PP2A modulation overcomes multidrug resistance in chronic lymphocytic leukemia via mPTP-dependent apoptosis

Abstract

Targeted therapies such as venetoclax (VEN) (Bcl-2 inhibitor) have revolutionized the treatment of chronic lymphocytic leukemia (CLL). We previously reported that persister CLL cells in treated patients overexpress multiple antiapoptotic proteins and display resistance to proapoptotic agents. Here, we demonstrated that multidrug-resistant CLL cells in vivo exhibited apoptosis restriction at a pre-mitochondrial level due to insufficient activation of the Bax and Bak (Bax/Bak) proteins. Co-immunoprecipitation analyses with selective BH domain antagonists revealed that the pleiotropic proapoptotic protein (Bim) was prevented from activating Bax/Bak by "switching" interactions to other upregulated antiapoptotic proteins (Mcl-1, Bcl-xL, Bcl-2). Hence, treatments that bypass Bax/Bak restriction are required to deplete these resistant cells in patients. Protein phosphatase 2A (PP2A) contributes to oncogenesis and treatment resistance. We observed that small-molecule activator of PP2A (SMAP) induced cytotoxicity in multiple cancer cell lines and CLL samples, including multidrug-resistant leukemia and lymphoma cells. The SMAP (DT-061) activated apoptosis in multidrug-resistant CLL cells through induction of mitochondrial permeability transition pores, independent of Bax/Bak. DT-061 inhibited the growth of wild-type and Bax/Bak double-knockout, multidrug-resistant CLL cells in a xenograft mouse model. Collectively, we discovered multidrug-resistant CLL cells in patients and validated a pharmacologically tractable pathway to deplete this reservoir.

Trial registration: ClinicalTrials.gov NCT02419560.

Keywords: Apoptosis pathways; Cancer; Cell Biology; Oncology; Phosphoprotein phosphatases.

Figure 1. Circulating CLL cells with the CD69^Pos activation phenotype in vivo display pre-mitochondrial apoptosis restriction due to defective activation of Bax and Bak proteins.

(A–C) Freshly frozen PBMCs from various CLL patients were screened in an ATA by incubation with an inhibitor of Bcl-2 (VEN: 12.5, 25, 50, 100 nM), Mcl-1 (S63845: 0.61, 0.91, 1.35, 2.05 μM), or Bcl-xL (A1155463: 4, 8, 16, 32 μM) for 3 hours without added agonists. (A) Representative flow images showing the expression of active Bax and Bak proteins in CD69^Pos and CD69^Neg CLL (viability dye^NegCD5⁺CD19⁺) cells in a patient’s PBMCs (patient [Pt] 08) incubated with DMSO or S63845 (2.05 μM). SSC-A, side scatter area. (B and C) Data showing the percentage of CD69^Pos or CD69^Neg CLL cells positive for active Bax (B) or active Bak (C) from multiple patient samples exposed to various proapoptotic agents in the ATA. (D) PBMCs from patients with CLL isolated prior to or during treatment with VEN (Supplemental Table 3) were analyzed in the ATA by incubation ex vivo with inhibitors of Bcl-2 (VEN: 25, 50, 100, 200 nM) or Bcl-xL (A1155463: 8, 16, 32, 64 μM) for 3 hours without agonists. Data show the percentage of CLL (viability dye^NegCD5⁺CD19⁺) cells positive for the active form of Bax following ex vivo incubation with VEN or A1155463. Statistical significance was determined by ANOVA with Šidák’s post hoc test for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Data are presented as the mean ± SD.

Figure 2. Proapoptotic protein Bim swapping by antiapoptotic proteins establishes pre-mitochondrial apoptosis restriction in multidrug-resistant CLL cells.

(A) Diagram of the experimental design. PBMCs from patients with CLL were pretreated with the agonist mix for 12 hours to induce a multidrug-resistant state. Then, cells were treated with an inhibitor of Bcl-2 (VEN, 200 nM), Bcl-xL (A1155463, 16 μM), or Mcl-1 (S63845, 273 nM) as well as a second dose of agonist mix for an additional 12 hours. The proapoptotic protein Bim in cell lysates was immunoprecipitated using an anti-Bim antibody, and the antiapoptotic proteins Bcl-2, Bcl-xL, and Mcl-1 bound to Bim were analyzed by Western blotting using the corresponding antibodies. (B) WB images from a representative patient sample (Pt 33) showing a shift in Bim binding to Bcl-2, Mcl-1, and Bcl-xL in the presence of VEN, A1155463, or S63845 as compared with DMSO control. (C) Densitometric quantitation data from experiments involving 6 different patients with CLL demonstrating a shift in Bim recruitment to various antiapoptotic proteins in the presence of VEN, A1155463, or S63845 as compared with the DMSO control. Statistical significance was determined by Student’s t test. *P < 0.05, **P < 0.01, and ****P < 0.0001. Data are presented as the mean ± SD.

Figure 3. Activation of PP2A using SMAPs induces cytotoxicity in leukemia/lymphoma cells that exhibit apoptosis resistance.

