ABSTRACT

Mitochondria contribute to cell signaling by controlling store-operated Ca2+ entry (SOCE). SOCE is activated by Ca2+ release from the endoplasmic reticulum (ER), whereupon stromal interacting molecule 1 (STIM1) forms oligomers, redistributes to ER–plasma-membrane junctions and opens plasma membrane Ca2+ channels. The mechanisms by which mitochondria interfere with the complex process of SOCE are insufficiently clarified. In this study, we used an shRNA approach to investigate the direct involvement of mitochondrial Ca2+ buffering in SOCE. We demonstrate that knockdown of either of two proteins that are essential for mitochondrial Ca2+ uptake, the mitochondrial calcium uniporter (MCU) or uncoupling protein 2 (UCP2), results in decelerated STIM1 oligomerization and impaired SOCE following cell stimulation with an inositol-1,4,5-trisphosphate (IP3)-generating agonist. Upon artificially augmented cytosolic Ca2+ buffering or ER Ca2+ depletion by sarcoplasmic or endoplasmic reticulum Ca2+-ATPase (SERCA) inhibitors, STIM1 oligomerization did not rely on intact mitochondrial Ca2+ uptake. However, MCU-dependent mitochondrial sequestration of Ca2+ entering through the SOCE pathway was essential to prevent slow deactivation of SOCE. Our findings show a stimulus-specific contribution of mitochondrial Ca2+ uptake to the SOCE machinery, likely through a role in shaping cytosolic Ca2+ micro-domains.

INTRODUCTION

Store-operated Ca2+ entry (SOCE) is a common form of Ca2+ influx that is linked to important physiological functions of different cell types (Parekh and Putney, 2005). One characteristic feature of SOCE is its activation upon depletion of the endoplasmic reticulum (ER) Ca2+ store (Putney, 1986). With the identification of the key molecular constituents of SOCE – the stromal interacting molecule 1 (STIM1) (Zhang et al., 2005; Liou et al., 2005) and the plasma membrane Ca2+-pore-forming Orai1 (Vig et al., 2006; Zhang et al., 2006) – the clarification of the elusive molecular mechanism of SOCE has become possible (Soboloff et al., 2012). When the concentration of Ca2+ in the ER ([Ca2+]ER) is reduced, Ca2+ dissociates from the luminal EF-hand domain of the ER-membrane-spanning STIM1, initiating its oligomerization (Liou et al., 2007). Subsequently, STIM1 oligomers translocate to subplasmalemmal ER domains, where they form higher-order aggregates, which appear as the so-called ‘STIM1 punctae’ (Park et al., 2009). In this form, STIM1 couples with and activates Orai1 (Park et al., 2009) and other store-operated channels (Cheng et al., 2013), resulting in Ca2+ entry. Apart from this function, STIM1 has been shown to regulate the activity of ion pumps (Manjarrés et al., 2010; Ritchie et al., 2012), enzymes (Lefkimmiatis et al., 2009) and cell adhesion proteins (Shinde et al., 2013), pointing to a fundamental role of Ca2+-dependent STIM1 oligomerization in cell signaling.

Long before the identification of STIM and Orai proteins and their role in SOCE, mitochondria were shown to contribute to the regulation of SOCE in immune cells (Hoth et al., 1997). Although the exact mechanisms by which mitochondria facilitate SOCE are still unclear, it is assumed that the ability of mitochondria to buffer Ca2+ counteracts the Ca2+-dependent inactivation of this Ca2+-sensitive Ca2+ entry pathway (Demaurex et al., 2009; Parekh, 2008b). In addition, mitochondrial Ca2+ uptake upon cell stimulation has been suggested to cause a more pronounced depletion of Ca2+ from the ER that consequently facilitates SOCE (Demaurex et al., 2009). Because proteins that mediate mitochondrial Ca2+ uptake have been identified only recently, the contribution of mitochondrial Ca2+ uptake to SOCE has only been investigated indirectly until now. In many studies, mitochondrial Ca2+ uptake was diminished by chemical uncouplers, such as carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP) or antimycin A, an inhibitor of complex III of the respiratory chain, which resulted in a significant reduction in SOCE (Naghdi et al., 2010; Hoth et al., 1997; Glitsch et al., 2002). In line with these findings, energized mitochondria were shown to result in increased SOCE (Gilabert and Parekh, 2000). Although these findings point to the importance of mitochondria in SOCE regulation, the actual role of mitochondrial Ca2+ uptake in the process of SOCE activation and maintenance under physiological conditions remains elusive.

In this study, mitochondrial Ca2+ uptake in HeLa cells was strongly diminished by a stable knockdown of either MCU (Baughman et al., 2011; De Stefani et al., 2011) (MCUKD) or UCP2 (Trenker et al., 2007; Trenker et al., 2008) (UCP2KD), two proteins of the inner mitochondrial membrane, which have been shown to be involved in mitochondrial Ca2+ uptake (Waldeck-Weiermair et al., 2013). MCU, the proposed core component of a ubiquitous mitochondrial Ca2+ channel, contributes to mitochondrial Ca2+ uptake regardless of the source and mode of Ca2+ mobilization (Baughman et al., 2011; De Stefani et al., 2011), whereas UCP2 has been shown to mediate principally the uptake of Ca2+ from areas of high Ca2+ concentration – Ca2+ micro-domains, which are formed upon ER Ca2+ release (Waldeck-Weiermair et al., 2013). Accordingly, the individual knockdown of these proteins enabled us to distinguish whether mitochondrial uptake of intracellularly released Ca2+ or of Ca2+ entering the cell is determinant for SOCE. Our data demonstrate that UCP2- and MCU-dependent mitochondrial uptake of inositol-1,4,5-trisphosphate (IP3)-dependent intracellularly released Ca2+ represents an essential step in the activation of STIM1 and, hence, SOCE. Correlations between the dynamics of ER Ca2+ depletion and STIM1 oligomerization upon different modes of Ca2+ mobilization suggest that mitochondrial Ca2+ sequestration predominately shapes IP3-mediated cytosolic Ca2+ micro-domains, which facilitates STIM1 oligomerization under physiological conditions of cell stimulation. In addition, MCU-dependent mitochondrial buffering of entering Ca2+ is crucial for the maintenance of SERCA-inhibition-induced SOCE signals. In summary, we highlight a special and tight regulation of STIM1 activation and SOCE maintenance by mitochondrial Ca2+ uptake upon physiological and non-physiological stimuli.

RESULTS

Stable knock-down of either MCU or UCP2 inhibits mitochondrial Ca2+ uptake and impairs STIM1 oligomerization upon IP3-mediated Ca2+ release

In line with recent reports (Baughman et al., 2011; De Stefani et al., 2011; Waldeck-Weiermair et al., 2013) a stable knockdown of MCU in HeLa cells (supplementary material Fig. S1A) strongly reduced mitochondrial sequestration of intracellularly released (supplementary material Fig. S1B) and entering Ca2+ (supplementary material Fig. S1C). By contrast, HeLa cells stably depleted of UCP2 (supplementary material Fig. S1A) showed an impaired mitochondrial Ca2+ signal upon intracellular Ca2+ mobilization, whereas mitochondrial uptake of entering Ca2+ was not affected (supplementary material Fig. S1B,C). These results confirm our recent findings, which have revealed the exclusive involvement of UCP2 and UCP3 in mitochondrial Ca2+ sequestration upon IP3-mediated intracellular Ca2+ release (Waldeck-Weiermair et al., 2011; Trenker et al., 2007; ,Waldeck-Weiermair et al., 2010b). As mitochondria play a central role in cellular energy homeostasis (Graier et al., 2007; Duchen, 2004), we tested whether stable silencing of either MCU or UCP2 altered the basal metabolic status of HeLa cells. For this purpose, we measured the mitochondrial membrane potential (supplementary material Fig. S1D), oxygen consumption rate (supplementary material Fig. S1E) and cellular ATP content (supplementary material Fig. S1F) in stable knockdown cells. All of these variables remained unaffected by the knockdown of MCU or UCP2. Moreover, the morphology of mitochondria (supplementary material Fig. S1G,H), ER structure (supplementary material Fig. S1G) and the contact sites between these organelles (supplementary material Fig. S1I) were not altered by the stable knockdown of MCU or UCP2. In summary, these data demonstrate that the knockdown of MCU or UCP2 results in specific effects on distinct modes of mitochondrial Ca2+ uptake. Therefore, MCUKD and UCP2KD cells can serve as suitable models to specifically investigate the role of mitochondrial Ca2+ sequestration in SOCE regulation. In addition, due to the selective engagement of UCP2 in mitochondrial Ca2+ sequestration depending on the source of Ca2+ (supplementary material Fig. S1B,C) (,Waldeck-Weiermair et al., 2010b</emph>; Waldeck-Weiermair et al., 2011), these cell models allowed us to separately examine whether mitochondrial buffering of intracellularly released or entering Ca2+ contributes to SOCE activation.

The initial step of SOCE activation is accomplished by the formation of STIM1 oligomers upon reduction of [Ca2+]ER (Luik et al., 2008). By measuring Förster resonance energy transfer (FRET) between STIM1 proteins fused to either cyan or yellow fluorescent proteins (CFP or YFP, respectively), the dynamic process of STIM1 oligomerization can be visualized and quantified (Liou et al., 2007; Deak et al., 2013). We used this approach to investigate whether impaired mitochondrial Ca2+ uptake induced by the stable knockdown of MCU or UCP2 affects STIM1 oligomerization upon ER Ca2+ mobilization under physiological conditions (Fig. 1). Cell treatment with histamine in the presence of extracellular Ca2+ resulted in a moderate, but significant increase in the FRET signal between CFP–STIM1 and YFP–STIM1 in control cells (Fig. 1A). However, this increase was almost absent in MCUKD and UCP2KD cells (Fig. 1A). The measurement of [Ca2+]ER in response to histamine using D1ER, an ER-targeted genetically encoded Ca2+ probe (Palmer et al., 2004), revealed that the reduction in the global ER Ca2+ concentration was slightly slower in the knockdown cells compared with that of control cells (Fig. 1B). The subsequent addition of the SERCA inhibitor 2,5-di-t-butyl-1,4-benzohydroquinone (BHQ), which causes a strong depletion of the ER Ca2+ store, considerably augmented STIM1 oligomerization in all three cell types (Fig. 1C,D). These findings indicate that, upon IP3-mediated depletion of ER Ca2+, the subsequent oligomerization of STIM1 proteins is facilitated by MCU- and UCP2-dependent mitochondrial Ca2+ uptake. Notably, after ∼4 minutes of histamine treatment, the degree of ER Ca2+ depletion in MCUKD and UCP2KD cells reached that of control cells (Fig. 1B), whereas the STIM1 FRET signals in the knock-down cells remained strongly reduced (Fig. 1A). This suggests that mitochondrial Ca2+ sequestration does not just contribute to STIM1 activation by shaping the ER Ca2+ response.

