Mitochondria and Ca²⁺ signaling: old guests, new functions

Mitochondrial Ca²⁺ uptake

The phenomenon of mitochondrial Ca²⁺ uptake basically builds on two landmark observations that lay 30 years apart from each other. In the late 1960s, Naranjan S. Dhalla described for the first time strong Ca²⁺ accumulation in isolated respiring mitochondria [49]. However, due to the low sensitivity of the mitochondrial Ca²⁺ uptake machinery, it was assumed that Ca²⁺ sequestration by mitochondria in living cells is physiologically irrelevant. This false paradigm lasted until reliable measurements of mitochondrial Ca²⁺ uptake with _targeted protein based Ca²⁺ sensors clearly demonstrated that mitochondria in intact cells promptly sequester Ca²⁺ upon cell stimulation under physiological condition [184]. The apparent discrepancy between the striking mitochondrial Ca²⁺ response in intact cells and the low affinity of the mitochondrial Ca²⁺ uptake system of isolated mitochondria has not been solved entirely, but is most likely due to the exposure of the mitochondria to microdomains of high Ca²⁺ that meet the low Ca²⁺ affinity of the mitochondrial Ca²⁺ uptake system [187]. Such microdomains of high Ca²⁺ are thought to be achieved by a close proximity of sites of mitochondrial Ca²⁺ uptake with those of Ca²⁺ release at the endoplasmic reticulum (ER) and/or by functional coupling with Ca²⁺ entry channels at the plasma membrane [44, 126, 186]. Another aspect, which has been barely considered so far, but may also explain the obvious discrepancy between the prompt mitochondrial Ca²⁺ response in intact cells and the rather ponderous Ca²⁺ sequestration of isolated mitochondria, is that mitochondrial Ca²⁺ uptake in intact cells might be modulated by soluble proteins, which trail away upon cell permeabilization and/or isolation of the organelles. In line with this suggestion, mitochondrial Ca²⁺ homeostasis has been linked to protein kinase C family members [171], although the final proof is still missing.

Regardless of whether the organelles are isolated or reside within a cell, Ca²⁺ uptake into mitochondria critically depends on the electrical driving force for Ca²⁺ influx that is established by a huge negative membrane potential across the inner mitochondrial membrane (IMM). In respiring, energized mitochondria, a membrane potential of approximately −180 mV is generated by a large H⁺ gradient at the IMM that is established by continuous translocation of H⁺ from the matrix into the intermembrane space, which is accompanied by electron fluxes between the complexes of the respiratory chain [140]. According to Nernst equation, equilibrium would be reached only if Ca²⁺ inside the mitochondria reaches values 10⁶ higher than in the extramitochondrial area. Dissipation of the proton gradient across the IMM by chemical uncouplers such as carbonyl cyanide 4-(trifluoromethoxyl)s-phenyl-hydrazone (FCCP), which immediately causes a depolarization of the IMM, reduces the driving force for Ca²⁺ entry and thus strongly diminishes mitochondrial Ca²⁺ uptake (Fig. 1).

At this point it should be mentioned that protonophores like FCCP rapidly provoke a severe morphological alteration of the structural organization of the mitochondria in living cells. Notably, under physiological conditions mitochondria form a tubular highly interconnected network while mitochondrial structure turns toward disconnected, spherical, singular mitochondria upon FCCP (Fig. 2). Notably, such structural changes of mitochondria are accompanied with alterations in mitochondrial Ca²⁺ homeostasis independently of the initiator of such structural reorganization of this organelle [160].

