Identification of oligopeptides binding to peritoneal tumors of gastric cancer

Abstract

This is a report of in vivo intraperitoneal biopanning, and we successfully identified a novel peptide to target the multiple peritoneal tumors of gastric cancer. A phage display library was injected directly into the abdominal cavity of mice bearing peritoneal tumors of human gastric cancer, and phages associated with the tumors were subsequently reclaimed from isolated samples. The tumor-associated phages were amplified and the biopanning cycle was repeated five times to enrich for high affinity tumor-selective binding peptides. Finally, a tri-peptide motif, KLP, which showed homology with laminin 5 (a ligand for alpha3beta1 integrin), was identified as a binding peptide for peritoneal tumors of gastric cancer. Phage clones displaying the sequence KLP showed 64-fold higher binding to peritoneal tumors than control phage and were preferentially distributed in tumors rather than in normal organs after intraperitoneal injection into mice. In addition, the KLP phages were more likely to bind to cancer cells in malignant ascites derived from a patient with recurrent gastric cancer. Synthesized peptide containing the motif KLP (SWKLPPS) also showed a strong binding activity to peritoneal tumors without cancer growth effect. Liposomes conjugated with SWKLPPS peptide appeared significantly more often in tumors than control liposomes after intraperitoneal injection into mice. Furthermore, modification of liposomes with SWKLPPS peptide enhanced the antitumor activity of adriamycin on gastric cancer cells. The peptide motif KLP seems a potential targeting ligand for the treatment of peritoneal metastasis of gastric cancer.

Figure 1

In vivo binding activities of selected phage clones expressing the candidate peptides for binding to peritoneal tumors of gastric cancer. Each selected phage clone expressing the candidate peptides was injected intraperitoneally into model mice. The mice were killed 20 min after injection. Peritoneal tumor nodules were harvested from each mouse and homogenized, and the phages accumulated in the nodules were quantified by titering multiple dilutions of the homogenate. The results are expressed as p.f.u./g tissue, and a phage clone displaying no oligopeptide insert was used as a control.

Figure 2

Distribution of selected phage clones in model mice with peritoneal metastases. Twenty minutes after intraperitoneal injection of phage clones displaying the SWKLPPS sequence into model mice with peritoneal metastases, samples from the peritoneal tumors, normal stomach, liver and spleen were obtained. The distributions of the phages in the tumors and organs were quantified by titering and expressed as percentages of the accumulation in each organ compared to that in tumors (a). Simultaneously, the phage distributions were evaluated by immunohistochemistry (b–g). Phage accumulation is revealed by brown dots in each figure. (b, c) SWKLPPS phages in a tumor (magnification: ×40 and ×100, respectively). (d) Control phages in a tumor (magnfication: ×40). (e–g) SWKLPPS phages in the stomach, liver and spleen, respectively (magnification: ×100).

Figure 3

Competitive inhibition of synthesized peptides against phage accumulation. AZ‐P7a cells were pre‐incubated with 0.1 (□), 1 (▨) or 10 µM (▤) of the SWKLPPS or QPLLKLP peptide for 30 min at 4°C, followed by the addition of 5 × 10⁸ p.f.u. of the SWKLPPS phage in vitro. The inhibitory effects of the synthesized peptides on phage accumulation were examined by titering the phages bound to the cells (a). Similarly, the SWKLPPS phage (2 × 10¹¹ p.f.u.) and 10 µM (□) or 1 mM (▨) of each synthesized peptide were co‐injected intraperitoneally into model mice with peritoneal metastases. The mice were killed 20 min after injection. Peritoneal tumor nodules were harvested, and the phages accumulated in the tumors were quantified by titering to confirm the in vivo inhibitory effects of the synthesized peptides on phage accumulation (b). An irrelevant heptapeptide (TTPRDAY, 10 µM in vitro and 1 mM in vivo) was used as a control. *P < 0.05 compared to the control.

Figure 4

Assessment of the mitogenicity of the SWKLPPS peptide in AZ‐P7a cells. AZ‐P7a gastric cancer cells were incubated in 96‐well plates at 5 × 10³ cells/well in the presence of 1 µM (▪), 10 µM (▴) or 100 µM (×) of the SWKLPPS peptide or without the peptide (◆). The cell viability was monitored after 24, 48, 72 and 96 h using the MTS assay. The quantity of the formazan product present was determined by measuring the absorbance at 490 nm using a microplate autoreader.

Figure 5

Binding of the SWKLPPS phage to floating cells in malignant ascites from a patient with gastric cancer. The ex vivo binding activity of the SWKLPPS phage to floating cells in malignant ascites from a patient with gastric cancer was examined. The ascites from the patient was concentrated by centrifuge and co‐incubated with SWKLPPS phage or insertless phage in 6‐well plate. Then the number of phages binding to cells was determined by titering. *P < 0.05 compared to the control.

Figure 6

Biodistribution of SWKLPPS‐conjugated liposomes after intraperitoneal injection. Mice with peritoneal metastasis were anesthetized and injected with the radiolabeled liposomes containing [1á,2á(n)‐³H] cholesterol oleoyl ether with stearoyl 7 mer peptide SWKLPPS (▪) or without peptide conjugates (control, ▨) intraperitoneally. The mice were killed 24 h after injection, and blood was collected and centrifuged to obtain plasma. After the mice had been bled, the tumor and normal organs were removed, washed with saline and weighed. The radioactivity in samples was determined with a liquid scintillation counter. Data are represented as the percentage of the injected dose per 100 mg wet tumor tissue or 100 µL plasma. *P < 0.05 compared to the each control.

Figure 7

Anticancer activity of SWK‐LipADM. After AZ‐P7a cells were plated on a 96‐well plate and cultured in a CO₂ incubator at 37°C for 24 h, 20 µL LipADM or SWK‐LipADM was added to each well at the ADM concentration of 0.3, 1, 3, 10, 30 and 100 mg/mL and allowed to bind to the cells for 30 min at 37°C. The mediums were changed to RPMI containing 10% FCS and cells were cultured for further 24 h. Cell proliferation assay with TetraColor One was carried out. * P < 0.01 compared to LipADM.