Fig. 1.

IP3-mediated STIM1 oligomerization is impaired in MCUKD and UCP2KD cells. (A,B) Left-most panels, dynamic changes in the ratio of F535/F480 [indicating STIM1 oligomerization (A) or ER Ca2+-response (B)] over time, normalized to the basal ratio (R0). Middle-left panels; Curves were fitted with straight lines using linear regression to assess the maximal slope (s) of the increase (A) or decrease (B) of the ratio F535/F480/R0. Scale bar indicates 30 seconds (30 s). Middle-right panels; data show the mean±s.e.m. of the maximal slope normalized to control (set at 100%). Right-most panels; data show the mean±s.e.m. of the Δmax ratio F535/F480/R0 normalized to control. *P<0.05 (unpaired Student's t-test). (A) STIM1 oligomerization induced by 100 µM histamine in the presence of extracellular Ca2+ was visualized in control (black, n = 35), MCUKD (red, n = 37) and UCP2KD (blue, n = 33) cells by measuring intermolecular FRET between CFP–STIM1 and YFP–STIM. (B) Dynamics of ER Ca2+ depletion (ER Ca2+-response) were monitored in control, MCUKD and UCPKD (n = 33, 33 and 35, respectively) cells expressing D1ER in the same experimental protocol described in A. (C) Respective pseudocolor CFP/YFP FRET images of control, MCUKD and UCP2KD cells under basal conditions (i), after 100 µM histamine (ii) and after addition of 15 µM BHQ (iii). Increased FRET signals appear as red pixels. Scale bar: 10 µm. (D) Time-lapse FRET measurement of STIM1 oligomerization in the control, MCUKD and UCP2KD cells shown in B. The thin regions of the curves (0–6 minutes) are identical to those shown in the left-most panel of A, without y-axis magnification.

Fig. 1.

IP3-mediated STIM1 oligomerization is impaired in MCUKD and UCP2KD cells. (A,B) Left-most panels, dynamic changes in the ratio of F535/F480 [indicating STIM1 oligomerization (A) or ER Ca2+-response (B)] over time, normalized to the basal ratio (R0). Middle-left panels; Curves were fitted with straight lines using linear regression to assess the maximal slope (s) of the increase (A) or decrease (B) of the ratio F535/F480/R0. Scale bar indicates 30 seconds (30 s). Middle-right panels; data show the mean±s.e.m. of the maximal slope normalized to control (set at 100%). Right-most panels; data show the mean±s.e.m. of the Δmax ratio F535/F480/R0 normalized to control. *P<0.05 (unpaired Student's t-test). (A) STIM1 oligomerization induced by 100 µM histamine in the presence of extracellular Ca2+ was visualized in control (black, n = 35), MCUKD (red, n = 37) and UCP2KD (blue, n = 33) cells by measuring intermolecular FRET between CFP–STIM1 and YFP–STIM. (B) Dynamics of ER Ca2+ depletion (ER Ca2+-response) were monitored in control, MCUKD and UCPKD (n = 33, 33 and 35, respectively) cells expressing D1ER in the same experimental protocol described in A. (C) Respective pseudocolor CFP/YFP FRET images of control, MCUKD and UCP2KD cells under basal conditions (i), after 100 µM histamine (ii) and after addition of 15 µM BHQ (iii). Increased FRET signals appear as red pixels. Scale bar: 10 µm. (D) Time-lapse FRET measurement of STIM1 oligomerization in the control, MCUKD and UCP2KD cells shown in B. The thin regions of the curves (0–6 minutes) are identical to those shown in the left-most panel of A, without y-axis magnification.

The contribution of mitochondrial Ca2+ uptake to STIM1 oligomerization is determined by the mode of Ca2+ mobilization

To further characterize the role of mitochondrial Ca2+ buffering in IP3-mediated STIM1 activation and ER Ca2+ depletion, cells were stimulated with histamine in the absence of extracellular Ca2+. This led to an enhanced and rapid STIM1 oligomerization (Fig. 2A) due to a pronounced release of Ca2+ from the ER (Fig. 2B). Although the ER depleted at the same rate in all three cell lines under these conditions (Fig. 2B), STIM1 oligomerization was clearly delayed in MCUKD and UCP2KD cells compared with that of the control cells (Fig. 2A). However, neither STIM1 oligomerization (Fig. 2C) nor ER Ca2+ release (Fig. 2D) was affected by MCU or UCP2 knockdown when ER Ca2+ was mobilized by the SERCA inhibitor BHQ. These findings show a stimulus-specific contribution of mitochondrial Ca2+ buffering to STIM1 oligomerization dynamics. Moreover, ER Ca2+ responses were insensitive to MCU or UCP2 silencing under these conditions (Fig. 2B,D), confirming that mitochondrial Ca2+ uptake promotes STIM1 activation by a mechanism other than the control of ER Ca2+-responses only.

Fig. 2.

Stimulus-specific dependency of STIM1 oligomerization on mitochondrial Ca2+ uptake. Left-most panels, dynamic changes of the ratio F535/F480 [indicating STIM1 oligomerization (A,C) or ER Ca2+ response (B,D)] over time normalized to the basal ratio (R0). Middle-left panels, curves were fitted with straight lines using linear regression to assess the maximal slope (s) of the increase (A,C) or decrease (B,D) in the ratio F535/F480/R0. Scale bar indicates 30 seconds (30 s). Middle-right panels, data show the mean±s.e.m. of the maximal slope normalized to control (set at 100%). Right-most panels, data show the mean±s.e.m. of the Δmax ratio F535/F480/R0 normalized to control. (A) Control, MCUKD and UCP2KD cells (n = 65, 55 and 55, respectively) coexpressing YFP–STIM1 and CFP–STIM1 were stimulated with 100 µM histamine in the absence of extracellular Ca2+ (EGTA). *P<0.05 compared with control (unpaired Student's t-test). (B) The ER Ca2+ response of control, MCUKD and UCP2KD cells (n = 31, 36 and 27, respectively) expressing D1ER in the same experimental protocol as in A. (C) Control, MCUKD and UCP2KD cells (n = 22, 24 and 22, respectively) coexpressing YFP–STIM1 and CFP–STIM1 were stimulated with 15 µM BHQ in the absence of extracellular Ca2+ (EGTA). (D) The ER Ca2+ response of control, MCUKD and UCP2KD cells (n = 35, 45 and 38, respectively) expressing D1ER in the same experimental protocol as in C.

Fig. 2.

Stimulus-specific dependency of STIM1 oligomerization on mitochondrial Ca2+ uptake. Left-most panels, dynamic changes of the ratio F535/F480 [indicating STIM1 oligomerization (A,C) or ER Ca2+ response (B,D)] over time normalized to the basal ratio (R0). Middle-left panels, curves were fitted with straight lines using linear regression to assess the maximal slope (s) of the increase (A,C) or decrease (B,D) in the ratio F535/F480/R0. Scale bar indicates 30 seconds (30 s). Middle-right panels, data show the mean±s.e.m. of the maximal slope normalized to control (set at 100%). Right-most panels, data show the mean±s.e.m. of the Δmax ratio F535/F480/R0 normalized to control. (A) Control, MCUKD and UCP2KD cells (n = 65, 55 and 55, respectively) coexpressing YFP–STIM1 and CFP–STIM1 were stimulated with 100 µM histamine in the absence of extracellular Ca2+ (EGTA). *P<0.05 compared with control (unpaired Student's t-test). (B) The ER Ca2+ response of control, MCUKD and UCP2KD cells (n = 31, 36 and 27, respectively) expressing D1ER in the same experimental protocol as in A. (C) Control, MCUKD and UCP2KD cells (n = 22, 24 and 22, respectively) coexpressing YFP–STIM1 and CFP–STIM1 were stimulated with 15 µM BHQ in the absence of extracellular Ca2+ (EGTA). (D) The ER Ca2+ response of control, MCUKD and UCP2KD cells (n = 35, 45 and 38, respectively) expressing D1ER in the same experimental protocol as in C.

Increasing cytosolic Ca2+ buffering with BAPTA diminishes the contribution of mitochondrial Ca2+ uptake to IP3-induced STIM1 oligomerization

Global and local cytosolic Ca2+ elevations were shown to impair STIM1 activation independently of [Ca2+]ER (Malli et al., 2008; Shen et al., 2011b). Accordingly, upon IP3-mediated release of Ca2+ from the ER, mitochondrial buffering of cytosolic Ca2+ might counteract this negative feedback and, thus, facilitate STIM1 activation. In order to experimentally prove this hypothesis, we investigated histamine-induced STIM1 oligomerization and ER Ca2+ release in cells loaded with the chemical Ca2+ chelator 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid tetrakis(acetoxymethyl ester) (BAPTA-AM), which accumulates in the cytosol and prevents the formation of high-concentration cytosolic Ca2+ micro-domains (Parekh, 2008a). In cells loaded with BAPTA-AM, STIM1 oligomerization (Fig. 3A) and ER Ca2+ depletion (Fig. 3B) in response to histamine were rapid and occurred identically in MCUKD, UCP2KD and control cells. However, both STIM1 oligomerization and ER Ca2+ depletion were transient in BAPTA-AM-loaded cells during treatment with histamine (Fig. 3). The transient nature of both signals most likely reflects SERCA-dependent replenishment of ER Ca2+ from the BAPTA-chelated cytosolic Ca2+. Nevertheless, BAPTA-AM loading abolished the contribution of mitochondrial Ca2+ uptake to histamine-induced STIM1 oligomerization. This points to a possible role for mitochondria in shaping high-concentration cytosolic Ca2+ micro-domains that negatively regulate STIM1 activation upon IP3-mediated Ca2+ mobilization independently of the ER Ca2+ content. A relative independency of STIM1 activation from the speed of ER Ca2+ depletion becomes obvious by correlating the BHQ-induced initial maximal slope of STIM1 oligomerization with the ER Ca2+ depletion kinetics (Fig. 3C). Although the SERCA inhibitor evoked a slow ER Ca2+ depletion (Fig. 2D; Fig. 3C), the kinetics of STIM1 oligomerization were quite fast (Fig. 2C; Fig. 3C). This mismatch further points to a role of cytosolic Ca2+ in the regulation of STIM1 activation.

