The non-apoptotic action of Bcl-xL: regulating Ca(2+) signaling and bioenergetics at the ER-mitochondrion interface

Abstract

Bcl-2 family proteins are known to competitively regulate Ca(2+); however, the specific inter-organelle signaling pathways and related cellular functions are not fully elucidated. In this study, a portion of Bcl-xL was detected at the ER-mitochondrion interface or MAM (mitochondria-associated ER membrane) in association with type 3 inositol 1,4,5-trisphosphate receptors (IP3R3); an association facilitated by the BH4 and transmembrane domains of Bcl-xL. Moreover, increasing Bcl-xL expression enhanced transient mitochondrial Ca(2+) levels upon ER Ca(2+) depletion induced by short-term, non-apoptotic incubation with thapsigargin (Tg), while concomitantly reducing cytosolic Ca(2+) release. These mitochondrial changes appear to be IP3R3-dependent and resulted in decreased NAD/NADH ratios and higher electron transport chain oxidase activity. Interestingly, extended Tg exposure stimulated ER stress, but not apoptosis, and further enhanced TCA cycling. Indeed, confocal analysis indicated that Bcl-xL translocated to the MAM and increased its interaction with IP3R3 following extended Tg treatment. Thus, the MAM is a critical cell-signaling junction whereby Bcl-xL dynamically interacts with IP3R3 to coordinate mitochondrial Ca(2+) transfer and alters cellular metabolism in order to increase the cells' bioenergetic capacity, particularly during periods of stress.

Keywords: Bcl-xL; Bioenergetics; Calicum signaling; ER; IP3R3; MAM; Mitochondria.

Fig. 1

Bcl-x_L resides at the MAM. a Cell fractionation of CHO-K1 and CHO-Bcl-x_L with markers for known nuclear, mitochondrial, ER, and MAM proteins are used to validate the purity of fractions. P1, whole cell/nuclear; Mito, mitochondria; P3, microsome; Cyto, cytosol. b Bcl-x_L expression in CHO-K1 and CHO-Bcl-x_L cells. c Bcl-x_L is peripherally attached to cellular membranes, and is thereby released from membrane fractions following alkali treatment (100 mM NaCO₃ for 30 min). IP₃R3, a known integral membrane protein, was used for comparison. Mem, crude mitochondrial membrane

Fig. 2

Bcl-x_L physically interacts with IP₃R3. a IP₃R3 co-precipitates with Bcl-x_L in CHO-Bcl-x_L cells transiently overexpressing IP₃R3. b Full-length (FL), BH4-truncated (ΔBH4), and TM domain-truncated (ΔTM) Bcl-x_L constructs. c A comparison of IP₃R3/Bcl-x_L binding by nickel pull-down of Bcl-x_L in CHO-K1 cells transiently overexpressing FL, ΔBH4, or ΔTM his-tagged Bcl-x_L. Endogenous IP₃R3 interaction with FL His-Bcl-x_L is taken as one hundred percent. Data represented as mean ± SEM (n = 4). *P < 0.05 versus full-length Bcl-x_L

Fig. 3

Limited thapsigargin treatment does not induce apoptosis. a In comparison to known apoptosis-inducing compounds, staurosporine (3 uM STA, 4 h) or camptothecin (50 uM Camp, overnight), there is not a significant increase in apoptosis due to prolonged treatment with Tg as indicated by Annexin V positive cells (n = 3). b There is not a significant activation of caspase 3/7 activities in comparison to known apoptosis-inducing compounds (STA or Camp) following prolonged Tg treatment (n = 3). 500 nM Tg treatment was used for all experiments. *P < 0.05 versus control at identical time point. **P < 0.05 versus non-treated control of the same cell line

Fig. 4

MAM-associated Bcl-x_L modulates IP₃R-mediated Ca²⁺ signaling. Ca²⁺ signaling was compared in co-cultured control (CHOYFP or CHO-HcRed) and Bcl-x_L overexpressing cells (CHO-YFP-Bclx_L or CHO-HcRed-Bcl-x_L) in response to 500 nM Tg treatment. During experimentation, cells were perfused in Ca²⁺(+) Krebs-HEPES Buffer (KHB) for 5 min, followed by perfusion in Ca²⁺(−) KHB. a Net FRET (nF) measurement, FRET measurement corrected for background, of ER Ca²⁺ using D1ER Ca²⁺ cameleon shows no significant difference in ER Ca²⁺ depletion in Bcl-x_L overexpressing cells compared to control (n = 3). b ER to cytosol Ca²⁺ influx was measured using Fluo-4 AM Ca²⁺ dye (n = 5). A smaller increase in cytosolic Ca²⁺ influx was observed in Bclx_L overexpressing cells. c ER to mitochondria Ca²⁺ influx was measured using X-Rhod-5f AM Ca²⁺ dye (n = 5). A larger increase in mitochondrial Ca²⁺ influx was observed in Bcl-x_L overexpressing cells relative to control cells. d CHO-Bcl-x_L cells were transfected with non-active control siRNA (CNT siRNA) or siRNA targeting IP₃R3. Ca²⁺ signaling was compared in co-cultured null (non-siRNA treated) cells and siRNA treated (CNT or IP₃R3) cells. e Knocking down IP₃R3 does not affect ER to cytosolic Ca²⁺ transfer, but f significantly diminishes ER to mitochondrial Ca²⁺ transfer in CHO-Bcl-x_L cells (n = 3). All data represented as mean ± SEM. *P < 0.05 versus control. AUC, or area under the curve, is provided as a comparison of calcium transients between cell lines, as it takes into account differences in the fluorescence baseline fluorescence between cells