Mitochondrial Ca²⁺ buffering and Ca²⁺ storage

The ability of isolated mitochondria to buffer and store large amounts of Ca²⁺ depends on phosphate and physiological concentrations of adenine nucleotides [148, 150], and the level of the free matrix Ca²⁺ concentration is controlled by the formation of an instantly reversible Ca²⁺-phosphate complex in the mitochondrial matrix [232]. In the presence of sufficient phosphate concentrations in the buffer, the free matrix Ca²⁺ concentration of isolated mitochondria is surprisingly invariant from Ca²⁺ load and was found to change less than 50% when total matrix Ca²⁺ was increased 50-fold [31]. While there is considerable uncertainty on the actual structure of the Ca²⁺-phosphate complex formed, the importance of matrix pH in dissolving this complex is known. Notably, matrix acidification induced by protonophores counteract the formation of Ca²⁺-phosphate complexes within mitochondria and thus reduces the mitochondrial Ca²⁺ buffer capacity while initiating large mitochondrial Ca²⁺ release [15, 149]. The formation of Ca²⁺-phosphate complexes as a high capacity mechanism of strong mitochondrial Ca²⁺ buffering could also be observed in intact bovine adrenal chromaffin cells [53].

On the contrary, in intact human endothelial cells, no large mitochondrial Ca²⁺ release is observed upon mitochondrial acidification, and free matrix Ca²⁺ concentration is more likely under the control of the balance between the amount of Ca²⁺ uptake and the capacity of its extrusion [125–128, 215]. These data may suggest that the formation of Ca²⁺-phosphate complexes, at least in endothelial cells , is marginally involved in mitochondrial Ca²⁺ homeostasis under physiological conditions. In view of the signaling function of matrix Ca²⁺ to stimulate various metabolic and housekeeping enzymes and to facilitate a fast regulator and tunable regulatory function of mitochondrial Ca²⁺, initial mitochondrial Ca²⁺ elevations up to 1 to 2 μM free Ca²⁺ might not be buffered by the formation of Ca²⁺-phosphate. Nevertheless, it is reasonable to assume that under conditions of altered mitochondrial/cellular Ca²⁺ homeostasis, the formation of Ca²⁺-phosphate complex occurs and may counteract massive accumulation of free matrix Ca²⁺ that would initiate apoptosis. Notably, such mitochondrial Ca²⁺ buffer capacity via formation of Ca²⁺-phosphate critically depends on the availability of phosphate that enters the mitochondria as H₃PO₄ (${\text{H}}_2{\text{PO}}_4^{-}/{\text{OH}}^{-}$antiport) via the electroneutral phosphate transporter [149]. Once the Ca²⁺-phosphate complex is formed, the H⁺ release needs to be compensated by the respiratory chain. A limited disposal of phosphate under pathological conditions of mitochondrial Ca²⁺ overload might explain the brake of mitochondrial Ca²⁺ buffer capacity and initiation of Ca²⁺-triggered apoptosis. In line with this assumption, the environmental buffer phosphate concentration was described to enhance Ca²⁺ buffer capacity of isolated mitochondria [232].

Whether the appearance of H⁺ in the matrix that is associated with the deprotonation of H₃PO₄ to form ${\text{PO}}_4^{3-}$ prior complexing with Ca²⁺ exhibits a beneficial smooth uncoupling that might decrease excessive ROS production is unknown but would be a further possibility how UCP2 and UCP3, which facilitate Ca²⁺ entry into the mitochondria, trigger smooth uncoupling and attenuation of mitochondrial ROS production [25, 26].

Mitochondrial Ca²⁺ efflux and functional coupling with other organelles

Tuning mitochondrial Ca²⁺ signaling

Shaping cytosolic Ca²⁺ elevation

Ca²⁺ mobilization from the internal Ca²⁺ stores occurs primarily from the ER upon activation of the IP₃ receptor (IP₃R) or the RyR. IP₃R is activated following cell stimulation by extracellular agonists, activation of the phospholipase C and production of IP₃. On the other hand, the RyR can be activated by a conformational change of the plasma membrane voltage-operated Ca²⁺ channels in skeletal muscle, by Ca²⁺ itself leading to Ca²⁺-induced Ca²⁺-release or by cADP ribose or NAADP [16]. The latter two mechanisms are less well documented at present. Once Ca²⁺ is elevated in the cytosol as a result of Ca²⁺ mobilization, it is immediately taken up by several pumps or exchangers to counteract the Ca²⁺ rise.

