Mitochondrial ClpP-Mediated Proteolysis Induces Selective Cancer Cell Lethality

Abstract

The mitochondrial caseinolytic protease P (ClpP) plays a central role in mitochondrial protein quality control by degrading misfolded proteins. Using genetic and chemical approaches, we showed that hyperactivation of the protease selectively kills cancer cells, independently of p53 status, by selective degradation of its respiratory chain protein substrates and disrupts mitochondrial structure and function, while it does not affect non-malignant cells. We identified imipridones as potent activators of ClpP. Through biochemical studies and crystallography, we show that imipridones bind ClpP non-covalently and induce proteolysis by diverse structural changes. Imipridones are presently in clinical trials. Our findings suggest a general concept of inducing cancer cell lethality through activation of mitochondrial proteolysis.

Keywords: acute myeloid leukemia; cancer; imipridone; lymphoma; mitochondrial ClpP; mitochondrial proteolysis; oxidative phosphorylation; respiratory chain complex.

Figure 1.. Mitochondrial ClpP activation induces anti-tumor effects in vitro and in vivo.

(A) Quantification for apoptotic (annexin V-positive, left) and immunoblot analysis for the ClpP protein level (right) in OCI-AML3 and Z138 cells with tetracycline-inducible over-expression of WT or Y118A mutant ClpP treated with tetracycline at indicated concentrations for 144 hr. The results are expressed as the mean of triplicate samples ± SD (error bars). ***p < 0.001, ****p < 0.0001. (B) Survivals of xenograft mice using Z138 cells with tetracycline-inducible Y118A mutant ClpP over-expression. The mice (n = 10 each) were treated with or without tetracycline (2 mg/mL in drinking water). (C) Effects of ADEP1 on degradation of FITC-casein by recombinant WT ClpP. The results are expressed as the mean value of triplicate samples ± SD (error bars). (D) The viability of OCI-AML2 cells measured by alamar blue assay after a 72-hr exposure to ADEP1. The results are expressed as the mean value of triplicate samples ± SD (error bars). See also Figure S1.

Figure 2.. The imipridones ONC201 and ONC212 activate mitochondrial ClpP.

(A) Degradation rate of fluorogenic substrate FITC-casein by recombinant WT ClpP incubated with each compound of a chemical library of 747 molecules. (B) Effects of ONC201 and ONC212 on degradation of fluorogenic substrates AC-WLA-AMC and FITC-casein by recombinant WT ClpP. The results are expressed as the mean value of triplicate samples ± SD (error bars). (C) Degradation of α-casein by purified recombinant WT ClpP and ClpXP complexes treated for 3 hr with ONC201, ONC212 or vehicle control (DMSO) in FITC-casein assay buffer detected on SDS-PAGE. See also Figure S2.

Figure 3.. The imipridones bind to ClpP and are cytotoxic to leukemia and lymphoma cells.

(A) Isothermal calorimetry binding experiment of 100 μM ONC201 titrated into 20 μM ClpP (concentration of ClpP monomer). (B) The co-crystallography result focusing on the hydrophobic pocket between two ClpP subunits bound to ONC201. Hydrogen bonds are indicated by dashed lines and water molecule mediating hydrogen bonding as a red sphere. (C) Top (top) and side (bottom) views of human apo-ClpP (grey; PDB-ID:1TG6) and its ONC201 complex (blue) in surface representation; the ONC201 molecule is shown in magenta. Insert: ONC201 molecule in stick representation against ClpP surface. (D) Apo-ClpP (left) and its ONC201 complex (right) color-coded according to residue B-factors (low to high – blue to red). Insert color coding: carbon in green (ligand) and white (protein), nitrogen in dark blue, oxygen in red. (E) Cross-section through the human ONC201-complexed ClpP tetradecamer; position of side pores are indicated by black triangles. Close up of the pore (insert) between chains C (bottom left), D (bottom right) and symmetry-related chain K (top). Protein chains are indicated by ribbon colored based on residue B-factors (blue to red – low to high; protein surface in shades of gray). (F) Model of ONC212-binding to ClpP. The ligand is displayed as sticks and the surrounding protein is shown in surface representation. Color coding as in panel C but carbon in yellow (ligand); fluorine in light blue. (G) Concentration-dependent effects of treatment with ONC201 and ONC212 on thermal stability of endogenous ClpP in OCI-AML2 cells assessed using cellular thermal shift assays (CETSA). UHC: unheated control. (H) Effect of ONC201 removal from media on thermal stability of endogenous ClpP in intact OCI-AML2 cells. ONC201 (10 mM) treated cells were washed with PBS and re-incubated in fresh medium for up to 75 min prior to CETSA. w = wash. See also Figure S3.

Figure 4.. ClpP activation by imipridones kills malignant cells through a ClpP-dependent mechanism.

(A) Effects of ONC201 and ONC212 on viability of OCI-AML2, TEX, OCI-AML3 and Z138 cells. Data represent percent mean ± SD viable cells measured by alamar blue assay in OCI-AML2, TEX cells, or by annexin V assay in OCI-AML3 and Z138 cells after a 72-hr period of exposure to the drugs.(B) Changes in live cell number by ONC201 and ONC212 compared to untreated controls in primary AML and normal bone marrow mononuclear cells (BM-MNC). Cells were treated with ONC201 and ONC212 at indicated concentrations for 72 hr. Annexin V- and DAPI-negative cells were measured by flow cytometry and normalized to that in untreated controls. Data represent mean ± SD. #, ##: samples that were relatively resistant to ONC201. (C) Effects of ONC201 and ONC212 on viability in ClpP⁺/⁺ and ClpP^−/− T-REx HEK293 cells and protein expression of ClpP. Data represent percent mean ± SD viable cells measured by alamar blue assay after a 72-hr period of exposure to the drugs. (D) Correlation between pretreatment expression level of ClpP and the effects of ONC201 on viability of primary AML samples measured by annexin V assay after a 72-hr period of exposure to the drug (D). R= −0.82, p = 0.003. ClpP levels were quantified by immunoblot analysis of untreated samples. (E) The correlation in (D) is plotted by dichotomizing the ClpP expression. Low ClpP = samples with ClpP levels that were 1 SD below average. High ClpP = all other samples. The results are expressed as the mean value of triplicate samples ± SD (error bars). ***p < 0.001. See also Figure S4 and Tables S1 and S2.

