREENING A SYNTHETIC PEPTIDE COMBINATORIAL LIBRARY

We describe a method, using an automated multiple-column chromatograph ic approach, for identifying a ligand from a peptide library (containi ng greater than 2.48 x 10(6) unique peptides) with specificity for the sugar-binding site of the lectin Concanavalin A. The method used an i mmobilized target to capture moieties from the library as the latter f lowed through a chromatographic column. Due to the complexity of the i nitial library, it was not possible to select for individual peptide s equences with high affinity and specificity for the sugar binding site . However, identification of peptides which specifically bound to the target at this site was possible using subtractive pool sequencing of affinity captured material. The latter technique involved sequencing t he peptides retained (after washing the column for a fixed time) in th e presence and absence of an excess of the known ligand for the target , methyl alpha-D-mannopyranoside. Comparisons between the proportion o f each amino acid at each sequencing cycle in the absence or presence of an excess of sugar resulted in a peptide sequence of enriched amino acids of the formula HxxSx (where x represents any one of the natural amino acids except cysteine). This sublibrary (containing similar to 6859 individual peptides) was synthesized and rescreened. Two peptide sequences (HHRSY and HVVSV) were identified with relatively high affin ity for the sugar-binding site of Concanavalin A. The described techni que of solution-phase subtractive pool sequencing (Patent pending) can be employed for rapidly screening highly complex mixtures of peptides and obtaining information about the amino acids within the sequences that are essential for binding to a particular site on the target. Thi s technique could also be applied to other combinatorial mixtures (e,g ., PNAs, nucleic acids, or libraries composed of either non-natural or D-amino acids) where a defined number of discrete components are synt hesized in a variety of permutations.