In excitable cells, in which cytosolic Ca²⁺ elevation is due to Ca²⁺ entry through voltage-gated Ca²⁺channels, the impact of mitochondria is rather clear and consistent. For instance, in chromaffin cells, mitochondria rapidly accumulate Ca²⁺ in an initial phase and thus delivers Ca²⁺ back to the cytosol, thus prolonging the low plateau phase of the Ca²⁺ elevation [8, 9, 54, 89]. However, if the ER represents the source of Ca²⁺, the way how mitochondria shape the cytosolic Ca²⁺ is less clear. In that case, the release of Ca²⁺ is accompanied by v entry through store-operated Ca²⁺ channels, which are also modulated by mitochondria (see: Ca²⁺ entry pathways). Thus, to precisely assess the role of the mitochondria, experiments should be performed in the absence of extracellular Ca²⁺. In such circumstances, preventing mitochondrial Ca²⁺ uptake results in an increased cytosolic Ca²⁺ signal [5]. In case of intracellular Ca²⁺ wave propagation, mitochondria were shown to accelerate or slow down the wave, depending most likely on the IP₃R subtype [54], which exhibit isoform specific Ca²⁺ dependence. Thus, in case of a linear Ca²⁺ activation of IP₃R, mitochondria slowed down the wave propagation [20], whereas in case of a bell shape Ca²⁺ activation, mitochondria accelerated the process by preventing Ca²⁺-dependent inhibition of the IP₃R [103].

The inhibition of local Ca²⁺ elevation by mitochondrial Ca²⁺ uptake that yields suppression of Ca²⁺-triggered feedback to IP₃R has also been found to contribute to cytosolic Ca²⁺ oscillation [8, 62, 81, 103, 112, 182, 187]. Recent data show that Ca²⁺ concentrations in the mitochondria and the ER oscillate concomitantly with cytosolic Ca²⁺ oscillations and that the Ca²⁺ shuttling between these organelles exhibits a pacemaker role in cytosolic Ca²⁺ oscillation [96].

Another important aspect in modulating cytosolic Ca²⁺ signaling by mitochondria is their capacity to exhibit spatial Ca²⁺ buffering in distinct area of the cytosol. Notably, the property of mitochondria to act as an intracellular Ca²⁺ buffering mechanism in contractile tissue with regard to the organelle’s intracellular location was already described in 1980 [135]. Subsequently, numerous reports approved these early findings in many excitable cells, such like smooth muscle cells, in which mitochondria were found to modulate Ca²⁺ sparks and Ca²⁺-activated K⁺ currents [35]. In chromaffin cells mitochondria were found to modulate Ca²⁺ inhibition of various Ca²⁺ channels [88]. Moreover, mitochondria may also be responsible for the difference in Ca²⁺ signaling pattern in the perinuclear and subplasmalemmal cytosol [207]. In cardiac myocytes, mitochondrial Ca²⁺ buffering was found to be responsible for the delay between the peak of the Ca²⁺ transient and the peak shortening in myocytes [209]. Intriguingly, in myocytes as well as smooth muscle cells, mitochondrial Ca²⁺ buffering was found to be linked to the cellular Na⁺ homeostasis [123, 173], thus pointing to the strong inter-dependence of mitochondrial Ca²⁺ homeostasis with that of other ions.

In respect of spatial mitochondrial Ca²⁺ buffering, the group of Ole Petersen has conducted pioneer work in pancreatic acinar cells in which the mitochondria are distinctively located around the granular pole and form a belt that retains the Ca²⁺ signal in the granular pole [162, 166–168, 214]. This landmark observation was complemented by their findings on the focal Ca²⁺ delivery function of the ER in this cell type [22, 68, 162, 166, 167, 169].

This overview on the aspect of spatial mitochondrial Ca²⁺ buffering only merely touches the role of mitochondria as a local and/or global regulator of cytosolic Ca²⁺ signal, but indicates that the quality of the modification of the cytosolic Ca²⁺ signal by mitochondria is complex and highly depends on the cellular system investigated.