Fig. 3.

IP3-mediated STIM1 oligomerization does not require intact mitochondrial Ca2+ uptake in BAPTA-AM-loaded cells. (A,B) Left-most panels, dynamic changes in the ratio F535/F480 [indicating STIM1 oligomerization (A) or ER Ca2+ response (B)] over time normalized to the basal ratio (R0). Middle-left panels, curves were fitted with straight lines using linear regression to assess the maximal slope (s) of the increase (A) or decrease (B) of the ratio F535/F480/R0. Scale bar indicates 30 seconds (30 s). Middle-right panels, data show the mean±s.e.m. of the maximal slope normalized to control (set at 100%). Right-most panels, data show the mean±s.e.m. of the Δmax ratio F535/F480/R0 normalized to control. (A) Prior to experiments, control, MCUKD and UCP2KD cells (n = 16, 10 and 12, respectively) coexpressing YFP–STIM1 and CFP–STIM1 were exposed to 10 µM BAPTA-AM for 45 minutes and were stimulated with 100 µM histamine in a Ca2+-free solution (EGTA). (B) The ER Ca2+ response of control, MCUKD and UCP2KD cells (n = 16, 15 and 19, respectively) expressing D1ER in the same experimental protocol as in A. (C) Initial slopes of ER Ca2+ depletion (x-axis) were plotted against maximal slopes of STIM1 oligomerization (y-axis) of control cells that were treated with 100 µM histamine in the presence of 2 mM Ca2+ (grey triangle; n = 35 for STIM1 oligomerization, n = 33 for ER Ca2+ response), 100 µM histamine in EGTA (orange diamond; n = 65 for STIM1 oligomerization, n = 31 for ER Ca2+ depletion), 15 µM BHQ in EGTA (purple square; n = 22 for STIM1 oligomerization, n = 35 for ER Ca2+ response) and 100 µM histamine in BAPTA-AM-loaded cells (black circle; n = 16 for each). Data represent the mean±s.e.m., normalized to the maximal slopes in BAPTA-AM-loaded cells, which were defined as 100%.

Fig. 3.

IP3-mediated STIM1 oligomerization does not require intact mitochondrial Ca2+ uptake in BAPTA-AM-loaded cells. (A,B) Left-most panels, dynamic changes in the ratio F535/F480 [indicating STIM1 oligomerization (A) or ER Ca2+ response (B)] over time normalized to the basal ratio (R0). Middle-left panels, curves were fitted with straight lines using linear regression to assess the maximal slope (s) of the increase (A) or decrease (B) of the ratio F535/F480/R0. Scale bar indicates 30 seconds (30 s). Middle-right panels, data show the mean±s.e.m. of the maximal slope normalized to control (set at 100%). Right-most panels, data show the mean±s.e.m. of the Δmax ratio F535/F480/R0 normalized to control. (A) Prior to experiments, control, MCUKD and UCP2KD cells (n = 16, 10 and 12, respectively) coexpressing YFP–STIM1 and CFP–STIM1 were exposed to 10 µM BAPTA-AM for 45 minutes and were stimulated with 100 µM histamine in a Ca2+-free solution (EGTA). (B) The ER Ca2+ response of control, MCUKD and UCP2KD cells (n = 16, 15 and 19, respectively) expressing D1ER in the same experimental protocol as in A. (C) Initial slopes of ER Ca2+ depletion (x-axis) were plotted against maximal slopes of STIM1 oligomerization (y-axis) of control cells that were treated with 100 µM histamine in the presence of 2 mM Ca2+ (grey triangle; n = 35 for STIM1 oligomerization, n = 33 for ER Ca2+ response), 100 µM histamine in EGTA (orange diamond; n = 65 for STIM1 oligomerization, n = 31 for ER Ca2+ depletion), 15 µM BHQ in EGTA (purple square; n = 22 for STIM1 oligomerization, n = 35 for ER Ca2+ response) and 100 µM histamine in BAPTA-AM-loaded cells (black circle; n = 16 for each). Data represent the mean±s.e.m., normalized to the maximal slopes in BAPTA-AM-loaded cells, which were defined as 100%.

SOCE activity upon stimulation with an IP3-generating agonist is attenuated in cells depleted of MCU or UCP2

The impact of decelerated IP3-mediated STIM1 activation on Ca2+ entry was examined using the fura-2 technique in MCUKD and UCPKD cells. In wild-type cells, treatment with histamine in the presence of extracellular Ca2+ generated sustained cytosolic Ca2+ signals (Fig. 4A), whereas, in the absence of extracellular Ca2+, histamine-evoked Ca2+ signals became transient (Fig. 4B; supplementary material Fig. S2A); thus indicating that Ca2+ entry is required to maintain elevated cytosolic Ca2+ levels during stimulation. By contrast, in MCU- or UCP2-depleted cells, the SOCE-dependent second phase of cytosolic Ca2+ elevation was strongly reduced (Fig. 4A).

Fig. 4.

IP3-triggered Ca2+ entry is attenuated in MCUKD and UCP2KD cells. (A) Curves indicate single-cell cytosolic Ca2+ signals (thin lines) and their respective mean±s.e.m. (thick lines) upon addition of 100 µM histamine to fura-2/am-loaded control cells (left panel, n = 51), MCUKD cells (middle-left panel, n = 57) and UCP2KD cells (middle-right panel, n = 60). Right-most panel, data show the mean±s.e.m. of the Ca2+ responses of control (black), MCUKD (red) and UCPKD (blue) cells. (B) The area under the curve (AUC) corresponding to the respective cytosolic Ca2+ responses shown in A and in supplementary material Fig. S2 was calculated for each sample and normalized to the control cytosolic Ca2+ responses in the presence of 2 mM Ca2+ in the extracellular medium (set at 100%). The data show the mean±s.e.m. The number of control, MCUKD and UCP2KD cells for each condition are as follows; Ca2+ (n = 51, 57 and 60, respectively), EGTA (n = 76, 76 and 65, respectively), Orai(E106Q) (n = 43,39 and 36, respectively), FCCP/Oligomycin (n = 54, 40 and 56, respectively). (C) Perfusion with 100 µM Mn2+ quenched cytosolic fura-2 fluorescence in control, MCUKD and UCP2KD cells (n = 43, 56 and 52, respectively) upon treatment with 100 µM histamine. Data show the mean±s.e.m. of normalized fluorescence intensities measured at 360 nm excitation (F360/F0); *P<0.05 versus control (unpaired Student's t-test).

Fig. 4.

IP3-triggered Ca2+ entry is attenuated in MCUKD and UCP2KD cells. (A) Curves indicate single-cell cytosolic Ca2+ signals (thin lines) and their respective mean±s.e.m. (thick lines) upon addition of 100 µM histamine to fura-2/am-loaded control cells (left panel, n = 51), MCUKD cells (middle-left panel, n = 57) and UCP2KD cells (middle-right panel, n = 60). Right-most panel, data show the mean±s.e.m. of the Ca2+ responses of control (black), MCUKD (red) and UCPKD (blue) cells. (B) The area under the curve (AUC) corresponding to the respective cytosolic Ca2+ responses shown in A and in supplementary material Fig. S2 was calculated for each sample and normalized to the control cytosolic Ca2+ responses in the presence of 2 mM Ca2+ in the extracellular medium (set at 100%). The data show the mean±s.e.m. The number of control, MCUKD and UCP2KD cells for each condition are as follows; Ca2+ (n = 51, 57 and 60, respectively), EGTA (n = 76, 76 and 65, respectively), Orai(E106Q) (n = 43,39 and 36, respectively), FCCP/Oligomycin (n = 54, 40 and 56, respectively). (C) Perfusion with 100 µM Mn2+ quenched cytosolic fura-2 fluorescence in control, MCUKD and UCP2KD cells (n = 43, 56 and 52, respectively) upon treatment with 100 µM histamine. Data show the mean±s.e.m. of normalized fluorescence intensities measured at 360 nm excitation (F360/F0); *P<0.05 versus control (unpaired Student's t-test).

To verify whether the Ca2+ influx induced by IP3-mediated ER depletion is indeed accomplished by the STIM1–Orai1-driven SOCE pathway, HeLa cells were transfected with Orai1(E106Q), a dominant-negative form of the wild-type Orai1 protein (Vig et al., 2006). Ca2+ responses upon stimulation of control cells expressing Orai1(E106Q) showed a strongly reduced plateau phase in the presence of extracellular Ca2+ (Fig. 4B; supplementary material Fig. S2B). Notably, the already transient elevation of cytosolic Ca2+ in response to histamine in the MCUKD and UCP2KD cells remained unaffected by an additional expression of the dominant-negative Orai (Fig. 4B; supplementary material Fig. S2B). This indicates that STIM1–Orai1-dependent SOCE is not activated in cells lacking the mitochondrial Ca2+ sequestration of intracellularly released Ca2+. Similar results were obtained in all cell models (i.e. wild-type, MCUKD and UCP2KD cells) upon incubation with FCCP and oligomycin A prior to histamine stimulation (Fig. 4B; supplementary material Fig. S2C). Interestingly, the sustained phase of the cytosolic Ca2+ signal in MCU- and UCP2-silenced cells in the absence of FCCP and oligomycin A was reduced to the same level as in control cells pre-exposed to the drugs (Fig. 4B), indicating that mitochondrial Ca2+ uptake is the molecular mechanism by which mitochondria contribute to SOCE under these conditions. To additionally confirm the reduced SOCE in MCUKD and UCP2KD cells, the rate of Mn2+ influx was measured (Fig. 4C). Cytosolic fura-2 fluorescence in response to histamine was quenched at a significantly slower rate by Mn2+ in MCUKD and UCP2KD compared with controls cells (Fig. 4C), indicating that SOCE was indeed reduced in cells lacking mitochondrial Ca2+ uptake.

