Molecular modeling as a powerful tool for the mapping of fibroblast growthfactor receptor-1 ligand binding determinants

Molecular modeling has allowed us to propose that one main contact surface of the Fibroblast Growth Factor Receptor -1 (FGFR-1) to the ligand FGF-1 is formed by a 16 amino acid sequence comprised by the C-terminal region of t he domain II (DII) plus the hinge linking DII and DIII domains and the N-te rminal region of domain III (DIII). Therefore, this sequence was used to de sign the following three peptides: Ac-YQLDVVERS-NH2 (R1); Ac-YQLDVVERSPHRPI LQ-NH2 (R2) and Ac-RSPHRPILQ-NH2 (R3). The synthetic peptides were tested i n their ability to inhibit the mitogenic activity of FGF-1 and FGF-2 in cul tured Balb/c 3T3 fibroblasts. The results showed that R1 and R2 inhibited t he activity of FGF-1 (ID50 = 40 -50 mu M) but not that of FGF-2. Molecular modeling studies of R1 and its docking to FGF-1 suggested that this peptide could assume a conformation very similar to that found in the correspondin g segment of FGFR-1. All these results support our hypothesis that the C-te rminal residues of the DII domain, represented by peptide R1, are part of a surface responsible for the binding of FGF-1 to FGFR-1 but not of FGF-2. A lso, they indicate that peptide R1 may be useful for the development of sma ll selective peptide inhibitors of the FGF-1 biological activities.