Ca²⁺ entry pathways

The role of mitochondria as a regulator of Ca²⁺ entry has been mainly studied in the particular context of the activation of store-operated Ca²⁺ entry. This mechanism, originally described by Putney [175], is now evident in virtually every cell type. Upon store depletion, Ca²⁺ permeable channels open at the plasma membrane and allow Ca²⁺ entry that eventually refilled the store. The best characterized current supporting this influx was described in blood cells and termed CRAC for Ca²⁺ release activated Ca²⁺ current. One of the characteristics of this current is a feedback inhibition due to rise in intracellular Ca²⁺ concentration [161]. Mitochondria, by their ability to take up Ca²⁺, were shown to prevent Ca²⁺-dependent inhibition of the current, and thus allow a sustained Ca²⁺ entry through CRAC [94]. This mechanism of SOCE was also reported in other cell types, and mitochondria were shown to have the same effect. In endothelial cells, it was clearly shown that mitochondria are able to maintain low level of Ca²⁺ concentration in their vicinity, thus preventing Ca²⁺-depedent inhibition of SOCE [126].

In the case of Ca²⁺ entry, it is also postulated that mitochondria have to be located close to the membrane to efficiently buffer entering Ca²⁺. This was indeed confirmed by experiments where mitochondria were relocalized more distantly from the membrane after overexpression of dynamitin [216]. Intriguingly, a sustained activity of CRAC requires translocation of mitochondria to the plasma membrane [176]. In that particular case, the SOCE was significantly reduced. However, overexpression of hFis1 that also relocated mitochondria away from the plasma membrane, did not significantly impact on SOCE [65]. In this condition, mitochondria were still able to accumulate entering Ca²⁺ (even with a slower time course), and depolarizing mitochondria reduced the SOCE. Thus, mitochondria prevent Ca²⁺-dependent inhibition of SOCE by local Ca²⁺ buffering, but they may also release a diffusible compound that activates Ca²⁺ entry as shown in RBL cells [71], as well as in immune cells [6].

Until recently, neither the mechanism of SOCE activation nor the molecular identity of the channel was known. During the last 2 years, major advances in the field have been obtained with the identification of STIM 1 as a key regulator of SOCE [119, 189] and Orai1 as the channel supporting the CRAC current [218, 226]. With this in hand, the role of mitochondria in SOCE regulation will probably be reassessed and some remaining controversies clarified.

Ca²⁺ entry does not exclusively rely on SOCE. The arachidonic-activated Ca²⁺ entry pathway is a well-documented route of Ca²⁺ influx that is distinct from SOCE [199, 221]. In addition, the large family of TRP cation channels also constitute an important way of Ca²⁺ influx [141]. However, the role of mitochondria in these pathways has not been investigated so far. An interesting exception is the TRPM2 channel. This nonselective cation channel is Ca²⁺-permeable and activated by different compounds, among them oxidative stress and ADP-ribose. Recently, it was shown that oxidative stress stimulates mitochondria to produce ADP-ribose that subsequently activates TRPM2 [165]. Whether mitochondria are also involved in modulating the activation of other TRP channels is currently unknown.

Effect of Ca²⁺ on mitochondrial metabolism

In most cells, mitochondria provide most of the cell’s energy as ATP of which the synthesis from ADP is linked to a series of electron transport through the IMM. This complex endergonic reaction is powered by the oxidation of reduced cofactors (NADH⁺+H⁺, FADH₂), which are generated during nutrition catabolism via, e.g., the Krebs cycle, and deliver electrons to complex I and complex II of the respiratory chain in the IMM. The subsequent electron transfer to complex III and further to complex IV of the electron transport chain supplies energy to establish a proton gradient across the IMM, which is finally converted into the synthesis of ATP molecules at the ATP synthase complex (complex V). This vital process leading to mitochondrial ATP production has to be up- or downregulated according to the cellular needs of energy and the fluctuating supply of nutrition as well [21].