Re-expression of MCU restores both IP3-mediated mitochondrial Ca2+ uptake and SOCE

To correlate histamine-evoked cytosolic Ca2+ signals with changes in [Ca2+]mito on the single-cell level, Ca2+ responses in both compartments were measured simultaneously using a red-shifted mitochondrially targeted cameleon (a protein that allows the visualization of Ca2+ levels in live cells) in combination with fura-2 (Waldeck-Weiermair et al., 2012). This approach revealed a positive correlation between mitochondrial Ca2+ uptake and the plateau phase of [Ca2+]cyto in response to histamine (Fig. 5A,B). Notably, due to oscillating cytosolic Ca2+ responses, the correlation was rather weak. However, a transient re-expression of MCU in MCUKD cells efficiently rescued the mRNA levels of MCU (Fig. 5C) and restored both mitochondrial Ca2+ uptake and the plateau phase of the cytosolic Ca2+ signal in the same individual cells (Fig. 5A,B). Thus, these results confirm that mitochondrial Ca2+ uptake contributes to SOCE upon cell stimulation with histamine.

Fig. 5.

Re-expression of MCU in MCUKD cells restores both IP3-mediated mitochondrial Ca2+ uptake and SOCE. (A) Representative cytosolic (thick lines) and mitochondrial (thin lines) Ca2+ signals from the same individual 4mtD1GO-expressing and fura-2/am-loaded control cells (left-most panel, black), MCUKD cells (middle panel, red) and MCU-transfected MCUKD cells (right panel, green). (B) Scatter plot correlating cytosolic and mitochondrial Ca2+ responses (black dotted line) shown in A. The values of the area under the curve (AUC) of cytosolic Ca2+ signals (arbitrary units; y-axis) were plotted against peak mitochondrial Ca2+ uptake (Δmax Ratio F560/F510/R0; x-axis) derived from the same individual control (open black symbols, n = 17), MCUKD (open red symbols, n = 17) and MCUKD + MCU (open green symbols, n = 22) cells. Filled symbols represent the mean±s.e.m; *P<0.05 versus control; #P<0.05 versus control (unpaired Student's t-test). (C) Relative mRNA levels (arbitrary units, AU) of MCU in control cells (n = 9) and MCUKD cells (n = 9), as shown in supplementary material Fig. S1A, as well as MCUKD cells expressing MCU (n = 9). Data show the mean±s.e.m; *P<0.05 versus control.

Fig. 5.

Re-expression of MCU in MCUKD cells restores both IP3-mediated mitochondrial Ca2+ uptake and SOCE. (A) Representative cytosolic (thick lines) and mitochondrial (thin lines) Ca2+ signals from the same individual 4mtD1GO-expressing and fura-2/am-loaded control cells (left-most panel, black), MCUKD cells (middle panel, red) and MCU-transfected MCUKD cells (right panel, green). (B) Scatter plot correlating cytosolic and mitochondrial Ca2+ responses (black dotted line) shown in A. The values of the area under the curve (AUC) of cytosolic Ca2+ signals (arbitrary units; y-axis) were plotted against peak mitochondrial Ca2+ uptake (Δmax Ratio F560/F510/R0; x-axis) derived from the same individual control (open black symbols, n = 17), MCUKD (open red symbols, n = 17) and MCUKD + MCU (open green symbols, n = 22) cells. Filled symbols represent the mean±s.e.m; *P<0.05 versus control; #P<0.05 versus control (unpaired Student's t-test). (C) Relative mRNA levels (arbitrary units, AU) of MCU in control cells (n = 9) and MCUKD cells (n = 9), as shown in supplementary material Fig. S1A, as well as MCUKD cells expressing MCU (n = 9). Data show the mean±s.e.m; *P<0.05 versus control.

MCU-dependent mitochondrial buffering of entering Ca2+ is essential for the maintenance of thapsigargin-induced SOCE

A classical protocol to image SOCE is based on a complete ER Ca2+ depletion with a SERCA inhibitor in the absence of extracellular Ca2+ and the subsequent addition of Ca2+, which leads to a pronounced, SOCE-mediated elevation of [Ca2+]cyto within these pre-stimulated cells (Parekh and Putney, 2005). The initial phase of this cytosolic SOCE signal was affected in neither MCUKD nor UCP2KD cells (Fig. 6A), which is in line with the unaltered STIM1 oligomerization triggered by SERCA inhibition (Fig. 2C). By contrast, incubating the cells with FCCP and oligomycin A prior to the addition of Ca2+ in the same experimental protocol suppressed cytosolic Ca2+ elevations to the same level in all three cell types (Fig. 6B), indicating that SOCE triggered by SERCA inhibition requires intact mitochondria, but not mitochondrial Ca2+ buffering, for activation. However, exclusively in MCUKD cells, which show reduced mitochondrial accumulation of entering Ca2+ (supplementary material Fig. S1C), the thapsigargin-induced SOCE signal was almost completely terminated within several minutes (Fig. 6A,B). This finding suggests that intact MCU-dependent mitochondrial buffering of entering Ca2+ is required for the maintenance of SOCE, even if this Ca2+ entry pathway is maximally activated by irreversible SERCA inhibition.

Fig. 6.

The maintenance of thapsigargin-induced SOCE relies on MCU-dependent mitochondrial Ca2+ buffering. (A) Curves represent cytosolic Ca2+ responses of control, MCUKD and UCP2KD cells upon addition of 2 mM Ca2+. Cells were pre-stimulated with 1 µM thapsigargin to induce depletion of the ER Ca2+ store. (B) Left panel, curves were fitted with straight lines using linear regression to assess the maximal slope of the increase in [Ca2+]cyto (SOCE onset) upon Ca2+ addition. As a negative control, cells were exposed to FCCP and oligomycin (2 µM each) prior to the addition of Ca2+. The number of control, MCUKD and UCP2KD cells for each condition are: untreated, n = 104, 93 and 91, respectively; +FCCP/oligomycin, n = 90, 77 and 99, respectively. Right panel, Values of the area under the curve (AUC) of SOCE-driven Ca2+ signals were calculated to quantify inactivation (SOCE plateau). Data show the mean±s.e.m. normalized to control (set at 100%); *P<0.05 versus control; #P<0.05 versus MCUKD; §P<0.05 versus UCP2KD.

Fig. 6.

The maintenance of thapsigargin-induced SOCE relies on MCU-dependent mitochondrial Ca2+ buffering. (A) Curves represent cytosolic Ca2+ responses of control, MCUKD and UCP2KD cells upon addition of 2 mM Ca2+. Cells were pre-stimulated with 1 µM thapsigargin to induce depletion of the ER Ca2+ store. (B) Left panel, curves were fitted with straight lines using linear regression to assess the maximal slope of the increase in [Ca2+]cyto (SOCE onset) upon Ca2+ addition. As a negative control, cells were exposed to FCCP and oligomycin (2 µM each) prior to the addition of Ca2+. The number of control, MCUKD and UCP2KD cells for each condition are: untreated, n = 104, 93 and 91, respectively; +FCCP/oligomycin, n = 90, 77 and 99, respectively. Right panel, Values of the area under the curve (AUC) of SOCE-driven Ca2+ signals were calculated to quantify inactivation (SOCE plateau). Data show the mean±s.e.m. normalized to control (set at 100%); *P<0.05 versus control; #P<0.05 versus MCUKD; §P<0.05 versus UCP2KD.

DISCUSSION

In this study, the significance of mitochondrial Ca2+ uptake in SOCE activation and maintenance in HeLa cells was assessed. A shRNA-mediated stable knockdown of the confirmed key components of mitochondrial Ca2+ uptake allowed us to separately investigate the contribution of mitochondrial Ca2+ buffering from that of mitochondrial depolarization to SOCE. We also took advantage of recent findings showing that UCP2 is involved in mitochondrial Ca2+ sequestration upon intracellular Ca2+ release, but not in the buffering of entering Ca2+ (Waldeck-Weiermair et al., 2010b; Waldeck-Weiermair et al., 2011), whereas MCU is generally involved in mitochondrial Ca2+ uptake regardless the source of Ca2+ (Waldeck-Weiermair et al., 2013; De Stefani et al., 2011). Accordingly, the use of MCUKD and UCP2KD cell models enabled us to determine whether SOCE activation and maintenance is facilitated by local mitochondrial buffering of Ca2+ entering by SOCE or mitochondrial sequestration of Ca2+ released from ER.

In accordance with recent reports (Waldeck-Weiermair et al., 2011; Waldeck-Weiermair et al., 2013), mitochondrial Ca2+ uptake upon cell treatment with an IP3-generating agonist was strongly reduced in HeLa cells with a stable knockdown of UCP2 (supplementary material Fig. S1B), whereas the lack of this protein did not affect mitochondrial uptake of entering Ca2+ (supplementary material Fig. S1C). Although it has been hypothesized that UCP2 and UCP3 are preferentially located within sites of mitochondrial Ca2+ uptake that face areas of ER Ca2+ release (Waldeck-Weiermair et al., 2010a), it is still unknown why the knockdown of these proteins exclusively diminishes mitochondrial uptake of Ca2+ that is mobilized from the ER. Interestingly, UCP2 depletion reduced mitochondrial Ca2+ uptake upon treatment of the cells with histamine to the same extent as a stable knockdown of MCU (supplementary material Fig. S1B), the proposed pore-forming subunit of a mitochondrial Ca2+ channel (De Stefani et al., 2011). These findings show that actually both UCP2 and MCU are necessary to accomplish IP3-mediated mitochondrial Ca2+ uptake in HeLa cells, thus pointing to a possible functional interrelation of these proteins, which requires further investigation. Notably, our recent study identified three biophysically distinct mitochondrial Ca2+ channel currents, of which only the most frequently occurring one depends on the presence of MCU (Bondarenko et al., 2013). Our data, however, do not only confirm that both UCP2 and MCU are essential for mitochondrial Ca2+ uptake, but demonstrate that the two distinct proteins influence SOCE regulation in a similar manner, due to their common involvement in mitochondrial Ca2+ uptake.

