Mito-priming as a method to engineer Bcl-2 addiction

Abstract

Most apoptotic stimuli require mitochondrial outer membrane permeabilization (MOMP) in order to execute cell death. As such, MOMP is subject to tight control by Bcl-2 family proteins. We have developed a powerful new technique to investigate Bcl-2-mediated regulation of MOMP. This method, called mito-priming, uses co-expression of pro- and anti-apoptotic Bcl-2 proteins to engineer Bcl-2 addiction. On addition of Bcl-2 targeting BH3 mimetics, mito-primed cells undergo apoptosis in a rapid and synchronous manner. Using this method we have comprehensively surveyed the efficacy of BH3 mimetic compounds, identifying potent and specific MCL-1 inhibitors. Furthermore, by combining different pro- and anti-apoptotic Bcl-2 pairings together with CRISPR/Cas9-based genome editing, we find that tBID and PUMA can preferentially kill in a BAK-dependent manner. In summary, mito-priming represents a facile and robust means to trigger mitochondrial apoptosis.

Figure 1. Mito-priming as a method of induced Bcl-2 addiction.

(a) Method outline. (b) SVEC cells expressing eGFP-tBID 2A BCL-xL or other constructs were co-stained with MitoTracker Deep Red. Colocalization was quantified using the Pearson's coefficient. Representative images show eGFP-tBID expression and MitoTracker Deep Red staining from SVEC cells stably expressing eGFP-tBID 2A BCL-xL. Scale bar, 10 μm. (c) SVEC cells expressing eGFP-tBID 2A BCL-xL were treated with ABT-737 and analysed for cell viability using an IncuCyte imager and SYTOX Green exclusion. Percentage cell death was calculated by normalizing against maximal cell death (13-h treatment with 10 μmol l⁻¹ ABT-737). Error bars represent the standard error of the mean (s.e.m.) from three independent experiments. (d) SVEC cells stably expressing the indicated constructs were treated with increasing concentrations of ABT-737 (μmol l⁻¹; 24 h) and assayed for clonogenic survival. Error bars represent the s.d. of triplicate samples from a representative experiment carried out twice independently. (e) SVEC cells stably expressing indicated constructs were treated with increasing concentration of ABT-737 (μmol l⁻¹) for 6 h with 10 μmol l⁻¹ ABT-737 in presence of Q-VD-OPh. Cells were quantified for cytochrome c release by confocal microscopy. Error bars represent the s.e.m. of three independent experiments. (f) SVEC cells stably expressing the indicated constructs were treated for 6 h with ABT-737 (μmol l⁻¹). PARP and caspase-3 cleavage was determined by western blot. Actin was probed as a loading control. (g) SVEC cells stably expressing the indicated constructs were treated for 13 h with 10 μmol l⁻¹ ABT-737 or enantiomer (ENA). Cell viability was determined as in c. Percentage cell death was calculated by normalizing against 100% cell death (24-h treatment with 1 μmol l⁻¹ actinomycin D). (h) Cell lines stably expressing control vector (pLKO1) or two independent shRNA sequences targeting APAF-1 were treated for 16 h with enantiomer (ENA) or ABT-737. 100% death: 24-h treatment with 1 μmol l⁻¹ actinomycin D (for each line). Cell viability was determined as in c. Percentage cell death was calculated by normalizing against 100% cell death (24-h treatment with 1 μmol l⁻¹ actinomycin D). In g,h error bars represent the s.e.m. of three independent experiments. See also Supplementary Fig. 1 and Supplementary Movies 1–3. Q-VD-OPh, quinolyl-valyl-O-methylaspartyl-(2,6-difluoro- phenoxy)-methyl ketone.

Figure 2. Applying mito-priming to define the selectivity and potency of BH3 mimetics.