Fig. 5

Bcl-x expression facilitates Ca²⁺− dependent TCA cycle activity. a Metabolic consumption rates of key metabolites contained on PM-M1 Phenotype MicroArray plates were evaluated in CHO-K1 and CHO-Bcl-x_L cells (n = 3). Wells containing α-D-glucose serve as a positive control, while tricarballylic acid, an inhibitor of aconitase, functions as a negative control. All rates normalized to the slope of glucose consumption. b Lactate accumulation was reduced in Bcl-x_L overexpressing cells compared to controls (n = 3). c Relative NAD/NADH ratios were reduced in Bcl-x_L overexpressing in comparison to control cells. Incubating cells with Tg for 15 or 30 min causes a further reduction in NAD/NADH (n = 3). d Higher NAD/NADH (relative ratios) are observed when IP₃R3 is knocked down using siRNA compared to null cells, while control (CNT) NADH ratios/NADH ratios (n = 3). (Di) Electron transport chain (ETC.) oxidase activity is higher in Bcl-x_L overexpressing cells compared to controls in non-treated cells (n = 6). Pre-incubation with (Dii) Tg, (Diii) BAPTA-AM, a Ca²⁺ chelater, (Div) or a combination of both also alters NADH oxidation (n = 3). ROC, rate of change. All data is represented as mean ± SEM. *P < 0.05 versus control at identical time point. **P < 0.05 versus non-treated, or initial time point, control of the same cell line

Fig. 6

Bcl-x_L expression enhances TCA cycle activity during ER stress conditions. a Bcl- x_L protein levels are not significantly altered in control and Bcl-x_L overexpressing cells after 1 and 3 h treatment with Tg (n = 3). b Activation of ER stress, as indicated by the upregulation of phosphorylated PERK (pPERK), is detected in both endogenous and Bcl-x_L overexpressing cell lines, but is most notable in CHO-K1 cells. c Substantial increases in NADH production, as indicated by reduced NAD/NADH (relative ratios), are observed following 1 or 3 h Tg insult, particularly in Bcl-x_Loverexpressing cells. (Di) Electron transport chain (ETC.) oxidase activity in non-treated CHO-K1 and CHO-Bcl-x_L cells. (Dii) In comparison, ETC.oxidase activity is significantly enhanced when cells are incubated with Tg for 3 h. (Diii-iv) This effect is hindered by pre-treating cells with BAPTAAM. Overall, ETC. oxidase activity remains higher in Bcl-x_L expressing cells. ROC, rate of change. 500 nM Tg treatment was used for all experiments. All data represented as mean ± SEM for three independent experiments. *P < 0.05 versus control at identical time point. **P < 0.05 versus non-treated control of the same cell line

Fig. 7

Three distinct patterns of Bcl-x_L in relation to mitochondria. a Immunostaining of CHO-Bcl-x_L (red) transiently transfected with GFP-mito (green). Areas in which Bcl-x_L and mitochondria co-localize are observed in yellow. Scale bar = 10 μm. b Bcl-x_L was categories into three distribution patterns based on its relation to mitochondria—Type 1: clustered residing on mitochondria (≥50 % Bcl-x_L colocalization), Type 2: clustered adjacent to mitochondria (<50 % Bcl-xL colocalization), or Type 3: evenly distributed on mitochondria. c Mitochondria with exclusively Type 1 Bcl-x_L were the most prominent, followed by mitochondria exhibiting Bcl-x_L exclusively in a Type 2 pattern. Only a small percentage of mitochondria were observed exclusively with Type 3 Bcl-x_L. Data represented as mean ± SEM (n = 300). Scale bar = 2 μm. d Immunostaining and co-localization of Type 1 and Type 2 Bcl-x_L (red) with GFP-mito (green) and MAM marker proteins (blue)—IP₃R3 and Mfn2 (a MAM-mitochondria tethering protein). Data represented as mean ± SEM (n = 6). Scale bar = 2 μm. e Schematic representation of mitochondria with Type 1 and Type 2 Bcl-x_L and its association with the MAM are depicted

Fig. 8

Bcl-x_L translocates to MAM following thapsigargin treatment. a Tg treatment (3 h) resulted in an increase in the number of Type 2 Bcl-x_L, and a corresponding decrease in Type 1 Bcl-x_L in immunostained CHOBcl-x_L cells (n = 300). b Decreased Bcl-x_L co-localization with mitochondria was observed during live-cell imaging of CHO-YFP-Bclx_L cells transiently transfected with DsRed-mito and incubated with Tg over a 3 h time-span (n = 3). Bcl-x_L is depicted in green and mitochondria is depicted in red, while overlapping is depicted in yellow. Bcl-x_L co-localization with mitochondria decreases after Tg treatment, and more distinct green and red fluorescence is observed. c Cell fractionation studies showed decreases in overexpressed Bcl-x_L at mitochondria-enriched fractions, and corresponding increases at MAM-enriched fractions after prolonged Tg treatment (1 and 3 h). Additionally, Tg treatment increased interaction of Bcl-x_L with IP₃R3 over a 3 h time-span as indicated by d immunoprecipitation and e nickel pull-down. 500 nM Tg treatment was used for all experiments. Data is represented as mean ± SEM (n = 3). *P < 0.05 versus non-treated control