The endoplasmic reticulum pool of Bcl-xL prevents cell death through IP3R-dependent calcium release

Abstract

Apoptosis plays a role in cell homeostasis in both normal development and disease. Bcl-xL, a member of the Bcl-2 family of proteins, regulates the intrinsic mitochondrial pathway of apoptosis. It is overexpressed in several cancers. Bcl-xL has a dual subcellular localisation and is found at the mitochondria as well as the endoplasmic reticulum (ER). However, the biological significance of its ER localisation is unclear. In order to decipher the functional contributions of the mitochondrial and reticular pools of Bcl-xL, we generated genetically modified mice expressing exclusively Bcl-xL at the ER, referred to as ER-xL, or the mitochondria, referred to as Mt-xL. By performing cell death assays, we demonstrated that ER-xL MEFs show increased vulnerability to apoptotic stimuli but are more resistant to ER stress. Furthermore, ER-xL MEFs displayed reduced 1,4,5-inositol trisphosphate receptor (IP3R)-mediated ER calcium release downstream of Phospholipase C activation. Collectively, our data indicate that upon ER stress, Bcl-xL negatively regulates IP3R-mediated calcium flux from the ER, which prevents ER calcium depletion and maintains the UPR and subsequent cell death in check. This work reveals a moonlighting function of Bcl-xL at the level of the ER, in addition to its well-known role in regulating apoptosis through the mitochondria.

Fig. 1. Bcl-xL protects from ER stress.

A Western blotting analysis of Bcl-xL endogenous levels in WT and bclx KO MEFs. α-Tubulin was used as a loading control. B–D Cell death quantification in WT and bclx KO MEFs treated with 1 µM Staurosporine for 16 h (B), 10 µM Thapsigargin for 48 h (C) or 2.5 µM Tunicamycin for 48 h (D). The difference between treated cells and negative controls (DMSO-treated) is shown on the vertical axis of each panel (Sytox GREEN^TM dots). Left panels: histograms showing the results from three independent experiments. A Student’s t test was performed with N_WT = 3, N_KO = 3 (mean ± SD; *, p < 0.05; **p < 0.01; ***p < 0.001). Right panels: representative kinetics. E. Western blotting analysis of the expression of ATF4 & CHOP, used as UPR markers, in the presence of increasing amounts of Thapsigargin (TG) for 24 h in WT and bclx KO MEFs. Vinculin was used as a loading control.

Fig. 2. Caracterization of ER-xL and Mt-xL MEFs.

A Diagram of Bcl-xL products in WT, ER-xL and Mt-xL MEFs. These cells respectively express Bcl-xL with intact TM domain, CB5 targeting sequence and ActA targeting sequence. MEFs were obtained from E13 embryos. Generation of recombinant mice is described in [21]. B Western blotting. Detection of Bcl-xL in WT, ER-xL and Mt-xL MEFs. Vinculin was used as loading control. C Immunofluorescence. Bcl-xL subcellular localisation in WT, ER-xL and Mt-xL MEFs. Co-localisation was assessed using ER-EGFP transfection and Mitotracker staining for ER and mitochondrial localization, respectively. Scale bar: 10 µm. D Profile plots of fluorescence signals. Fluorescence intensity along the white segments on merged images (last two right panels, see “C” above) were quantified by ImageJ software. E Western blotting. Detection of Bcl-xL in subcellular fractions from ER-xL MEFs (top panels) and Mt-xL MEFs (lower panels). (Tot) whole-cell lysates; (Mito) mitochondria; (Cyto) cytosol; (ER) endoplasmic reticulum. Vinculin was used as a cytosol marker, Calnexin as an ER marker and F0F1 ATPase as a mitochondrial marker.

Fig. 3. ER-xL MEFs resist Ca²⁺-dependent cell death.

A–C Cell death quantification (Sytox GREEN^TM-positive cells) in WT, ER-xL and Mt-xL MEFs treated with 1 µM Staurosporine (A), 10 µM Thapsigargin (B) or 2.5 µM Tunicamycin (C). The difference between treated cells and negative controls (DMSO-treated) is shown on the vertical axis of each panel (Sytox GREEN^TM dots). Left panels: histograms showing the results from three independent experiments. A one way ANOVA test was performed accordingly with N_WT = 3, N_ER-xL = 3, N_Mt-xL = 3 (mean ± SD; ns, non-statistically significant, p > 0.05; **, p < 0.01; ***, p < 0.001). Cell death was measured at 24 h for Staurosporine and at 48 h for Thapsigargin and Tunicamycin. Right panels: representative kinetics. D Detection of cleaved Caspase 3 through fluorescence microscope analysis in WT, ER-xL and Mt-xL MEFs after treatment with 250 nM Staurosporine (STS) for 6 h. Cells treated with DMSO were used as negative controls. Histograms show cleaved Caspase 3 quantification. A two-way ANOVA was performed with N_{WT Control} = 16, N_{ER-xL control} = 17, N_{Mt-xL control} = 17, N_{WT STS} = 18 N_{ER-xL STS} = 18, N_{Mt-xL STS} = 18. (mean ± SEM; ns, non-statistically significant, ***, p < 0.001). E. Detection of cleaved Caspase 3 through fluorescence microscope analysis in WT, ER-xL and Mt-xL MEFs after treatment with 4 µM Thapsigargin (TG) for 24 h. Cells treated with DMSO were used as negative controls. Histograms (left panel) show cleaved Caspase 3 quantification. A two-way ANOVA was performed with N_WT = 15, N_KO = 15, N_WT = 15, N_KO = 15, N_WT = 15, N_KO = 15. (mean ± SEM; ns, non-statistically significant, ***, p < 0.001).