(a) Predicted pattern of inhibition by different BH3 mimetics tested. (b) Western blot analysis of SVEC cells stably expressing eGFP-tBID 2A BCL-xL, eGFP-tBID 2A BCL-2 or eGFP-tBID 2A MCL-1. * denotes ectopically expressed MCL-1. (c,d) SVEC cells stably expressing eGFP-tBID 2A BCL-xL (BCL-xL-dependent line) or eGFP-tBID 2A BCL-2 (BCL2-dependent line) were treated with increasing concentrations of different BH3 mimetics. Cell viability was determined by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager. Error bars represent the s.e.m. of three independent experiments. (e,f) BCL-xL (e) or BCL-2-dependent lines (f) were treated with increasing concentrations of different BH3 mimetics and analysed for long-term survival by clonogenic assay. Percentage clonogenic survival is plotted relative to untreated cells. Error bars represent the s.d. of triplicate samples from a representative experiment carried out twice independently. (g) BCL-xL-dependent line was treated with increasing concentrations of BCL-xL-directed BH3 mimetics. Cell viability was determined by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager. Errors bars represent the s.e.m. of three independent experiments. See also Supplementary Fig. 2.

Figure 3. Mito-priming defines potent and specific MCL-1 targeting BH3 mimetics.

(a) SVEC cells stably expressing FLAG-tBID 2A GFP-BCL-xL, FLAG-tBID 2A GFP-BCL-2 or FLAG-tBID 2A GFP-MCL-1 were transiently transfected with NOXA. Cell viability was analysed 24 h post-transfection by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager. Error bars represent the s.e.m. of three independent experiments. (b) SVEC cells stably expressing FLAG-tBID 2A GFP-MCL-1 (MCL-1-dependent line) were treated with increasing concentrations of putative MCL-1 inhibitors UMI-77 or A-1210477. Cell viability was analysed 24 h post-treatment by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager. Error bars represent the s.e.m. of three independent experiments. (c) MCL-1-dependent line was treated with UMI-77 or A-1210477 (both 10 μmol l⁻¹) and analysed over time for cell viability by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager. Error bars represent the s.e.m. of three independent experiments. (d,e) SVEC cells stably expressing FLAG-tBID 2A GFP-BCL-xL, FLAG-tBID 2A GFP-BCL-2 or FLAG-tBID 2A GFP-MCL-1 were treated with UMI-77 or A-1210477 (10 μM for 24 h) then cell viability was analysed by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager (d) or by clonogenic survival assay (e). Error bars represent the s.e.m. of three independent experiments for d and s.d. of triplicate samples from a representative experiment carried out twice independently for e. In all cases, cells were treated with MCL-1 inhibitors in 3% FBS containing DMEM.

Figure 4. Mito-priming reveals a dependence of tBID and PUMA for BAK over BAX.

(a) Overview of cell lines generated. (b) Western blot analysis of BAX and BAK expression in CRISPR/CAS9 genome edited cell lines. Actin was probed as a loading control. (c,d) SVEC cells stably expressing tBID-, BIM- or PUMA 2A GFP-BCL-xL were treated with ABT-737 (10 μmol l⁻¹) then cell viability was analysed by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager (c) or by clonogenic survival assay (d). In c cell death was normalized to treated, empty CRISPR control for each line (Control, each 100% cell death). Error bars represent the s.e.m. of 3 independent experiments (c) or s.d. of triplicate samples from a representative experiment carried out twice independently (d). (e) BAK-deleted or reconstituted SVEC cells stably expressing FLAG-tBID 2A BCL-xL or FLAG-PUMA 2A BCL-xL were treated with ABT-737 (10 μmol l⁻¹ for 16 h). Cell viability was analysed by SYTOX Green dye exclusion and IncuCyte imaging. Percentage cell death was calculated following normalization to treated control for each line (each 100% cell death). Errors bars represent the s.e.m. of three independent experiments. (f) SVEC cells stably expressing FLAG-PUMA 2A BCL-xL were treated for 4 h with increasing concentrations of ABT-737. Cleavage of caspase-3 and PARP was assessed by western blot. Actin was probed as a loading control. (g) Genome-edited SVEC cells stably expressing eGFP-tBID 2A BCL-2 were treated with ABT-199 (10 μmol l⁻¹) and analysed for cell viability using SYTOX Green exclusion and IncuCyte imaging. Percentage cell death was calculated following normalization to ABT-199 treated empty CRISPR vector control line for 16 h. Errors bars represent the s.e.m. of three independent experiments. (h) Genome-edited HeLa cells stably expressing either eGFP-tBID, BIM or PUMA 2A BCL-xL were treated with 10 μmol l⁻¹ ABT-737 for 24 h and analysed for cell viability by SYTOX Green exclusion and IncuCyte imaging. 100% death was set-up as the maximal ABT-737-induced death in the empty control line. Errors bars represent the s.e.m. of three independent experiments. See also Supplementary Fig. 3.

