Inhibitors of Tyrosine Kinase-Dependent Signaling as Anti-Cancer Agents

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

Objective One: Plk1 is composed of an N-terminal catalytic kinase domain (KD) and a C-terminal polo-box domain (PBD). The PBD recognizes phosphoserine (pS) and phosphothreonine (pT)-containing sequences, facilitating protein-protein interactions (PPIs). The KD is tethered to the PBD by means of a flexible interdomain linker. Intramolecular engagement of the KD with the PBD can result in auto-inhibition that serves to regulate kinase function spatiotemporally. Bivalency represents one potential approach toward enhancing target selectivity. Recently, we reported bivalent inhibitors that show extremely high PBD affinities. These inhibitors are constructed by PEG linker tethering the Plk1 KD-binding inhibitors BI2536 or Wortmannin, to the PBD-binding peptide PLH*SpT, where H* represents a -(CH2)8Ph group on the histidine side chain pi-nitrogen. This modification improves PBD-binding affinity more than 1000-fold relative to the parent PLHSpT peptide. These bivalent inhibitors show improved kinase selectivity as compared to BI2536 when examined against a panel of eight kinases. Due to the high affinity of the bivalent inhibitors, ELISA assays can "bottom out" and it may not be possible to accurately determine affinities. To overcome these limitations, we devised a fluorescence polarization (FP) assay that uses a FITC-labeled PLH*SpT fluorescent probe. This assay enabled us to evaluate PLH*SpT-based bivalent inhibitors that had bottomed out ELISA assays. However, the affinity of the fluorescent probe is so high that moderate-affinity ligands such as PLHSpT and the bivalent inhibitors cannot compete with probe for binding to the PBD and no affinity could be measured. Due to strong intramolecular KD - PBD dependent auto-inhibition of full-length Plk1, lower-affinity fluorescent probes, such as 5-carboxy fluorescein (5CF)-GPMQSpTPLNG, which are suitable for use in FP assays against isolated Plk1 PBD, cannot overcome auto-inhibition and they fail to bind when the full-length Plk1 is used. Therefore, we developed FITC-miniPEG-FDPPLHSpTA as a new fluorescent probe having moderate Plk1-binding affinity, which is suitable for use in FP assays with full-length Plk1. The new probe has enabled us to successfully determine full-length Plk1 PBD affinities of bivalent ligands that could not be measured using earlier FP probe. Objective Two: TDP1 is a member of the phospholipase D family that can downregulate the anticancer effects of inhibitors of the type I topoisomerase (TOP1) by hydrolyzing the 3'-phosphodiester bond between DNA and the TOP1 residue Y723 residue in the critical stalled intermediate that is the foundation of TOP1 inhibitor mechanism of action. Thus, developing effective inhibitors of TDP1 remains an unmet need. In one aspect of our efforts to develop TDP1 inhibitors we utilized our recently reported several crystal structures of TDP1 with small molecules bound within the catalytic pocket. These molecules bind by forming hydrogen bonds with residues of the catalytic HKN motifs. Guided by these interactions, we used the MolSoft ICM Pro suite of software to perform a virtual screen of the publicly available DrugBank 5.0 (3449 structures) for the ability to bind to the TDP1 catalytic pocket. Among compounds identified as giving good binding scores were several beta-lactams. The beta-lactam pharmacophore serves as a key component in a range of antibiotics. We subjected a subset of the beta-lactam hits to gel-based TDP1 fluorescence catalytic assays and established that certain members showed micromolar TDP1 inhibition. In follow-up, we evaluated a commercially available library of 90 beta-lactam antibiotics. This led to our identification of additional beta-lactams having micromolar TDP1 inhibitory potencies. In particular, cephalosporin C showed single-digit micromolar TDP1 IC50 values. Since beta-lactams can form covalent bonds with serine residues in target penicillin-binding proteins (PBPs), we performed docking studies with cephalosporin C, which showed that it bound within the catalytic pocket and extended into the DNA substrate binding channel. Importantly, the modeling indicated that both noncovalent and covalent binding modes were theoretically possible. Surface plasmon resonance analysis demonstrated its non-covalent binding mode. Thus, beta-lactams may serve as a new and potentially useful platform to design TDP1-binding ligands that interact with the catalytic pocket and extend into the DNA substrate binding channel. In an alternate approach, we took advantage of the observation that TDP1 could be significantly enhanced by introducing the ability to selectively eliminate TDP1 using protein degraders. Using crystal structures of lead inhibitors bound to TDP1, we designed and synthesized a series of bivalent proteolysis-targeting chimeras (PROTACs) intended introduce proximity between TDP1 and E3 ligases and thereby induce proteolytic degradation. The focus this work is to explore synthetic approaches to modifying TDP1 inhibitors that would permit installation of E3 ligase-targeting functionality, while retaining the ability to bind to TDP1. We employed copper-catalyzed azide-alkyne cycloaddition (CuAAC) "click" reactions to assemble PROTAC constituents. Starting from terminal azide and alkyne precursors, we were able to incorporate in facile fashion 1,2,3-triazole-containing linkers. Although the added linkers and E3-targeting moieties are quite large, we retained an ability to inhibit TDP1 and in this way our work achieved its objectives. Despite poor cellular activity that was consistent with limited cellular uptake, our work sheds light on potential directions to modify TDP1 inhibitors that could lead to optimizing PROTAC biological activity. Such agents would represent a new therapeutic class that could potentially enhance the efficacy and selectivity of TOP1 inhibitors.