Mitochondrial Ca²⁺ signaling appears to be fundamental in the control of mitochondrial metabolism. Forty years ago, Hansford and Chappell reported that the oxidation of glycerol phosphate is activated by Ca²⁺ [85] and later on, McCormack and Denton precisely described that pyruvate dehydrogenase is activated by Ca²⁺-dependent dephosphorylation, while the NAD⁺-isocitrate deyhydrogenase and the 2-oxoglutarate dehydrogenase are directly activated by Ca²⁺ [133]. Thus, an increase in matrix Ca²⁺ concentration in the mitochondria accelerates the enzymatic activities of these three dehydrogenases leading to increased NADH levels [52, 82, 174], and subsequently, this Ca²⁺-triggered activation of Krebs cycle dehydrogenases actually culminates in an augmentation of the mitochondrial ATP production in intact living cells. That was shown in elegant experiments using the luciferin luciferase reaction [104]. Moreover, they could also describe a so far not further characterized mitochondrial memory that allows a long-term activation of mitochondrial ATP production up to 60 min after agonist-induced Ca²⁺ signal elevation. In line with these reports, we could demonstrate that an augmentation of mitochondrial Ca²⁺ uptake capacity upon cell stimulation with an IP3 generating agonist by overexpression of UCP2 or UCP3 resulted in a significant enhancement of the agonist-induced mitochondrial ATP generation in endothelial cells [215]. In addition, there are several reports indicating that the Ca²⁺-sensitive dehydrogenases are not the only _targets of the mitochondrial metabolic pathways for Ca²⁺-dependent activation of mitochondrial ATP production. Ca²⁺-sensitivity was also reported for the F₀/F₁-ATPase (heart, [87, 211]), the adenine nucleotide translocase (liver, [144]), complexes of the respiratory chain [188], and EF-hand containing substrate carriers of the IMM [113, 158].

All these reports on the stimulatory effect of Ca²⁺ for mitochondrial ATP production illustrate the functional significance of mitochondrial Ca²⁺ sequestration and explain how a cell can accommodate the increasing demand of energy during stimulated states. However, the effects of matrix Ca²⁺ that exceed its regulatory function on oxidative phosphorylation might be multiple and await further investigation.

Interestingly, there is a limited number of reports that are contrary to the dogma that mitochondrial Ca²⁺ accumulation accelerates mitochondrial metabolism [17, 122]. In these reports, conditions are described in which the depolarizing effect of Ca²⁺ on the IMM exceeds its stimulatory effect on the respiratory chain/dehydrogenases, and consequently leads to a Ca²⁺-induced decrease in NADH levels. Such a scenario seems conceivable under conditions of excessive activation of respiration by substrate overload and/or under conditions of pathologically increased mitochondrial Ca²⁺ sequestration as elaborated in a mathematical model by Bertram et al. [17]. Notably, under conditions of substrate or Ca²⁺ overload of the mitochondria the formation of reactive oxygen species (ROS) is assumed to play an important role in mitochondrial degeneration/dysfunction.

Impact of mitochondrial Ca²⁺ on ROS generation and ROS defense

In most cells mitochondria represent the main source of the physiological production of ROS. Basically, mitochondrial-generated ROS come from an interaction between oxygen (O₂) with unpaired electrons, which occur along the respiration. In average 1 to 2% of the total electrons transported through the respiratory chain leak to produce superoxide anions $\left(•{\text{O}}_2^{-}\right)$ [24], which can be rapidly converted into the more reactive H₂O₂ and its aggressive derivative the hydroxyl radicals (•OH^–). Among the actual sites of ROS generation in mitochondria complex I and complex III have attracted most attention [10, 114]. However, very recently, matrix enzymes such as the alpha-ketoglutarate dehydrogenase and other components of dehydrogenase complexes of the Krebs cycle have been also considered as important sources of ROS within mitochondria [36]. To balance the permanent production of ROS, mitochondria also house a sophisticated defense network of enzymatic and nonenzymatic antioxidants against ROS [2]. However, although ROS have been accused to exhibit numerous pathological effects in mitochondria and can lead to lipid peroxidation, which subsequently causes oxidative damage of mitochondrial DNA, RNA and enzymes, mitochondrial ROS generation might constitute an important signaling molecule to modulate cellular signal transduction in health and disease.

The impact of mitochondrial Ca²⁺ signaling on mitochondrial ROS homeostasis is in most instances ambiguous. There are various reports implying that an excessive mitochondrial Ca²⁺ sequestration facilitate the generation of ROS within mitochondria [57, 179, 180]. However, the underlying mechanisms for Ca²⁺-induced mitochondrial ROS production has not been clarified entirely but might include activation of the electron transport chain as well as Ca²⁺-induced opening of the mitochondrial permeability transition pore (PTP).