Figure 5. Selective dependence of tBID and PUMA for BAK is independent of prosurvival Bcl-2 function and requires a cellular environment.

(a) SVEC cell lines were transiently transfected with GFP–tBID or GFP–PUMA. Sixteen hours post transfection the percentage of Annexin V-positive cells in the GFP-expressing population (AnnV+/GFP+) was quantified by flow-cytometry. Error bars represent the s.d. of triplicate samples from a representative experiment carried out twice independently. (b) Control, BAX and/or BAK^{CRISPR−/−} cells stably expressing eGFP-tBID 2A BCL-xL were treated with for 1 h with 10 μmol l⁻¹ ABT-199 and A-1210477 before the addition of A-1331852 (10 μmol l⁻¹). Cell viability was analysed by SYTOX Green dye exclusion and live-cell imaging using an IncuCyte imager. Percentage cell death was calculated following normalization to treated control for each line (each 100% cell death). Error bars represent the s.e.m. from three independent experiments. (c) Control, BAX and/or BAK^{CRISPR−/−} SVEC cells stably expressing eGFP-PUMA 2A BCL-xL were treated as in b. Errors bars represent the s.e.m. from three independent experiments. (d) Isolated mitochondria from BAX and/or BAK^{CRISPR−/−} SVEC cells stably expressing eGFP-tBid 2A BCL-x_L were treated in the presence or absence of 100 nM recombinant wild-type Bax and/or 10 μM ABT-737 or 10 nM recombinant tBID. Mitochondrial permeabilization was monitored by western blot analysis of cytochrome c into the supernatant (S). The mitochondrial fraction (P), Bax and VDAC serve as loading controls.

Figure 6. BH3-only protein dependency for BAK is dictated by the BH3-domain.

(a) SVEC cells stably expressing GFP-BIM^{tBID BH3} or GFP-tBID^{BIM BH3} together with BCL-xL were analysed by Western blot for construct expression. (b,c) Cell death and long-term clonogenic survival of GFP-BIM^{tBID BH3} and GFP-tBID^{BIM BH3} 2A BCL-xL CRISPR lines treated with 10 μmol l⁻¹ ABT-737. (b) Error bars represent the s.e.m. of three independent experiments. Percentage cell death was calculated by normalizing to ABT-737-induced death in the empty CRISPR control line (100% death). (c) Error bars represent the s.d. of triplicate samples from a representative experiment carried out twice independently. (d) SVEC cells stably expressing GFP–BIM^{PUMA BH3} or GFP–PUMA^{BIM BH3} together with BCL-xL were analysed by western blot for construct expression. (e) Cell death of GFP–BIM^{PUMA BH3} and GFP–PUMA^{BIM BH3} 2A BCL-xL CRISPR lines treated with 10 μmol l⁻¹ ABT-737. Error bars represent the s.e.m. of three independent experiments. Percentage cell death was calculated by normalizing to ABT-737-induced death in the empty CRISPR control line (100% death). See also Supplementary Fig. 4.