(A) Samples from patients with CLL were pretreated or not with the agonist mix, and various leukemia/lymphoma cell lines were analyzed for cytotoxicity with DT-061 or DBK-766 (8, 12, 16, 20, and 24 μM) using an alamarBlue assay. Samples were treated with drugs for 24 hours. The cytotoxicity data for cell lines were confirmed in 2 independent experiments, and multiple patient samples were screened. (B) Samples from patients with CLL were pretreated or not with the agonist mix as described in B, and various leukemia/lymphoma cell lines were analyzed for cytotoxicity with inhibitors of Bcl-2 (VEN; 12.5, 25, 50, 100, and 200 nM), Mcl-1 (S63845; 0.0625, 0.125, 0.25, 0.5, and 1 μM), Bcl-xL (A1155463; 0.5, 1, 2, 4, and 8 μM), and the combination (VAS) in an alamarBlue assay. Samples were treated with the drugs for 24 hours. The cytotoxicity data for the cell lines were confirmed in 2 independent experiments, and multiple patient samples were screened in an independent experiment. The average cell survival values are presented in a heatmap. (C) Sensitive or IBR- and VEN-resistant MCL cell lines were treated with DT-061 (8, 12, 16, 20, and 24 μM) for 24 hours, and drug-induced cytotoxicity was determined by alamarBlue assay. Statistical significance was determined by ANOVA with -Šidák’s post hoc test for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. The data were confirmed in multiple experiments as indicated and are presented as the mean ± SD.

Figure 4. PP2A activation by the small-molecule agonist DT-061 induces Bax/Bak-independent apoptosis in CLL cells.

(A) PBMCs from patients with CLL were preincubated with the agonist mix for 12 hours. Samples were treated with a second dose of the agonist mix as well as DT-061 (12, 16, and 20 μM) for 18 hours. Apoptosis induction (Bax activation and cleaved caspase 9 and cleaved PARP) and viability dye staining in CLL (CD5⁺CD19⁺) cells were analyzed by flow cytometry. (B) PBMCs from patients with CLL were screened by flow cytometry for Bax activation as well as cleaved PARP following incubation with DT-061 (9, 12, 15, and 18 μM) for 9 hours without added agonists. Data show the percentage CD69^Pos or CD69^Neg CLL (CD5⁺CD19⁺) cells positive for active Bax or cleaved PARP, after subtraction of the spontaneous apoptosis values from the DMSO-treated controls. (C) The Bax/Bak-DKO CLL cell line MEC1 was developed using the CRISPR/Cas9 system, as described in Methods. WB data show the expression of Bax and Bak proteins in WT and Bax/Bak-DKO clones. (D) The parent MEC1 cell line as well as WT and Bax/Bak-DKO clones were treated with DT-061 (12, 16, and 20 μM) or a combination of VEN (0.2 μM), S63845 (2 μM), and A1155463 (1.6 μM) (SVA) for 12 hours. Cleaved PARP was analyzed by flow cytometry. The average data from 3 independent experiments are presented as a bar graph, which shows the percentage of MEC1 cells positive for cleaved PARP. Statistical significance was determined by ANOVA with Šidák’s post hoc test for multiple comparisons. ***P < 0.001 and ****P < 0.0001. Data are presented as the mean ± SD.

Figure 5. PP2A modulation by DT-061 activates apoptosis in CLL cells by releasing mPTPs.

(A and B) PBMCs from patients with CLL pretreated with increasing concentrations of the CypD inhibitor NIM811 or CspA for 1 hour were incubated with DT-061 (16 μM) or VEN (25 nM) for 12 hours. Apoptosis induction was determined by analyzing cleaved caspase 9, cleaved PARP, and viability dye staining in CLL cells using flow cytometry (cleaved PARP and viability dye data are included in Supplemental Figure 7). Data are presented as box plots showing the percentage of CLL (CD5⁺CD19⁺) cells positive for cleaved caspase-9. (C) Samples from patients with CLL pretreated with the CypD inhibitor NIM811 (10 μM) or CspA (10 μM) were incubated with DT-061 (12, 16, and 20 μM) for 6 hours, and mPTP opening in CLL (CD5⁺CD19⁺) cells was assessed using flow cytometry as described in Methods. Stacked histograms show calcein AM staining in CLL cells subjected to various treatments (left panel). Box plots show the percentage CLL cells positive for Calcein AM staining in multiple patient samples treated with DT-061 with or without NIM811 or CspA pretreatment (right panel). Statistical significance was determined by ANOVA with Šidák’s post hoc test for multiple comparisons. *P < 0.05 and **P < 0.01. Data are presented as the mean ± SD.

Figure 6. DT-061 overcomes antiapoptotic multidrug resistance in CLL xenograft mouse model in vivo.

(A) Tumor growth in mice subcutaneously inoculated with the WT MEC1 cell line and treated with vehicle, DT-061, or the combination of IBR and VEN as indicated. (B) Tumor growth in mice inoculated with Bax/Bak-DKO MEC1 cells and treated with vehicle, DT-061, or the combination of IBR and VEN. (C and D) Percentage of body weight change during drug treatment in mice inoculated with WT (C) or Bax/Bak-DKO (D) MEC1 cells. Data are presented as the mean ± SEM. Statistical significance was determined by 2-way ANOVA with Dunnet’s post hoc test for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.