Chemoproteomics Identifies State-Dependent and Proteoform-Selective Caspase-2 Inhibitors

Abstract

Caspases are a highly conserved family of cysteine-aspartyl proteases known for their essential roles in regulating apoptosis, inflammation, cell differentiation, and proliferation. Complementary to genetic approaches, small-molecule probes have emerged as useful tools for modulating caspase activity. However, due to the high sequence and structure homology of all 12 human caspases, achieving selectivity remains a central challenge for caspase-directed small-molecule inhibitor development efforts. Here, using mass spectrometry-based chemoproteomics, we first identify a highly reactive noncatalytic cysteine that is unique to caspase-2. By combining both gel-based activity-based protein profiling (ABPP) and a tobacco etch virus (TEV) protease activation assay, we then identify covalent lead compounds that react preferentially with this cysteine and afford a complete blockade of caspase-2 activity. Inhibitory activity is restricted to the zymogen or precursor form of monomeric caspase-2. Focused analogue synthesis combined with chemoproteomic target engagement analysis in cellular lysates and in cells yielded both pan-caspase-reactive molecules and caspase-2 selective lead compounds together with a structurally matched inactive control. Application of this focused set of tool compounds to stratify the functions of the zymogen and partially processed (p32) forms of caspase-2 provide evidence to support that caspase-2-mediated response to DNA damage is largely driven by the partially processed p32 form of the enzyme. More broadly, our study highlights future opportunities for the development of proteoform-selective caspase inhibitors that target nonconserved and noncatalytic cysteine residues.

Figure 1.

Isotopic tandem orthogonal proteolysis–activity-based protein profiling (isoTOP-ABPP) to stratify the reactivity of caspase cysteine residues. (A) IsoTOP-ABPP workflow used here was modified from the previously published methods^, to incorporate single-pot, solid-phase-enhanced sample-preparation (SP3) cleanup. Lysates are treated with either 100 μM or 10 μM IAA followed by copper-catalyzed azide–alkyne cycloaddition (CuAAC) conjugation to the previously reported isotopically labeled tobacco etch virus (TEV)-cleavable biotinylated peptide tags. The samples are then combined, subjected to SP3 cleanup and on-resin trypsin digest, enrichment on streptavidin resin followed by TEV proteolysis, and liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis. isoTOP-ABPP ratios (log₂ H:L) are calculated from MS1 ion intensity ratios for heavy- versus light-labeled peptides for peptides from nonapoptotic Jurkat cells. Ratio thresholds: “high” reactivity: Log₂(R_heavy:light) = −1.0–1.0, “medium” reactivity: Log₂(R_heavy:light) = = 1–2.32, “low” reactivity Log₂(R_heavy:light) = > 2.32. ND-Not detected. Workflow was adapted with permission from ref . Copyright [2019][Springer Nature]. (B) Waterfall plot showing isoTOP-ABPP reactivity analysis of cell lysates derived from viable nonapoptotic Jurkat cells. (C) Comparison of the measured isoTOP-ABPP reactivity ratios for caspase cysteines identified in the analysis of vehicle-treated (DMSO) and apoptotic (50 ng/μL FasL, 4 h) cellular lysates. Nonapoptotic experiments (n = 6) and apoptotic experiments (n = 5). (D) Comparison of the activity of recombinant activeCASP2 and activeCASP2_C370A using fluorogenic substrate Ac-VDVAD-AFC monitored using a multimodal plate reader with percentage activity calculated from the linear range of the reaction curves. (E) Crystal structure of activeCASP2 (PDB: 1PYO; complexed with Acetyl-Leu-Asp-Glu-Ser-Asp-cho) in cyan highlighting active-site adjacent loop (Loop 1) overlaid with the predicted (I-TASSER^–6768 structure of procaspase-2 in green. (F) 180° flip of loop 1 repositions the C370 sulfhydryl. Figures were generated using Pymol. For D data represents mean activity ± STDEV for two technical replicate experiments. Statistical significance was calculated with unpaired Student’s t tests, comparing activeCASP2 to activeCASP2_C370A; not significant (n.s.) p > 0.05. All MS data can be found in Table S1. Cys shown in Red are catalytic nucleophiles.

Figure 2.