Interestingly, a knockdown of UCP2, which, in contrast to MCU, is not essential for mitochondrial sequestration of entering Ca2+ (Waldeck-Weiermair et al., 2013), mimicked the inhibitory effect of MCU knockdown on IP3-triggered STIM1 activation (Fig. 1) and, in turn, SOCE (Fig. 4). These findings indicate that mitochondrial Ca2+ buffering at the mouth of IP3 receptors is the major mechanism by which mitochondria contribute to SOCE activation under conditions of physiological stimulation. Accordingly, our findings partially confirm a model of the mechanism by which mitochondria regulate SOCE in immune cells, introduced by Anant Parekh before STIM1 and Orai1 were described (Gilabert et al., 2001). According to this study, energized mitochondria and, hence, increased mitochondrial Ca2+ buffering, augments Ca2+-release-activated Ca2+ current (ICRAC), the electrophysiological correlate of SOCE (Hoth and Penner, 1992). They concluded that local mitochondrial Ca2+ buffering at sites of ER Ca2+ release facilitates store depletion and, hence, significantly supports SOCE activation by IP3 (Gilabert et al., 2001). Several (whole-cell) patch-clamp studies in immune cells have indicated that Ca2+-chelating compounds, such as BAPTA or EGTA, need to be added to the pipette solution to evoke a pronounced ICRAC (Zweifach and Lewis, 1995; Glitsch et al., 2002). However, under such conditions of artificially increased intracellular Ca2+-buffering capacity, the mitochondrial contribution is not relevant for SOCE/ICRAC (Gilabert et al., 2001). In agreement with these findings, our data confirm that STIM1 oligomerization, which represents a key step in SOCE activation, becomes independent of MCU- and UCP2-mediated mitochondrial Ca2+ buffering upon histamine stimulation in cells loaded with BAPTA-AM (Fig. 3). When Ca2+ was mobilized by SERCA inhibition, which is known to efficiently deplete the store through Ca2+ leakage, STIM1 oligomerization remained largely unaffected by the knockdown of MCU or UCP2 (Fig. 2C), indicating that intact mitochondrial Ca2+ uptake is not required for SOCE activation under these conditions. SERCA inhibition does not promote the formation of high-concentration Ca2+ micro-domains (Waldeck-Weiermair et al., 2013; Giacomello et al., 2010), which are pivotal to activate the low-Ca2+-sensitive mitochondrial Ca2+-uptake pathway (Rizzuto et al., 1999). Studies using both chemical indicators (Collins et al., 2001) and genetically encoded Ca2+ probes (Malli et al., 2005) have revealed that SERCA inhibition evokes only small and delayed mitochondrial Ca2+ signals. Accordingly, we can assume that, due to the lack of an efficient activation of mitochondrial Ca2+ uptake under these conditions, the contribution of MCU and UCP2 to STIM1 oligomerization is irrelevant.

Experiments using genetically encoded, organelle-targeted Ca2+ probes have revealed that the ER Ca2+ homeostasis is controlled by mitochondria in a rather complex manner. Upon IP3-mediated cell stimulation, mitochondria are able to recycle Ca2+ that has been released from intracellular stores (Arnaudeau et al., 2001) and shuttle entering Ca2+ to ER Ca2+ reuptake sites (Malli et al., 2005), thus facilitating Ca2+ refilling of the ER, which counteracts store depletion and, hence, SOCE activation. In this study, we used D1ER, an ER-targeted cameleon (Palmer et al., 2004), to investigate the impact of MCU and UCP2 knockdown on the Ca2+ homeostasis of the ER. Our data show that D1ER signals are only slightly affected in MCUKD or UCP2KD cells when Ca2+ is mobilized with histamine in the presence of extracellular Ca2+ (Fig. 1B), indicating that reduced mitochondrial Ca2+ uptake to some extent attenuates the kinetics of IP3-mediated ER Ca2+ depletion. This observation conflicts with the idea that mitochondria recycle released Ca2+ back to the ER (Arnaudeau et al., 2001), whereas it confirms the assumption that local mitochondrial Ca2+ buffering facilitates IP3-mediated ER Ca2+ depletion (Gilabert et al., 2001). Interestingly, we found no differences in the kinetics of ER Ca2+ depletion between the three cell lines when ER Ca2+ was mobilized with histamine in the absence of extracellular Ca2+ (Fig. 2B). However, STIM1 oligomerization was clearly delayed in MCUKD and UCP2KD cells under these conditions (Fig. 2A), raising the question of why a difference in STIM1 activation occurs despite the same rate of ER Ca2+ depletion. This discrepancy might depend on the different Ca2+-binding affinities of D1ER and the EF-hand motif of STIM1. In vitro calibration of D1ER revealed a dissociation constant (Kd) of 60 µM (Palmer et al., 2004), whereas the Kd of the Ca2+ probe was estimated to be ∼220 µM in vivo (Rudolf et al., 2006). The Kd of the EF-hand domain of STIM1 proteins to detect ER Ca2+ fluctuations has been calculated to be between 200 and 600 µM (Stathopulos et al., 2006). Based on these clear differences, it appears feasible that small changes in [Ca2+]ER are sensed by STIM1, but not by D1ER. Alternatively, mitochondrial Ca2+ buffering facilitates the reduction of [Ca2+]ER primarily within subcompartments of the ER that are sensed by STIM1, whereas the global ER Ca2+ concentration remains unaffected. Such specialized compartments of the ER have recently been described and named as the pre-cortical ER. These structures display thin ER tubules linked to microtubules and are enriched in STIM1 proteins (Shen et al., 2011a). In addition, cortical ER sections, which are also enriched in STIM1 proteins, have been observed close to the plasma membrane (Shen et al., 2011a). Interestingly, these thin ER compartments do not contain Ca2+-binding chaperones, indicating a different mechanism of Ca2+ homeostasis within the cortical ER compared with that of the bulk ER. Accordingly, it is tempting to speculate that mitochondria, which have been shown to be tethered to certain ER areas (de Brito and Scorrano, 2008), preferentially interact with those STIM1-enriched specialized ER compartments. In addition, our data unveiled a clear discrepancy between the kinetics of ER Ca2+ depletion and STIM1 oligomerization in response to BHQ (Fig. 3C). Under these conditions, STIM1 oligomerization was fast, despite slow ER Ca2+ depletion. We, and others, have suggested previously that STIM1 punctae formation and oligomerization is indeed under the control of cytosolic Ca2+ (Malli et al., 2008; Shen et al., 2011b). As slow ER Ca2+ depletion with a SERCA inhibitor does not generate high-concentration cytosolic Ca2+ micro-domains on the ER surface, STIM1 oligomerization under these conditions is independent of mitochondrial Ca2+ uptake. This finding indicates that the lack of cytosolic Ca2+ hot-spot formation upon SERCA inhibition promotes STIM1 activation. In accordance with these observations, our findings highlight that, upon ER Ca2+ mobilization with a physiological IP3-generating agonist, UCP2- and MCU-dependent mitochondrial Ca2+ buffering is essential for STIM1 oligomerization. Therefore, we conclude that mitochondrial Ca2+ buffering in the vicinity of the ER predominately shapes cytosolic Ca2+ micro-domains, thus facilitating STIM1 activation (Fig. 7A). However, our data demonstrate that, upon irreversible SERCA inhibition with thapsigargin, a sustained Ca2+ entry through the SOCE pathway requires MCU-dependent mitochondrial Ca2+ uptake (Fig. 6B; Fig. 7B). This is a striking finding in view of an earlier study that questioned the formation of high-concentration Ca2+ micro-domains on mitochondria upon SOCE in HeLa cells (Giacomello et al., 2010). Therefore, it is tempting to speculate that the MCU-dependent transfer of entering Ca2+ into mitochondria controls the maintenance of SOCE by a mechanism other than by shaping subplasmalemmal Ca2+ hot spots.

Fig. 7.

Schematic showing the importance of mitochondrial Ca2+ uptake for STIM1 oligomerization and SOCE maintenance in HeLa cells. (A) The impact of UCP2- and MCU-dependent mitochondrial Ca2+ uptake on STIM1 activation. Dashed arrow, STIM1 oligomerization; solid arrows, Ca2+ movements. (1) IP3-mediated Ca2+ release causes a reduction in the Ca2+ content of the ER, which triggers STIM1 oligomerization. However, local high-concentration Ca2+ micro-domains on the cytosolic side of the ER membrane restrain the formation of STIM1 oligomers. (2) Mitochondria (mito.) adjacent to sites of ER Ca2+ release act to sequester Ca2+ by a UCP2- and MCU-dependent pathway, which dissipates high-concentration Ca2+ micro-domains and, hence, facilitates STIM1 oligomerization. (3) Accordingly, STIM1 oligomerization upon IP3-mediated Ca2+ mobilization requires mitochondrial Ca2+ uptake, as Ca2+ buffering by the mitochondria results in both an efficient depletion of Ca2+ from the ER and, importantly, the dissipation of local cytosolic Ca2+ micro-domains at the ER surface. (B) MCU-dependent mitochondrial Ca2+ uptake counteracts Ca2+-dependent SOCE inactivation. Solid arrows, Ca2+ movements. (1) SERCA inhibition with thapsigargin induces strong ER Ca2+ depletion, leading to STIM1 oligomerization and punctae formation in superficial ER domains. (2) STIM1 aggregates interfere with plasma membrane (PM) Ca2+ channels such as Orai1, which leads to strong SOCE activation. (3) Upon Ca2+ addition to thapsigargin-treated cells, SOCE induces an increase in global and local cytosolic Ca2+ levels, and Ca2+ negatively impacts on SOCE. (4) MCU-dependent mitochondrial sequestration of entering Ca2+ is irrelevant for thapsigargin-induced SOCE activation but essential to maintain SOCE – probably by abrogating the (slow) Ca2+-dependent inactivation of this Ca2+ entry pathway.

Fig. 7.

Schematic showing the importance of mitochondrial Ca2+ uptake for STIM1 oligomerization and SOCE maintenance in HeLa cells. (A) The impact of UCP2- and MCU-dependent mitochondrial Ca2+ uptake on STIM1 activation. Dashed arrow, STIM1 oligomerization; solid arrows, Ca2+ movements. (1) IP3-mediated Ca2+ release causes a reduction in the Ca2+ content of the ER, which triggers STIM1 oligomerization. However, local high-concentration Ca2+ micro-domains on the cytosolic side of the ER membrane restrain the formation of STIM1 oligomers. (2) Mitochondria (mito.) adjacent to sites of ER Ca2+ release act to sequester Ca2+ by a UCP2- and MCU-dependent pathway, which dissipates high-concentration Ca2+ micro-domains and, hence, facilitates STIM1 oligomerization. (3) Accordingly, STIM1 oligomerization upon IP3-mediated Ca2+ mobilization requires mitochondrial Ca2+ uptake, as Ca2+ buffering by the mitochondria results in both an efficient depletion of Ca2+ from the ER and, importantly, the dissipation of local cytosolic Ca2+ micro-domains at the ER surface. (B) MCU-dependent mitochondrial Ca2+ uptake counteracts Ca2+-dependent SOCE inactivation. Solid arrows, Ca2+ movements. (1) SERCA inhibition with thapsigargin induces strong ER Ca2+ depletion, leading to STIM1 oligomerization and punctae formation in superficial ER domains. (2) STIM1 aggregates interfere with plasma membrane (PM) Ca2+ channels such as Orai1, which leads to strong SOCE activation. (3) Upon Ca2+ addition to thapsigargin-treated cells, SOCE induces an increase in global and local cytosolic Ca2+ levels, and Ca2+ negatively impacts on SOCE. (4) MCU-dependent mitochondrial sequestration of entering Ca2+ is irrelevant for thapsigargin-induced SOCE activation but essential to maintain SOCE – probably by abrogating the (slow) Ca2+-dependent inactivation of this Ca2+ entry pathway.

