A novel multi-modal drug repurposing approach for identification of potent ACK1 inhibitors

Abstract

Exploiting drug polypharmacology to identify novel modes of actions for drug repurposing has gained significant attentions in the current era of weak drug pipelines. From a serendipitous to systematic or rational ways, a variety of unimodal computational approaches have been developed but the complexity of the problem clearly needs multi-modal approaches for better solutions. In this study, we propose an integrative computational framework based on classical structure-based drug design and chemical-genomic similarity methods, combined with molecular graph theories for this task. Briefly, a pharmacophore modeling method was employed to guide the selection of docked poses resulting from our high-throughput virtual screening. We then evaluated if complementary results (hits missed by docking) can be obtained by using a novel chemo-genomic similarity approach based on chemical/sequence information. Finally, we developed a bipartite-graph based on the extensive data curation of DrugBank, PDB, and UniProt. This drug-target bipartite graph was used to assess similarity of different inhibitors based on their connections to other compounds and targets. The approaches were applied to the repurposing of existing drugs against ACK1, a novel cancer target significantly overexpressed in breast and prostate cancers during their progression. Upon screening of ∼1,447 marketed drugs, a final set of 10 hits were selected for experimental testing. Among them, four drugs were identified as potent ACK1 inhibitors. Especially the inhibition of ACK1 by Dasatinib was as strong as IC(50)=1nM. We anticipate that our novel, integrative strategy can be easily extended to other biological targets with a more comprehensive coverage of known bio-chemical space for repurposing studies.

Fig. 1

The schematic Diagram of our modeling workflow. The first step is the construction of the drug-target bipartite graph. Drugs and targets are represented as circles and rectangles, respectively. Node sizes and color are proportional to the degree of each node. The larger shapes and the red color represent nodes with higher degrees. After three steps: A. high-throughput docking; B. Chemical similarity search using AIM-100, a known inhibitor of ACK1; C. Genomic similarity search of ACK1 against proteins in the drug-target graph to identify similar proteins and only the corresponding interacting drugs are selected; D. Using only the drug-target graph to identify drugs similar to those identified from steps A–C.

Fig. 2

Dasatinib (magenta sticks) docked into ACK1 (ribbon display). It was ranked top and has reasonable interactions with ACK1. The gray lines are critical residues in the active site. Hydrogen bonds are in magenta dashed lines. The spheres are pharmacophores: gray for hydrophobic, cyan for hydrogen bonds, and yellow for solvent exposed groups.

Fig. 3

A. The graph was derived from the drug chemical similarity and target genomic similarity. It represents the inhibitor AIM-100 (red square) and ACK1 (red circle) and those drugs obtained from the chemical similarity search (non-red squares) and proteins similar to ACK1 (green circles). B. The enlarged portion of graph A shows Gefitinib is similar to AIM-100 and its target (P00533) has significant genomic similarity to ACK1.

Fig. 4

A representative heat-map of purely graph-based cosine similarities of Flavopiridol against drugs identified from docking and chem-genomic similarity. The higher values (darker red) means higher graph-based similarity.