Alkyne probe 3 preferentially labels C370 over other caspase cysteines to afford partial blockade in procaspase activity. (A) Structures of cysteine-reactive clickable probes including previously reported IAA and KB18, KB19, and KB61, as well as compounds 1–4, unique to this study. (B) Gel-based ABPP analysis of recombinant caspase-2 constructs harboring the indicated mutations in Jurkat whole cell lysates treated for 1 h with the indicated compounds (10 μM for all click probes except for IAA, which was analyzed at 1 μM, and Rho-DEVD-AOMK at 2 μM). (C) Assessment of compound-induced changes in protease activity for proCASP2 and proCASP2_C370A proteins subjected to 3 (100 μM, 2 h, 30 °C) followed by multimode plate reader fluorescence analysis of caspase activity using the fluorogenic substrate Ac-VDVAD-AFC (1 μM) under kosmotropic conditions (sodium citrate).^, (D) Relative activities of activeCASP2 in the absence (Black) or presence of 3 (Gray) (100 μM) assessed using fluorogenic substrate Ac-VDVAD-AFC. (E) Gel-based ABPP analysis of recombinant procaspase-8 (proCASP8) and procaspase-10 (proCASP10) subjected to the indicated compounds (10 μM for all click probes except for IAA, which was analyzed at 1 μM, and Rho-DEVD-AOMK at 2 μM). (F) Relative protease activities of activeCASP8 and activeCASP10 in the absence (Black) or presence of 3 (Gray) (100 μM, 2 h, 30 °C) assessed using fluorogenic substrate Ac-VDVAD-AFC. For Figure 2D-F, experiments were performed in triplicate. Full-length gels shown in Figures S7 and S8. For C, D, and F Data represent mean values and standard deviation. Statistical significance was calculated with unpaired Student’s t tests, n.s. * p < 0.05, ** p < 0.01, ** *p < 0.001, NS p > 0.05.

Figure 3.

Fragment electrophiles selectively label proCASP2 at C370. (A) Structures of the electrophilic compound library of 3 and P01 analogues. (B) Gel-based ABPP analysis of the indicated recombinant proteins in whole cell lysates treated for 1 h with the indicated compounds followed by click conjugation to compound 3 (10 μM) for proCASP2 and KB61 (10 μM) for proCASP8 and proCASP10. Full-length gel images in Supporting Information Figure S11 (C) Apparent IC50 curve for blockade of 3-labeling of proCASP2 by the indicated compounds. CI, 95% confidence intervals. (D) Caspase cysteines quantified in competitive isoTOP-ABPP analysis of Jurkat cell lysates treated with P01, 5 and 6 (25 μM) and 7, 8, 9, 11, 12, 14, 15, and 16 (100 μM). Competition ratios are calculated from the mean quantified precursor intensity (log₂H/L) for DMSO (H, heavy) versus compound treatment (L, light). Cysteines shown in Red are catalytic residues. ND-Not detected. For C, data represent mean values ± standard deviation of the mean for at least three independent experiments. For D, experiments were conducted in at least two biological replicates. All MS data can be found in Table S3.

Figure 4.

TEV-cleavable proCASP2 is activated by TEV protease with minimal effects on enzyme kinetics. (A) Replacement of caspase-2 cleavage sites (D333 and D347) with TEV (ENLYFQG) cleavage motifs affords engineered proCASP2TEV and proCASP2TEV_C370A proteins, which exhibit TEV-dependent increase in caspase activity toward the Ac-VDVAD-AFC fluorogenic substrate. (B) Gel-based ABPP analysis of (B) proCASP2TEV and (C) proCASP2TEV_C370A, which were treated with either Rho-DEVD-AOMK (1 μM final concentration) or compound 3 (10 μM final concentration) for 1 h, followed by activation with TEV protease at increasing concentrations and click conjugation to Rho-Azide for 3-treated samples. For quantitation and Coomassie staining refer to Figures S20 and S21. (D) Calculated Michaelis–Menten kinetic parameters using the Ac-VDVAD-AFC substrate, including 95% confidence intervals, for proCASP2TEV following activation with the TEV protease and activeCASP2. (E) Relative activity of proCASP2TEV and proCASP2TEV_C370W proteins assessed using the Ac-VDVAD-AFC substrate following activation with TEV protease. For E, data represents mean values ± standard deviation of the mean for three independent experiments. Statistical significance was calculated using unpaired Student’s t tests, **** p < 0.005. n.s. p > 0.05.

