Identification of oligopeptide binding to colon cancer cells separated from patients using laser capture...

The development of intravascular conjugates that efficiently deliver genes or drugs to tumors is limited by the lack of efficacious targeting ligands. Small targeting peptides, such as those iterated by phage display technology, offer enormous potential for these applications. The majority of reports published to date have focused on the identification of peptides isolated for their ability to bind to human cancer cell lines in vitro, and have failed to account for the loss of polarization and de-differentiation of such cells from their in vivo state. Here, we report a novel approach for the identification of peptides capable of binding specifically to cancer cells derived from clinically resected human colon cancer. In this strategy, laser capture microdissection (LCM) is performed on a surgically resected colon cancer specimen to separate only cancer cells from the specimen. Subsequently, biopanning was performed on the LCM-selected colon cancer cells to identify peptide sequences that bound specifically to them. A peptide containing the SPT motif was selected as the most promising consensus sequence binding specifically to the LCM-selected colon cancer cells. Phage clones displaying the SPT motif demonstrated 9-fold higher binding to colon cancer cells derived from a patient than insertless phage (p < 0.05), while, recovery of the SPT phage from the colon cancer cell lines DLD-1 and HCT-15 was 7-fold higher than that of the control insertless phage (p < 0.05). The binding of SPT phage to colon cancer cells from the patient was confirmed by immunofluorescence. Additionally, a synthesized SPT-containing peptide (SPTKSNS) showed binding activity in the absence of mitogenic effects on colon cancer cells in vitro. In summary, we have introduced LCM into a biopanning procedure and identified a small peptide that binds preferentially to colon cancer cells derived from a clinically resected sample. This procedure could be applicable for the design of customized cancer cell targeting methodologies using clinical biopsy samples from human subjects.