A different conclusion of the consequence of mitochondrial Ca²⁺ uptake for mitochondrial ROS generation can be obtained if the electrical effect of Ca²⁺ on the potential of the IMM (Δψ_mito) is considered. Ca²⁺ uptake into mitochondria reduces Δψ_mito, which has been shown to counteract generation of $•O_2^{-}$ during oxidative phosphorylation [202, 219]. Moreover, Ca²⁺ directly or via calmodulin modulates the activity of enzymes of the antioxidant defense systems (e,g, MnSOD), indicating that Ca²⁺ exhibits dual contribution in the regulation and control of the mitochondrial ROS homeostasis [91, 225]. Additionally, mitochondrial Ca²⁺ inhibits the removal of hydrogen peroxide and its succinate-fueled production, thus indicating that mitochondria function as intracellular Ca²⁺-modulated peroxide sinks [231].

Our understanding on mitochondrial ROS production is controversial and rather confused. This is not surprising as the methods to measure ROS production lack specificity and mitochondria from different tissue origin might metabolically not be comparable. Notably, UCP2 and UCP3 have been shown to suppress mitochondrial ROS production under various experimental conditions presumably via their uncoupling function (for a recent review see Dlaskova et al. [50]). We have recently demonstrated in intact as well as isolated mitochondria that UCP2 and UCP3 are fundamental for mitochondrial Ca²⁺ uniport while no obvious evidence for H⁺ conductance was found [215]. Consequently, in addition to the dogma of smooth uncoupling via UCP2/UCP3-mediated H⁺ flux, the discovered Ca²⁺ function of these proteins might be, at least partly, responsible for the reported reduction in mitochondrial ROS production by these UCP proteins. Since the correlation of UCP2/3-dependent mitochondrial Ca²⁺ fluxes with mitochondrial ROS production has not been elucidated so far, the molecular mechanisms beyond UCP2/UCP3-associated attenuation of mitochondrial ROS production awaits further investigation.

On the other hand, extramitochondrial-produced ROS might affect mitochondria as well. In recent studies, exposure of mitochondria to ROS have been shown to affect the structural organization and Ca²⁺ homeostasis [160] and trigger IP₃-linked apoptotic cascade [124], thus indicating that this organelle can be affected by cytosolic-/plasma membrane-originated ROS also.

Ca²⁺ as regulator of mitochondrial phosphoproteom

Reversible phosphorylation of enzymes, receptors, ion channels, transcription factors and cytoskeletal elements is a hallmark of cell signaling (Greengard, Science, 1976, 146–152). It has recently emerged that the role of phosphorylated proteins as physiological effectors is not restricted to the cytosol, but is also fundamentally involved in the homeostasis of mitochondrial functions. With the development of improved phosphoproteomic screens, the existence of multiple phosphoproteins in mitochondria could be confirmed [91]. Importantly, there are protein kinases and phosphatases that are exclusively located within the mitochondrial matrix [132], in the IMM [107], the intermembrane space [194], and at the cytoplasmic surface of mitochondria [34]. Moreover, cytosolic protein kinases are known to be imported into mitochondria by a yet not fully understood mechanisms [97, 156].

Although the Ca²⁺ dependent dephosphorylation of the pyruvate dehydrogenase complex within mitochondria was already described in the late 1960s [118], only limited knowledge is available about the impact of Ca²⁺ on posttranslational modification of mitochondrial proteins. Very recently, a dynamic change in the phosphorylation state of multiple mitochondrial proteins has been shown in porcine heart mitochondria in response to Ca²⁺ elevation [91]. The mitochondrial proteins that got phosphorylated in response to matrix Ca²⁺ elevation included components of the electron transport chain and the Krebs cycle, indicating that Ca²⁺ impacts mitochondrial bioenergetics on various levels. Moreover, a Ca²⁺-induced dephosphorylation of the MnSOD was also described, which was associated with an increase in its enzymatic activity, while neither the kinase(s) nor the phosphatase(s) involved in this process has been identified so far. Similar findings were obtained in rat brain mitochondria, in which several, so far unidentified, low molecular mass proteins were found to be phosphorylated in a Ca²⁺-dependent manner [7]. Interestingly, this Ca²⁺-induced protein phosphorylation depended on the opening of the mitochondrial PTP while its particular function in Ca²⁺-activated phosphorylation of mitochondrial proteins remained unresolved.

