Chemoproteomics-Enabled Covalent Ligand Screening Reveals a Thioredoxin-Caspase 3 Interaction Disruptor That Impairs Breast Cancer Pathogenicity

Abstract

Covalent ligand discovery is a promising strategy to develop small-molecule effectors against therapeutic targets. Recent studies have shown that dichlorotriazines are promising reactive scaffolds that preferentially react with lysines. Here, we have synthesized a series of dichlorotriazine-based covalent ligands and have screened this library to reveal small molecules that impair triple-negative breast cancer cell survival. Upon identifying a lead hit from this screen KEA1-97, we used activity-based protein profiling (ABPP)-based chemoproteomic platforms to identify that this compound targets lysine 72 of thioredoxin-a site previously shown to be important in protein interactions with caspase 3 to inhibit caspase 3 activity and suppress apoptosis. We show that KEA1-97 disrupts the interaction of thioredoxin with caspase 3, activates caspases, and induces apoptosis without affecting thioredoxin activity. Moreover, KEA1-97 impairs in vivo breast tumor xenograft growth. Our study showcases how the screening of covalent ligands can be coupled with ABPP platforms to identify unique anticancer lead and target pairs.

Figure 1.

Screening of dichlorotriazine library in breast cancer cells. (A) Dichlorotriazines have been shown to have preferred reactivity with lysines. A library of 58 dichlorotriazines was synthesized. (B) This library was screened (10 μM) in 231MFP breast cancer cells for impairments in serumfree cell survival after 48 h compared to ACN-treated controls. The top hit from this screen was KEA1–97. Detailed data can be found in Table S1. (C) KEA1–97 (10 μM) impairs 231MFP serum-free cell survival and proliferation after 48 h. (D) Counterscreen of hits in MCF10A cells (10 μM, 48 h). Data are shown as average ± SEM, n = 3/group. Significance is expressed as *p < 0.05 compared to control.

Figure 2.

KEA1–97 targets thioredoxin. (A) IsoTOP-ABPP analysis of KEA1–97 (10 μM) in 231MFP breast cancer proteomes using DCT-alkyne probe. 231MFP breast cancer proteomes were pretreated with acetonitrile vehicle or KEA1–97 prior to labeling with DCT-alkyne (100 μM) for 1 h, followed by CuAAC-mediated appendage of light or heavy enrichment handles, and analysis of eluted probe-modified peptides by LC-LC/MS/MS. Structures of KEA1–97 and DCT-alkyne are shown on the left. Shown on the right are isotopically light (acetonitrile-treated) to heavy (KEA1–97treated) probe-modified peptide ratios for peptides identified in two out of three biological replicates. Detailed data can be found in Table S2. (B) Gel-based ABPP studies showing competition of KEA1–97 against DCT-alkyne labeling (10 μM) of pure human TXN. Shown is a representative gel and a dose−response curve. (C) TXN expression in GFP control or TXN-overexpressing 231MFP cells as assessed by qPCR. (D) Cell survival of 231MFP control or TXN-overexpressing cells treated with acetonitrile vehicle or KEA1–97 (10 μM) for 24 h as assessed by Hoechst stain. (E) TXN activity assay. (F) Disruption of TXN interaction with caspase 3 upon treatment with KEA1–97 or mutation of TXN K72 to alanine. In the upper blot and left bar graph, pure His-tagged TXN and caspase 3 were preincubated with acetonitrile or KEA1–97 (100 μM) prior to anti-His pulldown, SDS/PAGE, and blotting for caspase 3. In the lower blot and right bar graph, pure His-tagged TXN or TXN K72A mutant protein and caspase 3 were preincubated, subjected to anti-His pulldown, SDS/PAGE, and blotting for caspase 3. Data in B−E are shown as average ± SEM; data in A−F are n = 3/group. Significance is expressed as *p < 0.05 compared to controls and ^#p < 0.05 compared to KEA1–97-treated controls.

Figure 3.

Apoptosis and antitumorigenic effects induced by KEA1–97 in 231MFP breast cancer cells. (A) Caspase 3/7 activation using a CellEvent Caspase 3/7 Green Detection Reagent. (B) KEA1–97 (10 μM) induces apoptosis in 231MFP breast cancer cells assessed by propidium iodine and FITC Annexin-V staining and quantified by flow cytometry. (C) KEA1–97 treatment (5 mg/kg ip once per day) was initiated 16 days after the initiation of 231MFP tumor xenografts in immune-deficient SCID mice. Data in A−C are shown as average ± SEM. Data are n = 3/group in A and B, and n = 8 mice/group in C. Significance is expressed as *p < 0.05 compared to control.