Recently, an MCU knockout mouse was generated (Pan et al., 2013). Interestingly, MCU depletion resulted in a rather inconspicuous phenotype, which is comparable with that of UCP2 or UCP3 knockout mice (Arsenijevic et al., 2000). The lack of MCU was, however, associated with reduced Ca2+-stimulated mitochondrial ATP generation and skeletal muscle work (Pan et al., 2013). Typical STIM1-related functions, such as T-cell activation and skeletal muscle physiology (Frischauf et al., 2008), have not been specifically examined in MCU knockout mice thus far. However, both the clear inability of mitochondria to respond to increased levels of Ca2+ with increased ATP generation in MCU knockout cells and our findings demonstrating the importance of MCU in SOCE activation and maintenance point to mitochondrial ATP as a potential regulator of SOCE. Indeed, it has been suggested that ATP facilitates SOCE by activating certain kinases (Lang et al., 2012) and/or increasing the local Ca2+-buffering capacity (Montalvo et al., 2006). Additional studies are necessary to entirely characterize the molecular mechanisms by which mitochondrial Ca2+ handling and organelle metabolism interfere with SOCE activation, maintenance and termination in different cell types.

Our study, however, reveals that, under physiological conditions of IP3-mediated Ca2+ mobilization, the transfer of Ca2+ from the ER into mitochondria is essential for efficient STIM1 oligomerization and SOCE activation. Notably, the mitochondria–SOCE interaction we describe herein might be a cell-type-specific phenomenon. However, considering the versatile functions of STIM1 (Soboloff et al., 2012), the identification of any molecular mechanism by which mitochondria regulate the activation of this protein under physiological conditions of cell stimulation will help to improve our understanding of other STIM1-dependent cell signaling events as well.

MATERIALS AND METHODS

Chemicals and buffer solutions

Cell culture materials were obtained from PAA laboratories (Pasching, Austria). FCCP, oligomycin A, histamine, BHQ and EGTA were purchased from Sigma Aldrich (Vienna, Austria). Thapsigargin and BAPTA-AM were from Abcam® (London, UK). Prior to experiments, cells were washed and maintained for 20 minutes in a HEPES-buffered solution containing 138 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 1 mM HEPES, 2.6 mM NaHCO3, 0.44 mM KH2PO4, 0.34 mM Na2HPO4, 10 mM d-glucose, 0.1% vitamins, 0.2% essential amino acids and 1% penicillin-streptomycin, the pH of which was adjusted to 7.4 with NaOH. During the experiments, cells were perfused with a Ca2+-containing buffer, which consisted of 145 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM d-glucose and 10 mM HEPES, the pH of which was adjusted to 7.4 with NaOH. During experiments in which a Ca2+-free solution was applied to the cells, the CaCl2 was replaced with 1 mM EGTA.

Stable knockdown cell lines

HeLa SilenciX® knockdown cell lines were purchased from Tebu-bio (Le-Perray-en-Yvelines, France). We used HeLa cells stably expressing scrambled shRNA (control) or shRNA against MCU (MCUKD) (Bondarenko et al., 2013) or the UCP2 (UCP2KD).

Cell culture and transfection

HeLa cells were grown in Dulbeccos's Modified Eagle Medium (Sigma Aldrich) containing 10% fetal bovine serum, 100 U/ml penicillin and 100 µg/ml streptomycin, and they were plated on 30-mm glass coverslips. At 60–80% confluence, cells were transfected with 1.5 µg (per 30-mm well) of plasmid DNA encoding the appropriate sensor/fluorophore using TransFastTM transfection reagent at 3 µg/well (Promega, Madison, WI) in 1 ml of serum- and antibiotic-free medium. Cells were maintained in a humidified incubator (37°C, 5% CO2, 95% air) for 16–20 hours prior to changing the culture medium. All experiments were performed either 24 hours or 48 hours after transfection.

Validation of shRNA-mediated stable knockdown of MCU and UCP2 using RT-PCR

RNA was isolated from SilenceX control, MCUKD and UCP2KD HeLa cells using a Total RNA isolation kit (Peqlab Biotechnologie, Erlangen, Germany). For reverse transcription, a High Capacity cDNA Reverse Transcription Kit from Applied Biosystems (Live Technologies Corporation, Vienna, Austria) was used. The analysis of the expression of MCU and UCP2 was performed by conventional PCR using GoTaq Green master mix (Promega) and by real-time (RT)-PCR using QuantiFast SYBR Green RT-PCR kit (Qiagen, Germantown, MD) on a LightCycler 480 (Roche Diagnostics Deutschland, Mannheim, Germany). RNA polymerase II (RPOL2) was used as the housekeeping control. Primers for RPOL2 and MCU were obtained from Invitrogen (Live Technologies Corporation, Vienna, Austria), and their sequences were as follows: RPOL2 forward, 5′-CATTGACTTGCGTTTCCACC-3′; RPOL2 reverse, 5′-ACATTTTGTGCAGAGTTGGC-3′; MCU forward, 5′-TTCCTGGCAGAATTTGGGAG-3′; and MCU reverse, 5′-AGAGATAGGCTTGAGTGTGAAC-3′. For human UCP2, the QuantiTect®Primer Assay (Qiagen, QT00014140) was used.

Measurement of mitochondrial membrane potential

HeLa cells were loaded with 1 µM of the ratiometric dye JC-1 (Invitrogen) in culture medium at 37°C for 40 minutes, washed with PBS, dissociated by trypsinization, centrifuged at 262 g for 5 minutes and resuspended in the Ca2+-containing buffer described above. JC-1 fluorescence was detected using a fluorescence spectrophotometer (Hitachi F-4500, Hitachi, Inula, Austria). JC-1 was excited at 490 nm and 540 nm and emission was collected at 540 nm and 590 nm, respectively. The basal fluorescence ratio was normalized to the ratio after dissipation of mitochondrial membrane by treatment with 10 µM FCCP.

Measurement of cellular oxygen consumption rate

Cells were plated in XF96 polystyrene cell culture microplates (Seahorse Bioscience) at a density of 30,000 cells per well. After an overnight incubation, cells were washed and preincubated for 30 minutes in unbuffered XF assay medium (Seahorse Bioscience) supplemented with 5.5 mM d-glucose and 1 mM sodium pyruvate at 37°C in a non-CO2 environment. Oxygen consumption rates were subsequently measured using an XF96 extracellular flux analyzer.

Quantification of total cellular ATP content using HPLC

Separation of adenine nucleotides (neutralized supernatant) after cell lysis was performed on a Hypersil ODS column (5 µm, 250×4-mm inner diameter), using a L2200 autosampler, two L-2130 HTA pumps and a L2450 diode array detector (all from Hitachi), as described previously (Khan et al., 2012).

Intra- and intermolecular FRET measurements

Dynamic changes in [Ca2+]ER and [Ca2+]mito were followed in cells expressing the ER- or mitochondria-targeted cameleons D1ER (Palmer et al., 2004) and 4mtD3cpv (Jean-Quartier et al., 2012), respectively. STIM1-oligomerization was monitored by measuring intermolecular FRET between CFP–STIM1 and YFP–STIM1 (Deak et al., 2013). Experiments were performed on an inverted microscope (Axio Observer.A1, Zeiss, Göttingen, Germany) equipped with a polychromator illumination system (VisiChrome, Visitron Systems, Puchheim, Germany) and a thermoelectric-cooled CCD camera (Photometrics CoolSNAP HQ, Visitron Systems). Cells were imaged with a 40× oil-immersion objective (Zeiss). Excitation of the fluorophores was at 440±10 nm (440AF21, Omega Optical, Brattleboro, VT), and emission was recorded at 480 and 535 nm using emission filters (480AF30 and 535AF26, Omega Optical) mounted on a Ludl filterwheel. Results of FRET measurements are shown as the ratio of (F535/F480)/R0 (where R0 is the basal ratio), to correct for photobleaching and/or photochromism, as described previously (Waldeck-Weiermair et al., 2012).

Single-cell Ca2+ imaging using fura-2 and 4mtD1GO-Cam

Cytosolic Ca2+ signals of single cells were monitored using the classical fura-2 technique on a digital imaging system, as described previously (Deak et al., 2013). Briefly, prior to experiments, cells were incubated with 2 µM Fura-2/am (TEFLabs, Austin, TX) for 45 minutes and were alternately illuminated at 340 and 380 nm, while fluorescence emission was gathered at 510 nm. For simultaneous cytosolic and mitochondrial Ca2+ measurements, 4mtD1GO-Cam transfected HeLa cells loaded with Fura-2/am were used. Co-imaging of the different fluorophores was achieved with a digital wide-field imaging system, the Till iMIC (Till Photonics, Gräfelfing, Germany), using a 40× objective (Zeiss). Fura-2 and the 4mtD1GO-Cam were alternately excited at 340 nm or 380 nm and at 477 nm, respectively, with an ultra-fast switching monochromator, the Polychrome V (Till Photonics), equipped with an excitation filter (E500spuv) and a dichroic filter (495dcxru, Chroma Technology Corp., VT). Emitted light was simultaneously collected at 510 nm (Fura-2 and GFP of GO-Cam) and at 560 nm (FRET-channel of GO-Cam) using a single beam splitter design (Dichrotome, Till Photonics) that was equipped with a dual band emission filter (59004m ET Fitc/Tritc Dual Emitter) and a second dichroic filter (560dcxr, Chroma Technology Corp.). Images were recorded with a CCD camera (AVT Stringray F145B, Till Photonics). The digital imaging system was controlled by the live-acquisition software v2.0.0.12 (Till Photonics), as described previously (Waldeck-Weiermair et al., 2012).

Mn2+ quench experiments

Fura-2/am-loaded HeLa cells were perfused with the Ca2+-containing experimental buffer supplemented with 100 µM MnCl2. Mn2+ quenching of cytosolic fura-2 fluorescence upon the addition of 100 µM histamine was measured by using the Ca2+ imaging system described above at an excitation wavelength of 360 nm.