In summary, convincing evidence points to an importance of mitochondrial Ca²⁺ fluxes for protein phosphorylation and dephosphorylation as a crucial mechanism in the dynamic regulation of various mitochondrial functions. However, this remains an emerging field, which requires further investigation to be fully integrated into a complex understanding of the molecular mechanisms of the regulation of mitochondrial functions.

Effect of Ca²⁺ on mitochondrial motility and morphology

Mitochondria display a very complex architectural organization that varies continuously over time depending on their metabolic activity, the level of cell activation [224], the cell cycle status [14, 64], and the impact of physical or chemical environmental signals. In most cells, mitochondria are largely interconnected tubular structures [190] (Fig. 6a), whereas mitochondria with round punctuated morphology are less frequently described [38] and may represent an early indication of mitochondrial stress initiated by either metabolic overload [160], chemical depolarization (Fig. 2), uncontrolled fission processes [1, 163, 164], or physical disturbance (R. Malli and W.F. Graier, unpublished observations). In addition, in certain cell types, mitochondrial structure reflects their function, such like spatial Ca²⁺ buffering in pancreas acinar cells [22, 68, 162, 166, 167, 169], cardiac muscle [135, 209], chromaffin cells [88], or nerves [29, 32, 151].

Besides the complexity of their architectural organization, mitochondria exhibit complex dynamics (Fig. 6b) that have been shown to depend on cytoskeleton-based transport mainly along microtubule via motor proteins such as the mitochondria linked kinesin Kif1B [210]. Hence, mitochondrial structure continuously changes by processes of fusion, branching, or fission [14, 224] that results in a permanent rearrangement of the mitochondria (Fig. 6b). In contrast to the molecular mechanisms responsible for mitochondrial motility that are poorly understood so far, major proteins that participate in the regulation of the organelles fission, fusion, and transport along cytoskeleton elements have already been identified (for review, see [190, 224]). The processes of fission and fusion are mainly under the control of guanosine triphosphatases (GTPases) such Drp1, Mitofusins (Mfn) and OPA-1 [23], and their associated adaptor proteins like hFis1 [98] as well as Rho family proteins [95, 204].

Both, mitochondrial motility and morphology are significantly affected by Ca²⁺. Particularly, mitochondrial motility is regulated by Ca²⁺ in the physiological range with maximal movement at low resting cytosolic Ca²⁺ levels and a complete arrest at 1–2 μM free Ca²⁺ in the cytosol [227]. The molecular mechanism behind this Ca²⁺-triggered arresting of mitochondria is unknown, but indicates that mitochondria might be recruited in a Ca²⁺-dependent manner to enhance local Ca²⁺ buffering and/or ATP supply on distinct cellular demands. Hence, the elevation of intracellular Ca²⁺ concentration initiates the translocation of Drp1 to the mitochondria and thus initiation of fission processes that consequently results in mitochondrial fragmentation [27]. In this regard, a reversible Ca²⁺-dependent fragmentation of interconnected tubular mitochondria upon stimulation with an IP₃-generating agonist was described in HeLa cells [204]. Whether this Ca²⁺ induced mitochondrial fission is due to alterations of the potential of the inner mitochondrial membrane and alteration of the matrix pH or depends on an activation of GTPases most likely of Drp-1 [116] remains however elusive.

In summary, mitochondria are very mobile organelles, which move, fuse, branch, and divide to create a flexible and highly dynamic tubular network, whose morphology, motility, and intracellular distribution is evidently affected by Ca²⁺. Although the molecular processes, pathways of regulation as well as the physiological function of this striking aspect of mitochondrial Ca²⁺ signaling are not fully understood yet, one can expect that along their exploration we will learn further exciting and so far unknown functions of these old guests.