Figure 2. with four supplements: The impact of small-molecule inhibitors and BH3-sequence mutations on the interactions between anti-apoptotic proteins and BH3-only pro-apoptotic proteins.

(A) Values of R can be used to define binding interactions from FLIM-FRET data. Sample calculation of R_ABT-263 for ^CBcl-XL:^VBad binding in MCF-7 cells. To maximize the accuracy and dynamic range of the assay to quantify the impact of ABT-263 on the ^VBH3-proteins binding to ^CBcl-XL, we interpolated the FLIM-FRET efficiency from the fitted binding curves at an intensity ratio of Venus to mCer3 of 0.5 (points A, (B and C). The non-binding mutant ^VBad4E, in which h1, h2, h3 and h4 in the BH3 region were all mutated to glutamic acid, served as a control for FRET due to random collisions (red line, point C) and was subtracted from A and B as background. The signal remaining after the addition of ABT-263 (cyan line, point B) expressed as a percentage of the signal with DMSO added instead of the drug (green line, point A) is defined as R_ABT-263. (B) R_ABT-263 and R_BH3-2A for binding of the indicated ^VBH3-proteins to ^CBcl-XL (black) from binding curves in shown in Figure 1and ^CBcl-2 (blue) in MCF-7 cells. R_BH3-2A is calculated similarly to R_ABT-263 except the FLIM-FRET efficiency of the mutant is substituted for the value after adding ABT-263. Binned data, binding curves and sample images for Bcl-2 are shown in Figure 2—figure supplement 1. (C) The R_ABT-263 and R_BH3-2A values for ^CBcl-XL:^VBH3 proteins interactions in BMK-DKO cells. Binned data, binding curves and sample images are shown in Figure 2—figure supplement 2. Control experiments showing that the morphology changes that accompany cell death do not change the lifetime values determined by FLIM are shown in Figure 2—figure supplement 3. (D) Bcl-XL inhibitors displace ^VBad efficiently but ^VBimEL poorly from ^CBcl-XL in live cells. The dose-dependent inhibition curves due to the indicated concentrations of ABT-263 (black), and the Bcl-XL inhibitors A-1155463 (red) and A-1331852 (green) shown for ^CBcl-XL:^VBad (left) and ^CBcl-XL:^VBimEL (right) complexes in live MCF-7 cells. R values are ±95% confidence intervals from binned data and binding curves shown in Figure 2—figure supplement 4 (E) ABT-263 displaces tBid (10 nM) but not Bim (10 nM) from Bcl-XL (40 nM) in vitro. Percent of BH3 protein bound to Bcl-XL (BH3 bound %) measured by loss of FRET for Bcl-XL:tBid and Bcl-XL:Bim quantified for purified full-length dye labeled proteins incubated with liposomes and the indicated concentrations of drug. Data are from three experimental replicates, not all points are visible due to overlap. In all figures R_ABT-263 and R_BH3-2A data points for Bcl-XL and Bcl-2 are black and blue respectively.

Figure 2—figure supplement 1. FLIM-FRET binding curves for Bcl-2 binding to BH3 proteins in MCF-7 cells.

(A–C) Both BH3-2A mutation and ABT-263 displace (A) ^VBad, (B) ^VtBid but not (C) ^VBimEL from ^CBcl-2 in MCF-7 cells. FLIM FRET binding curves are shown above representative images. Scale bar 10 μm. FLIM-FRET binding curves for protein pairs as indicated illustrate the extent of binding for untreated or DMSO treated controls (black and green, respectively), cells treated with 20 μM ABT-263 (blue) or cells expressing BH3-2A mutant BH3-proteins (red). Data points in FLIM-FRET binding curves correspond to the average FLIM-FRET efficiency for binned data (20 < cells/bin < 3,000). Error bars, SEM; shadowed area, 95% confidence interval for the best fit of the model to the data for each binding curve.

Figure 2—figure supplement 2. FLIM-FRET binding curves for Bcl-XL binding to BH3 proteins in BMK DKO cells.

(A–C) Both BH3-2A mutation and ABT-263 displace (A) ^VBad, (B) ^VtBid but not (C) ^VBimEL from ^CBcl-XL in BMK cells in which the genes for Bax and Bak have been deleted (DKO). FLIM FRET binding curves are shown above representative images. Scale bar 10 μm. FLIM-FRET binding curves for protein pairs as indicated illustrate the extent of binding for untreated or DMSO treated controls (black and green, respectively), cells treated with 20 μM ABT-263 (blue) or cells expressing BH3-2A mutant BH3-proteins (red). Data points in FLIM-FRET binding curves correspond to the average FLIM-FRET efficiency for binned data (20 < cells/bin < 3,000). Error bars, SEM; shadowed area, 95% confidence interval for the best fit of the model to the data for each binding curve.

Figure 2—figure supplement 3. FLIM-FRET measurements for ^CBcl-XL binding to ^VBimEL-Bad in alive and dying MCF-7 cells.

Representative micrographs selected showing mCer3 and Venus intensities, Intensity weighted FLIM and selected regions of interest (ROIs) for (A) alive and (B) dying cells. Lifetime determinations for mCer3 for the identified ROIs are presented in tabular form at the left. Ch1 – Venus intensity, Ch2 – mCer3 intensity, ratio of Venus to mCer3 – ratio, Lifetime for ROI – Tau. MCF-7 overexpressing mCer3-Bcl-XL were transiently transfected to express vBimEL-Bad and after treatment with DMSO for >16 hr images of live cells were acquired by imaging adherent flat cells, dying cells (spherical) were imaged from a higher focal plane one hour later due to the time needed to acquire the first set of data.

Figure 2—figure supplement 4. FLIM-FRET binding curves for ^VBad (left) and ^VBimEL (right) to ^CBcl-XL in MCF-7 cells in the presence of different concentrations of BH3-mimetics.

The curves illustrate the extent of binding for DMSO treated controls (black), or cells treated with 1.25 μM (pink), 2.5 μM (light blue), 5 μM (green), 10 μM (red), and 20 μM (dark blue) BH3-mimetics: (a and d) A-1155463, (b and e) A-1331852, and (c and f) ABT-263. Data points in FLIM-FRET binding curves correspond to the average FLIM-FRET efficiency for binned data (20 < cells/bin < 3,000). Error bars, SEM; shadowed area, 95% confidence interval for the best fit of the model to the data.