Figure 5.

State-dependent inhibition of caspase-2. (A–D) Gel filtration analysis of (A, B) proCASP2TEV and (C, D) proCASP2TEV_C370A showing (A, C) the UV absorbance at 280 nm (A280) for each sample and (B, D) gel-based ABPP analysis of the indicated fractions labeled with Rho-DEVD-AOMK (1 μM). (E) Percent of maximum dimeric and monomeric fractions of proCASP2TEV and proCASP2TEV C370A. Relative band intensities were determined for coomassie bands (black) and Rho-DEVD-AOMK-labeled bands (red). Percent components of total dimer and monomer forms calculated from the area under the curve (AUC) of dimeric column volume fractions (0.40–0.46) and monomeric column volume fractions (0.50–0.55) for proCASP2TEV activity assessed by labeling with Rho-DEVD-AOMK and total protein assessed with coomassie staining. (F) Gel-based ABPP analysis of the indicated gel filtration fractions of proCASP2TEV and proCASP2TEV_ C370A with either Rho-DEVD-AOMK or 3, with the latter samples conjugated to Rho-azide by click chemistry. (G) Relative activity of monomeric versus dimeric fractions of proCASP2TEV subjected to the indicated treatment by either compound 9 (100 μM, 1 h, 30 °C) or vehicle (DMSO) followed by addition of TEV protease or vehicle and activity analysis under kosmotropic conditions using the Ac-VDVAD-AFC fluorogenic substrate. (H) Measured percent activity for monomeric fractions of proCASP2TEV treated with the indicated compounds (100 μM, 1 h, 30 °C) followed by the addition of TEV protease and kosmotropic buffer. For B, D, and F, arrows indicate cleaved or processed caspase. For G and H data represent mean values and ± standard deviation of the mean, n = 3, and statistical significance was calculated with unpaired Student’s t tests, * p < 0.05, ** p < 0.01.

Figure 6.

In situ labeling of caspase-2 by electrophilic lead compounds and blockade of extrinsic and intrinsic apoptosis by promiscuous caspase inhibitors. (A) Caspase cysteines quantified in competitive isoTOP-ABPP analysis of Jurkat cells treated with P01, 5, 6 (25 μM) and 8, 9, 10, 11, and 12 (100 μM). Competition ratios are calculated from the mean quantified precursor intensity (log₂ H/L) for DMSO (H, heavy) versus compound treatment (L, light). * indicates catalytic cysteine residue. Experiments were conducted in at least two biological replicates. ND-Not detected. (B) Protein-directed ABPP analysis of U2OS cells treated with compound 9 (100 μM) or vehicle (DMSO) for 1 h followed by compound 3 (20 μM) for 1 h with proteomic analysis using LFQ and FragPipe Analyst. (C) Comparison of effects of the indicated compounds [1 h pretreatment with P01, 5, and 6 (25 μM) and 8, 9, 10, 11, and 12 (100 μM)] on staurosporine (STS, 1 μM, 4 h)-induced apoptosis of Jurkat cells assayed using CellTiter-Glo. Viability experiments were performed in biological triplicate. (D) Immunoblot analysis of caspase-9 shRNA knockdown U2OS cell line JC1 (see Figure S34 for confirmation of knockdown) subjected to etoposide (10 μM) for the indicated treatment times. (E) CETSA analysis of U2OS cell lysate treated with compound 9 or compound 12 (100 μM) for 1 h followed by immunoblotting. (F) U2OS cells were treated with either vehicle (DMSO) or compound 9 (100 μM) for 1 h followed by compound 3 (20 μM) for 1 h. Following lysis, treated samples were subjected to click-enabled pulldown on streptavidin resin followed by immunoblot analysis. (G) U2OS cells treated with the indicated compounds (100 μM) for 1 h followed by etoposide (10 μM) for 3 h were stained for y-H2AX and relative intensity was quantified using ImageJ. All MS data can be found in Table S5. C data represent mean values and standard deviation, n = 3. Statistical significance was calculated with unpaired Student’s t tests, * p < 0.05, ** p < 0.01, *** p < 0.005. ****p < 0.001. n.s. p > 0.05.