(447c) Engineering Cystine Knot Peptides As a New Class of Molecular Imaging Agents

Cystine knots, also known as knottins, are a class of peptides that consist of a structural core of three disulfide bonds interwoven into a molecular “knot” confirmation. This disulfide-bonded structure confers knottins with high chemical, thermal, and proteolytic stability. We are using knottins as novel molecular scaffolds to engineer peptides that can target and illuminate tumors. Using yeast surface display, our lab demonstrated that two knottin scaffolds, based on the Ecballium elaterium trypsin inhibitor (EETI) and the Agouti-related protein (AgRP), can be engineered to bind with high affinity to αvβ3 integrins expressed on tumor vasculature. Non-invasive in vivo optical imaging of the engineered knottin peptides based on both EETI and AgRP scaffolds elicited high tumor signals and rapid clearance from non-target tissues. However, AgRP-based peptides had high kidney uptake and retention, while EETI-based peptides showed minimal kidney uptake. To further investigate and elucidate these differences amongst knottin scaffolds, we engineered a third knottin from spider venom, Agatoxin (AgTx), to bind to αvβ3 integrins using a loop grafting approach. We also found mutations in the AgTx scaffold that assisted in folding of the engineered protein. We then compared the tissue biodistribution observed between wild-type scaffolds and integrin-binding EETI, AgRP, and AgTx knottin peptides in mouse tumor models. We showed that desirable in vivo biodistribution properties are inherent to the three knottin scaffolds tested, but particular amino acid sequences within the engineered binding loop of AgRP and AgTx result in high imaging signals in the kidneys. Thus, we demonstrate that AgTx can be engineered for new molecular recognition properties, and that tissue biodistribution of fluorescent knottin peptides can be tuned by modifying their amino acid composition.
Funded by NIH NCI P50 CA114747: In Vivo Cellular and Molecular Imaging Center at Stanford