Fig. 4. Bcl-xL at the ER affects intracellular Ca²⁺ fluxes.

A Representative images showing the endogenous interactions between IP3R - C terminus and Bcl-xL in WT and KO MEFs through a Proximity Ligation Assay (PLA). DAPI was used to stain nuclei (blue dots). Red dots correspond to PLA-positive signals, indicating close proximity between Bcl-xL and IP3R (middle upper panel and right upper panel). The middle bottom panel (PLA/KO, negative control) is unstained, confirming the specificity of the labelling observed in the top panel (PLA/WT). White arrows point to red dots. Scale bar: 30 µm. B Quantification of PLA experiments. A Mann Whitney’s test was performed with N_WT = 30, N_KO = 31 (mean ± SEM; ***, p < 0.001). C Representative curves of ER passive Ca²⁺ leakage in WT, ER-xL, and Mt-xL MEFs transfected with CEPIA-1er after 10 μM Thapsigargin injection. D Quantification of the slope coefficient of ER passive calcium leakage in MEFs. A one way ANOVA was performed with N_WT = 14, N_ER-xL = 16, N_Mt-xL = 17 (mean ± SEM; ns, non-statistically significant). E. Representative curves of Store Operated Calcium Entry (SOCE) assessed with 5 μM Fluoforte after draining the ER from calcium by 10 μM Thapsigargin injection, followed by 2 mM CaCl₂ injection, in WT, ER-xL and Mt-xL MEFs. F The ratio of fluorescence (F/F0) indicating the maximal calcium uptake after CaCl₂ injection is shown. A one way ANOVA test was performed with N_WT = 3, N_ER-xL = 3, N_Mt-xL = 3 (mean ± SD; ns, non-statistically significant).

Fig. 5. Bcl-xL at the ER interacts with IP3R and decreases IP3R Ca²⁺ permeability.

A Representative curves of ER calcium release in WT, ER-xL and Mt-xL MEFs transfected with CEPIA-1er after 25μM m-3M3FBS (PLC agonist) injection. B Quantification of the slope coefficient of ER calcium release after PLC activation in MEFs. A Kruskall Wallis ‘s test was performed with N_WT = 46, N_ER-xL = 51, N_Mt-xL = 43 (mean ± SEM; **, p < 0.01; ***, p < 0.001). C Representative curves of cytosolic Ca²⁺ increase assessed with 5 μM Fluoforte after Ionomycin injection in WT, ER-xL and Mt-xL MEFs. D Quantification of Ionomycin response. The ratio of fluorescence, indicating Ca²⁺ peak amplitude relative to WT, is shown. A Kruskall Wallis’s test was performed with N_WT = 26, N_ER-xL = 26, N_Mt-xL = 26 (mean ± SEM; *, p < 0.05; **, p < 0.01; ***, p < 0.001). E-F. Quantification of ER calcium release normalized to A23187 maximal ionophore-induced Ca²⁺ release response through ⁴⁵Ca²⁺ flux analysis. Dose-response curves are shown in E. Quantification performed at 10μM IP3 is shown in F. A one way ANOVA test was performed with N_WT = 5, N_ER-xL = 5, N_Mt-xL = 5 (mean ± SEM; ns, non-statistically significant; *, p < 0.05; **, p < 0.01). IP3-induced Ca²⁺ release in Mt-Bcl-xL MEFs is increased, compared to wild-type MEFs and ER-Bcl-xL.

Fig. 6. Bcl-xL reduces Ca²⁺-dependent ER stress by closing the IP3R channel.

Proposed model for the role of Bcl-xL at the ER. Left panel: ER stress might induce ER Ca²⁺ release in the cytosol (yellow dots), promoting Ca²⁺ uptake by the mitochondria and subsequent apoptosis initiation through mPTP opening and cytochrome C release in the cytosol (red diamonds). Right panel: ER-targeted Bcl-xL interacts with IP3R and closes the channel, abrogating ER Ca²⁺ depletion and downstream apoptosis. This model highlights a new indirect anti-apoptotic function of Bcl-xL upon ER stress.