The use of one-bead one-compound combinatorial library technology to discover high-affinity αvβ3 integrin and cancer targeting arginine-glycine-aspartic acid ligands with a built-in handle

Abstract

The αvβ3 integrin, expressed on the surface of various normal and cancer cells, is involved in numerous physiologic processes such as angiogenesis, apoptosis, and bone resorption. Because this integrin plays a key role in angiogenesis and metastasis of human tumors, αvβ3 integrin ligands are of great interest to advances in targeted therapy and cancer imaging. In this report, one-bead one-compound (OBOC) combinatorial libraries containing the arginine-glycine-aspartic acid (RGD) motif were designed and screened against K562 myeloid leukemia cells that had been transfected with the human αvβ3 integrin gene. Cyclic peptide LXW7 was identified as a leading ligand with a built-in handle that binds specifically to αvβ3 and showed comparable binding affinity (IC(50) = 0.68 ± 0.08 μmol/L) to some of the well-known RGD "head-to-tail" cyclic pentapeptide ligands reported in the literature. The biotinylated form of LXW7 ligand showed similar binding strength as LXW7 against αvβ3 integrin, whereas biotinylated RGD cyclopentapeptide ligands revealed a 2- to 8-fold weaker binding affinity than their free forms. LXW7 was able to bind to both U-87MG glioblastoma and A375M melanoma cell lines, both of which express high levels of αvβ3 integrin. In vivo and ex vivo optical imaging studies with the biotinylated ligand/streptavidin-Cy5.5 complex in nude mice bearing U-87MG or A375M xenografts revealed preferential uptake of biotinylated LXW7 in tumor. When compared with biotinylated RGD cyclopentapeptide ligands, biotinylated LXW7 showed higher tumor uptake but lower liver uptake.

Figure 1

a: Structure of LXW7 b: Structure of biotinylated LXW7 c: Series of integrin transfected K562 cells were stained with LXW7. Samples with black curves were treated with 1μM LXW7-Biotin and detected with Flow Cytometry. Red curves depict samples without treatment of LXW7-Biotin and as negative controls. LXW7 showed remarkable positive binding with αvβ3, very weak cross-reactivity with αvβ5 and αIIbβ3, no binding with α5β1.

Figure 1

a: Structure of LXW7 b: Structure of biotinylated LXW7 c: Series of integrin transfected K562 cells were stained with LXW7. Samples with black curves were treated with 1μM LXW7-Biotin and detected with Flow Cytometry. Red curves depict samples without treatment of LXW7-Biotin and as negative controls. LXW7 showed remarkable positive binding with αvβ3, very weak cross-reactivity with αvβ5 and αIIbβ3, no binding with α5β1.

Figure 1

a: Structure of LXW7 b: Structure of biotinylated LXW7 c: Series of integrin transfected K562 cells were stained with LXW7. Samples with black curves were treated with 1μM LXW7-Biotin and detected with Flow Cytometry. Red curves depict samples without treatment of LXW7-Biotin and as negative controls. LXW7 showed remarkable positive binding with αvβ3, very weak cross-reactivity with αvβ5 and αIIbβ3, no binding with α5β1.

Figure 2

Docking simulation studies on LXW7. Panel a displays the docked conformation of cyclo(RGDfV) and LXW7 (yellow and cyan, respectively) on the surface of αvβ3 integrin. Panel b indicates an interaction between the carboxylate side chain of Asp of both ligands with the Mg²⁺ within the MIDAS domain of the β-subunit. The salt bridge formed between D-Asp of LXW7 and Arg214 of the β-subunit is shown in Panel c. Panel d illustrates the salt bridge formed between Arg of LXW7 and Asp218 of the α-subunit. The guanidinium side chain of Arg in cyclo(RGDfV) is able to interact with both carboxylate oxygens while the side chain of Arg in LXW7 only interacts with one of the carboxylate oxygens.

Figure 3

The bindings of LXW7 to U-87 MG and A375M cells were blocked by anti-αvβ3 antibody (LM609). Both glioblastoma U-87MG (a) and melanoma A375M (c) displayed expression of αvβ3 integrin. LXW7 bound to both tumor cells, and the binding was markedly blocked by anti-human αvβ3 antibody (b, d).

