Targeting Angiogenesis: Mono- and Bi-Specific Proteins That Target VEGF Receptor-2 and αvβ3 Integrin
Adam P. Silverman1, Aron M. Levin2, Jennifer L. Lahti2 and Jennifer R. Cochran2, (1)Stanford University, Stanford, CA, (2)Bioengineering, Stanford University, Stanford, CA

Vascular endothelial growth factor receptor-2 (VEGFR2) and αvβ3 integrin integrin have attracted clinical interest as anti-cancer targets because of their roles in tumor angiogenesis and metastasis. Both receptors are highly overexpressed in the tumor neovasculature but are weakly expressed in resting endothelium and most normal tissues. Consequentially, there has been great interest in developing antibodies and proteins that will selectively target these receptors and block binding of the native ligand to elicit an antagonistic response. Since multiple receptors contribute to angiogenesis and metastasis, cocktails of molecules may be needed to fully block these processes, increasing potential side effects and cost of therapy. The prospect of a bispecific agent capable of targeting both of these proteins is a particularly promising approach, as such a compound would have enhanced ability to target tumors and block angiogenesis.

First, we are developing polypeptides that specifically target integrin receptors for use in therapeutic and imaging applications. We employed cystine-knot proteins as a scaffold for engineering new molecular recognition properties. Using yeast surface display, we have engineered proteins that bind to αvβ3 integrin or αiibβ3 integrin with single digit- to sub-nanomolar affinities and very high specificity over other related integrins. In addition, consensus residues were identified that appear to confer integrin binding specificity. Selected peptides were prepared recombinantly in Pichia pastoris and their properties were studied in vitro using cells expressing the appropriate receptors as well as cancer cell lines. The soluble αvβ3-binding peptides were found to have KD values against U87MG glioblastoma cancer cells in the range of 0.78-13 nM.

Next, we describe the use of rational engineering and combinatorial methods to develop single protein domains capable of binding to both VEGFR2 and αvβ3 integrin with single digit-nanomolar to picomolar affinities. Preliminary in vitro studies of the effects of these proteins on cancer cells and competition studies with wild-type ligands will be discussed.

Extended Abstract Status: Not Uploaded

Poster Session

The Preliminary Program for SBE's 2nd International Conference on Biomolecular Engineering