Confocal analysis and 3D rendering

High resolution Z-scan imaging of subcellular structures was performed in cells coexpressing D1ER and the mitochondria-targeted DsRed. Images were acquired with an array confocal laser scanning microscope, built on an inverse fully automatic microscope equipped with VoxCell Scan® (VisiTech, Visitron Systems) and a 100× objective (Plan-Fluor 100×/1.45 oil, Zeiss). Fluorophores were illuminated at 488 nm (120 mW diode laser, Visitron Systems) and 515 nm (50 mW, VSLaserModul, Visitron Systems). Emitted light was acquired with a CCD camera (CoolSNAP-HQ, Photometrics,) using the emission filters ET535/30m and E570LPv2 (Chroma Technology) mounted on a computer-controlled fast-filter wheel (Ludl Electronic Products). All devices were controlled by VisiView Premier Acquisition software (Visitron Systems). The ER and mitochondria z-stacks were deconvoluted using the iterative quick maximum likelihood estimation algorithm (QMLE) of Huygens 2.4.1p3 (SVI, Hilversum, Netherlands). Subsequently, combined three-dimensional rendering of the organelles was performed with Imaris 3.3 software (Bitplane AG, Zurich, Switzerland). Quantitative mitochondrial shape analysis and colocalization computations were performed with the integrated morphometric analysis plug-in of MetaMorph 7.7.0.0 software (Visitron). The intensity threshold values of shape analysis and colocalization computations did not significantly differ within all the samples analyzed and were determined over a range that completely eliminated background fluorescence but preserved organelle structures.

Statistics

Data shown represent the mean±s.e.m., where n reflects the number of cells. Statistical analyses were performed by using the unpaired Student's t-test, and P<0.05 was considered to be significant.

Acknowledgements

We thank Therese Macher (Institute of Molecular Biology and Biochemistry, Graz, Austria) and Rene Rost (Institute of Molecular Biology and Biochemistry, Graz, Austria) for their excellent technical assistance. We also thank Klaus Groschner (Institute of Biophysics, Graz, Austria) for sending us the YFP-tagged Orai (E106Q) construct, Christoph Romanin (Institute of Biophysics, Linz, Austria), who provided us with YFP–STIM1 and CFP–STIM1, and Roger Y. Tsien (Department of Chemistry & Biochemistry, San Diego, USA) for sending us D1ER and 4mtD3cpv.

Author contributions

A.T.D. performed and analyzed experiments and wrote the manuscript. S.B. performed and analyzed Ca2+ imaging experiments. M.J.K. validated shRNA-mediated knockdown efficiencies. L.N.G. performed imaging and Seahorse experiments. M.W.-W. performed and analyzed multichannel imaging experiments. S.H. quantified cellular ATP levels. W.F.G. conceived the study and wrote the manuscript. R.M. analyzed data, conceived the study and wrote the manuscript.

Funding

This work was supported by the Austrian Science Fund (FWF) [grant numbers P20181-B05, P21857-B18 and P22553-B18]. A.T.D. is funded by the FWF within the Neuroscience PhD program at the Medical University of Graz. Deposited in PMC for immediate release.

