Structure-based optimization of GRP78-binding peptides that enhances efficacy in cancer imaging and therapy.

It is more challenging to design peptide drugs than small molecules through molecular docking and in silico analysis. Here, we developed a structure-based approach with various computational and analytical techniques to optimize cancer-targeting peptides for molecular imaging and therapy. We first utilized a peptide-binding protein database to identify GRP78, a specific cancer cell-surface marker, as a target protein for the lead, L-peptide. Subsequently, we used homologous modeling and molecular docking to identify a peptide-binding domain within GRP78 and optimized a series of peptides with a new protein-ligand scoring program, HotLig. Binding of these peptides to GRP78 was confirmed using an oriented immobilization technique for the Biacore system. We further examined the ability of the peptides to target cancer cells through in vitro binding studies with cell lines and clinical cancer specimens, and in vivo tumor imaging and targeted chemotherapeutic studies. MicroSPECT/CT imaging revealed significantly greater uptake of (188)Re-liposomes linked to these peptides as compared with non-targeting (188)Re-liposomes. Conjugation with these peptides also significantly increased the therapeutic efficacy of Lipo-Dox. Notably, peptide-conjugated Lipo-Dox significantly reduced stem-cell subpopulation in xenografts of breast cancer. The structure-based optimization strategy for peptides described here may be useful for developing peptide drugs for cancer imaging and therapy.