Figure 5.. Cytotoxicity of imipridones is ClpP-dependent.

(A) Sensitivity of ONC201-sensitive and ONC201-resistant Z138 to ONC201 and ONC212 was assessed by Annexin V assays. Data represent percent mean of triplicate samples ± SD (error bars) for viable (annexin V and PI double negative) cells. (B) Effects of WT and D190A-ClpP on degradation of fluorogenic AC-WLA-AMC. The results are expressed as the mean value of triplicate samples ± SD (error bars). (C) Effects of ONC201 and ONC212 on degradation of the fluorogenic substrates AC-WLA-AMC and FITC-casein by D190A ClpP. The results are expressed as the mean value of triplicate samples ± SD (error bars). (D) ITC data for ONC201 (100 μM) titrated into D190A-ClpP (20 μM; concentration of ClpP monomer). (E) The location of D190 and R226 at the interface of two heptamer rings in an apparently closed conformation of human mitochondrial ClpP. (F) Apoptosis induction by imipridones when WT ClpP is overexpressed (O/E) in ONC-R Z138 cells carrying D190A mutant ClpP. Cells were treated with ONC201 and ONC212 at indicated concentrations for 72 hr. E/V; empty vector as control. Protein expression levels of ClpP was assessed by immunoblotting. (G) Apoptosis induction by imipridones when WT or D190A-ClpP is overexpressed (O/E) in parental Z138 and OCI-AML3 cells. Cells were treated with ONC201 and ONC212 at indicated concentrations for 72 hr. Protein expression levels of ClpP were assessed by immunoblotting. The results are expressed as the mean value of triplicate samples ± SD (error bars) for annexin V-positive cells. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (F-G) See also Figure S5, S6.

Figure 6.. ClpP hyperactivation induces apoptosis following reduction of respiratory chain complex subunits.

(A) A subset of ClpP mitochondrial interactors was identified using BioID-MS and categorized according to selected gene ontology biological processes. Decreases in spectral counts following ONC201 treatment are illustrated and proportional to the decreases in color intensity. (B-E) Immunoblot analysis of respiratory chain complex subunits in parental (ONC-sensitive) Z138 cells and ONC-R Z138 cells (the clone #2) with over-expression of WT ClpP or an empty vector (B); in parental Z138 cells with over-expression of WT ClpP, D190A mutant ClpP, or an empty vector. (C); in Z138 cells with tetracycline-inducible Y118A mutant ClpP (D); and in primary AML cells and normal bone marrow (NBM) cells (E). AML#3_1 and #3_2 are from the same patient but at different time points in relapse. Cells were treated with ONC201 at indicated concentrations for 24 hr. See also Figure S7 and Table S3.

Figure 7.. ClpP hyperactivation by ONC201 impairs oxidative phosphorylation.

(A) Effect of ONC201 on oxygen consumption rate (OCR) per 1 × 10⁵ cells in Z138 and Z138 D190A ClpP cells (measured by Seahorse Analyzer). The results are expressed as the mean value of triplicate samples ± SD (error bars). (B) Effects of ONC201 treatment on activity of respiratory chain complexes I, II, and IV in OCI-AML2 cells. The results are expressed as the mean value of triplicate samples ± SD (error bars). *p < 0.05 and **p < 0.01. (C) Effect of ONC201 treatment on mitochondrial ROS production in Z138 and Z138 D190A ClpP cells. The results are expressed as the mean value of triplicate samples ± SD (error bars). ***p < 0.001. (D) A transmission electron microscopy image of mitochondria in OCI-AML3 cells treated with or without 5 mM ONC201 for 24 hr. (E) Immunoblot of ATF4, p-eIF2a and eIF2a in Z138 cells with tetracycline-inducible Y118A ClpP treated with tetracycline for 48 hr at indicated concentrations.

Figure 8.. ClpP activation exerts anti-tumor effects in vivo.

(A) Tumor burden measured by luciferase activity using IVIS imaging in xenograft mice with WT or D190A-mutant ClpP overexpressing Z138 cells treated with ONC212 (50 mg/kg every other day, oral gavage) or vehicle (n = 7 each) after confirming engraftment. Color scale bars represent relative luminescence intensity (photons/sec). (B) Intensities of luminescence detected by IVIS imaging in the mice in Figure 6A. The results are expressed as the mean value of triplicate samples ± SD (error bars). ***p < 0.001. (C) Survivals of xenograft mice using Z138 cells over-expressed with WT or D190A ClpP. (D) Tumor volumes of xenograft mice using OCI-AML2 cells treated with ONC201 (100 mg/kg twice daily, oral gavage) or vehicle (n = 10 each) from 5 days after transplantation for 13 days. The results are expressed as the mean value of n = 10 ± SD (error bars). **p < 0.01 and ***p < 0.001. (E) Survivals of NSG mice injected with PDX cells [t(9;11)(p22; q23), CEBPA and ATM mutants] treated with or without 250 nM ONC212 for 36 hr (n = 10 each). See also Figure S8.