References

References
Arnaudeau
S.
,
Kelley
W. L.
,
Walsh
J. V.
 Jr
,
Demaurex
N.
(
2001
).
Mitochondria recycle Ca(2+) to the endoplasmic reticulum and prevent the depletion of neighboring endoplasmic reticulum regions.
J. Biol. Chem.
276
,
29430
29439
.
Arsenijevic
D.
,
Onuma
H.
,
Pecqueur
C.
,
Raimbault
S.
,
Manning
B. S.
,
Miroux
B.
,
Couplan
E.
,
Alves-Guerra
M. C.
,
Goubern
M.
,
Surwit
R.
 et al. (
2000
).
Disruption of the uncoupling protein-2 gene in mice reveals a role in immunity and reactive oxygen species production.
Nat. Genet.
26
,
435
439
.
Baughman
J. M.
,
Perocchi
F.
,
Girgis
H. S.
,
Plovanich
M.
,
Belcher-Timme
C. A.
,
Sancak
Y.
,
Bao
X. R.
,
Strittmatter
L.
,
Goldberger
O.
,
Bogorad
R. L.
 et al. (
2011
).
Integrative genomics identifies MCU as an essential component of the mitochondrial calcium uniporter.
Nature
476
,
341
345
.
Bondarenko
A. I.
,
Jean-Quartier
C.
,
Parichatikanond
W.
,
Alam
M. R.
,
Waldeck-Weiermair
M.
,
Malli
R.
,
Graier
W. F.
(
2013
).
Mitochondrial Ca2+ uniporter (MCU)-dependent and MCU-independent Ca2+ channels coexist in the inner mitochondrial membrane.
Pflugers Arch.
doi: 10.1007/s00424-013-1383-0
Cheng
K. T.
,
Ong
H. L.
,
Liu
X.
,
Ambudkar
I. S.
(
2013
).
Contribution and regulation of TRPC channels in store-operated Ca2+ entry.
Curr. Top. Mem.
71
,
149
179
.
Collins
T. J.
,
Lipp
P.
,
Berridge
M. J.
,
Bootman
M. D.
(
2001
).
Mitochondrial Ca(2+) uptake depends on the spatial and temporal profile of cytosolic Ca(2+) signals.
J. Biol. Chem.
276
,
26411
26420
.
de Brito
O. M.
,
Scorrano
L.
(
2008
).
Mitofusin 2 tethers endoplasmic reticulum to mitochondria.
Nature
456
,
605
610
.
De Stefani
D.
,
Raffaello
A.
,
Teardo
E.
,
Szabò
I.
,
Rizzuto
R.
(
2011
).
A forty-kilodalton protein of the inner membrane is the mitochondrial calcium uniporter.
Nature
476
,
336
340
.
Deak
A. T.
,
Groschner
L. N.
,
Alam
M. R.
,
Seles
E.
,
Bondarenko
A. I.
,
Graier
W. F.
,
Malli
R.
(
2013
).
The endocannabinoid N-arachidonoyl glycine (NAGly) inhibits store-operated Ca2+ entry by preventing STIM1-Orai1 interaction.
J. Cell Sci.
126
,
879
888
.
Demaurex
N.
,
Poburko
D.
,
Frieden
M.
(
2009
).
Regulation of plasma membrane calcium fluxes by mitochondria.
Biochim. Biophys. Acta
1787
,
1383
1394
.
Duchen
M. R.
(
2004
).
Mitochondria in health and disease: perspectives on a new mitochondrial biology.
Mol. Aspects Med.
25
,
365
451
.
Frischauf
I.
,
Schindl
R.
,
Derler
I.
,
Bergsmann
J.
,
Fahrner
M.
,
Romanin
C.
(
2008
).
The STIM/Orai coupling machinery.
Channels (Austin)
2
,
261
268
.
Giacomello
M.
,
Drago
I.
,
Bortolozzi
M.
,
Scorzeto
M.
,
Gianelle
A.
,
Pizzo
P.
,
Pozzan
T.
(
2010
).
Ca2+ hot spots on the mitochondrial surface are generated by Ca2+ mobilization from stores, but not by activation of store-operated Ca2+ channels.
Mol. Cell
38
,
280
290
.
Gilabert
J. A.
,
Parekh
A. B.
(
2000
).
Respiring mitochondria determine the pattern of activation and inactivation of the store-operated Ca(2+) current I(CRAC).
EMBO J.
19
,
6401
6407
.
Gilabert
J. A.
,
Bakowski
D.
,
Parekh
A. B.
(
2001
).
Energized mitochondria increase the dynamic range over which inositol 1,4,5-trisphosphate activates store-operated calcium influx.
EMBO J.
20
,
2672
2679
.
Glitsch
M. D.
,
Bakowski
D.
,
Parekh
A. B.
(
2002
).
Store-operated Ca2+ entry depends on mitochondrial Ca2+ uptake.
EMBO J.
21
,
6744
6754
.
Graier
W. F.
,
Frieden
M.
,
Malli
R.
(
2007
).
Mitochondria and Ca(2+) signaling: old guests, new functions.
Pflugers Arch.
455
,
375
396
.
Hoth
M.
,
Penner
R.
(
1992
).
Depletion of intracellular calcium stores activates a calcium current in mast cells.
Nature
355
,
353
356
.
Hoth
M.
,
Fanger
C. M.
,
Lewis
R. S.
(
1997
).
Mitochondrial regulation of store-operated calcium signaling in T lymphocytes.
J. Cell Biol.
137
,
633
648
.
Jean-Quartier
C.
,
Bondarenko
A. I.
,
Alam
M. R.
,
Trenker
M.
,
Waldeck-Weiermair
M.
,
Malli
R.
,
Graier
W. F.
(
2012
).
Studying mitochondrial Ca(2+) uptake - a revisit.
Mol. Cell. Endocrinol.
353
,
114
127
.
Khan
M. J.
,
Rizwan Alam
M.
,
Waldeck-Weiermair
M.
,
Karsten
F.
,
Groschner
L.
,
Riederer
M.
,
Hallström
S.
,
Rockenfeller
P.
,
Konya
V.
,
Heinemann
A.
 et al. (
2012
).
Inhibition of autophagy rescues palmitic acid-induced necroptosis of endothelial cells.
J. Biol. Chem.
287
,
21110
21120
.
Lang
F.
,
Eylenstein
A.
,
Shumilina
E.
(
2012
).
Regulation of Orai1/STIM1 by the kinases SGK1 and AMPK.
Cell Calcium
52
,
347
354
.
Lefkimmiatis
K.
,
Srikanthan
M.
,
Maiellaro
I.
,
Moyer
M. P.
,
Curci
S.
,
Hofer
A. M.
(
2009
).
Store-operated cyclic AMP signalling mediated by STIM1.
Nat. Cell Biol.
11
,
433
442
.
Liou
J.
,
Kim
M. L.
,
Heo
W. D.
,
Jones
J. T.
,
Myers
J. W.
,
Ferrell
J. E.
 Jr
,
Meyer
T.
(
2005
).
STIM is a Ca2+ sensor essential for Ca2+-store-depletion-triggered Ca2+ influx.
Curr. Biol.
15
,
1235
1241
.
Liou
J.
,
Fivaz
M.
,
Inoue
T.
,
Meyer
T.
(
2007
).
Live-cell imaging reveals sequential oligomerization and local plasma membrane targeting of stromal interaction molecule 1 after Ca2+ store depletion.
Proc. Natl. Acad. Sci. USA
104
,
9301
9306
.
Luik
R. M.
,
Wang
B.
,
Prakriya
M.
,
Wu
M. M.
,
Lewis
R. S.
(
2008
).
Oligomerization of STIM1 couples ER calcium depletion to CRAC channel activation.
Nature
454
,
538
542
.
Malli
R.
,
Frieden
M.
,
Trenker
M.
,
Graier
W. F.
(
2005
).
The role of mitochondria for Ca2+ refilling of the endoplasmic reticulum.
J. Biol. Chem.
280
,
12114
12122
.
Malli
R.
,
Naghdi
S.
,
Romanin
C.
,
Graier
W. F.
(
2008
).
Cytosolic Ca2+ prevents the subplasmalemmal clustering of STIM1: an intrinsic mechanism to avoid Ca2+ overload.
J. Cell Sci.
121
,
3133
3139
.
Manjarrés
I. M.
,
Rodríguez-García
A.
,
Alonso
M. T.
,
García-Sancho
J.
(
2010
).
The sarco/endoplasmic reticulum Ca(2+) ATPase (SERCA) is the third element in capacitative calcium entry.
Cell Calcium
47
,
412
418
.
Montalvo
G. B.
,
Artalejo
A. R.
,
Gilabert
J. A.
(
2006
).
ATP from subplasmalemmal mitochondria controls Ca2+-dependent inactivation of CRAC channels.
J. Biol. Chem.
281
,
35616
35623
.
Naghdi
S.
,
Waldeck-Weiermair
M.
,
Fertschai
I.
,
Poteser
M.
,
Graier
W. F.
,
Malli
R.
(
2010
).
Mitochondrial Ca2+ uptake and not mitochondrial motility is required for STIM1-Orai1-dependent store-operated Ca2+ entry.
J. Cell Sci.
123
,
2553
2564
.
Palmer
A. E.
,
Jin
C.
,
Reed
J. C.
,
Tsien
R. Y.
(
2004
).
Bcl-2-mediated alterations in endoplasmic reticulum Ca2+ analyzed with an improved genetically encoded fluorescent sensor.
Proc. Natl. Acad. Sci. USA
101
,
17404
17409
.
Pan
X.
,
Liu
J.
,
Nguyen
T.
,
Liu
C.
,
Sun
J.
,
Teng
Y.
,
Fergusson
M. M.
,
Rovira
I. I.
,
Allen
M.
,
Springer
D. A.
 et al. (
2013
).
The physiological role of mitochondrial calcium revealed by mice lacking the mitochondrial calcium uniporter.
Nat. Cell Biol.
15
,
1464
1472
.
Parekh
A. B.
(
2008a
).
Ca2+ microdomains near plasma membrane Ca2+ channels: impact on cell function.
J. Physiol.
586
,
3043
3054
.
Parekh
A. B.
(
2008b
).
Mitochondrial regulation of store-operated CRAC channels.
Cell Calcium
44
,
6
13
.
Parekh
A. B.
,
Putney
J. W.
 Jr
(
2005
).
Store-operated calcium channels.
Physiol. Rev.
85
,
757
810
.
Park
C. Y.
,
Hoover
P. J.
,
Mullins
F. M.
,
Bachhawat
P.
,
Covington
E. D.
,
Raunser
S.
,
Walz
T.
,
Garcia
K. C.
,
Dolmetsch
R. E.
,
Lewis
R. S.
(
2009
).
STIM1 clusters and activates CRAC channels via direct binding of a cytosolic domain to Orai1.
Cell
136
,
876
890
.
Putney
J. W.
 Jr
(
1986
).
A model for receptor-regulated calcium entry.
Cell Calcium
7
,
1
12
.
Ritchie
M. F.
,
Samakai
E.
,
Soboloff
J.
(
2012
).
STIM1 is required for attenuation of PMCA-mediated Ca2+ clearance during T-cell activation.
EMBO J.
31
,
1123
1133
.
Rizzuto
R.
,
Pinton
P.
,
Brini
M.
,
Chiesa
A.
,
Filippin
L.
,
Pozzan
T.
(
1999
).
Mitochondria as biosensors of calcium microdomains.
Cell Calcium
26
,
193
200
.
Rudolf
R.
,
Magalhães
P. J.
,
Pozzan
T.
(
2006
).
Direct in vivo monitoring of sarcoplasmic reticulum Ca2+ and cytosolic cAMP dynamics in mouse skeletal muscle.
J. Cell Biol.
173
,
187
193
.
Shen
W. W.
,
Frieden
M.
,
Demaurex
N.
(
2011a
).
Remodelling of the endoplasmic reticulum during store-operated calcium entry.
Biol. Cell
103
,
365
380
.
Shen
W. W.
,
Frieden
M.
,
Demaurex
N.
(
2011b
).
Local cytosolic Ca2+ elevations are required for stromal interaction molecule 1 (STIM1) de-oligomerization and termination of store-operated Ca2+ entry.
J. Biol. Chem.
286
,
36448
36459
.
Shinde
A. V.
,
Motiani
R. K.
,
Zhang
X.
,
Abdullaev
I. F.
,
Adam
A. P.
,
González-Cobos
J. C.
,
Zhang
W.
,
Matrougui
K.
,
Vincent
P. A.
,
Trebak
M.
(
2013
).
STIM1 controls endothelial barrier function independently of Orai1 and Ca2+ entry.
Sci. Signal.
6
,
ra18
.
Soboloff
J.
,
Rothberg
B. S.
,
Madesh
M.
,
Gill
D. L.
(
2012
).
STIM proteins: dynamic calcium signal transducers.
Nat. Rev. Mol. Cell Biol.
13
,
549
565
.
Stathopulos
P. B.
,
Li
G. Y.
,
Plevin
M. J.
,
Ames
J. B.
,
Ikura
M.
(
2006
).
Stored Ca2+ depletion-induced oligomerization of stromal interaction molecule 1 (STIM1) via the EF-SAM region: An initiation mechanism for capacitive Ca2+ entry.
J. Biol. Chem.
281
,
35855
35862
.
Trenker
M.
,
Malli
R.
,
Fertschai
I.
,
Levak-Frank
S.
,
Graier
W. F.
(
2007
).
Uncoupling proteins 2 and 3 are fundamental for mitochondrial Ca2+ uniport.
Nat. Cell Biol.
9
,
445
452
.
Trenker
M.
,
Fertschai
I.
,
Malli
R.
,
Graier
W.
(
2008
).
UCP2/3 – likely to be fundamental for mitochondrial Ca2+ uniport.
Nat. Cell Biol.
10
,
1237
1240
.
Vig
M.
,
Peinelt
C.
,
Beck
A.
,
Koomoa
D. L.
,
Rabah
D.
,
Koblan-Huberson
M.
,
Kraft
S.
,
Turner
H.
,
Fleig
A.
,
Penner
R.
 et al. (
2006
).
CRACM1 is a plasma membrane protein essential for store-operated Ca2+ entry.
Science
312
,
1220
1223
.
Waldeck-Weiermair
M.
,
Duan
X.
,
Naghdi
S.
,
Khan
M. J.
,
Trenker
M.
,
Malli
R.
,
Graier
W. F.
(
2010a
).
Uncoupling protein 3 adjusts mitochondrial Ca(2+) uptake to high and low Ca(2+) signals.
Cell Calcium
48
,
288
301
.
Waldeck-Weiermair
M.
,
Malli
R.
,
Naghdi
S.
,
Trenker
M.
,
Kahn
M. J.
,
Graier
W. F.
(
2010b
).
The contribution of UCP2 and UCP3 to mitochondrial Ca(2+) uptake is differentially determined by the source of supplied Ca(2+).
Cell Calcium
47
,
433
440
.
Waldeck-Weiermair
M.
,
Jean-Quartier
C.
,
Rost
R.
,
Khan
M. J.
,
Vishnu
N.
,
Bondarenko
A. I.
,
Imamura
H.
,
Malli
R.
,
Graier
W. F.
(
2011
).
Leucine zipper EF hand-containing transmembrane protein 1 (Letm1) and uncoupling proteins 2 and 3 (UCP2/3) contribute to two distinct mitochondrial Ca2+ uptake pathways.
J. Biol. Chem.
286
,
28444
28455
.
Waldeck-Weiermair
M.
,
Alam
M. R.
,
Khan
M. J.
,
Deak
A. T.
,
Vishnu
N.
,
Karsten
F.
,
Imamura
H.
,
Graier
W. F.
,
Malli
R.
(
2012
).
Spatiotemporal correlations between cytosolic and mitochondrial Ca(2+) signals using a novel red-shifted mitochondrial targeted cameleon.
PLoS ONE
7
,
e45917
.
Waldeck-Weiermair
M.
,
Deak
A. T.
,
Groschner
L. N.
,
Alam
M. R.
,
Jean-Quartier
C.
,
Malli
R.
,
Graier
W. F.
(
2013
).
Molecularly distinct routes of mitochondrial Ca2+ uptake are activated depending on the activity of the sarco/endoplasmic reticulum Ca2+ ATPase (SERCA).
J. Biol. Chem.
288
,
15367
15379
.
Zhang
S. L.
,
Yu
Y.
,
Roos
J.
,
Kozak
J. A.
,
Deerinck
T. J.
,
Ellisman
M. H.
,
Stauderman
K. A.
,
Cahalan
M. D.
(
2005
).
STIM1 is a Ca2+ sensor that activates CRAC channels and migrates from the Ca2+ store to the plasma membrane.
Nature
437
,
902
905
.
Zhang
S. L.
,
Yeromin
A. V.
,
Zhang
X. H. F.
,
Yu
Y.
,
Safrina
O.
,
Penna
A.
,
Roos
J.
,
Stauderman
K. A.
,
Cahalan
M. D.
(
2006
).
Genome-wide RNAi screen of Ca(2+) influx identifies genes that regulate Ca(2+) release-activated Ca(2+) channel activity.
Proc. Natl. Acad. Sci. USA
103
,
9357
9362
.
Zweifach
A.
,
Lewis
R. S.
(
1995
).
Rapid inactivation of depletion-activated calcium current (ICRAC) due to local calcium feedback.
J. Gen. Physiol.
105
,
209
226
.

Competing interests

The authors declare no competing interests.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

Supplementary information