Searching for Novel Noncovalent Nuclear Export Inhibitors through a Drug Repurposing Approach

Abstract

Chromosomal region maintenance protein 1 (CRM1) is a validated anticancer drug target, and its covalent inhibitor KPT-330 has been approved for marketing. However, the development of CRM1 inhibitors, especially the noncovalent ones, is still very limited. Drug repurposing is an effective strategy to develop drug leads for new targets. In this work, we virtually screened a library of marketed drugs and identified zafirlukast as a new CRM1 inhibitor. Biochemical and structural analysis revealed that zafirlukast was a noncovalent CRM1 inhibitor that bound to four subpockets in the nuclear-export-signal (NES) groove. Methylation of the sulfonamide group rendered zafirlukast completely inactive against CRM1. Zafirlukast inhibited the growth of a variety of cancer cells and worked synergistically with the drug doxorubicin. Taken together, these works laid a solid foundation for reshaping zafirlukast as a valuable lead compound for further design of noncovalent, specific, and potent CRM1 inhibitors toward the treatment of various cancers.

Discovery of Aminoratjadone Derivatives as Potent Noncovalent CRM1 Inhibitors

“…Two previously reported noncovalent CRM1 inhibitors are insensitive to the CRM1 C528 mutation. 21,22 Here, the CRM1 C528S mutation partially impaired the activity of compounds in pull down and completely abolished the inhibitory effect of KL1 and KL2 on nuclear export (Figures S8 and S9). Structural analysis showed that C528 formed substantial hydrophobic interactions with aminoratjadone derivatives, regardless of compound hydrolysis (Figures 5C and S7).…”

“…In this work, we identified two novel CRM1 inhibitors, KL1 and KL2, that inhibited nuclear export much more potently than the reported noncovalent CRM1 inhibitors. 21,22 In most yeast species, the cysteine of CRM1 that is conjugated to LMB (1) is mutated to a threonine residue, which confers resistance to LMB (1) produced by neighboring competing cells. 27 Unlike LMB, the binding of several aminoratjadone derivatives to CRM1 did not rely on intermolecular covalent bonds.…”

“…These facts render it difficult to develop noncovalent CRM1 inhibitors. Nonetheless, we discovered NCI-1 ( 7 ) and zafirlukast ( 6 ) as noncovalent CRM1 inhibitors by fragment linking and drug repurposing approaches, respectively. , In other work, we demonstrated that extracellular-targeting drug conjugates of 16 R -aminoratjadone ( 4 , Figure ) are potent anticancer agents . The compound was semi-synthetically derived from the naturally occurring ratjadone A ( 5 , Figure ) which was obtained from cultures of the of myxobacterium Sorangium cellulosum (Soce 1047 = DSM 32464).…”

Discovery of a Hidden Pocket beneath the NES Groove by Novel Noncovalent CRM1 Inhibitors

“…Due to the plasticity of the NES groove and the high binding affinity of NES, the development of noncovalent inhibitors remains a considerable challenge. , We previously reported two noncovalent inhibitors, NCI-1 and zafirlukast, but these compounds suffered from problems such as low stability or low potency. , In this work, we discovered a series of novel noncovalent CRM1 inhibitors with a new scaffold by virtually screening a library of 1.4 million compounds. Structural and biochemical studies established the preliminary structure–activity relationship (SAR) of these compounds.…”

“…Leptomycin B and KPT-330 are covalent inhibitors, whereas NCI-1 and zafirlukast are noncovalent inhibitors. The reported cell viability IC 50 values − are indicated in the figure.…”