Figure 4

In vivo and ex vivo near infra red fluorescent imaging on nude mice implanted with xenografts. Six hours after tail vein injection with (1) Streptavidin-Cy5.5 alone or (2) Biotinylated LXW7-Streptavidin-Cy5.5 complex, Kidney uptake was high in both mice but preferential uptake into the U-87MG tumor (a) and A375M (b) were noted in mice given biotinylated LXW7-Streptavidin-Cy5.5 complex. The images shown represent one of three groups. Ex vivo imaging of U-87MG xenograft bearing nude mice after injection of Streptavidin-Cy5.5-biotinylated LXW7 and cyclic RGD pentapeptides (cyclo(RGDfE), cyclo(RGDfK) and cyclo(RGDyK)) was shown in (c). LXW7 exhibited high tumor uptake with low liver uptake (c). The reverse was true for the other three cyclic RGD pentapeptides. Relative fluorescence uptake by tumor and liver (c) was quantified in (d). Streptavidin-Cy5.5 alone without peptide was utilized as control.

Figure 4

In vivo and ex vivo near infra red fluorescent imaging on nude mice implanted with xenografts. Six hours after tail vein injection with (1) Streptavidin-Cy5.5 alone or (2) Biotinylated LXW7-Streptavidin-Cy5.5 complex, Kidney uptake was high in both mice but preferential uptake into the U-87MG tumor (a) and A375M (b) were noted in mice given biotinylated LXW7-Streptavidin-Cy5.5 complex. The images shown represent one of three groups. Ex vivo imaging of U-87MG xenograft bearing nude mice after injection of Streptavidin-Cy5.5-biotinylated LXW7 and cyclic RGD pentapeptides (cyclo(RGDfE), cyclo(RGDfK) and cyclo(RGDyK)) was shown in (c). LXW7 exhibited high tumor uptake with low liver uptake (c). The reverse was true for the other three cyclic RGD pentapeptides. Relative fluorescence uptake by tumor and liver (c) was quantified in (d). Streptavidin-Cy5.5 alone without peptide was utilized as control.

Figure 4

In vivo and ex vivo near infra red fluorescent imaging on nude mice implanted with xenografts. Six hours after tail vein injection with (1) Streptavidin-Cy5.5 alone or (2) Biotinylated LXW7-Streptavidin-Cy5.5 complex, Kidney uptake was high in both mice but preferential uptake into the U-87MG tumor (a) and A375M (b) were noted in mice given biotinylated LXW7-Streptavidin-Cy5.5 complex. The images shown represent one of three groups. Ex vivo imaging of U-87MG xenograft bearing nude mice after injection of Streptavidin-Cy5.5-biotinylated LXW7 and cyclic RGD pentapeptides (cyclo(RGDfE), cyclo(RGDfK) and cyclo(RGDyK)) was shown in (c). LXW7 exhibited high tumor uptake with low liver uptake (c). The reverse was true for the other three cyclic RGD pentapeptides. Relative fluorescence uptake by tumor and liver (c) was quantified in (d). Streptavidin-Cy5.5 alone without peptide was utilized as control.

Figure 4

In vivo and ex vivo near infra red fluorescent imaging on nude mice implanted with xenografts. Six hours after tail vein injection with (1) Streptavidin-Cy5.5 alone or (2) Biotinylated LXW7-Streptavidin-Cy5.5 complex, Kidney uptake was high in both mice but preferential uptake into the U-87MG tumor (a) and A375M (b) were noted in mice given biotinylated LXW7-Streptavidin-Cy5.5 complex. The images shown represent one of three groups. Ex vivo imaging of U-87MG xenograft bearing nude mice after injection of Streptavidin-Cy5.5-biotinylated LXW7 and cyclic RGD pentapeptides (cyclo(RGDfE), cyclo(RGDfK) and cyclo(RGDyK)) was shown in (c). LXW7 exhibited high tumor uptake with low liver uptake (c). The reverse was true for the other three cyclic RGD pentapeptides. Relative fluorescence uptake by tumor and liver (c) was quantified in (d). Streptavidin-Cy5.5 alone without peptide was utilized as control.