Overexpression of the serine/threonine polo-like kinase 1 (Plk1) is tightly associated with oncogenesis in several human cancers. Upregulated Plk1 activity appears to be closely associated with the aggressiveness and poor prognosis of cancers, which are often addicted to Plk1 overexpression for their viability. Interference with Plk1 function induces apoptosis in tumor cells but not in normal cells. Targeting Plk1 may permit induction of cancer cell-selective mitotic block and apoptotic cell death in Plk1-addicted cancers. Accordingly, Plk1 is a potentially attractive anticancer chemotherapeutic target. Plk1 possesses a unique phosphopeptide-binding polo box domain (PBD), which functions by recognizing and binding to to phosphothreonine (pT)/phosphoserine (pS)-containing protein sequences. This recognition and binding is essential for intracellular localization and mitotic functions of Plk1. Unlike kinase domains, PBDs are only found among the Plks. Therefore, PBDs represent attractive targets for selectively down-regulating Plk function. We are engaged in efforts to develop Plk1 PBD-binding inhibitors. Starting from the 5-mer phosphopeptide PLHSpT (which specifically interacts with the Plk1 PBD, while failing to significantly interact with the PBDs of two closely-related kinases, Plk2 and Plk3), we recently identified three families of peptidic inhibitors that showed from 1000- to more than 10,000-fold improved PBDbinding affinity. In collaboration with Dr. Michael Yaffe (MIT), X-ray co-crystal structures of these peptides bound to Plk1 PBD indicated unanticipated modes of binding that take advantage of a cryptic binding channel that is not present in the non-liganded PBD or engaged by the parent pentamer phosphopeptide. Although critical elements in the high affinity recognition of peptides and proteins by PBD are derived from pT/pS-residues, the use of these residues in therapeutics is potentially limited by poor cellular uptake, in part due to high anionic charge of the phosphoryl moiety. We have recently discovered new synthetic transformations that reduce the overall peptide anionic charge by intramolecular charge masking, which provides peptides with enhanced efficacy in cellular assays. We have further modified these peptides by introducing bio-reversible prodrug protection of one phosphoryl acidic hydroxyl. This yielded neutral peptides that show even greater cellular efficacies. We have also explored the application of conformational constraint (in which binding entropy penalties are reduced by reducing ligand flexibility). We synthesized several families of macrocycles using methodologies that have not previously been reported for peptide macrocyclization. Efforts in this project are currently focused on attaining more drug-like compounds that are suitable for examination in rodent models of cancer. In further work, we are developing proteins that merge properties of antibodies with biologically active small molecules. This aspect of the Project deals with antibody-drug conjugates (ADCs), which link cytotoxic payloads to antibodies or antibody fragments to achieve a highly specific delivery of the cytotoxins to desired targets. Fundamentally, this Project is based on the hypothesis that combining synthetically diverse peptides, peptidomimetics, and small molecules together with the immunological characteristics of antibodies can lead to therapeutic agents with improved properties. This work is being done in collaboration with Dr. Christoph Rader (Scripps Florida). We are exploring the structures of linkers as well as payloads and methods of attaching these constructs to proteins. Key considerations include the ability for controlled, selective attachment of the drug-linker payload to the antibody; a mechanism for intracellular release of the drug cargo and a highly potent drug payload. One focus of this work is the development of next-generation antibody therapeutics with molecularly defined payload:carrier ratios. Many clinically established ADCs rely on non-specific conjugation to Cys or Lys residues, which yields heterogeneous mixtures of ADCs having from 0 - 8 drugs per antibody and each of these ADCs can exhibit distinct pharmacological profiles. We are striving to develop ADCs with defined loading. In one approach, we initially examined ADCs that employ engineered selenocysteine residues (Sec). The selenol side chain of Sec exhibits higher nucleophilicity than Cys thiols and allows for site-selective attachment of drug cargo at acidic pH (5). We have combined a variety of bio-cleavable linkers together with derivatives of highly cytotoxic cemadotin and monomethyl auristatin F (MMAF) and both mono- and divalent constructs that employed the integrin (alpha4beta1) antagonist LLP2A. We applied this technology to the B-cell malignancy, chronic lymphocytic leukemia (CLL). The recently identified IgM receptor Fc(mu)R is overexpressed on malignant B cells in CLL and mediates the rapid internalization and lysosomal shuttling of IgM via its Fc fragment (Fc(mu)). To exploit this internalization and trafficking pathway for targeted drug delivery, we engineered an IgM-derived protein scaffold (Fc(mu)) and linked it with the cytotoxic agent MMAF. This Fc(mu)-drug conjugate was selectively toxic for Fc(mu)R-expressing cell lines in vitro and for primary CLL cells but not autologous normal T cells ex vivo. In collaboration with Dr. Adrian Wiestner (NHLBI, NIH), we tested the possible therapeutic application of the Fc(mu)-drug conjugate in immunodeficient (NSG) mice engrafted with peripheral blood mononuclear cells from patients with leukemia. Three intravenous injections of the Fc(mu)-drug conjugate over a 10-day period were well tolerated and selectively killed the human CLL cells but not the co-engrafted autologous human T cells, thus providing proof of concept for Fc(mu)R as a valuable therapeutic target in CLL and for IgM-based antibody-drug conjugates as a new targeting platform. We also designed and synthesized three monomethyl auristatin F (MMAF) derivatives with non-cleavable iodoacetamido-caproyl linkers at their N-terminus and variable C-termini. The three cytotoxic payloads were subjected to Sec-selective conjugation to anti-HER2 scFv-Fc-Sec and all three conjugates were found to exhibit potent and specific activity toward HER2-high breast cancer cell lines. In another approach, we have used the humanized anti-hapten monoclonal antibody, h38C2, which contains a natural Lys at the bottom of an 11-A philic at physiological pH (pKa 6) and it is capable of performing catalytic reactions or being selectively conjugated to 1,3-diketone and beta-lactam derivatives. Therefore, although approximately 40 surface exposed Lys residues may be available for conjugation, this methodology affords a uniquely robust, clean, and facile platform for making homogenous ADCs. We utilized h38C2's unique cargo attachment point for generating covalent conjugation by means of 1,3-diketone or beta-lactam functionality with small molecules that bind to receptors of interest. We are applying this in a format coined DART, which separates variable domains of heavy and light chains of two antigen or hapten binding specificities on separate polypeptide chains that are stabilized via a C-terminal disulfide bridge. We combined h38C2 with the anti-CD3 mAb v9 for T-cell recruitment and activation. In one application, h38C2 x v9 DARTs were chemically programmed with hapten-folate, which endowed the assembly with folate receptor 1 (FOLR1) binding capability. FOLR1 is a clinically investigated target for both mAbs and small molecules in cancer therapy. We found that these potently killed FOLR1-expressing human ovarian cancer cell lines in vitro and in vivo.
Showing the most recent 10 out of 48 publications
