Fig. 3: The K87 residue is critical for Bcl-xL interaction with IP₃R.

a In silico representations of Bcl-2 and Bcl-xL three-dimensional structures. The lysine residues of interest (K17 in Bcl-2 and K87 in Bcl-xL) are indicated. Image taken from our previously published work [28]; this work is licensed under a Creative Commons Attribution 4. 0 International License. b Alignment of the conserved amino acid motifs for Bcl-xL’s BH3 domain in vertebrates. The conserved lysine (K87 in human) is highlighted (red rectangle). The number of species used for each motif construction is shown in parentheses. “Z” means glutamic acid or glutamine. Top numbers represent amino acid numbers in human Bcl-xL sequence. c Representative co-immunoprecipitation (coIP) experiments using anti-FLAG performed in lysates from HeLa cells transiently overexpressing 3xFLAG-Bcl-xL or 3xFLAG-Bcl-xL^K87D. The samples were analyzed via western blot using antibodies against IP₃R1, FLAG and Bax. Total HeLa lysates were used as input (10 µg). PD: pull down; IB: immunoblot. The immunoreactive bands from independent coIP experiments, using each time independently transfected cells and freshly prepared lysates, were quantified and normalized to the binding of IP₃R1 (d) and Bax (e) to 3xFLAG-Bcl-xL. Data represent mean ± SD (N = 5). Statistically significant differences were determined using a one-way ANOVA (*P < 0.05). f Representative GST-pull down experiment comparing the binding of 3xFLAG-Bcl-xL vs. 3xFLAG-Bcl-xL^K87D from COS-7 cell lysate to purified GST-fused IP₃R1 fragments and parental GST control. The samples were analyzed via western blot using anti-FLAG. Total COS-7 lysates were used as input (0.1 µg). The corresponding western blot for the GST-IP₃R fragments is shown in Fig. S2b. g The immunoreactive bands from independent GST-pull down experiments, using each time independently transfected cells and freshly prepared lysates, were quantified and normalized to the binding of 3xFLAG-Bcl-xL and 3xFLAG-Bcl-xL^K87D to parental GST control, which was set as 1 for each experiment. The data are plotted as mean ± SD (N = 5). Statistically significant differences were determined using paired t test (*P < 0.05). h Binding curves showing the interaction of purified 6xHis-Bcl-xL and 6xHis-Bcl-xL^K87D with titrated GST-fused IP₃R domains generated by MST. Concentration of the 6xHis-Bcl-xL and 6xHis-Bcl-xL^K87D targets was kept constant at 50 nM, whereas the GST-LBD and GST-Fragment 3 were titrated down from 15 µM to 5 nM. The unit of the left axis (ΔF_norm) is a ratio of normalized fluorescence. The binding curves of wild-type 6xHis-Bcl-xL with GST-fused proteins are represented from Fig. 2d and are shown as reference. The binding curve of 6xHis-Bcl-xL^K87D with parental GST is shown in Fig. S4. Data points represent mean ± SD from triplicate measurements. i Left: Representative example of a MAPPIT experiment. The binding is shown as fold induction, calculated by dividing the average luciferase activity of erythropoietin-stimulated cells by the average obtained in non-stimulated cells. Binding of Bcl-xL, the Bcl-xL^K87D mutant or irrelevant prey control (SV40 large T antigen) to the IP₃R Fragment 3 and as negative control the bait vector without Fragment 3 are shown. Fold induction values at least four times higher than the irrelevant prey control are considered as bona fide protein-protein interactions. Values are represented as the mean of triplicates ± SEM within one representative experiment. The experiment was independently performed four times (N = 4). Statistically significant differences were determined using a one-way ANOVA (*P < 0.05). Right: Odyssey western blot analyses for the FLAG tag of the prey vector containing Bcl-xL or the Bcl-xL^K87D mutant fusion proteins (green) or for β-actin (red).